Advances in Botanical Research, Volume 7

Advances in

BOTANICAL RESEARCH VOLUME 7

This Page Intentionally Left Blank

Advances in

BOTANICAL RESEARCH Edited by

H. W. WOOLHOUSE Department of Plant Sciences, The University, Leeds, England

VOLUME 7

1979

ACADEMIC PRESS London

New York

Toronto

Sydney

San Francisco

A Subsidiary of Harcourt Brace Jovanovich, Publishers

ACADEMIC PRESS INC. (LONDON) LTD. 24/28 Oval Road, London NWl 7DX

US.Edition published by ACADEMIC PRESS INC. 111 Fifth Avenue New York. New York 10003

Copyright 0 1979 by Academic Press Inc. (London) Ltd

All Rights Reserved

No part of this book may be reproduced in any form by photostat, microfilm, or any other means, without written permission from the publishers

British Library Cataloguing in Publication Data

Advances in botanical research. Vol. 7 1. Botany I. Woolhouse, Harold William 581 QK45.2

62-21 144

ISBN 0-12-005907-X

Printed in Great Britain by Latimer Trend & Company Ltd, Plymouth

CONTRIBUTORS TO VOLUME 7 J. ARDITTI, Department of Developmental and Cell Biology, School of Biological Sciences, University of California, Irvine, California 92717, U.S.A. W. ARMSTRONG, Department of Plant Biology, University of Hull, Hull HU6 7 R X , England P. F . BROWNELL, Department of Botany, James Cook University of North Queensland, Townsville, Queensland, Australia R. DOUCE, Laboratoire de Biologie VPge‘tale, Commissariat ci I’Energie Atomique, Centre d’ Etudes Nucleaires de Grenoble, 85 X , 38041 Grenoble Cedex, France J. JOYARD, Laboratoire de Biologie Vkgitale, Commissariat Ci I’Energie Atomique, Centre #Etudes Nucleaires de Grenoble, 85 X , 38041 Grenoble Cedex, France A. D. M. RAYNER, School of Biological Sciences, University of Bath, Claverton Down, Bath BA2 7 A Y , England N. K . TODD, Department of Biological Sciences, University of Exeter, Hatherly Laboratories, Prince of Wales Road, Exeter EX4 4PS, England

V

This Page Intentionally Left Blank

PREFACE During the past decade progress has been made in the matter of separating and characterizing the various types of membranes which occur in plant cells. The subject is still very much in its infancy, however, and can perhaps best be put in perspective by considering the fact that more work has been published on the erythrocyte membrane than all the types of plant cell membranes put together. In this volume Douce provides an introduction to the study of chloroplast envelope membranes and supplies fascinating evidence of their complex structure and diverse functions. In 1959 Brownell provided the first definitive evidence that sodium was an essential element for the growth of a flowering plant, Atriplex vesicaria. Subsequent work revealed a situation unique in the mineral nutrition of higher plants, in that sodium was shown to be essential for some species but not others. Brownell and his group went on to show that species possessing the C , pathway of photosynthesis are the ones which require sodium. In this volume Brownell gives an account of this technically exacting line of work and goes on to discuss the recent work from his group which suggests that to understand the biochemistry of sodium action in C , species is likely to prove a difficult problem. Armstrong discusses the development of an electrical analogue model used to solve quantitative problems in the internal movements of oxygen in wetland plants and casts doubt on some of the recent work purporting to show biochemical adaptation to growth under anaerobic conditions. This is the type of study which encourages the hope that physiological work of a good standard is at last infiltrating the field of experimental ecology. Again within the context of a more rigorous analytical ecology, Todd and Rayner provide a fascinating account of how genetic analysis may be brought to bear on the complex problem of fungal decay processes in timber. It is a matter of editorial policy to include accounts of progress in the biology of specific plant groups; in the current volume Arditti, an eloquent advocate of the orchid in all its forms, reviews progress in the rapidly expanding subject of the physiology of orchids. It remains for me to thank the authors for their efforts in minimizing the editor’s task; my indexers for their patient efforts and Miss J. M. Denison for invaluable secretarial assistance.

H. W. Woolhouse

Leeds 1979 vii

This Page Intentionally Left Blank

CONTENTS CONTRIBUTORS TO VOLUME 7

.

V

PREFACE

.

vii

.

2

.

3

.

12

Structure and Function of the Plastid Envelope ROLAND DOUCE and JACQUES JOYARD I. Introduction

.

11. Structure of the Plastid Envelope

.

111. Relationships Between the Plastid Envelope and Other Cell

Membranes . A. The Inner Envelope Membrane and the Internal Membranes of Plastids . . B. The Outer Envelope Membrane and the Extra-chloroplastic . Membranes . C. The Inner and Outer Envelope Membranes . .

IV.

Isolation of the Chloroplast Envelope A. Principles . B. Procedure .

.

V. Chemical Composition of the Chloroplast Envelope . A. Polar Lipid Composition of the Chloroplast Envelope B. Pigment Composition of the Chloroplast Envelope . C. Chloroplast Envelope Polypeptides VI. Enzymic Activities and Functions of the Chloroplast Envelope A. Metabolite Transport in Intact Chloroplasts . B. Protein Transport Through the Envelope Membranes C. Lipid Synthesis by the Chloroplast Envelope Membranes VII. Origin of the Chloroplast Envelope Membranes VIII. Summary , Acknowledgements References .

.

18

22

. . .

25 25 32

. . . .

37 39 4 4

. . .

53

.

49

54 74

81

.

95

. .

97 100 100

. ix

12

CONTENTS

X

Sodium as an Essential Micronutrient Element for Plants and its Possible Role in Metabolism P. F. BROWNELL 1. Introduction . A. Scope . B. Historical Perspective-1860 11. Low-sodium Culture Conditions A. Definition . B. Determination of Sodium C. Culture Techniques .

. to 1977

.

.

.

.

126 126 127 132

.

111. Responses to Sodium at Low Concentrations

118 118 119

.

144 144 149

Lowerplants . Atriplex vesicuria Heward ex Benth . Other Species having the C, Dicarboxylic Photosynthetic Pathway . Response by a Species having Crassulacean Acid Metabolism Discussion .

157 159 168

Metabolic and Physiological Effects of Sodium at Low Concentrations. . A. General Strategies B. Anubaenu cylindrica . C. Other Lower Plants . D. C,andCAMPlants .

172 172 172 184 186

V. Tentative Schemes for the Role of Sodium in C4 and CAM Plants and in Blue-green Algae . A. C,and CAMplants . . . B. Anabaenu cylindrica

207 207 212

A. B. C.

D. E. IV.

VI. Summary and Conclusions Acknowledgements . References .

. .

215 219 219

Aeration in Higher Plants W. ARMSTRONG I. Introduction

.

226

TI. Principles of Aeration and Aeration Modelling A. Entry and Dispersal of Respiratory Gases B. Diffusion and the Ventilating Process C. The Oxygen Source D. The Aeration Model

.

111. The Cylindrical Platinum Electrode Technique

. ,

.

227 227 242 260 265 272

xi

CONTENTS

IV.

Aeration in the Wetland Condition A. The Wetland Plant . B. The Non-Wetland Plant. C. Trees .

.

278 278 298 305 313 324

V. Root Aeration in Unsaturated Soil Acknowledgements . Appendix 1

The Transport of Diffusible Species in Media Moving by Mass . Flow

Appendix 2 Radial Diffusion into Respiring Spherical Bodies References Note Added in Proof .

.

325

.

326 328 332

Population and Community Structure and Dynamics of Fungi in Decaying Wood A. D. M. RAYNER and N. K. TODD I. Introduction . A. General . B. Types of Decay and Fungi Inhabiting Wood C. Scope of the Present Article , 11. Direct Methods of Analysis . A. General . B. Recognition of Interactional Patterns C. Zone Lines .

334 334 335 336

.

337 337 339 343

.

111. Intraspecific Antagonism : The Delimitation of Individual Mycelia . A. General B. Basis of Antagonism . C. Significance and Potential Use of Antagonism

346 346 359 375

IV. Interspecific Interactions: Their Role in the Development and Maintenance of Community Structure . A. Theories of Succession and Community Development B. Interactions and Their Significance C. Results from Laboratory-based Studies D. Interactions in Nature . E. Concluding Comments .

380 380 384 388 399 401

.

V. Ecological Roles and Spatial Distribution Within Communities of Fungi in Decaying Wood . A. Introduction B. Factors Influencing Mode and Pattern of Growth of Fungi in Decaying Wood . C. Discussion . VI. Schema for Fungal Community Development in Decaying Sapwood of Hardwoods after Felling . .

403 403 404 414 415

xii

CONTENTS

VII. Concluding Comment Acknowledgements References .

. .

.

. .

417 417 417

Aspects of the Physiology of Orchids JOSEPH ARDITTI I. Introduction

11. Seeds A. B. C. D. E. F. G.

.

422

,

History . External Morphology . Structure and Ultrastructure Longevity . Asymbiotic Germination Symbiotic Germination . Summation

.

.

423 423 438 438 441 441 489 506

111. Phytoalexins . A. History B. Chemistry Production and Distribution . C. Action Spectrum and Activity D. Biological Role .

508 508 510 512 517

IV. Carbon Fixation . A. History . B. Stomata1 Rhythms . C. Crassulacean Acid Meatabolism . D. C, Photosynthesis. . E. C, Photosynthesis Carbon Fixation by Different Plant Organs F. G. Photorespiration . H. Summation

519 519 521 521 529 529 530 532 532

V.

VI.

Flowers A. History . B. Introduction C. Pollination . D. Post-pollination Phenomena E. Induction of Phenomena Tissue Culture

VII. Epilogue . Acknowledgements References .

.

.

.

533 534 534 552 566 617 634

.

637 637 638

AUTHORINDEX

.

657

SUBJECTINDEX

.

677

Structure and Function of the Plastid Envelope

R O L A N D D O U C E and JACQUES J O Y A R D Laboratoire de Biologie Vigirale. C.E.A.-Centre d'Etudes Nucleaires de Grenoble. 85 X . 38041 Grenoble Cedex. France

I . Introduction

. . . . . . . . . . . . . . . . .

I1. Structure of the Plastid Envelope .

. . . . . . . . . .

2 3

111. Relationships Between the Plastid Envelope and Other Cell Membranes . . . . . . . . . . . . . . . . . A. The Inner Envelope Membrane and the Internal Membranes of Plastids . . . . . . . . . . . . . . . B. The Outer Envelope Membrane and the Extra-chloroplastic Membranes . . . . . . . . . . . . . . . C. The Inner and Outer Envelope Membranes . . . . . .

18 22

IV . Isolation of the Chloroplast Envelope . A. Principles . . . . . . . B. Procedure . . . . . .

25 25 32

. . . . . . . . . . . . . . . . . . . . . . . . . . .

V . Chemical Composition of the Chloroplast Envelope . . Polar Lipid Composition of the Chloroplast Envelope A. Pigment Composition of the Chloroplast Envelope . B. C. Chloroplast Envelope Polypeptides . . . . .

12 12

. . . .

. . . .

. . . .

37 39 44 49

VI . Enzymic Activities and Functions of the Chloroplast Envelope A. Metabolite Transport in Intact Chloroplasts . . . B. Protein Transport Through the Envelope Membranes . C. Lipid Synthesis by the Chloroplast Envelope Membranes

. . . .

. . . .

53 54 74 81

. . . . . .

95

VII. Origin of the Chloroplast Envelope Membranes

2

R. DOUCE AND J. JOYARD

VIII. Summary

. . . . . . . . . . . . . . . . . .

97

Acknowledgements . . . . . . . . . . . . . . . 100 References . . . . . . . . . . . . . . . . . 100

I. INTRODUCTION In higher plants, the process of photosynthesis occurs within specific membrane bounded organelles called chloroplasts. It is now well established that most of the chloroplasts, which are probably the descendants of prokaryotic ancestors, are found in the cytoplasm of the spongy and palisade mesophyll leaf cells in apposition to the cytoplasmic and tonoplastic membranes. Most published electron micrographs of higher plant leaf cells show that all the chloroplasts examined present the three following major structural regions (Fig. 1): a. Highly organized internal sac-like flat compressed vesicles called thylakoi'ds (Menke, 1961). b. An amorphous background rich in soluble proteins called stroma. c. A pair of outer membranes known as the envelope.

Fig. 1. A chloroplast is surrounded by two continuous membranes called the envelope. This swollen, whole spinach chloroplast was isolated in 0.33 M sorbitol. Micrograph provided by Drs B. Sprey and w. M. Laetsch; reproduced with permission from 2.Pflphys. (1975) 75, 38-52.

STRUCTURE AND FUNCTION OF THE PLASTID ENVELOPE

3

It is now well established that the light-dependent biophysical reactions of photosynthesis are associated with the thylako’ids whereas the dark biochemical reactions of CO, assimilation into starch and phosphorylated sugars are dependent on the soluble proteins of the stroma. It is also evident that the envelope regulates the transfer of chemical energy, of reducing power and of fixed-carbon into and out of the chloroplasts (Douce and Joyard, 1977). The structure, chemical composition, function and the origin of the internal membrane system (thylakolds) has been described and discussed in a number of recent reviews and monographs. These include description of the structure of thylakolds (Thomson, 1974; Coombs and Greenwood, 1976; Muhlethaler, 1977;Sane, 1977),relationships between this structureandphotosynthetic functions (Arntzen and Briantais, 1975; Anderson, 1975a) and aspects of thylakold development (Kirk and Tilney-Bassett, 1967;Boardman, 1977). Unfortunately the envelope has received little attention despite its importance in the functional and structural integrity of the chloroplast. In this review, we shall discuss the structure, isolation, chemical composition and origin of the higher plant chloroplast envelope. We shall also examine the multiple functions of this important membranous system involved in the regulation of the inflow of raw materials for photosynthesis and the outflow of photosynthetic products (Heber, 1974; Walker, 1974; Heldt, 1976a). 11. STRUCTURE OF THE PLASTID ENVELOPE The typical higher plant chloroplast is shaped like a lens. In transverse section, the chloroplast is elliptical or circular. The longitudinal diameter of the chloroplast varies from 3 to 10 pm (Gunning and Steer, 1975). Except for a few early reports (Heitz, 1936; Wieler, 1936; Weier, 1938; Granick, 1949; Strugger, 1951; Weier and Stocking, 1952) the presence of an envelope around the Chloroplast was not clearly seen until the advent of electron microscopy. In electron micrographs, after KMnO, or OsO, fixation, the envelope appears at low magnification (Fig. 1) as two electron dense lines separated by an electron transparent space. The envelope seemed to illustrate perfectly the “unit membrane” concept of Robertson (1959) and the well known molecular membrane theory of Danielli and Davson (1935). Indeed, the “unit membrane” is clearly visualized after fixation with KMnO, or OsO, as a triple-layered structure, approximately 7.5 nm thick, consisting of two dense lines each about 2 nm wide separated by a lighter space of 3.5 nm. For this simple reason, initially, the envelope was considered to be a single membrane. For example, McLean (1956) interpreted the electron dense lines not as two separate membranes but as components of a single membrane. Unfortunately, early interpretations of electron micrographs were hampered by incomplete understanding of the action of the fixatives and especially by relatively poor electron microscope resolution. Furthermore, the thickness of

4

R. DOUCE AND J. JOYARD

the envelope is much higher (above 20nm) than that expected for a “unit membrane”. Subsequently, it has been clearly established that these electron dense lines are due to two distinct membranes. Thus, Heitz (1957) showed a short stretch of double membranes around a chloroplast of Aneura pinguis. Von Wettstein (1958) also indicated the presence of a double membrane bounding the chloroplast and Buvat (1958) reported the presence of double membranes around chloroplasts of Elodea canadensis. Plastids in leaves of Chrysanthemum segetum were reported by Lance (1958) to be bounded by a double membrane. Most investigators now appear to be convinced that the chloroplast envelope of higher plants consists of two morphologically and topologically distinct membranes (Fig. 2). This applies even to highly senescent plastids from which all other membranous structures have disappeared (Barton, 1966; Butler and Simon, 1971; Priestley, 1977). It is also the accepted view that most algae chloroplasts examined so far (Bisalputra, 1974) have a pair of outer membranes. However, Wildman et al. (1962) and Wildman (1967, 1971) using information derived from light microscopy and cinematographic studies in vivo indicated that the entire system of thylakoids (“the stationary component”) is surrounded and interpenetrated by a “mobile phase”. Electron microscopy indicated also that union of plastids (fusion) by their mobile phase could occur temporarily in vivo (Lang and Potrykus, 1971 ; Esau, 1972). According to our studies, the mobile phase represents the chloroplast stroma which probably has a gelatinous consistency. Indeed, the stroma is rich in soluble proteins (about 0.4 g ml-l). The outer surface of the mobile phase which displays osmotic properties represents the double membranes of the envelope delineating the “biphasic” chloroplast from the cytoplasm of the cell. The total thickness of the component envelope membranes of higher plant chloroplasts is reported as 6 nm (Gunning and Steer, 1975). The two membranes generally stain with approximately equal electron-density (Fig. 2) although occasional exceptions may be noted (Weier and Thomson, 1962; Ohad et al., 1967; Robards and Humpherson, 1967). In a recent report, Kagan-Zur et al. (1977) have suggested that the overall affinity of chloroplast envelope membranes for heavy metal stains may be influenced by a period of illumination before fixation. The envelope membranes are separated by a region about 10-20 nm thick which appears electron-translucent (Gunning and Steer, 1975). Very often the envelope appears as two membranes which apparently diverge and converge at random producing small areas of membranes in close contact. This effect is enhanced by their tendency to twist within the thickness of the section giving a diffuse cross-sectional image (Weier, 1961). Considering the importance of the envelope it is surprising that more attention has not been given to elucidating its substructure. Furthermore, literature dealing with the fine structure of the plastid envelope is widely

STRUCTURE AND FUNCTION OF THE PLASTID ENVELOPE

5

Fig. 2. The chloroplast envelope of higher plants consists of two morphologically and topologically distinct membranes. (Portion of a cell of spinach, Spinaciu olerucea L.) Note that the two membranes stain with approximately equal electron-density. Micrograph provided by Dr J. P. Carde (unpublished data).

scattered and extremely diverse in character. Consequently, it has lacked detailed attention in recent reviews on plastid and cell ultrastructure (Thomson, 1974; Bisalputra, 1974; Arntzen and Briantais, 1975; Gunning and Steer, 1975; Coombs and Greenwood, 1976). Both membranes of the chloroplast envelope are shown to have a characteristic three-layered structure appearing in electron micrographs of permanganate-fixed sectioned material as two

6

R. DOUCE AND J. JOYARD

dense lines enclosing a lighter space. The organization is comparable with the “unit membrane” of Robertson (1959). However, in their paper on chloroplast membranes, Miihlethaler et al. (1965) advocate the view that the “unit membrane” is a fixation artifact. They explain the absence of the globular substructure in the “unit membrane” after chemical fixation by the following theory : during the fixation process the globular proteins are denatured which causes them to uncoil and spread out. The artificial surface lamellae thus formed take up heavy metal stains and appear as dark lines in cross-section. In a classic paper, Singer and Nicolson (1972) have postulated that the lipids and the proteins of intact membranes are organized as follows: a. The polar and ionic head groups of the lipid molecules together with all of the ionic side chains of the amphipathic globular proteins (integral proteins) are on the exterior surfaces of the membrane. b. The non-polar side chains of the integral proteins are in the interior of the membrane, together with the hydrocarbon tails of the polar lipids. c. The polar lipids are largely arranged in bilayer form. The matrix of biological membranes appears to exist as a bilayer of mobile lipids, the relative motion of which determines the fluidity or viscosity of the membrane interior. Several experiments indicate clearly that the structure of the envelope membranes is consistent with the lipid-globular protein mosaic model of membranes proposed by Singer and Nicolson (1972). The arguments in favour of this conclusion are as follows: a. With glutaraldehyde fixation and then staining with lead, the envelope membranes of chloroplasts of the green alga Scenedesmus quadricauda (Weier, 1966) and of higher plant chloroplasts (Weier et al., 1965) appear to have globular subunits at high magnification. b. Freeze-cleaving of the outer and inner membrane of the chloroplast envelope from the red alga Bangia fusco-purpurea (Bisalputra and Bailey, 1973), Euglena (Miller and Staehelin, 1973), spinach (Sprey and Laetsch, 1976b), and of the amyloplast envelope from root-tip cells (Fineran, 1975) reveal four fracture faces (Fig. 3). The outer membrane of the chloroplast envelope shows a low density (283 pm-2) of 9 nm particles in both complementary fracture faces. The inner membrane of the chloroplast envelope displays a significantly higher particle density (Fig. 4). The face of the inner membrane-half in close contact with the chloroplast stroma contains more particles (1820 pm-2) than that of the half closer to the cytoplasm (980 pm-”. Their size in the platinum shadowed replica was 9 nm. It has to be stressed that freeze-etching (Moor and Miihlethaler, 1963) is likely to be one of the most promising electron microscope techniques for studying membranes. Its main advantage is that surface views as well as crosssections of chemically unaltered membranes in their natural surroundings can be obtained (Staehelin and Probine, 1970). The technique also offers a

STRUCTURE AND FUNCTION OF THE PLASTID ENVELOPE

7

st roma thy lakoids inner

membrane

envelope

\

outer

membrane-

I OE

OOE

Fig. 3. Diagram showing the four fracture faces of freeze-fractured chloroplast envelope membranes. I1 E means inner face of the inner membrane of the envelope 0 1 E means outer face of the inner membrane of the envelope I 0 E means inner face of the outer membrane of the envelope 00 E means outer face of the outer membrane of the envelope When seen from the cytoplasm, the two fractured membranes show their inner faces; when seen from the stroma, the two fractured membranes show their outer faces.

unique opportunity to examine structural features within the membrane. The particles observed in both envelope membranes therefore represent indubitably the integral proteins described by Singer and Nicolson (1972). It is extremely likely that the differences present in the ultrastructure of inner and outer membranes of the envelope, as revealed by freezefracturing, reflect functional and compositional differences. c. Billecocq et al. (1972) and Billecocq (1974, 1975) by means of specific antibodies prepared according to Radunz (1969, 1972) and Radunz and Berzborn (1970) have shown that the polar head groups of the sulpholipid and the galactolipids are on the exterior surfaces of the envelope membranes. In other words, the polar heads of the envelope lipid molecules are exposed to the aqueous phases facing the membranes (Fig. 5). d. Neuburger et al. (1977) have demonstrated that binding of cationic ferricytochrome c to the envelope membranes is electrostatic and that the envelope membrane surfaces are strongly negatively charged. These authors also provide direct evidence that the outer surface of the outer envelope membrane is highly negatively charged (Figs 6 and 7). In accordance with these results, recent interesting studies have led to the proposal that the isoelectric point of intact chloroplasts determined by cross-

Fig. 4. Fracture faces of the inner membrane and the outer membrane of chloroplast envelopes seen from the cytoplasm (upper figure) or the stroma (lower figure). Note that the freeze-fractured inner membrane contains particles with higher density. For explanations see Fig. 3. Micrographs provided by Drs B. Sprey and W. M. Laetsch; reproduced with permission from Z . Pflphys. (1976) 78, 360-371.

Fig. 5. Galactolipids and sulpholipid of the chloroplast envelope are accessible to specific antibodies (an antiserum to the glycolipids was obtained by immunization of rabbits). The antibodies are directed towards the polar head of the glycolipids. The reaction is visualized by treatment with peroxidase according to the method of Avrameas and Ternynck (1971). It shows that both glycolipids are present in the outer envelope membrane and that the polar head groups are uniformly distributed on the outer face of the outer envelope membrane. A : reaction of the outer envelope membrane (om) to specific antibodies anti-galactosyldiglyceride (im: inner envelope membrane). Micrograph provided by Dr A. Billecocq, reproduced with permission from Biochim. biophys. Actu (1974) 352, 245-251, B: reaction of envelope (e) to specific antibodies anti-sulpholipid. Micrograph provided by Dr A. Billecocq, reproduced with permission from Ann. Immunol. (Inst. Pusteur) (1975), 126c, 337-352.

Fig. 6. Appearance of isolated intact washed chloroplasts in 0.33 M sorbitol under light ) . microscopy. A: without yeast cytochrome c ; B: with yeast cytochrome c ( 5 0 ~ ~The suspension medium is devoid of MgCI,.

Fig. 7. Mechanism which explains the strong agglutination of intact chloroplasts induced by cyt. c in a Mgzf free medium. A: the medium is devoid of Mgz+. The outer surface of the outer envelope membrane is strongly negatively charged. The cationic cyt. c molecule is a 34 x 34 x 30 8, prolate spheroid and most of the hydrophilic lysines which carry positive charges are distributed on the major axis of the molecule. In these conditions, once bound to a chloroplast, the cyt. c still carries free positive charges which, in turn, are able to interact with other chloroplasts. This explains the strong and fast agglutination observed. B: the electrostatic binding of cyt. c to the chloroplast envelope is prevented by the addition of Mg".

STRUCTURE AND FUNCTION OF THE PLASTID ENVELOPE

11

Fig. 8. The structure of both envelope membranes is consistent with the lipid-globular protein mosaic model of membrane proposed by Singer and Nicolson (1972). Note that the polar and ionic heads of the lipid molecules together with all the charged side chains of the amphipathic globular proteins are on the surfaces of the membrane exposed to the aqueous phase.

partition is acidic (Akerlund et al., 1975; Westrin et al., 1976). Therefore, the ionic heads of the acidic lipid molecules (sulpholipid and phosphatidylglycerol) and the acidic polar group of the integral proteins are exposed at the exterior surfaces of the envelope membranes. For these reasons, we believe that the envelope membrane substructure agrees well with the lipid-globular protein mosaic model of membrane proposed by Singer and Nicolson (1972) (Fig. 8). It is important to bear in mind, however, that a single rigidly defined model may eventually be difficult to reconcile with all biological membrane structures (Benson and Jokela, 1976).

12

R. DOUCE AND J. JOYARD

111. RELATIONSHIPS BETWEEN THE PLASTID ENVELOPE

AND OTHER CELL MEMBRANES A. THE INNER ENVELOPE MEMBRANE A N D THE INTERNAL MEMBRANES OF PLASTIDS

The inner envelope membrane is rarely completely smooth but possesses frequent discrete folds which invaginate into the plastid stroma (Gunning and Steer, 1975) or evaginate into the periplastidal space (Schotz and Diers, 1967; Weier and Thomson, 1962). In some cases much longer invaginations, which lie generally in the peripheral region of the stroma, occur (Thomson, 1974). There appears to be no free connection between the intermembrane space of the envelope and the thylakold space. This follows from the observation that, in intact spinach chloroplasts, a large pH gradient between the thylakoid space and the external medium can be maintained. Upon illumination, a massive influx of protons into the thylakold space occurs : the stroma becomes alkaline and the pH of the suspending medium decreases slightly (Heber and Krause, 1971 ; Heldt et al., 1975). If connections between the thylakoids and the inner envelope membrane really do exist, leakage from the acidic thylakoid through any connections should result in a considerable decrease of the pH of the external medium. However, in greening plants, continuities between the inner envelope membrane and the internal system are frequently seen and this phenomenon accompanies the development of chloroplasts in young non-etiolated tissue (Fig. 9). For example, in the proplastids of Oenothera hookeri leaf meristem, the inner envelope membrane invaginates in places into the stroma. These intrusions, about 0-5 pm long, produce flattened sacs joined to the inner envelope membrane by a narrow neck (Menke, 1964). As differentiation proceeds further, the connections between the inner envelope membrane and the thylakofds are no longer visible. More recent observations have been made by Cran and Possingham (1974), Whatley (1974, 1977), Damsz and Mikulska (1976), Platt-Aloia and Thomson (1977) and Priestley (1977). Inner membrane invagination may also be a feature of the development of etioplasts and the etioplast-chloroplast transformation (Kirk and Tilney-Bassett, 1967; Rosinski and Rosen, 1972; Boasson et al., 1972; Cran and Possinghan, 1973; Kasemir et al., 1975), (Fig. 10). Consequently, it has been suggested that the thylakoid membranes are formed by continued invaginations from the inner sheet of the plastid envelope (Miihlethaler and Frey-Wyssling, 1959; Von Wettstein, 1959; Menke, 1960, 1962; Kirk and Tilney-Bassett, 1967; Miihlethaler, 1971). There is, however, no convincing evidence for this. Several investigators have suggested that membrane invaginations in young plastids may also take the form of isolated sac-like “inpushings” which, usually, are noticeably less electron-dense than other chloroplast membranes (Israel and

STRUCTURE AND FUNCTION OF THE PLASTID ENVELOPE

13

Steward, 1967; Stetler and Laetsch, 1969; Sjolund and Weier, 1971 ; Rosinski and Rosen, 1972). In a mutant of Chlamydomonas reinhardtii, Ohad et al. (1967) found no evidence of inner envelope membrane invagination during rapid thylakoi‘d biogenesis. Analysis of images of freeze-factured membranes of mature chloroplasts of Portulaca oleracea show clearly that the thylakoi’ds which are supposed to have originated from the inner envelope membrane do not conserve the particle size and distribution of the inner envelope membrane (Sprey and Laetsch, 1978). As we shall see later, the chemical composition (pigments, lipids and proteins) of the plastid envelope is qualitatively

Fig. 9. Proplastid in Viciu fuhu root tip cell. This proplastid displays the two classical concentric membranes (outer and inner) of the envelope, with occasional imaginations of the inner membrane (arrow). The internal membrane system (thy1: thylakoi‘d) is sparse and not organized. A small starch grain (S) is present in the section. Micrograph provided by Drs B. E. S. Gunning and M. W. Steer; reproduced with permission from Edward Arnold Publishers, “Ultrastructure and the Biology of Plant Cells” (1975).

14

R. DOUCE AND J. JOYARD

and quantitatively very different from the composition of thylakolds. Clearly, if this form of “inpushing” frequently observed represents a rudimentary lamellar membrane, considerable compositional modification must take place following the initial step of invagination. Under these conditions, it is likely that thylakold growth is not restricted to any one small area of the inner envelope membrane such as at a point of invagination but occurs all over the thylakofd membrane (Park and Sane, 1971; Arntzen and Briantais, 1975) in a multistep assembly process (Siekevitz, 1972). It must be pointed out that although the great majority of reports of envelope invagination refer to work with developing chloroplasts, this feature undoubtedly occurs at other stages. Particularly interesting micrographs of this phenomenon in mature chloroplasts have been published by Sitte (1962) and Weier and Thomson (1962) and invagination of the inner membrane has also been reported for chromoplasts and for chloroplasts undergoing transformation into chromoplasts and vice versa (Thomson et al., 1967; Rosso, 1968; Catesson, 1970; Sjolund and Weier, 1971 ; Laborde and Spurr, 1973; Gronegress, 1974; Sitte, 1974; Huber and Newman, 1976), (Fig. 11). Direct connections between the plasmalemma of blue-green algae (equiv-

Fig. 10. Greening of an etiochloroplast from maize leaf mesophyll. The inner membrane (im) of the envelope is connected (arrow) with peripheral reticulum (pr). Note (circle) the active role of peripheral reticulum in thylakold (thyl) formation (om: outer membrane, pb: prolamellar body). Micrograph provided by J. Farineau, T. Guillot-Salomon and R. Popovic; reproduced with permission from Bio/.,-Ce//u/uire (1 978) 32, in press.

Fig. 11. Chromoplasts at different developmental stages. A: young chromoplast from a tomato fruit. Numerous undulating membranes containing lycopene crystals (stars) seem to be derived from the inner membrane of the envelope. The double envelope membranes are clearly visible (small circles) (V: vacuole, T: tonoplast). B: mature chromoplasts from the corolla tube of a Narcissus poeticus flower. Their internal membranes equivalent t o the inner membrane of the plastid envelope are convoluted. The clear zones between them represent probably crystals of B-carotene. Micrographs provided by Drs B. E. S. Gunning and M. W. Steer, reproduced with permission from Edward Arnold Publishers, “Ultrastructure and the Biology of Plant Cells” (1975).

16

R. DOUCE AND J. JOYARD

Fig. 12. Mesophyll chloroplast of maize (Zea mays L.) showing a well developed peripheral reticulum and an extensive grana fretwork system. The lower figure shows numerous continuities between the inner membrane (im) of the plastid envelope and the peripheral reticulum (pr), (om: outer membrane). Note the absence of connections between the grana fretwork system and the peripheral reticulum. Micrograph provided by Drs R. Chollet and D. J. Paolillo, reproduced with permission from Z. Pfiphys. (1972) 68, 304.

alent to the inner membrane of the plastid envelope) and outer thylakolds have also been reported in a range of species (for a review, see Lang and Whitton, 1973) although Echlin (1964) and Whitton et al. (1971) were unable to confirm any such connections in Anacystis montana and Chlorogloea fritschii. Stanier (1 970) suggested that direct connections between the

STRUCTURE AND FUNCTION OF THE PLASTID ENVELOPE

17

plasmalemma and thylakoi'ds are probably very rare in blue-green algae. Consequently, we believe that the sac-like intrusions of the inner membrane of the plastid envelope, which may have arisen in response to a need for more efficient transport, facilitate the passage of low molecular weight metabolites and also proteins into the developing organelle.

Fig. 13. Portion of a bundle sheath cell of maize (Zea mays L.). The chloroplasts contain mostly non-appressed, parallel lamellae and a well developed peripheral reticulum. The inset shows numerous continuities between the inner membrane (im) of the plastid envelope and the peripheral reticulum (pr) (om: outer membrane). Micrograph provided by Drs R. Chollet and D. J. Paolillo; reproduced with permission from Z . P'phys. (1972) 68, 30-44.

A fascinating variation in the morphology of the inner envelope membrane has been described in all chloroplasts of C, plant mesophyll (Fig. 12) and bundle sheath cells (Fig. 13). In the peripheral stroma of these chloroplasts, Shumway and Weier (1967) and Laetsch (1968) reported for the first time an extensive system of anastomosing tubules contiguous with the inner membrane of the chloroplast envelope (Chollet and Paolillo, 1972; Lunney et a/., 1975) and called the peripheral reticulum (Laetsch, 1968; Rosado-Alberio et al., 1968; Laetsch, 1974). The appearance of peripheral reticulum is variable in vivo, since the anastomoses are dynamic (Lyttleton et al., 1971 ; Laetsch, 1974). The presence of the peripheral reticulum in chloroplasts of C, plants has been reported in a few cases (Bisalputra et a/., 1969; Hilliard and West,

18

R. DOUCE AND J. JOYARD

1971; Taylor and Craig, 1971; Mache and Loiseaux, 1973; Pallas and Mollenhauer, 1972; Laetsch and Kortschak, 1972; Valanne, 1975). In this case, and in contrast with the C, plants, the peripheral reticulum is not highly developed except perhaps for guard cell chloroplasts (Pallas and Mollenhauer, 1972). The peripheral reticulum has been reported to be continuous with the thylakolds by Rosado-Alberio et al. (1968). However, many of the fixation images in that study were produced by KMnO, which alters peripheral reticulum morphology (Laetsch, 1971). Furthermore, in an interesting study of serially-sectioned chloroplasts of Portulaca oleracea, connections between peripheral reticulum and the thylakofds membrane systems have not been observed (Sprey and Laetsch, 1978). Other studies of peripheral reticulum have failed to find such connections (Laetsch and Price, 1969; Osmond et al., 1969; Laetsch, 1971; Chapman et al., 1975). Freeze-fracture of the peripheral reticulum and envelope membranes of chloroplasts in the mesophyll cells from Portulaca oleracea (Sprey and Laetsch, 1978) shows a good correlation in particle size and particle distribution between corresponding fracture halves of peripheral reticulum membranes and the inner membrane of the chloroplast. Consequently, these results demonstrate that the peripheral reticulum, in contrast to the thylakolds, is derived from the inner envelope membrane (Laetsch, 1974), (Figs 10, 12 and 13). The functional role of peripheral reticulum in C, photosynthesis is a matter of speculation. It is likely that peripheral reticulum which greatly increases the surface of the inner plastid envelope of the mesophyll cell chloroplasts may facilitate the transport of metabolites from cytoplasm to chloroplasts and vice versa (Laetsch, 1974). B. THE OUTER ENVELOPE MEMBRANE AND THE EXTRA-CHLOROPLASTIC MEMBRANES

In contrast to the inner membrane, the outer membrane of the chloroplast envelope appears completely smooth in most electron micrograph pictures (Gunning and Steer, 1975). Yet the outer membrane has frequently been reported to be in close association with the smooth endoplasmic reticulum (Wooding and Northcote, 1965; Schnepf, 1969; Camefort, 1970; Dumas, 1974; Galatis et al., 1974; Northcote, 1974; Whatley, 1977). For example Priestley (1977) in an interesting study reports sheaths of endoplasmic reticulum around developing plastids of Phaseolus vulgaris. In the brown algae (Phaeophyceae) and Chrysophyceae,chloroplasts or phaeoplasts are shrouded in a layer of smooth endoplasmic reticulum (“chloroplast endoplasmic reticulum”, Bouck, 1965) lying just external to the chloroplast envelope (Fig. 14). In this case the envelope seems to be three-layered (Gibbs, 1962a, 1962b, 1970; Loiseaux et al., 1974). In algae, this endoplasmic reticulum system joins with the nuclear external membrane (Bisalputra, 1974), (Fig. 14).

STRUCTURE AND FUNCTION OF THE PLASTID ENVELOPE

19

Fig. 14. In the brown algae (Phaeophyceae) plastids (phaeoplasts) are shrouded in a layer of smooth endoplasmic reticulum (chloroplast endoplasmic reticulum). Arrows indicate connections between chloroplast endoplasmic reticulum and outer nuclear membrane. Micrograph provided by Dr S. Loiseaux; reproduced with permission from Masson (Atlas Micrographique de Cytologie Vegetale, by S. Loiseaux, F. Nurit and R. Mache).

An obvious interpretation is that the sheaths could serve in the efficient collection of raw materials produced by plastids and passed along the cisternae to sites of lipid and protein production. It is also possible that the periplastid endoplasmic reticulum especially in some specialized gland cells helps in transporting material moving to and from the plastids (Wooding and Northcote, 1965). For example it has been clearly shown that various products of secondary metabolism such as terpenoid compounds (Rogers et al., 1968; Schnepf, 1969; Amelunxen and Gronau, 1969; Heinrich, 1970; Loomis and Croteau, 1973; Carde, 1973) and hydrophobic flavonoid aglycone compounds (Schnepf and Klasova, 1972 ; Charriere-Ladreix, 1977), which are probably synthesized on the plastid envelope, are exported via the smooth endoplasmic reticulum. It is also possible that during the various stages of intracellular membrane breakdown, plastids become encircled by smooth endoplasmic reticulum which presumably then protects the organelle from autolytic activity within the cell (Northcote and Wooding, 1965). In a careful study of cell membranes in the fern Pteris vittata, Crotty and Ledbetter (1973) found membrane continuities between the outer envelope

20

R. DOUCE AND J. JOYARD

membrane of the chloroplast and the smooth endoplasmic reticulum (Fig. 15). Cran and Dyer (1973) described the same thing in the gametophyte of the fern Dryopteris borreri. An intimate association between endoplasmic reticulum and plastids during microsporogenesis in Lycopersicum esculentum, Lycopersicum peruvianum and Solanum tuberosum has been recently reported (Pacini and Cresti, 1976; Abreu and Santos, 1977). Schotz and Diers (1975) have also described membrane continuities in an abnormal hybrid of Oenotkera. It is worthy of note that the continuity of the outer mitochondria1 membrane with the smooth endoplasmic reticulum membrane has already been established in both plant and animal cells (MorrC et al., 1971). However, the infrequency with which direct endoplasmic reticulum-chloroplast envelope associations have been observed has led Bourdu (1975) to question the significance of membrane continuities between plastids and endoplasmic

Fig. 15. Direct continuity between the outer membrane of the chloroplast and the outer membrane of the mitochondria (arrow at left), and a more complex interconnection between the outer chloroplast membrane, the endoplasmic reticulum, and the outer mitochondria] membrane (arrows at right) in the fern Pferis viffutu L. Micrograph provided by Drs W. J. Crotty and M. C. Ledbetter; reproduced by permission from Science (19731, 182, 839-841. Copyright 1973 by the American Association for the Advancement of Science.

STRUCTURE AND FUNCTION OF THE PLASTID ENVELOPE

21

reticulum. Furthermore, the outer envelope membrane and the endoplasmic reticulum are quite dissimilar (Douce, 1974) and some local specialization at the molecular level may be needed to permit the fusion of the two. A further complication is the possibility of artifactual membrane fusions in the course of fixation for electron microscopy (Haussmann, 1977). This clouds the issue of whether the endoplasmic reticulum cisternae can fuse, even temporarily with the outer envelope membrane in order to allow the transfer of membrane components from the cytoplasm to the chloroplast and vice versa. Direct continuities of the outer envelope membrane with other cytoplasmic membranes may occur in some specialized tissues, but the documentation of this phenomenon is probably restricted to the interesting report of Crotty and Ledbetter (1973), (Fig. 15) and recent accounts from Kursanov and Paramonova (1976) and Paramonova (1977). Of special interest are the continuities between the outer chloroplast membrane and the plasma membrane. If such continuities really occur, the products of photosynthesis could be directly and rapidly discharged to the outside of the cell. Direct junctions between the plastids and the plasma membrane are most unlikely however, because chloroplasts of various origin are mobile inside cells (Honda et al., 1964; Wildman et al., 1966; Mayer, 1971) by passive and/or active movements within the streaming cytoplasm. Several reports have described close associations of mitochondria with the chloroplast outer envelope membrane (Jacobson et al., 1963; Dolzmann and Ullrich, 1966; Gunning and Steer, 1975; Montes and Bradbeer, 1976). A report based on light microscopy of living cells by Wildman et al. (1966) suggested that long envelope-bound protuberances could detach from the chloroplast to form mitochondria. However, this concept has been criticized (Possingham et al., 1964; Silayeva, 1969; Chapman et al., 1975). More recently, Wildman et al. (1974) have proposed that mitochondria may fuse completely with the chloroplast envelope and that subsequently starch grain formation occurs at the site of the fusion. It seems likely that at least some of the chloroplast mitochondria interactions reported by Wildman’s group may be easily explained by the formation and loss of deep mitochondria-containing invaginations in the chloroplast envelope (Montes and Bradbeer, 1976), (Fig. 16). In the intact leaves, the peroxysomes are usually found closely associated with the outer membrane of the chloroplast envelope (Gruber et a[., 1970; Frederick and Newcomb, 1969a, 1971; Vigil, 1973; Frederick et al., 1975). Recently, Schnarrenberger and Burkhard (1977) have reported an in vitro association of chloroplast and peroxysomes. They also reported evidence that inorganic phosphate can cause an interaction between chloroplasts and peroxysomes during isolation by isopycnic centrifugation in sucrose gradients, which implies that a form of membrane recognition is involved. Although inference of a physiological relationship from results of this sort may be

22

R. DOUCE AND J. JOYARD

Fig. 16. An association of chloroplasts and mitochondria in Funaria hygrometrica. Mitochondria are sometimes found in deep invaginations situated near the periphery of chloroplasts. Micrograph provided by F. Nurit (unpublished data).

premature in the absence of more conclusive information, it is possible that transfer of metabolites between organelles could be greatly facilitated by close proximity of boundary membranes (Fig. 17). Furthermore, the close association between chloroplasts and various cell organelles observed in most electron micrograph pictures may facilitate the cooperative action of enzymes in chloroplasts, peroxysomes, mitochondria and the cytoplasm in which they lie and so allow sufficiently high rates of carbon flow for photorespiration (Tolbert, 1971; Chollet and Ogren, 1975; Schnarrenberger and Fock, 1976; Chollet, 1977; Woolhouse, 1978). C. THE INNER AND OUTER ENVELOPE MEMBRANES

The presence of areas in which the inner and outer membrane of the chloroplast envelope have fused together has been reported recently by Priestley (1977). Similar electron dense areas in the periplastidal space may be seen in micrographs published by Surzycki et al. (1970) and Newcomb and

STRUCTURE AND FUNCTION OF THE PLASTID ENVELOPE

23

Fig. 17. Peroxysomes are bounded by a single membrane, they are very often appressed to the outer membrane of the chloroplasts. Arrow indicates contact point between the outer and the inner membranes of the chloroplast envelope. Micrograph provided by Drs S. E. Frederick and E. H. Newcomb; reproduced by permission from the Rockefeller University Press, J. Cell. Biol. (1969) 43, 343-353.

Frederick (1971), (Figs 17 and 18). The osmiophilic features reported by Priestley (1977) are obviously lipid rich. Studies with mammalian membranes have already indicated that membrane fusion occurs in protein depleted zones (Lucy, 1975). It is possible that the zones of fusion may serve to maintain the overall structure of the envelope (Silayeva and Shiryayev, 1966). A further function may be that of protein transport. It is also well known that an intermembrane space exists between the two membranes of the intact mitochondria. Scholte (1973) has reviewed the difficulties encountered by himself and others in isolating the outer membrane of heart mitochondria and has offered the suggestion that certain proteins are involved in binding together the outer and inner membranes of these mitochondria. Several authors (Hackenbrock, 1968; Huber and Morrison, 1973; Hackenbrock and Miller, 1975) have clearly demonstrated that a number of contact sites occur between

Fig. 18. Contact points occur between the inner membrane (im) and the outer membrane (om) of the chloroplast envelope (Spinaciu oleruceu L.). A: in a hypertonic medium, the outer envelope membrane of the chloroplasts appears to be loosely attached to the inner envelope membrane with large empty spaces in between. B: Separation of the two envelope membranes at a higher magnification. Note that the two layers of the outer membrane are clearly visible. Micrograph provided by Dr J. P. Carde (unpublished data).

STRUCTURE AND FUNCTION OF THE PLASTID ENVELOPE

25

the two mitochondrial membranes. Hackenbrock and Miller (1975) have also shown that the outer mitochondrial membrane, partially disrupted by treatment with digitonin, remains attached to the inner membrane at these contact sites as inverted vesicles. It is clear that a variety of fixation and staining conditions are needed in order to establish firmly the existence of such fusions in the chloroplast envelope. IV. ISOLATION O F THE CHLOROPLAST ENVELOPE A. PRINCIPLES

Mackender and Leech (1970) first reported a method for the isolation of a fraction “enriched” in envelope membranes from Vicia faba chloroplasts. This method involved the initial preparation of chloroplasts by differential centrifugation with a good degree of structural integrity. These chloroplasts were then burst by hypotonic shock and the “envelope” was separated from the bulk of the thylakold system by differential centrifugation. These authors reported that in the hypotonic medium the intact chloroplasts were initially opaque and shiny in appearance but after a short time the suspension contained many chloroplasts with greyish halos (Fig. 19). In each of these

Fig. 19. “Balloon” derived from spinach chloroplasts maintained in a low osmolarity medium. A and B: C type chloroplasts, swollen thylakolds (seen as greyish halos) give the impression that the chloroplast is intact and surrounded by the envelope. Under UV light, this balloon gives a red fluorescence indicating the presence of chlorophyll. C: after some minutes, numerous pieces of thylakolds are clearly visible around the balloon.

chloroplasts the thylakold system was excentric within “the greyish balloonlike structure” (TrCcul, 1858; Weier et al., 1967). According to Mackender and Leech (1970) the balloons were assumed to derive from the chloroplast envelopes. However, the swellings illustrated by Mackender and Leech take the form of spheres with a diameter of 8-15 pm (Fig. 19) and a consequent area of approximately 200 pm2. If the shape of a chloroplast approximates to a hemisphere of 30 pm3 a surface area of about 55 pm2 may be predicted (Priestley, 1977). In these conditions, the area covered by the balloon-like swelling (200 pm2) is several times greater than the area of the envelope in its normal state. It is most unlikely that stretching of the envelope membranes

26

R. DOUCE AND J. JOYARD

could account for this disparity because once the shape of a sphere is reached, further swelling is limited by the low elasticity of the membranes and bursting occurs (Burton, 1970). Demonstrable connections between the envelope inner membrane and the thylakoid system in mature chloroplasts have never been reported. Consequently, and in contrast to the inner mitochondria1 membrane (Moreau and Lance, 1972; Douce et al., 1973a; Day and Wiskich,

Fig. 20. Mechanism which explains the bursting of intact mitochondria in a low osmolarity medium. 1: intact mitochondria, in 0.3 M mannitol; 2: intact mitochondria in low osmolarity medium; water enters very quickly into the matrix space and the mitochondria swell. The inner membrane, owing to the internal cristae, ,easily supports the swelling in contrast to the chloroplast envelope (see Fig. 24). On the contrary, the outer membrane is unable to swell and bursts in a few seconds (arrows: points of rupture of the outer membrane); 3: the ruptured outer membranes are present as vesicles in the medium.

1975), (Fig. 20), the stretching of the envelope cannot originate from the addition of materials coming from the lamellar system. Moreover, the timecourse of the liberation of stromal proteins from hypotonically-shocked intact chloroplasts of Spinacia oleracea indicates that the majority have lysed within 10-15 s and that bursting occurs much earlier than the balloon formation (Joyard and Douce, 1976a). Finally, it has been shown that the balloon observed by light microscopy under ultraviolet light gives a strong red fluorescence indicating the presence of chlorophyll (Billecocq et al., 1972). This last result is in contrast with the fact that the envelope membranes are devoid of chlorophyll, an interpretation derived from light microscopy of chloroplasts in vivo (Honda et al., 1966). Consequently, it is more likely that the single membrane-bound balloons observed by Mackender and Leech (1970) are derived from the stromal lamellae (Hoshina and Nishida, 1970). The first crude envelope preparation reported by Mackender and Leech (1970) identified only by the use of phase contrast and electron microscopy was strongly contaminated by thylakoids (8-16 % of the membrane proteins in their envelope fraction are lamellar in origin) and by various extra-chloroplastic membranes such as mitochondria. In order to reduce the granal

STRUCTURE AND FUNCTION OF THE PLASTID ENVELOPE

27

fragments in the envelope fraction, the crude envelope was purified by density gradient centrifugation which reduced lamellar contamination by 75 % (Mackender, 1971; Mackender and Leech, 1972). Unfortunately, the envelope pellet obtained in this way appeared dark green with a yellow-green periphery and the yield of purified envelope membranes obtained (250 pg envelope protein per 100 g fresh weight leaves) was considered by the authors to be impracticably small for further analysis. In fact, the preparation of chloroplast envelopes is extremely difficult for the following reasons : a. The chloroplast is, in general, a delicate structure. Skill and expertise are required in order to avoid large-scale rupture of the two membranes of the envelope during chloroplast isolation (Walker, 1971). “It is difficult at present, if not impossible to prepare 100% intact and fully functional chloroplasts; it is a simple matter to prepare 100 % envelope free chloroplasts. Regrettably, many methods which are currently used yield preparations containing an appreciable proportion of intact chloroplasts and may lead, in consequence, to ambiguous and possibly misleading results.” (Walker and Crofts, 1970). The crucial point of the whole procedure is the grinding of the leaves. In order to obtain intact chloroplasts it is absolutely necessary to restrict the grinding procedure to a minimum (2-5 s of blending with a Waring blendor). Longer blending improves the yield of recovered chlorophyll but increases the proportion of broken chloroplasts (Classes C and D according to the nomenclature of Hall, 1972). The criterion for determining the percentage of intact chloroplasts is the comparison of the uncoupled rate of ferricyanide reduction before and after subjecting the preparation to osmotic shock. Since the chloroplast envelope is impermeable to ferricyanide only the broken chloroplasts in a suspension will be able to reduce this electron acceptor (Heber and Santarius, 1965; Cockburn ef al., 1967c; Mathieu, 1967; Heber and Santarius, 1970; Lilley et al., 1975). The two types of chloroplasts can also be distinguished by phase contrast microscopy (Leech, 1964; Spencer and Unt, 1965; Walker, 1965a), (Fig. 21). All preparations containing a high proportion of damaged chloroplasts must be discarded, otherwise the final envelope pellet will be contaminated with broken thylakoyds. Under these circumstances, the final yield of intact chloroplasts capable of C0,-dependent 0, evolution (average rate 80 pmol 0, mg-l chlorophyll h-l) is very low. For example, when the excellent classical methods of Walker (1964), Bucke et al. (1966), Jensen and Bassham (1966), Cockburn et al. (1967b), Heber (1973) and Nakatani and Barber (1977) are used, we have shown (Joyard and Douce, 1976a) that the maximum yield of intact chloroplasts obtained is roughly 1-2 % of the total leaf chloroplasts. b. Even though the final preparation contains more than 95 % of intact chloroplasts, the material is not necessarily devoid of various extra- and intra-chloroplastic membranes. For example, a careful examination by

28

R. DOUCE AND J. JOYARD

Fig. 21. Appearance of chloroplasts under phase contrast microscopy. The intact chloroplasts (class A) are highly reflective and present a halo. The broken chloroplasts (class C) are dark and granular.

phase contrast and electron microscopy of the intact chloroplast pellet obtained by differential centrifugation shows that the chloroplasts are more or less contaminated by smooth vesicles of unknown origin, intact and broken mitochondria, peroxysomes and both intact and broken cells (Joyard and Douce, 1976a). Moreover, it is clear that metabolically active preparations capable of high rates of C0,-dependent 0, evolution are systematically contaminated by a variable proportion of damaged or envelope-free chloroplasts. The worst contamination encountered is most certainly small pieces of broken thylakoid rich in osmiophilic globules (Fig. 22). These globules or plastoglobuli (Greenwood et al., 1963; Lichtenthaler and Sprey, 1966; Lichtenthaler, 1968) have constantly been observed in sections of osmium-fixed chloroplasts from all types of plants (Figs 1 and 17). The circular profiles of plastoglobuli vary from 10 to 500 nm in diameter and are embedded in the stroma between the thylakoid membranes but never within a granum or in contact with the inner envelope membrane. The plastoglobuli are generally considered to represent a reservoir of excess lipid (Lichtenthaler, 1968). The broken thylakoids associated with plastoglobuli possess a similar density as the envelope membranes (Sprey and Laetsch, 1976a). Finally, the form of contamination depicted by Larsson et al. (1971, 1977) in which individual

STRUCTURE AND FUNCTION OF THE PLASTID ENVELOPE

29

Fig. 22. Very often, osmiophilic globules are associated with small pieces of swollen thylakoids. Micrograph provided by Drs B. Sprey and W. M. Laetsch; reproduced by permission from 2. Pjphys. (1976) 78, 146-163.

chloroplasts are surrounded by a jacket of cytoplasm containing various cell organelles is sporadically observed. In order to avoid the danger of cross-contamination during the preparation of envelope membranes, the intact chloroplasts obtained by differential centrifugation must be purified by isopycnic centrifugation in a sucrose gradient (Leech, 1964; Tuquet, 1972) or silica sol (Lyttleton et a/., 1971; Morgenthaler et al., 1974) or by passage through a loosely packed Sephadex column (Wellburn and Wellburn, 1971 ; Evans and Smith, 1976a). c. The envelope membrane proteins represent a small proportion of the total chloroplastic proteins. Joyard and Douce (1976a) have shown that in the case of spinach chloroplasts the envelope proteins represent only 0.7-1 % of the total chloroplastic proteins. In one gram of fresh weight spinach leaves the total amount of chlorophyll and chloroplastic proteins are respectively 1 and 18.4 mg (Lilley et al., 1975). Therefore, one gram of spinach leaves contains about 180pg of envelope protein. If we assume a maximal yield of intact purified chloroplasts (see above) of 2 %, the total amount of envelope we could expect is roughly 3-4 pg of protein per gram of fresh weight leaves. Higher yields obtained in the literature must be interpreted with caution. Taking all these problems into consideration Douce et al. (1973b) reported a method for the isolation of the chloroplast envelope membranes from Spinacia oleracea. This material was chosen because : it is easily macerated

Chop 1 kg ( 7 w chlwcphyll) ~ olderibbed spinach leaves Homogenize in a

Waring Blendor

one gallon

at low speed lor 5 sec with 2 litrer 01 grinding medium ( sucrose 0 3 M I IricinwNaOH 30 mM. PH 7 8 , bovine serum albumin 1 9 per litrc) Pass through 8 layers 01 muslin and one of nylon blUteK ( 5 0 vm

*Idth) Cmntrltuwtions as

O IO I W~

I

I

s1

Pl

(discard)

Resuspended in 2 0 0 m l 01 suspension medlum (sucrose:0.3M ;trlcine-NaOH: 1OmM.pH 7.8 ).

I

w I

1

P2

s2 (discard)

Pelleis lrom 2 or 3 extractions resuspendad In 120 ml 01 suspension medlum.30ml 01 tho mixture layered

on

top 01 a sucrose gradient (4tubes.150ml each)

made of 3 layers

( 30

ml each)

containing: 1.5 , 1.0 end 0.75 M sucrow ;trlcine~NaOH10 mM . p n 7.6

4 input

4 0.75 M 4 1.0 M

4

1.5 M

I

s’3 (discard)

s”3(discard)

P3

s3

(discard)

0 S3 mcwered and dilulul very carnlully with eu6pe~mmedium

I

w I

1

P4

s4 Idiscard)

INTACT AND PURIFIED CHLOROPLASTS

Intact and purilied chloroplasts ( 6 to 12 mg chlorophyll per ml ) diluted in 90 ml of swelling medium ( t r i c i n e - N a O H : 10 m M , p H 7.6 ; Mg CI2: 4 m M ). 15 ml of the mixture

of a sucrose gradient ( 6 tubes,40

ml

layered on top

each) made of 2 layers ( 12 ml each ) Containing

0.93 and 0.6 M sucrose, tricine-Na OH l O m M and Mg Clp

4 mM.

72 000 g ( R max)

4 4 4

1 hour rotor:SW 27 Beckman

input 0.6 M 0.93 M

I

t

s'5

s5

P5

ENVE LOPE

CHLOROPLAST

EXTRACT

THYLAKOIDS

0. Diluted 2 fold with tricine-Na OH 10 mM ,pH 7 . 6

113ooog (Rmax ) 45 min rotor: S W 27 Beckman I

r

1-1

1

P6

s6 Idiscard ) (about 10 mg proteins)

Wsuspended in 0.3M sucrose containing

l O m M tricine-NaOH

Fig. 23. Diagram summarizing the preparation of the envelope membranes of spinach chloroplasts (P: pellet, S: supernatant).

32

R. DOUCE AND J. JOYARD

with a simple salad mixer; its vacuoles are practically free of endogenous inhibitors and calcium oxalate; very often the chloroplasts are devoid of starch and it is freely available during most of the year almost universally. Furthermore, spinach chloroplasts are physiologically well characterized (Heber, 1974; Walker, 1974; Heldt, 1976a). Finally, intact chloroplasts from the majority of species assimilate COz at about 10% or less of the rate of comparable preparation from spinach (Walker, 1971). B. PROCEDURE

The procedure is fully described in Fig. 23 and can be divided into three steps (Douce et al., 1973b; Joyard and Douce, 1976a). 1. Isolation of Chloroplasts Chloroplasts are prepared from washed eight week old spinach leaves. From 2 kg of leaves the yield of intact and purified chloroplasts (P, pellet) is equivalent t o about 30-50 mg of chlorophyll (500-900 mg of chloroplastic proteins). As already mentioned, this low recovery represents 2 % of the total chlorophyll contained in the leaves. During the course of the purification of intact chloroplasts the use of a cation free medium is of prime importance (Joyard and Douce, 1976a). We have observed, in good agreement with Nakatani and Barber (1977), that when the salt content of the medium is negligible there is a better centrifugal separation of intact and broken chloroplasts. 2. Bursting of Intact Chloroplasts The intact and purified chloroplasts are suspended in a medium of low osmolarity (swelling medium). The presence of magnesium ions in the medium serves to maintain the association of the thylakoid membranes and prevents breakage of thylakolds during swelling (Murakami and Packer, 1971). Under these conditions, immediate swelling of intact chloroplasts causes rupture and total detachment of the envelope membranes with the liberation of the stroma material within ten seconds (Joyard and Douce, 1976a), (Fig. 24). The electron-dense plastoglobuli are still present between the thylakoids after this procedure. In order to facilitate the total envelope detachment, the use of a homogenizer (e.g. Ten-Brock ground-glass homogenizer) must be avoided especially if the clearance between the plunger and the homogenization vessel is small. Thus, for each “up and down stroke” applied, the fragile thylakofd network becomes more and more disorganized and very rapidly the suspension is enriched by small pieces of thylakoids rich in plastoglobuli which are afterwards difficult to separate from the envelope membranes. Furthermore, the rupture of intact chloroplasts by drastic procedures such as sonication or slowly forcing the chloroplasts through a very small aperture under high nitrogen pressure, yields small envelope membrane vesicles which are difficult to separate from broken

STRUCTURE AND FUNCTION OF THE PLASTID ENVELOPE

., '

..

.,sucrose,

..

.

" 0 . .

0 .^. v , .

33

0 - 0

.

0 0

O.-

k

0

0 h

Fig. 24. Mechanism which explains the bursting of intact chloroplasts in a low osmolarity medium. 1 : intact chloroplast in 0.3 M sucrose; 2: intact chloroplast in low osmolarity medium; water enters very quickly into the chloroplast, the envelope is unable to support the high pressure and bursts in a few seconds (arrows: points of rupture of the envelope); 3: the envelope is ruptured and is present as vesicles in the medium; then, water enters slowly into the internal space of the thylakolds. A swollen thylakoid looks like a balloon; 4: a detached balloon formed by thylakolds (equivalent to Fig. 19, c).

thylakolds. Finally, chaotropic agents such as digitonin and cholate should be avoided so that meaningful investigations into the chemical composition and physical structure of these membranes can be undertaken later. It is noteworthy that the chloroplasts of certain marine green and brown algae have an envelope which resists both mechanical and chemical breakage (Trench et af., 1973; Giles and Sarafis, 1974; Grant et al., 1976; S . Loiseaux, personal communication). For example, isolated chloroplasts from the siphonous green algal genus Caulerpa are able to retain their integrity and their ability to carry out C0,-dependent 0, evolution even after incubation

34

R. DOUCE AND J. JOYARD

for ten minutes in 0.1 % sodium dodecyl sulphate (Grant et al., 1977). In this case, a mechanical system should be useful to disrupt the envelope membranes. 3. Isolation of Chloroplast Envelope Membranes Chloroplast components are then separated by a sucrose density gradient procedure (Fig. 24). After centrifugation, three subfractions can be distinguished: a tightly packed, dark green pellet (P,) at the bottom of the tube (thylako‘id subfraction); a yellow band (S,) at the interface of the two sucrose layers (envelope membranes subfraction) and a clear brown supernatant (S’,, soluble subfraction or chloroplast extract) (Douce et al., 1973b; Poincelot, 1973; Douce and Benson, 1974; Poincelot and Day, 1974; Mendiola-Morgenthaler and Morgenthaler, 1974; Hashimoto and Murakami, 1975; Tuquet and de Lubac, 1975; Sprey and Laetsch, 1975; Joyard and Douce, 1976a; Priestley, 1977; Priestley and Woolhouse, 1977). In contrast to recent reports of Poincelot and Day (1974), there is no membranous fraction at the top interface although occasionally a whitish band rich in nucleic acid material derived from the stroma subfraction occurs. When the initial preparation of intact chloroplasts is contaminated by smooth vesicles of unknown origin or by tiny pieces of thylakoid containingplastoglobuli (unpurified chloroplasts and/or damaged chloroplasts), an additional band, visible as a milky yellow-green band when observed by lighting from behind, is observed at the top interface. This band which is merely a simple artifact is due to the “damaged or separated envelope membranes” (d = 1.08 g cm-3) described by Poincelot and Day (1974). The envelope fraction (the 0.6-0.93 M sucrose interface) corresponds to the “double membrane fraction” described by Poincelot and Day (1974). Consequently, discontinuous density gradient centrifugation for the isolation of envelope membranes is a very “dangerous procedure in that it creates the illusion of a clear-cut separation” (de Duve, 1971). It is interesting to note that the envelope membranes of Euglena grucilis chloroplasts (Vasconcelos et al., 1976), Triticum sativum etioplasts (Bahl et al., 1976), Avena sativa etioplasts (Cobb and Wellburn, 1974) and Narcissus pseudonarcissus chromoplasts (Liedvogel et al., 1976) are also found at the 0.6-0.93 M sucrose interface. According to Joyard and Douce (1976a) and Priestley (1977), low levels of magnesium in sucrose layers (< 0.5 mM) lead to massive contamination of the envelope fraction with thylakoid fragments. In contrast, high levels of this cation (> 5 mM) cause the envelopes to sediment towards denser regions of the gradient. Magnesium must be used with caution and its concentration established for each species studied. The effect of magnesium on the envelope membranes is still obscure. However, magnesium ions are thought to exert effects on membranes by interacting with both anionic lipids and proteins (Hauser et al., 1976). An excessively high concentration of magnesium may

STRUCTURE AND FUNCTION OF THE PLASTID ENVELOPE

35

lead to adhesion of dissimilar membranes and a consequent change in their properties during centrifugation (Price, 1974). The envelope preparation can be satisfactorily checked for cross-contamination by several methods. First, the colour of the envelope membranes is examined by transmitted light. If free of thylakofd membranes they are clear and yellow (Neuburger et al., 1977). In contrast, in the case of contaminated envelope preparations a tiny green “pin-head’’ spot occurs at the base of the centrifuge tube embedded at the centre of the yellow pellet (P6, see Fig. 23). Secondly, electron micrographs of the purified envelope fraction show relatively large vesicles (average diameter: 0.5 pm) or elongated profiles bordered by a single or a double membrane (Fig. 25). In many instances, the vesicles contain smaller round vesicles bordered by a single membrane of the same thickness as the surrounding vesicle. No plastoglobuli are trapped in the network of the envelope membranes (Fig. 25). As already stated, the envelope preparations obtained from unpurified chloroplasts are heavily contaminated by other chloroplast constituents particularly by small pieces of thylakoid containing plastoglobule clusters (Sprey and Laetsch, 1976a). Thirdly, NADH : cytochrome c oxidoreductase activity is negligible in the envelope fraction and the envelope is devoid of b-type cytochrome. This is in contrast to the outer membrane of plant mitochondria which contains at least one NADHreducible b-type cytochrome with a double peaked a band centred at 553 nm a t low temperature (Douce et al., 1973a; Moreau, 1978). Hence, the chloroplast envelope membranes when carefully prepared are essentially free of microsomal and mitochondria1 membrane contamination (Douce er al., 1973b). Poincelot and Day (1974) have suggested that double membrane-bound vesicles are an important feature of the isolated chloroplast envelope being indicative of its state in vivo. Unfortunately the poor quality of the electron micrographs given by Poincelot and Day (1974) show only three incomplete double membrane-bound vesicles. In addition, very often isolated membranes in close juxtaposition when examined by electron microscopy give the appearance of double membrane vesicles. Double membrane vesicles do occur in the Spinacia oleracea envelope fraction (Douce et al., 1973b; Sprey and Laetsch, 1975; Joyard and Douce, 1976a), Vicia faba envelope fraction (Mackender and Leech, 1974) and in the Phaseolus vulgaris envelope fraction (Priestley, 1977) but we believe that these may have arisen fortuitously during membrane fractionation. Under these conditions, it is not clear whether both the inner and the outer membranes are retained in the preparation. Likewise, we have at present no useful knowledge about the orientation of these isolated vesicles. It is possible that the unnatural process of hypotonic shock markedly affects membrane structure. It may also introduce serious modifications in the molecular architecture of membranes. The envelope membranes of class A chloroplasts, for example, are resistant to

Fig. 25. Electron micrographs of thin sections prepared from the following material. A: intact and purified spinach chloroplasts (P4pellet, Fig. 23); B: envelope pellet (Pa pellet, Fig. 23); C and D: envelope vesicles (S5, Fig. 23).

STRUCTURE AND FUNCTION OF THE PLASTID ENVELOPE

37

attack by phospholipase A, (EC 3.1.1.4.) whereas the isolated envelope vesicles are deacylated very rapidly (Tuquet and de Lubac, 1975). According to Sprey and Laetsch (1976b), the unique particle distribution of split inner and outer membranes of the envelope present in isolated intact chloroplasts could serve as a marker to identify the envelope membranes in the isolated envelope fraction (Fig. 4). On the evidence of these simple morphological criteria they conclude that the envelope-membrane fraction consists mostly of outer envelope membranes and that possibly most of the inner envelope membrane vesicles, enriched in proteins, co-sediment with the main bulk of thylako'ids during the centrifugation process. This is unlikely for the following reasons : a. Flugge and Heldt (1976, 1977) have shown that the major protein of the envelope fraction with a molecular weight of 29 000 contains a binding site essential for the transport of phosphate and is involved in specific phosphate transport across the envelope membranes. This interesting result indicates that the inner plastid envelope must be present in our envelope fraction since the phosphate translocator (Heldt and Rapley, 1970), which is an important part of the CO, fixation system, has been shown clearly and specifically to be associated with the inner membrane of the plastid envelope (Heldt and Sauer, 1971). b. Points of fusion between inner and outer envelope membranes noted above (Fig. 18) may, if they exist, form a basis for the association of the two membranes when the envelope is disrupted. Previous methods used for the isolation of envelope membranes depended primarily on morphological criteria. Although considerable progress was made by early workers in this field using such criteria, critical evaluation of the purification procedure used was not possible since no unique enzymatic activity characteristic of this membrane system was known. The recent observation that in spinach chloroplasts galactosyltransferase activities are localized exclusively in envelope membrane vesicles (Douce, 1974) paved the way for the isolation of this membrane system in large quantities. Finally, all results which have been reported so far do not provide additional information about the separation of the inner and outer membranes of the chloroplast envelope and all attempts which have been made in our laboratory in order to separate these membranes have failed. V. CHEMICAL COMPOSITION O F THE CHLOROPLAST ENVELOPE

The chloroplast envelope membranes, when compared with other cell membranes analysed so far, exhibit a most unusual chemical composition. It must be emphasized, as mentioned before, that the chemical properties attributed to the isolated envelope membranes vary considerably from one

TABLE I Lipid Composition of Isolated Chloroplast Envelope Membranes. Data are Expressed as Dry Weight Percentage of Total Lipids (Excluding Pigments). (-: not recorded, tr: traces < 0.5) Species

MGDGDGDG TGDG TTGDG

SL

PC

PG

PI

PE

8 6 13

1 1

tr

0

-

-

tr

tr -

Spinacia oleracea Spinacia oleracea Spinacia oleracea (experiment I) Viciafaba

20 27 8

30 33 29

4 1 5

6 tr 6

20 25 27

27

37

-

-

2

Narcissus pseudonarcissus Triticum safivum Helianfhusannus Zea mays (rnesophyll)

24 22 31 46

26 43 26 18

-

5 2 3 9 14 10 11 1 2 9 5 3 7 2

-

tr

tr

8

8

3 0 1 1

PS

DPG

-

0 0 2 1

0 0

-

0 0 1

References Douce et al. (1973b) Poincelot (1973, 1976) Hashimoto and Murakami (1975) Mackender and Leech (1972, 1974) Liedvogel et al. (1976) Bahl et al. (1976) Poincelot (1976) Poincelot (1976)

STRUCTURE AND FUNCTION OF THE PLASTID ENVELOPE

39

group of workers to another (see for example Table I). Therefore, the reported disparities may arise due to contamination of envelope preparations by extra- and intra-chloroplastic membranes. Up to now, practically all analyses have been carried out on higher plant chloroplast or etioplast envelope membranes. It is probable, however, that there is a high degree of uniformity in the chemical compositions of the envelopes of various types of plastids in higher plants. A. POLAR LIPID COMPOSITION OF THE CHLOROPLAST ENVELOPE

Douce et al. (1973b) and Mackender and Leech (1972, 1974) were the first to report the polar lipid and fatty acid composition of the chloroplast envelope membranes of Spinacia oleracea and Viciafaba (Fig. 26). These investigators concluded that qualitatively the polar lipids of both types of chloroplast membrane, thylako’id and envelope are identical but the proportion in which they are present is different (Table 11). In the thylakoi’d membranes the two main galactolipids, monogalactosyldiacylglycerol and digalactosyldiacylglycerol, are present in a ratio of 2 : 1 and the two main phospholipids, phosphatidylcholine and phosphatidylglycerol, are present in a ratio of 1 : 3. This situation is totally reversed in the envelope fraction. In this case, monogalactosyldiacylglycerol and digalactosyldiacylglycerol are present in a ratio of 0.3-0.8 : 1 and phosphatidylcholine and phosphatidylglycerol are present in a ratio of 3 : 1. Although the concentration of all polar lipids is greater in the envelope fraction than in the thylakoi’d fraction (Douce et al., 1973b) this is especially true for the relative concentration of phosphatidylcholine in the two types of membranes. The significance of these differences has yet to be resolved although they almost certainly reflect the difference in function of each of the membranous systems. Both groups of investigators concluded also that the fatty acids are more saturated in the envelope membranes than in the thylakoids (Table 111) but the difference found is not as great as reported by Poincelot (1 976). It is possible that the increased degree of saturation in the envelope membranes may explain their higher thermostability (above 35°C) when compared to the thylakoi’d membranes (Krause and Santarius, 1975; Poincelot, 1976). It could also explain the fact that after freezing for a short time all chloroplasts lose their envelopes without apparent damage to the thylakoi’ds (Krause and Santarius, 1975). Membranes exhibit a melting transition over a particular temperature range which depends critically on the nature of the fatty acid residues of the polar lipids (Lyons, 1973; Raison and Chapman, 1976). It is low for unsaturated polar lipids and much higher for saturated ones. Below the melting temperature the hydrocarbon chains are rigid whereas above it they are free to move. However, recently Champigny and Joyard (1978) failed to show a melting transition temperature with purified lipids (polar lipids diglyceride) from envelope membranes.

+

~ E U T R A LLIPIDS

-

I

GALACTOLIPIDS

MONOGALACTOWLDIGLYCERIDE (MOOG) 0

DIGALACTOSYLDIGLYCERIDE (DGDG 1

SULFOLIPID

SULFOOUINWSVLDIGLYCERIDE (SL)

PHOSPHOLIPIDS I

~HOSPHATIWLCHOLINE (PC)

PHOSPHATIWLGLVCEROL (PG)

Fig. 26. Structure of the main l i p i d s found in t h e c h l o r o p l a s t envelope. Note the specific p o s i t i o n of the main fatty a c i d s on each polar lipid.

TABLE I1 Polar Lipid Composition of Isolated Plastid Membranes and Mitochondria1 Membranes. Data are Expressed as Dry Weight Percentage of Total Lipids (Excluding Pigments). (-: not recorded) ~

~

~~

~

~

~

~

M G D G DGDG TGDG T T G D G

Membranes

-

~~

Mitochondria outer membrane inner membrane inner membrane Chloroplast envelope membranes thylakoids Etioplast envelope membranes prolamellar bodies

20 3

8 9

1 1

tr 0

14 4

9 9

0 0

0 0

1 -

6 7

10 10

{

20 51

30 26

22 36

44

-

-

41

-

~~~

PE

21 5 5

4

0 0 0

PI

10 3 4

-

O* O* 0

PG

~

0 42 0 4 1 0 3 3

0 0 0

.~

PC

0 0 0

{

{

SL

~

~

24 3 3

DPG ~~

~~~~

-

~~~

References ~

-

3 14 23

Moreau et al. (1974) Moreau et al. (1974) McCarty et al. (1973)

0 0

0 0

Douce et al. (1973b) Douceet al. (1973b)

0 0

0

Bahl et al. (1976) Bahl et al. (1976)

-

7 2

~~~

~~~

PS

-

~~~~~~

0

__

~

* In some species rich in proplastids and amyloplasts (cauliflower buds, potatoes and sycamore cells) the outer and the inner mitochondrial membrane

contains variable amounts of DGDG (R. Bligny and R. Douce, unpublished data). However, this lipid derives probably from the proplastids which are systematically present in the purified mitochondrial preparation. It is interesting to note that the mung bean mitochondria1 preparation devoid of proplastids does not contain DGDG.

TABLE 111 Fatty Acid Composition (% by Weight) of Isolated Envelopes ( E ) and Thylakofds ( T )from Chloroplasts Unsaturated Species

C16:O

(T"

15

E ]

41

Viciufaba

{!

13 6

Triticum sativum Helianthus

{:

Spinacia oleracea Spinacia oleracea

annus

Zea mays (mesophyll) Phaseolus vulgaris }E

8

C16:l 3 5

C16:3

C18:O

C18:l

9 13

tr

tr

6 2

C18:2

C18:3

Saturated

References

10 2

57 70

5.7 12.2

Douce et al. (1973b)

2

10

5

8

29

0.9

Poincelot (1976)

-

4

-

12 5

63 83

5.3 12.3

Mackender and Leech (1974)

-

2 1

6 3 2

8

67

4.0

Bahl et al. (1976)

19

3 2 1 2

36

3

1

5

2

14

40

1.4

Poincelot (1976)

0

3

14

2

4

13

0.3

Poincelot (1976)

6

-

4

1

6

62

3

Priestley and Woolhouse (1977)

21

STRUCTURE AND FUNCTION OF THE PLASTID ENVELOPE

43

All these results have been confirmed by several groups (Hashimoto and Murakami, 1975; Bahl et al., 1976; Liedvogel et al., 1976) and it is interesting to note that the conclusions obtained with mature chloroplasts (Douce et al., 1973b; Mackender and Leech, 1974) are also valid with etioplasts and greening etioplasts (Bahl et al., 1976), (Table 11). According to Hashimoto and Murakami (1975) the high glycerophospholipid content observed in the chloroplast envelope, which is comparable to that of cytoplasmic membranes (Mazliak, 1977), reflects the chemical nature of the outer membrane of the envelope. However, there is no evidence for this. For example, Billecocq et al. (1972) and Billecocq (1974, 1975), by means of specific antibodies have clearly shown that the outer membrane of the chloroplast envelope contains galactolipids and sulpholipids (Fig. 5). Furthermore, phosphatidylethanolamine, which is a major constituent of all the cytoplasmic membranes examined so far (Mazliak, 1977) is barely detectable in both envelope membranes and, in addition, phosphatidylethanolamine is considered now to be a negative lipid marker for purified intact chloroplasts (Dubacq and Kader, 1978). In a review on biomembrane polar lipids in higher plants, Mazliak (1977) wrote “The overwhelming impression is one of great uniformity of lipid composition among the different cell membranes”. This remark is in contradiction to Table I1 which shows clearly that the polar lipid content of the envelope membranes is strikingly different from that of mitochondrial membranes. Thus the envelope, carefully prepared, is devoid of phosphatidylethanolamine and diphosphatidylglycerol but contains large amounts of galactolipids. Conversely, the pure mitochondrial membranes rich in phosphatidylethanolamine are practically devoid of galactolipids (Kader, 1972; Donaldson et al., 1972; McCarty et al., 1973; Ohmori and Yamada, 1974; Moreau et al., 1974). Consequently such a result precludes a structural relationship between the plastidal and the mitochondrial structures. It may, however, be noted that the presence of galactolipids in mitochondrial preparations from avocado fruits (Schwertner and Biale, 1973), cauliflower buds (Douce et al., 1968) and Viciafuba leaves (Mackender and Leech, 1974) has been reported. Mackender and Leech (1974) have also determined striking differences in the fatty acid composition of galactolipids from different organelles making it apparent that the finding of mono- and digalactosyldiglycerides in other organelles cannot be attributed to contamination by chloroplasts. However, plastids in a plant cell, especially chloroplasts, may represent a large amount of the total membrane area (Forde and Steer, 1976) and it is known that their fragmentation during isolation leads to a massive contamination of the mitochondrial fractions by a variable proportion of plastid subfractions (envelope, clusters of plastoglobuli, intact and broken thylakoids). For these reasons, we believe that the galactolipids found by several authors in mitochondrial membranes are entirely attributable to

44

R. DOUCE AND J. JOYARD

contaminating plastid fragments (Hurkman et al., 1976; J. Joyard and R. Douce, unpublished data). Only one fatty acid, trans-d3-hexadecanoic acid (c16:1), is found specifically in photosynthetic tissues (James and Nichols, 1966) where it is esterified to phosphatidylglycerol (Haverkate and van Deenen, 1964), (Fig. 26). The presence of this fatty acid in the envelope is still open to debate (Mackender and Leech, 1974; Tuquet and de Lubac, 1975; Bahl and Moneger, 1975; Bahl et al., 1976). Heinz et al. (unpublished) have found that the c 1 6 : 1 from phosphatidylglycerol was indistinguishable by gas liquid chromatography from the usual c 1 6 : I trans-d3 and was specifically localized at the 2 position of the glycerol molecule. It is now well established that membrane polar lipids are distributed asymmetrically between the two halves of the bilayer (Bretscher, 1972). The biological function of lipid asymmetry is unknown although certain species of lipids may be important for the activity of particular enzymes which are also asymmetrically arranged. It will be interesting to see if such an asymmetry exists for the envelope membranes. The proportion of lipids and proteins present in the chloroplast envelope is different from other membranes. Douce et al. (1973b) were the first to report a high ratio of acyl lipids to protein (1 : 2) in the chloroplast envelope of Spinacia oleracea. The complete chloroplast envelope compositional analysis given by Joyard and Douce (1976a) indicates that 58 % of the membrane dry weight is attributable to lipids. This value is much higher than that of mitochondrial and microsomal membranes (Moreau et al., 1974). It may be pertinent to point out here that myelin is the only membrane containing a higher proportion of lipid (Kotyk and JanAcek, 1977). The high ratio of acyl lipids to proteins in the chloroplast envelope membranes isolated from various species has been confirmed recently by several groups of investigators (Poincelot and Day, 1974; Poincelot, 1976; Liedvogel et al., 1976). Such a high ratio could easily explain the great fragility of higher plant chloroplast envelopes. Recently, Bahl(l977) reported a low lipid toprotein ratio (0.23 : 1) for the chloroplast envelope of Triticum sativum. It is now well established that there is a srong correlation between membrane protein content and buoyant density (Scanu, 1972; Sitte, 1977). For example, it has been shown (Neuburger et al., 1977) that the buoyant density of the purified chloroplast ~ ; lipid/protein = 1.2) differs markedly envelope fraction (d = 1.12 g ~ m - acyl from envelope loaded with cytochrome c by electrostatic binding (d = 1.14 g cm-3; acyl lipid/protein = 0.85). Therefore, it is surprising that the proteinrich membranes described by Bahl(l977) could be obtained from the interface between two sucrose layers equivalent to 1.08 and 1-12 g ~ m respectively. - ~ B. PIGMENT COMPOSITION OF THE CHLOROPLAST ENVELOPE

The chloroplast envelope membranes are devoid of chlorophylls (Mackender and Leech, 1970, 1972, 1974; Douce et al., 1973b; Poincelot, 1973),

45

STRUCTURE AND FUNCTION OF THE PLASTID ENVELOPE

438

08

THYLAUOIDS , 10%

I

08

878 06

06

B

B

1 0 4

4 0 4

02

02

0 350

400

450

500

550

600

650

700

0

750

350

WAVELENGTH, nni

400

450

500

A

550

600

650

703

750

700

750

WAVELENGTH nm

473 4?7

08

08

06 ~

Y

f

f

p

04

:

O4

<

02

02

ENVELOPE

196°C

387 I

0

0 350

400

450

500

550

800

WAVELENGTH nm

650

700

750

350

400

450

500

550

800

WAVELENGTH

650

nm

Fig. 27. Absorption spectra of isolated thylakolds and envelope membranes. Note that isolated envelopes do not contain chlorophylls.

(Fig. 27). Consequently these pigments are considered now to be a negative marker for highly purified envelope membranes (Poincelot, 1974, 1976; Joy and Ellis, 1975; Mackender and Leech, 1974). Chloroplast envelopes are deep yellow (Neuburger et al., 1977) and “yellowish” when the yield is low (Poincelot, 1973; Sprey and Laetsch, 1975). Douce et al. (1973b) were the first to report the presence of carotenoids in the envelope membranes (Fig. 28). They concluded that qualitatively the carotenoids of both types of chloroplast membranes, thylako’id and envelope, are identical but the proportion in which they are present is different (Table IV). /%carotene accounted for a higher proportion of the thylakoidal carotenoid content when compared with the envelope fraction. On the other hand, violaxanthin accounted for a higher proportion of the envelope carotenoid content when compared with the thylakoids. These differences may be characterized by the xanthophyll : carotene ratio which is much higher ( N 6) in the envelope fraction than in the thylako‘ids (2: 3). Criticism, however, has been levelled against the purity of membranes prepared by the method used.

46

R. DOUCE AND J. JOYARD OH

HO VIOLAXANTHIN OH

HO

LUTEIN

OH

HO

ZEAXANTHIN

HO OH

NEOXANTHIN

R-CAROTENE

Fig. 28. Structure of the main carotenolds found in the chloroplast envelope.

According to Sprey and Laetsch (1976a) and Goodwin (1977) the carotenofds could be present in trapped plastoglobule clusters and are probably absent from the envelope membranes. As already stated, this criticism is valid only if one starts with unpurified chloroplasts. In this case, the envelope fraction is contaminated by plastoglobule clusters embedded in swollen grana. In our case, we believe that the carotenolds are genuine constituents of the envelope membranes for the following reasons : a. The envelope membranes carefully prepared are devoid of plastoglobule clusters (see Fig. 25). b. The concentration ratio of violaxanthin in the envelope is reproducible for various experimental procedures.

STRUCTURE AND FUNCTION OF THE PLASTID ENVELOPE

47

TABLE IV Carotenof d Composition of Isolated Envelopes and ThylakoIds from Dark-treated Spinach Chloroplasts. Data are Expressed in pg Pigment per mg o j Envelope or ThylakofdProtein. (-: not determined)

Carotenold ~.

_

_

_

Violaxanthin Lutein + zeaxanthin 8-Carotene Antheraxanthin Neoxanthin

_

~ .. ~

~

Envelope

Thylakolds

6.50

4.52

2.50

6.05

14 2

-

0.79 0.40

c. The very low level of chlorophyll present in the Spinacia oleracea envelope fraction is also inconsistent with a significant contribution of carotenoids from fragments of the thylakoidal system (Fig. 27). d. The carotenoid content in envelope membranes is not a contamination by carotenoids randomly separated from the thylakoids during osmotic shock and redissolved in the highly lipophilic envelope membranes (Siefermann-Harms et a/., 1978). These observations rather suggest that the carotenoids found in unpurified preparations of plastoglobuli are attributable to a contamination by envelope membranes. The presence of carotenoids in the envelope membranes has been confirmed by Hashimoto and Murakami (1975) for spinach leaves and by Priestley and Woolhouse (1977) for Phaseolus vulgaris leaves. Jeffrey et al. (1974) demonstrated that envelope membranes prepared from dark-treated leaves had a violaxanthin content up to 3.5 times the lutein plus zeaxanthin content, whereas in chloroplast envelopes from illuminated leaves this ratio was only 0.75. Based on these data the authors suggested that the light-induced pigment changes in envelopes were probably due to deepoxidation of violaxanthin to zeaxanthin (Siefermann-Harms, 1977). Recently, working with intact spinach chloroplasts, Siefermann-Harms et al. (1978) have confirmed these results and demonstrated that the light-induced violaxanthin decrease, in the envelope, is not caused by an envelope or stromal de-epoxidase but results from a violaxanthin exchange between envelope and thylakoids. It has been suggested that the role of the envelope carotenoids might be to protect the thylakoidal chlorophylls from harmful photooxidations at high light intensities since the envelope is a yellow membrane system with absorption properties akin to a blue light filter (Jeffrey et al., 1974). This hypothesis is most unlikely, however, because the carotenoid content of the envelope fraction is very low (12 pg mg-1 protein). It has also been postulated that carotenoids in envelope membranes may act by stabilizing protein con-

48

R. DOUCE AND J. JOYARD

formation (Krinsky, 1971). Such a role may be particularly important in a membrane system which is continually synthesizing all the galactolipids of the thylakoids (Douce, 1974; Joyard and Douce, 1977) and which is involved in the transport of several specific anions (Heber, 1974; Walker, 1974; Heldt, 1976a). It is likely that the envelope membranes are the site of carotenoid synthesis in intact chloroplasts. This hypothesis is supported by observations that suggest a carotenoid exchange between envelope and thylakoids (Siefermann-Harms et al., 1978). In addition, theenvelope has been found in all plastids examined so far (Gunning and Steer, 1975 and Figs 1, 9-14) and the existence of the envelope structure precedes the thylakoids (Figs 9, 10,40). Furthermore, Moore and Shepard (1977) have indicated that in Acetabularia both the pattern and the rates of pigment synthesis are comparable in vivo and in vitro. They conclude, therefore, that the complete pigment-synthesis pathways and their control mechanisms reside within the chloroplasts. According to Goodwin (1958) and Bickel and Schultz (1976), intact, isolated spinach chloroplasts incorporated 14C from 14C0, into ,8carotene under photosynthetic conditions. Since synthesis of carotenoids also takes place in chromoplasts (Qureshi et al., 1974; Sitte, 1977) which lack thylakoids (Fig. 11) it seems reasonable to suppose that the envelope membranes of chloroplasts, in conjunction with soluble enzymes of the stroma, are active in carotenoid biosynthesis. For example, Costes et al. (1979) have demonstrated that spinach chloroplast envelopes are the site of violaxanthin synthesis. Finally, Neupert et al. (1972), Costes et al. (1976) and Moreau (1978) have shown the presence of substantial amounts of carotenoids in the outer membranes of plant mitochondria. In this case, it is possible that the carotenoids found in this membrane fraction could be entirely attributable to contamination by envelope membranes. Phytochrome is a chromoprotein photoreceptor, existing in two photoconvertible forms (Pr and Pfr) and is responsible for various metabolic and developmental responses of higher plants to light (Briggs and Rice, 1972; Smith, 1975). A suggested interpretation of the action of phytochrome is a reversible binding of the physiological active form (Pfr) to a receptor site probably located in a membranous fraction (Quail et al., 1973; Boisard et al., 1974; Marme et al., 1974; Mackenzie et al., 1975). It has been shown by Roux and Yguerabide (1973) that phytochrome applied to a black lipid membrane made from oxidized cholesterol increased the conductance of the phytochrome-modified membrane by red light irradiation but decreased conductance on irradiation with far-red light. Consequently, it is possible that phytochrome may be a normal component of the chloroplast envelope membranes where it could act as a permease to regulate the inter-compartmental movement of several critical metabolites such as gibberellins (Evans, 1975; Cooke and Kendrick, 1976; Browning and Saunders, 1977), abscisic acids

STRUCTURE AND FUNCTION OF THE PLASTID ENVELOPE

49

(Milborrow, 1976; Wellburn and Hampp, 1976a; Loveys, 1977), adenosine 3’-5’-cyclic phosphate (Wellburn and Hampp, 1976a), mevalonate (Wellburn and Hampp, 1976b; Schneider et al., 1977) and acetate (Wellburn and Hampp, 1976b; Schneider et al., 1977). Recently, Evans and Smith (1976a, b) and Cook and Kendrick (1976) reported that although only a small proportion of the total cell phytochrome is associated with etioplasts, this chromoprotein seems to be concentrated in the envelope membranes. On the other hand, Quail et al. (1976) in studies on hypocotyl hooks excisedfromfour-dayold etiolated Cucurbita pep0 found no evidence to support this observation. They found that the phytochrome and plastid membrane marker carotenoid profiles on sucrose gradient were non-coincident under the conditions used. We have also failed to detect phytochrome pigments in chloroplast envelope membranes obtained from 4 kg of young spinach leaves (Beggs, Joyard and Douce, 1976, unpublished data). C. CHLOROPLAST ENVELOPE POLYPEPTIDES

Several reports have appeared describing sodium dodecylsulphate polyacrylamide gel electrophoresis of chloroplast envelope proteins (Pineau and Douce, 1974; Cobb and Wellburn, 1974; Mendiola-Morgenthaler and Morgenthaler, 1974; Joy and Ellis, 1975; Cobb, 1975; Sprey and Laetsch, 1975; Liedvogel et al., 1976; Fliigge and Heldt, 1976; Vasconcelos, 1976; Vasconcelos et al., 1976; Priestley, 1977). Practically all workers reported a series of high molecular weight bands (above 70 000 daltons) and two predominant polypeptides at approximately 52 000 and 29 000 molecular weight (Table V, Fig. 29). The 52 000 band is electrophoretically non-coincident with the large subunit of ribulose bisphosphate carboxylase (Pineau and Douce, 1974) and thus could not be accounted for by a ribulose bisphosphate carboxylase contamination of the pure envelope fraction. Fliigge and Heldt (1976) have shown clearly that the 29 000 molecular weight envelope polypeptide binds 35S-p-(diazonium)-benzenesulphonic acid, an inhibitor of the phosphate translocator. Moreover, when phosphate or 3-phosphoglycerate was added during the incubation with the inhibitor, the labelling of the 29 000 dalton polypeptide was considerably decreased. In a further communication, Flugge and Heldt (1977) have shown that this polypeptide incorporates pyridoxal-5‘phosphate, a molecule which forms a Schiff base with lysine and inhibits the phosphate translocator. For these reasons, Fliigge and Heldt (1977) concluded that the 29 000 daltons polypeptide, which represents about 20-25 % of the total envelope protein, plays a role in the functioning of the phosphate translocator. The molecular weight distribution of the envelope polypeptides is markedly different from that of the thylakoid polypeptides (Pineau and Douce, 1974; Mendiola-Morgenthaler and Morgenthaler, 1974; Sprey and Laetsch, 1975; Joy and Ellis, 1975). Racusen and Poincelot (1976) have suggested that a

Fig. 29. Sodium dodecylsulphate polyacrylamide gel electrophoretic patterns of polypeptides of spinach chloroplast subfractions (envelope, stroma, thylakoids). Note the presence in envelope membranes of a major polypeptide (29 OOO daltons). This polypeptide plays a role in the functioning of the phosphate translocator. A: densitometric pattern (absorbance at 620 nm). B: Coomassie blue stained gels. C: molecular weights (in kilodaltons). Adapted from Pineau and Douce (1974).

TABLE V Molecular Weights of the Main Chloroplast Envelope Polypeptides Separated by S DS-Polyacrylamide Spinacia oleracea Spinacia oleracea

14000

Spinacia oleracea Spinacia oleracea Avena sativa

? ?

Pisum sativum Narcissus pseudonarcissus Phaseolus vulgaris Euglena gracilis

?

29000 27 OOO

33000 32000

52 OOO

125 000

5OOOO

?

28 500 29 OOO 25 OOO

52 OOO 53 OOO

90 500

26 OOO 32 OOO 23 OOO

53 OOO 52 OOO 50 OOO 48 OOO

90 500

Gel Electrophoresis

Pineau and Douce (1974) Mendiola-Morgenthaler and Morgenthaler (1974) Sprey and Laetsch (1975) Fliigge and Heldt (1976) Cobb and Welburn (1974) Cobb (1975) Joy and Ellis (1975) Liedvogel et al. (1976) Priestley (1977) Vasconcelos (1976) Vasconcelos et al. (1976)

52

R. DOUCE AND J. JOYARD

further distinction can be made in that the envelope shows a much higher proportion of protein-bound hexosamine. According to these authors, between 60 and 90% of the total chloroplast hexosamine is associated with chloroplast envelope preparations from spinach, maize and sunflower leaves. However, the possible errors involved in arriving at these values are quite significant if one considers that these authors found as much as 18 % of the total chloroplast proteins in the envelope preparations (for comparison, Joyard and Douce, 1976a, find 1 % only of the total chloroplast proteins in the envelope membranes of Spinacia oleracea). The significance of these facts might be that the synthesis of the thylakolds is not merely an extension of the synthesis of the inner membrane but involves a major change in the types of proteins inserted into the growing membrane (Pineau and Douce, 1974; Joy and Ellis, 1975). We suggest that the envelope proteins are mainly concerned with the transport of many low molecular weight metabolites as well as with protein synthesized on cytoplasmic ribosomes, across the envelope. Therefore, the envelope proteins may be expected to differ from those of the thylakoids which are involved in the light-dependent reactions of photosynthesis. The most direct means of identifying the protein-biosynthesizing activities of chloroplasts is to study their ability to incorporate labelled amino acid precursors in vitro (Boardman et al., 1966; Margulies and Parenti, 1968; Chen and Wildman, 1970; Harris et al., 1973). Isolated chloroplasts have been shown to synthesize one or several soluble proteins and some of the proteins of the thylakold system (Blair and Ellis, 1973; Eaglesham and Ellis, 1974; Bottomley et al., 1974; Morgenthaler and Mendiola-Morgenthaler, 1976; Vasconcelos, 1976; Ellis, 1977; Ellis et al., 1978). There is no evidence to indicate that the range of proteins synthesized by etioplasts differs from those made by chloroplasts (Siddell and Ellis, 1975). Blair and Ellis (1973) have identified one of the major labelled polypeptides with the large subunit of ribulose bisphosphate carboxylase. Mendiola-Morgenthaler et al. (1976), contrary to previous reports (Eaglesham and Ellis, 1974) have identified the major labelled thylako'id polypeptides with the two largest subunits of the coupling factor (CF,; Jagendorf, 1975) and possibly the smallest inhibitory subunit. The identities of the other labelled polypeptides remain unknown. According to Siddell and Ellis (1975) the abundance of plastid ribosomes which represent up to 50 % of the total ribosomal complement of photosynthetic cells (Ellis et al., 1973) suggests that they may produce a few proteins in large quantities rather than many proteins in smaller amounts. The origin of the chloroplast envelope with respect to its protein components is particularly interesting since these double membranes form a barrier between the contents of the chloroplasts and the surrounding cytoplasm. Joy and Ellis (1975) found two labelled polypeptides (molecular weights: 32 000 and 65 OOO) in the crude enveloped fraction after supplying

STRUCTURE AND FUNCTION OF THE PLASTID ENVELOPE

53

isolated pea chloroplasts with 35S-methionine. One of these bands (32 000) was electrophoretically coincident with the major thylakoid protein synthesized by chloroplasts. Incorporation of label into both the 32 000 and 65 000 molecular weight polypeptides is totally light-dependent and was inhibited by chloramphenicol. According to these authors the simplest interpretation of the results is that only two of the chloroplast envelope polypeptides are synthesized by chloroplast ribosomes. This conclusion is supported by the results of an in vivo experiment in which detached pea shoots were fed with 35S-methioninewith and without cycloheximide. It may, therefore, be postulated that the remainder of envelope polypeptides are synthesized on cytoplasmic ribosomes. Unfortunately, the results obtained must be interpreted with caution due to the likelihood of stromal or lamellar contaminations in the envelope preparations used. In support of this criticism, Ellis (1975) has indicated in a further communication that labelling of the 32 000 molecular weight polypeptide was due to contaminating lamellar membranes. In similar experiments Morgenthaler et al. (1975) reported incorporation of radioactivity into two spinach envelope polypeptides of molecular weight 50000 and 35 000. In later work (Morgenthaler and Mendiola-Morgenthaler, 1976), apparent labelling of the lower molecular weight band was attributed to thylakoid contamination. These results reveal, as pointed out before, the potential hazards of contamination in the isolated envelope membranes which not only constitute a very small proportion of the total chloroplast proteins (Joyard and Douce, 1976a) but also represent a minute fraction of the total radioactivity incorporated into chloroplasts (0.1 %, Morgenthaler and Mendiola-Morgenthaler, 1976). Recent work with Euglena gracilis chloroplasts (Vasconcelos, 1976) which had been purified on gradients of silica sol has suggested that a major envelope polypeptide (molecular weight 48 000) which corresponds probably to the 50 kilodaltons peak of spinach and pea chloroplasts, may be synthesized by the plastid. This raises the question as to whether the envelope polypeptides which are synthesized inside the chloroplast are also encoded by the chloroplast DNA. Unfortunately, all these experiments have been carried out on mature chloroplasts and different results would perhaps be obtained if dividing plastids were used. VI. ENZYMIC ACTIVITIES AND FUNCTIONS O F THE CHLOROPLAST ENVELOPE It is clear that the chloroplast envelope maintains the soluble enzymes (stromal enzymes) involved in the Benson-Calvin cycle (Benson and Calvin, 1947; Calvin and Benson, 1948; Bassham, 1964) in close contact with the thylakoid network. Intact chloroplasts (Class A ; Hall, 1972) contain a large amount of soluble enzymes in the stromal compartment (about 9-5 g protein

54

R. DOUCE AND J. JOYARD

per g of chlorophyll, Lilley et al., 1975) which are rapidly released into the medium when the envelope membranes are damaged (Leech, 1964). It is now well established that ruptured chloroplasts (Class C ; Hall, 1972) are normally capable of O2evolution (Hill, 1937) and photophosphorylation (Arnon et al., 1954) in the presence of appropriate oxidants and cofactors but, in contrast with intact chloroplasts, are inactive in C0,-fixation, C0,-dependent 0, evolution and phosphoglycerate-dependent 0, evolution (Arnon, 1955; Whatley et al., 1956; Havir and Gibbs, 1963; Walker, 1964; Walker, 1965b; Bucke et al., 1966; Jensen and Bassham, 1966; Cockburn et al., 1967c; Walker and Hill, 1967). A most important advance was made by Walker (1965a) when he used a variety of techniques namely light microscopy, electron microscopy and biochemical studies on the same material for the first time. His work allowed the clear and simple conclusion to be drawn that osmotic shock stripped isolated chloroplasts of their outer envelopes and that this in turn led: a. To loss of stroma and consequently the loss of the C0,-fixation ability; b. To increased photophosphorylation possibly as a consequence of the removal of a permeability barrier (the envelope) to ADP (Hall, 1976). Conversely, when ruptured chloroplasts (Class C; Hall, 1972) are supplemented with stromal enzymes, Mg2+ and catalytic quantities of ferredoxin, ATP, NADP and phosphoglycerate the rates of C0,-dependent 0, evolution which were then observed were of the same order as those recorded with intact chloroplasts (Walker and Lilley, 1974; Bassham et al., 1974; Walker et al., 1976). In addition to this simple mechanical role the three main functions of the chloroplast envelope membranes are : a. Regulation of the inflow of raw materials for photosynthesis and the outflow of photosynthetic products. b. Regulation of the uptake of the many chloroplast proteins that are made by the cytoplasmic ribosomes. c. Regulation of the synthesis of galactolipids. A. METABOLITE TRANSPORT IN INTACT CHLOROPLASTS

The expanding literature on metabolite transport in intact chloroplasts isolated from plants having only the Calvin-Benson pathway of photosynthesis (C, plants) has been reviewed thoroughly by Heber (1974), Walker (1974, 1976a, 1976b, 1977), Heldt (1976a, 1976b), Strotmann and Murakami (1976) and Herold and Walker (1977). Shorter treatments have been given by Krause and Heber (1976), Leech and Murphy (1976) and Gregory (1977). In contrast, metabolite transport in intact chloroplasts isolated from plants having the C, dicarboxylic acid pathway of photosynthesis (C, plants) is practically unknown. The reason for this, and despite intensive trials, is that

55

STRUCTURE AND FUNCTION OF THE PLASTID ENVELOPE

the preparation of chloroplasts from C , plants retaining both structural and functional integrity has attained only limited success. A number of recent investigations have been encouraging, however, in demonstrating the potential usefulness of C , and C , plant protoplasts for physiological research (Huber and Edwards, 1975a; Kanai and Edwards, 1973a, 1973b; Nishimura and Akazawa, 1975). Cell wall materials are enzymically degraded during the preparation of protoplasts, thus gentle disruption of the protoplast plasmalemma is ideal for isolating intact chloroplasts (Nishimura et al., 1976). Following the promising work of Huber and Edwards (1975b, 1976), we are convinced that the preparation of intact chloroplasts from C , plant mesophyll cell protoplasts will help to provide a better understanding of metabolite transport through the envelope of these chloroplasts (Huber and Edwards, 1975a, 1975b, 1976, 1977). Interesting experiments with isolated spinach chloroplasts have demonstrated that the two membranes of the envelope delineate two compartments differing in their accessibility to compounds of different molecular weights. These two compartments are the stromal space and the intermembrane space of the envelope (Heldt and Sauer, 1971 ; Heldt et al., 1972), (Fig. 30). The

..

.....

-...*

-

.. . ... . .. * . . .- ... .

.. , *

*

- '.

*

'

.

0

0

0.

v

*

0 - 0 :

.

a

. *.

WATER SPACE

.. '

0 0

SORBITOL

* * O 0

SPACE

0

0 0 0 . 0 0 DEXTRAN

0 . SPACE

Fig. 30. Transport studies across the chloroplast envelope require the use of different molecules having different permeability properties towards the two envelope membranes. Le): water penetrates very rapidly into all the different compartments of the choloroplast ; Centre: sorbitol penetrates the outer membrane but cannot pass through the inner membrane; Righr: dextran cannot pass through any of the envelope membranes. Comparison of the distribution of a given molecule with that of water, sorbitol or dextran allows the study of its transport (or absence of transport) across the chloroplast envelope.

membrane space situated between the inner and the outer membrane of the chloroplast envelope is found to be non-specifically permeable to sucrose and other molecules either charged or uncharged, up to a molecular weight of about 10000. In contrast, the inner envelope membrane of the envelope surrounding the stroma space which is impermeable to sucrose (Nobel, 1969; Wang and Nobel, 1971) is selectively permeable to a limited number of

56

R. DOUCE AND J. JOYARD

anions. The size of the sucrose permeable space depends on the tonicity of the medium. It is very small when the chloroplasts are kept in a hypotonic medium (0.2 M sorbitol) whereas in a hypertonic medium (0.6 M sorbitol) the outer envelope membrane appears to be loosely attached to the inner envelope membrane with large empty spaces in between (Fig. 18). The morphological responses of mitochondria to changes in the osmolarity of suspending media are also well documented (Parsons et al., 1965; Pfaff et al., 1968; Guillot-Salomon, 1972). The matrix space is usually very condensed in hypertonic media and expanded in hypotonic solutions. Under extreme hypotonic conditions, the inner membrane appears to unfold causing rupture of the outer membrane. The outer mitochondrial membrane itself appears osmotically inactive, an observation consistent with its high permeability t o low molecular weight compounds (O’Brien and Brierley, 1965; Parsons et al., 1965; Pfaff et al., 1968; Douce et al., 1972). It is interesting to note that the surfaces of non-fixed, negatively stained, outer membranes of plant and animal mitochondria have a pitted appearance (Parsons e f al., 1965; Stoeckenius, 1970). These dark (electron opaque) dots are about 3 nm in diameter and irregularly spaced with centre-to-centre distance of 4-5 nm. The presence of such presumptive pores in the outer mitochondrial membrane might be expected in view of the permeability of this membrane to low molecular weight solutes and its impermeability to ferricytochrome c (Wojtczak and Zaluska, 1969; Douce et al., 1973a) an oblate spheroid 3 x 3.4 x 3.4 nm. It will be interesting to see if such structures exist in the outer membrane of the plastid envelope. The selective permeability of the inner envelope membrane to a few species of anions is now known to be due to specific translocators (Heldt, 1976a, 1976b). As already stated (Douce et al., 1973b; Joyard and Douce, 1976a) the inner envelope membrane is devoid of chlorophyll and cytochrome. It therefore seems that there is no electron transport in the inner membrane of the envelope (Douce et al., 1973b). Consequently, in the chloroplasts in contrast to the mitochondria, the anion transport and the electron transport are found in two distinct membranes i.e. the inner membrane of the envelope and the thylakold membranes. So far, three specific translocators have been well characterized in the inner envelope membrane of the chloroplast isolated from plants having only the Calvin-Benson pathway of photosynthesis : a. The phosphate translocator is specific for the transport of inorganic phosphate, 3-phosphoglycerate, dihydroxyacetone phosphate and glyceraldehyde-3-phosphate (Heldt and Rapley, 1970); b. The dicarboxylate translocator is specific for the transport of malate, oxaloacetate, aspartate and glutamate (Heldt and Rapley, 1970); c. The ATP translocator has a high specificity for external ATP (Heldt, 1969; Strotmann and Berger, 1969).

STRUCTURE AND FUNCTION OF THE PLASTID ENVELOPE

57

Only the most important points will be summarized here. The reader is referred to the excellent reviews written by Heber (1 974); Walker (1 974, 1976a, 1976b) and Heldt (1976a), (Fig. 31). AT P

COP C 3p

Pi

Fig. 3 1. Summary of the main exchanges between chloroplast and its environment. Ease of penetration is represented by magnitude of arrows. Partially adapted from Walker (1976), p. 102.

I . The Phosphate Translocator Isolated intact spinach or pea chloroplasts behave as an orthophosphate and CO, consuming cell organelle (Walker, 1976a). In an elegant experiment, Cockburn et al. (1967a) were the first to demonstrate clearly with isolated chloroplasts, that dihydroxyacetone phosphate appears rapidly in the medium during illumination. These workers also demonstrated that, in the absence of exogenous orthophosphate, photosynthesis by intact isolated chloroplasts soon falls to a low level which may then be restored by the addition of catalytic amounts of orthophosphate. They demonstrated that three molecules of 0, are released for each molecule of orthophosphate added: 3 COz

+ 2 H,O + orthophosphate

-+

dihydroxyacetone phosphate

+ 3 0,

During photosynthesis, ATP and NADPH are produced and subsequently utilized to drive the photosynthetic carbon reduction cycle. Photosynthetically generated ATP and NADPH are not directly available for extra-chloroplastic

58

R. DOUCE AND J. JOYARD

reactions due to the quasi-impermeability of the inner envelope membrane t o these compounds (Heber and Santarius, 1965; Heber et al., 1967; Mathieu, 1967; Heber, 1974; Walker, 1974). Exogenous ATP cannot be utilized at high rates in 3-phosphoglycerate reduction by intact chloroplasts (Stokes and Walker, 1971). A shuttle system involving 3-phosphoglycerate and dihydroxyacetone phosphate, both of which move rapidly across the chloroplast envelope (Bamberger and Gibbs, 1965), has been reported to enable the indirect transfer of ATP and NAD(P)H from chloroplast to cytoplasm (Stocking and Larson, 1969; Heber and Santarius, 1970), (Fig. 32). Depending on the redox potential of pyridine nucleotides inside and outside of the

1

I DHAP

1,3PGA

PGA

I

cytoplasm

envelope

stroma

I

Fig. 32. Phosphate translocator facilitates the transport of different metabolites across the chloroplast envelope. 3-phosphoglycerate (PGA) is reduced in the stroma to dihydroxyacetone phosphate (DHAP) at the expense of ATP and NADPH. Dihydroxyacetone phosphate is exported into the cytoplasm in exchange for 3-phosphoglycerate. In the cytoplasm, dihydroxyacetone phosphate is reoxidized to 3-phosphoglycerate by the glycolytic pathway which produces ATP and NADH. These mechanisms allow the indirect transfer of reducing equivalents and nucleotides across the envelope. Adapted from Heldt (1976), p. 231.

chloroplasts, the shuttle can operate in both directions. This shuttle system requires the oxidation of dihydroxyacetone phosphate to 3-phosphoglycerate in the cytoplasm by a NADP-linked non-reversible glyceraldehyde-3-phosphate dehydrogenase (Kelly and Gibbs, 1973; Bamberger et al., 1975) or a NAD-linked reversible glyceraldehyde-3-phosphate dehydrogenase and 3phosphoglycerate kinase. It also requires the reduction of 3-phosphoglycerate to dihydroxyacetone phosphate in the stroma. Consequently, during photosynthesis, the principal imports passing into the chloroplasts are : water, CO,, orthophosphate and 3-phosphoglycerate.

STRUCTURE AND FUNCTION OF THE PLASTID ENVELOPE

59

The major exports to the cytoplasm are: 0,, dihydroxyacetone phosphate and/or glyceraldehyde-3-phosphate(Fig. 3 I). In contrast, hexose monophosphates, sedoheptulose and ribulose phosphate are not able to permeate the envelope (Bassham et al., 1968; Lilley et al., 1977). Heldt and Rapley (1970) by direct measurements (silicone layer filtering centrifugation) demonstrated that phosphate, dihydroxyacetone phosphate, 3-phosphoglycerate and glyceraldehyde-3-phosphate are specifically transported across the inner envelope membrane whereas hexose phosphate compounds are taken up only very slowly. This transport process is very active. At 20°, for example, the V,,, of uptake of 3-phosphoglycerate, dihydroxyacetone phosphate and phosphate was found to be in the range of 300 pmol mg-l chlorophyll h-’ (Heldt, 1976a). 3-Phosphoglycerate, dihydroxyacetone phosphate and phosphate were shown to compete with each other for transport (Werdan and Heldt, 1972). In all cases the Ki (0.15 mM) for inhibition of transport and the K, for transport (0.15420 mM) were identical (Heldt, 1976b). From these data, Heldt and Rapley (1970) rightly concluded that there is a specific carrier situated in the inner membrane of the chloroplast envelope which has been named the “phosphate translocator”. The phosphate translocator which is inhibited by low concentrations of mercurials such as p-chloromercuriphenyl sulphonic acid (Werdan and Heldt, 1972; Robinson and Wiskich, 1977a) catalyses a strict counterexchange of phosphate2-, 3-phosphoglycerate2-, glyceraldehyde 3-phosphate2- and dihydroxyacetone phosphate2-. It is interesting t o note that 3-phosphoglycerate3- is not transported. A counter-exchange of 3-phosphoglycerate with either dihydroxyacetone-phosphate or inorganic phosphate proceeds by electroneutral exchange i.e. the protons are consumed in the external medium for the formation of the lesser charged form of 3-phosphoglycerate (Fliege et al., 1978). For each molecule entering the chloroplast one molecule leaves the chloroplast and vice versa. In this way, the total pool of phosphate and phosphorylated compounds within the stroma is kept constant (Fig. 33). Phosphate alone never leaks out from intact isolated chloroplasts (Heldt, 1976a). The phosphate translocator plays an important role in the control of sucrose and starch synthesis (Heldt et al., 1977) in green leaf cells (Fig. 34). The triosephosphate molecules produced during photosynthesis in the stromal compartment and transported to the cytoplasm may be converted there to sucrose (Bird et al., 1974) and the phosphate thus released can reenter the chloroplast in exchange for more triosephosphate (Herold and Walker, 1977). As pointed out recently (Heldt et al., 1977), this allows a steady flux of fixed carbon from the chloroplast to the cytoplasm. If cytoplasmic phosphate is sequestered in the phosphorylation of hexose or ADP (Chen-She et al., 1975; Herold et al., 1976) or if utilization of triosephosphates in the cytoplasm is lower than its production, the level of phosphate in

60

R. DOUCE AND J. JOYARD

Fig. 33. Mechanism which explains the function of the phosphate translocator present in the envelope. This translocator allows the exchange of the internal phosphate pool of the chloroplast with exogenous phosphate added to the incubation medium. A: intact chloroplast with its internal phosphate pool; B: in the presence of exogenous phosphate ("Pi) the translocator allows the exchange of internal and exogenous phosphate; C: all the cold Pi molecules present in the chloroplast have been replaced by 3ePi.Ratio of chloroplast volume (30pl mg-l chlorophyll) to suspending medium volume is in the range of 1/500 to 1/700. The practical consequence of this situation in vitro is that substances lost from isolated chloroplasts (here, cold Pi molecules) will not accumulate in the medium very rapidly: the medium acts as a buffer volume (Coombs and Baldry, 1971). It is clear that the situation is entirely different in vivo, as in this case the ratio of chloroplast volume to cytoplasm volume is in the range of 0.4 to 1 (see Fig. 17).

the thin film of cytoplasm (Fig. 17) will be decreased. Under these conditions, the rate of triosephosphate export will be lowered because of the lack of exchangeable phosphate in the cytoplasm, and, more fixed carbon will accumulate inside the stroma as starch (Steup et al., 1976; Heldt et al., 1977). Furthermore, it is the 3-phosphoglycerate: Pi ratio which is important in the allosteric activation of ADP-glucose pyrophosphorylase (Preiss and Levi, 1978), an important enzyme in the starch biosynthetic pathway. The phosphate translocator also plays an important role in starch hydrolysis which occurs in the dark (Heldt et al., 1977; Peavey et al., 1977), (Fig. 34). Thus, starch mobilization (phosphorolysis) in the dark is promoted by orthophosphate and is inhibited by phosphoglycerate. The principal products of starch breakdown are the transport metabolites 3-phosphoglycerate and dihydroxyacetone phosphate (Embden-Meyerhof pathway) which are released from the chloroplasts via the phosphate translocator. It is interesting to note that the phosphate transport in chloroplasts appears to be entirely different from the phosphate transport in mitochondria since the carrier in the mitochondria does not transport 3-phosphoglycerate nor dihydroxyacetone phosphate. If we assume that in spinach chloroplasts : a. The COz fixation rate is 180 pmol CO, mg-l chlorophyll h-l; b. The average number of chloroplasts is 4 x lo8 mg-l chlorophyll (Walker, 1974);

Fig. 34. Regulation of triosephosphate export to the cytoplasm and storage within the chloroplast. A : in the cytoplasm, the sucrose synthesis from triosephosphate maintains Pi concentration at a high level in this compartment. So long as sucrose can be exported to the outside of the cell through the plasmalemma, triosephosphate synthesized within the chloroplast can be exchanged against Pi via the phosphate translocator. This mechanism maintains Pi concentration in the stroma at the level required by photophosphorylation and CO, fixation. Consequently, sucrose synthesis within the cytoplasm brings a major contribution to the recycling of Pi in photosynthetic carbon assimilation. B: sucrose cannot be exported to the outside; by feedback inhibition, sucrose synthesis stops and Pi concentration in the cytoplasm falls to a low level. Triosephosphate molecules are exported through the chloroplast envelope very slowly and are converted within the stroma to starch which accumulates. This mechanism maintains Pi concentration in the stroma at the level required by photophosphorylation and CO, fixation. C: during the night, starch is slowly converted into triosephosphate and/or glucose, these molecules are exported via different translocators and are metabolized into the cytoplasm.

62

R. DOUCE AND J. JOYARD

The total amount of protein per mg chlorophyll (or per 4 x lo8 chloroplasts) is 18.4 mg (Lilley et al., 1975); d. The inner membrane of the envelope represents roughly 1 % of the total chloroplast proteins (Joyard and Douce, 1976a); e. The phosphate translocator (molecular weight = 30 000) represents roughly 20% of the total envelope protein (Fliigge and Heldt, 1976); we can calculate that: The number of molecules of dihydroxyacetone phosphate exported per second and per chloroplast during photosynthesis is : 180 x lop6 x 6.02 x loz3= 25.10s 3 x 3600 x 4 x lo8

The approximate number of molecules of phosphate translocator per chloroplast is : 18.4 x

x

1 x 20 x 1 1 x (6.02 x loz3) x = 2.106 100 100 30 000 4*108

-

~~~

~

This allows us to calculate the turnover number per second: 25 x lo6 ______ = 13 2 x 106

Consequently, each molecule of phosphate translocator is able to export 13 molecules of dihydroxyacetone phosphate every second and import 13 molecules of phosphate every second. 2. The Dicarboxylate Translocator Intact well-washed chloroplasts (class A) are unable to reduce externally added NAD+ in the light. When intact chloroplasts are supplied with malate and NAD+-specific malate dehydrogenase, photosynthetic reduction of external NADf can be observed (Heber and Krause, 1971). Such a reduction can occur in vivo since malate dehydrogenase is present in the stroma as well as in the cytoplasm (Heber, 1960). A shuttle transfer of oxaloacetate/malate has been postulated to explain indirect transport of reduced pyridine nucleotide between the stroma and the cytoplasm (Heber and Krause, 1971), (Fig. 35). This shuttle, like the phosphate translocator, can operate in both directions and depends on the redox potential of pyridine nucleotides outside and inside the chloroplasts (Krause and Heber, 1976). It has also been suggested that oxaloacetate might undergo transamination with glutamate leading to the formation of aspartate and a-ketoglutarate since the enzyme involved (glutamate-oxaloacetate transaminase) is present in the stroma as well as in the cytoplasm (Heber, 1960; Santarius and Stocking, 1969). Aspartate and a-ketoglutarate are transported across the inner envelope membrane in ex-

STRUCTURE AND FUNCTION OF THE PLASTID ENVELOPE

I

cytoplasm

envelope

StrOma

63

I

Fig. 35. Dicarboxylate translocator facilitates the transport of different metabolites across the chloroplast envelope. A: glutamatela-ketoglutarateand aspartatelmalate shuttles; B: malate/oxaloacetate shuttle. These mechanisms allow the transfer of reducing equivalent across the envelope. NAD(P) is reduced in the stroma and NADH is oxidized in the cytoplasm by two isoenzymes: malic dehydrogenase. These shuttles can operate in both directions and depend on the redox potential of pyridine nucleotides outside and inside the chloroplasts. Adapted from Heldt (1976), p. 231.

change for glutamate and malate (Heber, 1974), (Fig. 35). This shuttle, which also allows an indirect transport of reduced pyridine nucleotides, is more efficient than the simpler malate/oxaloacetate shuttle since the concentration of oxaloacetate in the plant cell is very low (Heber, 1974). Again, Heldt and Rapley (1970) by direct measurements demonstrated that oxaloacetate, malate, aspartate and a-ketoglutarate (dicarboxylic acids) are transported across the inner membrane of the chloroplast envelope by a specific transporter called “dicarboxylate translocator”. This transport mechanism shows substrate saturation and all dicarboxylate species transported compete with each other for translocation into the stroma. For example, the transport of malate is inhibited by oxaloacetate and vice-versa. The V,,, of uptake of malate and aspartate was found to be in the range of 20-30 pmol mg-l chlorophyll h-l at 4°C (Heldt, 1976a). The dicarboxylate translocator catalyses a counter-exchange of dicarboxylic acids but this exchange is not as strictly coupled as transport via the phosphate translocator (Heldt et al., 1975). It is for this reason that dicarboxylic acids alone slowly

64

R. DOUCE AND J. JOYARD

leak out from intact isolated chloroplasts (Heldt, 1976b). The activation energy of transport of dicarboxylate as determined from its temperature dependence is about 7 kcal mol-l (Lehner and Heldt, 1978). This value is much lower than the activation energy of phosphate transport in chloroplasts. It is also possible, as suggested recently (Lehner and Heldt, 1978), that several carriers with overlapping specificity may be involved in dicarboxylate transport. In classical work Heber (1974) proposed a scheme for the transfer of phosphorylation energy in the light by co-operation of the 3-phosphoglycerate/dihydroxyacetone phosphate shuttle with the malate/aspartate and/or the malate/oxaloacetate shuttles. In this system, excess reducing equivalents which are transferred to the cytoplasm along with ATP by the 3-phosphoglycerate/dihydroxyacetone phosphate shuttle, are reimported into the chloroplasts for consumption in photosynthetic carbon reduction. This system explains why in vivo a higher phosphorylation potential along with a lower NADH/NAD+ ratio is maintained in the cytoplasm in the light (Heber and Santarius, 1965, 1970). 3. The Adenine Nucleotide Translocator Strotmann and Berger (1969) and Heldt (1969) were the first to report a counter-exchange of adenine nucleotides (ADP, ATP) in isolated intact chloroplasts. Robinson and Wiskich (1977b) have indicated that uptake of ATP analogues, which “closely mimic ATP” in their transport properties, decreased the internal ATP concentration and thus inhibited C0,-dependent 0,evolution. The average rate of adenine nucleotides through the spinach chloroplast envelope is too low ( 5 pmol mg-’ chlorophyll h-l) to account for the rapid increase of the ATP concentration in the cytoplasm which is caused by a dark to light transition in higher plant cells as measured by Santarius and Heber (1965) and Sellami (1976). The rate of translocation of ATP into chloroplasts is much higher than the transport of ADP (Heldt, 1969; Strotmann and Berger, 1969). Consequently, according to Heldt et al. (1972) the adenine nucleotide translocator in chloroplasts does not seem to be involved in the photophosphorylation of cytoplasmic ADP. This is in contrast to the situation in well coupled mitochondria which exhibit a preferential and very rapid uptake of external ADP rather than external ATP (Vignais, 1976). In mitochondria, the adenine nucleotide translocator plays an important role in the oxidative phosphorylation of external ADP by importing ADP and exporting ATP. It is clear that if movement of adenine nucleotides between the cytoplasm and chloroplast is not by direct transfer (via the adenine nucleotide translocator) but by indirect transfer (via shuttles of intermediates, see above) the role of the adenine nucleotide translocator in chloroplasts is to deliver ATP synthesized by respiration and/or glycolysis to the chloroplasts during the night phase (Heldt, 1969).

STRUCTURE AND FUNCTION OF THE PLASTID ENVELOPE

65

However, Robinson and Wiskich (1976, 1977c, 1977d) and Stankovic and Walker (1977) demonstrated that, in contrast to spinach chloroplasts, photosynthesis by chloroplasts from young pea shoots is affected by exogenous adenine nucleotides suggesting that the rate of adenine nucleotide translocation is higher in these chloroplasts. Nevertheless, it is not entirely clear whether this variation arises from a difference in species characteristics or from different developmental stages. Preliminary experiments carried out by Robinson and Wiskich (1977~)suggested that the rate of ATP uptake by chloroplasts isolated from young pea shoots is in excess of 20pmol mg-' chlorophyll h-l. Huber and Edwards (1976) also reported rates of ATP transport of 40 pmol mg-l chlorophyll h-l in mesophyll chloroplasts of a C4 plant (Digitaria sanguinalis). Murakami and Strotmann (1978) have shown that spinach chloroplasts contain an adenylate kinase bound to the envelope membranes. Sonication and repeated washing in a medium of high ionic strength do not release the enzyme from the envelope membranes. For these reasons, Murakami and Strotmann (1978) propose that this envelope-bound adenylate kinase may be involved in the control of the adenylate translocation and energy pool in chloroplasts as suggested before by Bomsel and Pradet (1968) and Bomsel and Sellami (1975). It must be noted, however, that the chloroplast-soluble fraction (stroma) also contains large amounts of adenylate kinase and that this constitutes the major portion of chloroplast adenylate kinase activity.

4. Miscellaneous Transport Mechanisms (a)Proton permeability of the chloroplast envelope is low. Illumination of intact chloroplasts causes a decrease of the pH in the thylakoid space of 1.5 pH units (Neumann and Jagendorf, 1964; Werdan et al., 1975) and an increase of the pH in the stroma by almost 1 pH unit (Werdan et al., 1975). If H+ and/or OH- could easily penetrate the chloroplast envelope, leakage from the alkaline stroma should increase the p H of the external medium. Instead, in a C0,-free medium, intact chloroplasts slowly excrete protons against the existing gradient although the total movement is very small compared to the quantities of protons passing through the thylakoid membranes (Heber and Krause, 1971; Heldt et al., 1973; Heber and Purczeld, 1978). There is evidence to suggest that the pH changes occurring in the stroma after illumination are sufficient to switch CO, fixation from zero to maximal activity (Werdan et al., 1975). The principal site of inhibition is found to be in the carbon cycle between fructose bisphosphate and sedoheptulose bisphosphate and the corresponding monophosphates (Hiller and Bassham, 1965; Purczeld et al., 1978) and also at the level of the ribulose 1,5-bisphosphate carboxylase (Bowes et al., 1975). Heber and Purczeld (1978) have clearly demonstrated that often both an anion and its neutral protonation product (e.g. NO,-/NO,H) can permeate the

66

R. DOUCE AND J. JOYARD

envelope inner membrane. Such a transfer constitutes a shuttle system enabling indirect transfer of protons across the envelope and significantly decreases the proton gradient between the stroma and the external medium. Under these conditions, such a transfer leads to an inhibition of CO, fixation. (b) Cation permeability. During illumination, proton uptake by isolated thylakoids is partially electrically compensated by an efflux of Mg2+ (Dilley and Vernon, 1965; Hind et al., 1974; Barber et al., 1974). As pointed out by Pfliiger (1973), a rapid exchange of Mg2+ between intact chloroplasts and cytoplasm would largely abolish the primary increase of Mg2+concentration in the stroma during illumination. There is good evidence, however, that the chloroplast envelope acts as an efficient barrier to free diffusion of cations (Pfliiger, 1973; Gimmler et al., 1974; Krause, 1974; Barber et al., 1974). Furthermore, it has been clearly shown by Portis and Heldt (1976), Krause (1977) and Miginiac-Maslow and Hoarau (1977) that illumination of intact chloroplasts causes an increase in the stromal free Mg2+concentration of 1-3 mM. Light-dependent extrusion of Mg2+out of the thylakoid space into the stroma could play an important role in the regulation of CO, fixation. In fact, Portis et al. (1977) have already concluded that the activities of sedoheptulose bisphosphatase and fructose bisphosphatase can be controlled by light-dependent changes of the stromal Mg2+ concentration. The effect on ribulose bisphosphate carboxylase is equally important. Thus, the reconstituted chloroplast system will support good 0, evolution with 3-phosphoglycerate as substrate and 1 mM MgCl,, but only negligible rates with ribulose 5-phosphate (or ribulose-l,5-bisphosphate)as substrates, until Mg2+ is raised to 2 mM (Lilley et al., 1974). It is surprising, however, that even though the envelope is impermeable t o various cations, the stroma of the chloroplasts contains large amounts (20-40 mM) of bound Ca2+ and Mg2+ (Joyard and Douce, 1976a); but it is not understood how and in what conditions these cations entered the chloroplasts. (c) CO, transport. At physiological pH both anionic HCO,- and molecular CO, are present in an aqueous medium (CO, equilibrates with water a t physiological pH values to give dissolved CO, and HC0,- anion). Under these conditions, it is difficult to say whether chloroplasts import CO,, HCO,or both. Werdan et al. (1972) concluded that CO, diffuses very rapidly across the chloroplast envelope membranes. CO, would then equilibrate with HC0,- in the stroma, the reaction is presumably catalysed by the large amounts of carbonic anhydrase found within this compartment (Everson and Slack, 1968; Everson, 1970; Poincelot, 1972). When intact isolated chloroplasts are illuminated, the accumulation of HC0,- in the stroma is considerably enhanced and the distribution of HC0,- between the stroma and the medium space (which is predicted from the Henderson/Hasselbach equation) is inversely proportional to the distribution of protons (Werdan et al., 1972). In other words, the equilibrium pH gradient across the envelope in

STRUCTURE AND FUNCTION OF THE PLASTID ENVELOPE

67

the light (see above) induces a rapid intake of CO, molecules into the stromal space. Werdan et a/. (1972) justifiably concluded that there is no requirement for transport of HC0,-. Using isolated envelope vesicles, Poincelot (1974, 1975) and Poincelot and Day (1 976) suggested that HCO,- is rapidly transported across the membrane of the envelope vesicles and then converted to CO, rather than direct passage of CO, as stated by Werdan et at. (1972). They suggested also that bicarbonate transport is directly correlated with the level of ATPase activity in the envelope membranes (Poincelot and Day, 1976). Such a result is in contrast to the fact that biological membranes do not present a significant barrier to dissolved CO, since molecular CO, diffuses readily through phospholipid bilayers (Blank and Roughton, 1960). In order to preserve electroneutrality a movement of HCO,- would also presumably involve cotransport with a cation or counter-transport against an anion. Furthermore, we believe that it is unwise to work with isolated envelope membranes because they are unstable and once isolated they fragment into smaller vesicles on ageing (Neuburger e t a / . , 1977). Finally, as already stated, we know nothing about the orientation of isolated envelope membrane vesicles and it cannot be assumed that transport properties, for example, are necessarily the same in both directions. Sabnis et al. (1970) by cytochemical methods demonstrated the presence of a Mg2+-dependentATPase on the chloroplast envelope in tendrils of Pisum sativum (Fig. 36). A cation stimulated ATPase associated with the peripheral reticulum of Amaranthus retroJexus (Amaranthaceae) has similarly been demonstrated (Bill’ et al., 1976). This ATPase has been characterized biochemically by Douce et al. (1973b) in the isolated envelope of spinach chloroplasts. This membrane-bound ATPase, specific for Mg2+and/or Mn2+ (Joyard and Douce, 1975), is distinct from the Mg2+-dependentATPase of the coupling factor 1 (because it is insensitive to N,N’-dicyclohexylcarbodiimide, to the antibiotic Dio-9 and to antisera against coupling factor 1). This ATPase presents a broad pH optimum between 7.4 and 8 and is not involved in cation transport through the envelope membranes (Joyard and Douce, 1975). The activity we have found is low. Higher values are obtained in spring and autumn (8-12 pmol phosphate formed mg-l protein h-l) and lower ones during winter and summer (2-4 pmol phosphate formed mg-l protein h-l). A non-latent Mg-2-ATPase of similar activity has been also noted for several Spinacia oleracea envelope preparations (Poincelot, 1973; Laetsch and Sprey, 1977; Champigny and Bismuth, 1977) and for fractions thought to be enriched in etioplast envelopes (Cobb and Wellburn, 1974; Cooke and Kendrick, 1976). Laetsch and Sprey (1977) also indicated that the chloroplast envelope fractions of Portulaca oleracea and Amaranthus gangeticus (C, plants) exhibit Mg2+-dependent ATPase activity which is up to five times greater than that of envelope-enriched fractions of C , plants. Poincelot and

Fig. 36. ATPase activity on chloroplast envelope. Glutaraldehyde-fixed tissue (tendril from pea, Pisum sativum L. Var. Alaska) is incubated with ATP and Mg2+at pH 7.2. The electron-dense reaction product is largely restricted to the envelope (arrows), with occasional spotty deposits on the mitochondria1 envelope. M: mitochondria; W: cell wall. Micrograph provided by Drs D. D. Sabnis, M. Gordon and W. A. Galston; reproduced by permission from Pi. Physiol(l970) 45,25-32.

STRUCTURE AND FUNCTION OF THE PLASTID ENVELOPE

69

Day (1974) claim, however, that this ATPase is present in the space between the outer and the inner membranes of the envelope and that this enzyme is probably released into the supernatant during the centrifugation especially if the membranes are “badly damaged”. They found also in their “double membrane vesicles” a huge ATPase activity (140 pmol phosphate formed mg-l protein h-l). In contrast, we showed that in C, plant species this ATPase is membrane bound and has low activity (Douce et al., 1973b; Joyard and Douce, 1975). In an interesting study Champigny and Bismuth (1977) by using the same method as Poincelot and Day (1974) for isolating the envelope membranes from spinach chloroplasts found that the ATPase activity in isolated envelope membranes is low and exhibits a biphasic response to increasing temperature. At the moment, the physiological function of the Mg2+ATPase of chloroplast envelopes cannot be fully explained. It is widely accepted that this enzyme is not involved in monovalent or divalent cation transport, but it may be involved in the slow excretion of protons in the external medium upon illumination of intact chloroplasts (Heber and Purczeld, 1978). It is also possible that this ATPase may be involved in the transport of bicarbonate, thus Champigny and Bismuth (1 977) demonstrated that the envelope ATPase activity is stimulated by the anion HC0,- and could be concerned with bicarbonate transport. The stimulation observed, however, is indirect and could be entirely attributable to an alkalinization of the stroma compartment during the course of ATP hydrolysis (Champigny and Joyard, 1978). Heber and Purczeld (1978) have criticized Poincelot’s work and in agreement with Werdan et d . (1972) they provide definitive evidence that the chloroplast envelope has a low permeability for the bicarbonate anion. The two major arguments in favour of this conclusion are: a. On addition of potassium bicarbonate, intact chloroplasts shrink osmotically. Valinomycin which is known to increase specifically the potassium permeability of biomembranes (Pressmann, 1973) does not induce chloroplast swelling, presumably because net influx of HCO,- is not possible. b. Carbonic anhydrase is present in the stroma and cannot be measured in intact chloroplasts. If there is a carrier capable of fast HC0,-/OHexchange in the envelope, it should transfer HCO,- (formed inside the chloroplasts from CO, by carbonic anhydrase) into the medium in exchange for OH- from the medium. This would neutralize the protons formed in the stroma and lead to an acidification of the medium. Obivously this is not the case. Finally, Douce et al. (1973b) and Poincelot (1973) found that isolated envelope membranes are devoid of carbonic anhydrase activity demonstrating that this enzyme is not involved in the passage of CO, through the envelope membranes. However, a crude chloroplast envelope preparation from a C,

70

R. DOUCE AND J. JOYARD

plant (Eleusine coracana, Gramineae) was found to show carbonic anhydrase activity (Rathnam and Das, 1974a). (d) N 0 , - transport. There is now some evidence which indicates that the light-dependent reduction of nitrite to amino-nitrogen occurs in isolated intact chloroplasts (Grant et al., 1970; Dalling et al., 1972; Miflin, 1974; Plaut et al., 1977). Rathnam and Das (1974b) have shown that over 80 % of the nitrate reductase activity of intact purified chloroplasts from Eleusine coracana is associated with the envelope membrane fraction. They believe that the chloroplast envelope appears to be a major site of nitrate reduction. In contrast, the pure chloroplast envelope fraction isolated from spinach leaves is totally devoid of nitrate reductase activity and the bulk of the activity is recovered in the soluble cell fraction (M. Neuberger, J. Joyard and R. Douce, unpublished data). Inside the green leaf cells, the nitrate reductase activity appears more likely to be associated with the thin film of cytoplasm (Ritenour et al., 1967; Swader and Stocking, 1971). Consequently, nitrite has to pass through the envelope in order to be reduced by ferredoxin and nitrite reductase present in the stromal space: 2 NO,-

+ 6 H2O --

-+

2 NH,+

+ 4 OH- + 3 0,

Nitrous acid has a pK of 3.4 and some HNO, will coexist with NO2- even at neutral pH. Heber and Purczeld (1978) demonstrated that NO2- and N0,H are able to traverse the spinach chloroplast envelope. Nevertheless, they suggest that it is the undissociated form which is preferentially transported. It is only when the amount of N0,H is low in the medium that NO2- uptake, in exchange for a transferable internal anion, appears to contribute significantly to the substrate flux (especially at high light intensities). In this case, the chemical nature of the transferable anion is not known. (e) SuIphate transport. It is generally accepted that assimilatory sulphate reduction takes place in the chloroplasts (Asahi, 1964; Schiff and Hodson, 1973; Schwenn and Trebst, 1976). This implies that the envelope membranes of chloroplasts are permeable to sulphate. By the use of a silicone layer filtering centrifugation technique, it was found that inorganic sulphate is slowly taken up into chloroplasts (Hampp and Ziegler, 1977; Mourioux and Douce, 1978). Sulphate transport has saturation kinetics of the Michaelis-Menten type (Fig. 37). The data presented by Mourioux and Douce (1978) show that sulphate could accumulate in the chloroplast against its own concentration gradient. The rate of sulphate uptake is increased when the sulphate concentration in the medium is increased from 0.3 mM to 6 mM (Mourioux and Douce, 1978). In marked contrast to Hampp and Ziegler (1977) we have found that the uptake of sulphate is lowered by addition of phosphate in the external medium: sulphate and phosphate compete for a common carrier (Fig. 37). Less inhibition is found

71

STRUCTURE AND FUNCTION OF THE PLASTID ENVELOPE

0.4

r

. J=

0.3

0 0)

f 0.2

?

‘*

-0 0)

E a 0.1

>

5 0

-1

0

1

2

3

l/SULPHATE, mM-l Fig. 37. Transport of sulphate across the chloroplast envelope. Concentration dependence of the transport of inorganic sulphate, inhibition by inorganic phosphate (Pi). Intact spinach chloroplasts were incubated in the dark at 20°C. Michaelis-Menten representation. Adapted from Mourioux and Douce (1978).

when 3-phosphoglycerate and dihydroxyacetone-phosphate are used. When the chloroplasts are preloaded with labelled sulphate, the radioactivity is released after addition of phosphate (Hampp and Ziegler, 1977; Mourioux and Douce, 1978), (Fig. 38). Consequently, the slow transport of sulphate occurs by a strict counter-exchange: for each molecule of sulphate entering the chloroplast, one molecule of phosphate leaves the chloroplast and vice-versa. From this it may be concluded that there is a specific carrier catalysing the transport of sulphate which has been named “sulphate translocator” (Mourioux and Douce, 1978). The activity of the sulphate carrier (about 20 pmol sulphate transported mg-l chlorophyll h-l) is much less than the activity of the phosphate translocator (about 300 pmol phosphate transported mg-l chlorophyll h-’ at 20”). The uptake of sulphate by isolated intact chloroplasts in exchange for endogenous free phosphate induces a lower rate of photophosphorylation which, in turn, inhibits C0,-dependent 0, evolution (Baldry et al., 1968). (f)Neutral amino acid transport. Some controversy exists over the permeability of chloroplasts to neutral amino acids. Two amino acid translocators have been implicated in the rapid transport of neutral amino acids into

72

R. DOUCE AND J. JOYARD

Fig. 38. Sulphate outflow from chloroplasts induced by phosphate. Intact spinach chloroplasts were preincubated with [J6S]-sulphate;then sulphate outflow was measured in presence or absence of phosphate in the incubation medium. Adapted from Mourioux and Douce (1978).

chloroplasts. Evidence for this observation was based on the osmotic response to added amino acids shown by preparations of chloroplasts which were thought to have intact envelopes (Nobel and Wang, 1970; Nobel and Cheung, 1972). A slight osmotic response was observed with amino acid solutions as high as 50 mM. Higher concentrations evoked osmotic shrinkage, which was interpreted as indicative of saturation of the transport system. In contrast, Gimmler et ul. (1974) and Heldt (1976a) reported very low rates of net uptake of neutral amino acids into isolated chloroplasts. Based on both uptake of labelled amino acids and osmotic volume changes they found no evidence for carrier-mediated transport. They did not investigate the transport of leucine. Recently, McLaren and Barber (1977) have found low rates for uptake of L-leucine and isoleucine comparable to those given by Gimmler et al. (1974) for other amino acids. However, labelled L-leucine, unlike other amino acids was accumulated against a concentration gradient which suggested, along with uptake kinetics and competition experiments, the involvement of a translocating system for L-leucine. It is noteworthy that isolated intact chloroplasts of Viciu fuba were found to be able to synthesize most amino acids provided that keto acids were available, but no synthesis of

STRUCTURE AND FUNCTION OF THE PLASTID ENVELOPE

73

leucine by the aminotransferase reaction could be detected (Kirk and Leech, 1972; Leech and Murphy, 1976). Consequently, according to McLaren and Barber (1977), the leucine carrier may be important in maintaining the leucine pool in chloroplasts especially during the course of protein synthesis which occurs in the stromal space. ( g ) Glucose transport. There are several indications that hexoses, especially glucose, pass through the chloroplast envelope. For example, it has been observed that feeding Chlorella (Kandler, 1954; Kandler and Gibbs, 1959), moss spores (Chevallier, 1974), tobacco leaves (McLachlan and Porter, 1959) or leaf discs of several plants (Kandler et a/., 1977) with l-[14C]-glucose yielded starch in which the glucose is still predominantly labelled in position 1. Since starch synthesis occurs in the plastids, these findings indicated that most of the glucose passes through the envelope without being transformed to trioses. Furthermore, it has been shown with isolated intact chloroplasts that during starch mobilization part of the carbon is released as triosephosphate (Heldt e t a / . , 1977; Peavey et a/., 1977) but some is also released as glucose (Fig. 34) and so, must be able to cross the chloroplast envelope. Schafer et a / . (1 977) found that D-glucose along with D-mannose, D-xylose and L-arabinose are slowly transported across the inner membrane of the chloroplast envelope. They concluded also that D-glucose and other hexoses are transported by carrier-mediated diffusion across the inner envelope membrane. According to these authors the glucose carrier may be directly involved in the release of glucose formed during starch hydrolysis in chloroplasts (Fig. 34). The exact molecular mechanics of transport are still open to debate. Models for transport across biological membranes fall into two general classes: the mobile carrier (Rosenberg and Wilbrandt, 1963) and the fixed pore (Vidaver, 1966). The mobile carrier is defined as a hydrophobic macromolecule which is able to shuttle between the two faces of the membrane when it is loaded with its substrate. The whole carrier macromolecule can move across the membrane by translation or by rotation. The mobile carrier exposes its binding site alternatively to the outer face and to the inner face of the membrane but not to both faces at the same time. Opponents of this hypothesis (Bretscher, 1973; Singer, 1974; Guidotti, 1976) suggest that the passage of ionic and polar molecules through the hydrophobic core of the membrane is thermodynamically a highly unfavourable process. The free energies of activation required are so large that the rates of such rotations or diffusions are negligibly slow. In the fixed pore or channel mechanism it is easy to imagine that a channel made of several polypeptide subunits which spans across the membrane may be asymmetric in structure and that the regions of the channel located on the inner and outer sides of the membrane may have different conformations and therefore different binding affinities. In active transport, a quaternary rearrangement in protein subunits could

74

R. DOUCE AND J. JOYARD

account for the transfer of metabolites between adjacent compartments. In this model, the energy requirements for active transport may serve to produce a quaternary rearrangement of the subunits forming the channel and so translocate the ligand. Orci et al. (1977) have observed intramembrane protein particles displaying small central pores. The extent to which such models of membrane architecture can be applied to the chloroplast inner envelope membrane is still largely a matter of conjecture. B. PROTEIN TRANSPORT THROUGH THE ENVELOPE MEMBRANES

The discovery of DNA and ribosomes in chloroplasts (Ris and Plaut, 1962; Lyttleton, 1962) supported the idea that these organelles may be genetically independent structures. Several attempts have been made to grow isolated envelope bounded chloroplasts in culture (Ridley and Leech, 1970; Giles and Sarafis, 1974). There is also at the present time a considerable amount of cytological evidence to support the concept of chloroplast continuity between cell generations in all of the major groups of plants (Guillermond, 1941; Kirk and Tilney-Bassett, 1967). Although chloroplasts contain all the components necessary for biological systems to be autonomous i.e. DNA, DNA polymerase, RNA polymerase and a protein-synthesizing machinery, these components neither synthesize nor code for all of the chloroplast proteins (Ellis, 1976). There is increasing evidence that many chloroplast proteins are synthesized on cytoplasmic ribosomes (Ellis et al., 1973). Furthermore, many genes directly involved in chloroplast structure and function are presumed to be located in the nucleus since they are inherited in a Mendelian fashion (Sager, 1972; Kirk, 1972). These data clearly show that the chloroplast is not as completely autonomous as the observations noted previously might imply. It now appears certain that a degree of integration exists between the action of the nucleus and that of the plastids. Consequently, it is obvious that during plastid division large amounts of cytoplasmically synthesized proteins must somehow cross the chloroplast envelope. For example, ribulose-l,5-bisphosphatecarboxylase is a major protein of the chloroplast stroma (half of the total stromal protein). It has a molecular weight of 550 000 and is composed of several copies of two nonidentical subunits (Kawashima and Wildman, 1970). It is well known that the large subunit is made within the chloroplast (Blair and Ellis, 1973; Bottomley et al., 1974; Morgenthaler and Mendiola-Morgenthaler, 1976; Coen et al., 1977) whereas the small subunit is synthesized on cytoplasmic ribosomes (Gray and Kekwick, 1974; Roy et al., 1976). Bradbeer (1973) has estimated that the maximum rate of synthesis of the large subunit in greening leaves of Phaseolus vulgaris is roughly lo4 molecules per plastid per hour. This means that small subunits which are encoded by the cytoplasmic genome are probably crossing the chloroplast envelope at a rate of lo4 molecules per plastid per hour. Cobb and Wellburn (1976), following an earlier suggestion of

STRUCTURE A N D FUNCTION OF THE PLASTID ENVELOPE

75

Pineau and Douce (1974), have reported the presence of the small subunit of ribulose bisphosphate carboxylase in well-washed plastid envelope fractions of Avena sativa. Evidence for this interpretation was based on pulse-chase experiments using 35S-methionine, tryptic digestion and peptide mapping. The presence of the small subunit in the envelope fraction was considered to have physiological significance, being indicative of its transport from the cytoplasm. The major chloroplast membrane polypeptides involved in the formation of the light-harvesting chlorophyll-protein complex were found to be synthesized also in the cytoplasm and thus have to be specifically transferred across the envelope membranes (Ohad, 1975). It is also the case of multimeric membrane-bound enzymes in mitochondria. For example, some subunits of mitochondria1 adenosine triphosphatase (Tzagoloff, 1971) and cytochrome oxidase (Mason and Schatz, 1973) in yeast are formed by mitochondrial and others by cytoplasmic ribosomes. There must, therefore, be mechanisms o n both envelope membranes for the selection of certain proteins (or protein subunits) which control their entry into the chloroplast. The mechanisms involved in transferring proteins from the cytoplasm into the chloroplast play a fundamental role in the interactions which necessarily exist between the nuclear and chloroplast genomes. So far, several mechanisms have been proposed in order to explain the passage of several proteins through the chloroplast envelope: direct injection, specific envelope carriers and pinocytosis (Fig. 39). It is necessary to remember that the chloroplast proteins encoded by the nuclear genome have to pass through two membranes since the outer membrane of the chloroplast envelope, as already stated, is not freely permeable to high molecular weight molecules (above 10 000 daltons). 1. Direct lnjection

This mechanism easily explains the passage of the proteins through the outer envelope membrane. It does not explain the passage of the proteins through the inner envelope membrane unless the passage occurs a t the point where both membranes are in contact (see above). According to this mechanism, nascent polypeptide chains destined to cross the outer membrane of the envelope and synthesized on membrane-bound ribosomes (80 S ) are vectorially discharged across the membrane in a manner similar to the trans-membrane vectorial transport of secretory proteins demonstrated to occur in animal cells (Blobel and Sabatini, 1971; Campbell and Blobel, 1976). The messenger RNA (mRNA) which is translated on bound (as opposed to free) ribosomes contains a unique sequence of codons named the signal codons. Translation of the signal codons results in a unique sequence of amino acid residues (signal peptide) at the amino terminal end of the nascent chain. The signal peptide triggers attachment of the ribosome to special membrane receptor proteins and in doing so determines which pro-

CYTOPLASM i'

pept ide

I ENVELOPE

c

3'

STROMA

THYLAKOIDS

I Fig. 39

receptor protein

STRUCTURE AND FUNCTION OF THE PLASTID ENVELOPE

77

teins may traverse the membrane via the ribosome membrane junction. Once the ribosome is bound to the membrane, the signal peptide and continuous sequence of the nascent chain passes through a proteinaceous tunnel in the membrane formed by the association of the signal sequence with specific membrane proteins. Campbell and Blobel (1976) have also shown that the signal sequence is removed by a membrane associated enzyme “signal peptidase or signalase” before translation of the chain is completed (Fig. 39). The validity of this mechanism for passage through the chloroplast outer envelope membrane was tested recently by electron microscopic studies and ribosome analysis of etiolated leaves greened in the light (Stocking et ul., 1977). An association of ribosomes appearing as groups of polysomes lying more or less at right angles to the outer envelope membrane was observed. Unfortunately, reports of extensive association of cytoplasmic ribosomes with chloroplasts have so far been very limited. When chloroplasts are isolated from greening leaves (Stocking et al., 1977) or green leaves (Lerbs and Wollgiehn, 1975; Laulhtre and Dorne, 1977), with or without the presence of inhibitors to prevent possible peptide release during the plastid isolation (Stocking et ul., 1977), a significant low level of 80 S ribosomes remained associated with the isolated washed chloroplasts. All these results suggest, but do not prove, that the envelope-bound ribosomes function in a manner similar to ribosomes bound to the endoplasmic reticulum of secretory animal cells. It is interesting to note that electron microscopic examination of growing spheroplasts prepared from Succharomyces cerevisiue also revealed ribosome-like particles aligned along the outer mitochondria1 membrane (Kellems and Butow, 1972; Keyani, 1973). These electron micrographs show a striking similarity between the association of cytoplasmic ribosomes with yeast mitochondria and the association of cytoplasmic ribosomes with the endoplasmic reticulum in secretory cells of higher eukaryotes. However, it is difficult to visualize how membrane proteins can adopt their

Fig. 39. Summary of different hypotheses related to protein integration within chloroplast membranes. Chloroplast proteins can be synthesized either by cytoribosomes (1 and 2) or plastid ribosomes (3). Integration into membranes (envelope or thylakolds) occurs when nascent proteins are either coiled or uncoiled. When proteins are synthesized by cytoribosomes, they must pass through the two envelope membranes. In this case, cytoribosomes which can be either free in the cytoplasm (1) or bound to the chloroplast envelope (2) could synthesize proteins with a signal peptide involved in the recognition of the nascent protein by a receptor protein located on the outer envelope membrane. This mechanism would allow the transfer of protein through this first barrier. The signal peptide would then separate from the nascent chain by an enzyme (signalase) located on the inner face of outer membrane. Later, proteins (coiled or uncoiled) would be integrated into the inner envelope membrane. Transport to thylako‘ids could be facilitated by pinocytosis. Contact points between outer and inner membrane (see Fig. 18) could also allow direct transfer through both membranes.

78

R. DOUCE AND J. JOYARD

correct polarity in the membrane if they are extruded through the membrane during synthesis. This problem is probably more acute for proteins which are only partially buried in the inner half of the bilayer. Therefore, we suggest it is more likely that the membrane proteins take up their correct location by diffusion to the membrane and insert themselves into it after they are synthesized in the cytoplasm (Bretscher, 1973).

2. Specific Envelope Carriers This hypothesis implies that a special class of proteins exists in the envelope of chloroplasts which catalyse the unidirectional influx of all those proteins (coiled or uncoiled) made on cytoplasmic ribosomes but which are destined to function in the chloroplast (Blair and Ellis, 1973). It is clear that a recognition event is necessary between the chloroplast envelope and the entering protein to ensure that the proteins synthesized in the cytoplasm and destined for the chloroplasts enter the correct compartment (i.e. the space between the outer and the inner membrane of the envelope). In an interesting study, Dobberstein et al. (1977) provided evidence that translation of mRNA of Chlamydomonas reinhardtii in a cell-free wheat germ system resulted in the synthesis of a major polypeptide with a molecular weight of 20000. This polypeptide was specifically immunoprecipitated by antibodies raised against the small subunit (16 500 daltons) of ribulose-l,5-bisphosphatecarboxylase. From this it was concluded that the small subunit is synthesized with an additional sequence (3500 daltons) by free ribosomes in the cytoplasm. In contrast to secretory proteins, the additional sequence does not trigger attachment of ribosomes to the chloroplast outer envelope membrane and is cleaved after chain completion by a soluble or envelope membrane bound endoprotease. It is most likely that this additional sequence contains the information necessary for specific binding to a receptor in the chloroplast envelope (the envelope protein carrier) possibly localized in the outer envelope membrane, or in regions where the inner and outer envelope membranes are in contact, to mediate transfer of the small subunit from the cytosol into the chloroplast stroma (Fig. 39). This work has been confirmed by Ellis et al. (1978) and Highfield and Ellis (1978). These authors also gave evidence that the small subunit of ribulose-l,5-bisphosphatecarboxylase is synthesized as a precursor of higher molecular weight when mRNA from pea cytoplasmic polysomes is translated in a cell-free wheat germ system. This precursor is taken up into isolated intact chloroplasts by the specific envelope carriers, and cleaved to its final size in the absence of protein synthesis by a specific envelope bound endoprotease. Feierabend and Schrader-Reichhardt (1976) have also shown that chloroplast proteins made in the cytoplasm can enter the chloroplast in the absence of chloroplast ribosomes. This hypothesis is strengthened by the fact that the small subunit of ribulose-l,5-bisphosphatecarboxylase prepared

STRUCTURE AND FUNCTION OF THE PLASTID ENVELOPE

79

from [14C0,]-labelled young spinach leaves (M. Neuburger, B. Pineau and R. Douce, unpublished data), is unable, as such to pass through the spinach chloroplast envelope although Marra et al. (1978) have shown the direct transport of several mitochondrial enzymes through both outer and inner mitochondrial membranes. It is, therefore, clear that further work has to be done in order to provide definitive evidence for the existence of specific protein carriers in the outer membrane of the envelope. In addition, information on intramembranous particles in the outer envelope membrane could give important clues to the possible operation of such mechanisms. It is also conceivable that the mechanism involved in the passage of the hydrophobic polypeptides (intrinsic proteins) through the envelope could be entirely different to that involved in the passage of the hydrophilic proteins (e.g. ferredoxin).

3.Pinocytosis The space between the outer and the inner membrane of the chloroplast envelope is electron transparent and devoid of ribosomes. Consequently, both previous mechanisms which readily explain the passage of the polypeptides through the outer envelope membrane are no longer valid for the inner membrane unless specific envelope carrier proteins are localized in regions where the inner and the outer envelope membranes are in contact (see above). As already discussed, observations made during electron microscopic studies of developing plastids indicate that the inner membrane of the plastid envelope invaginates (figure of pinocytosis) and forms vesicles, which could then fuse with the growing internal membrane system (Fig. 40). It is possible that some proteins which pass through the outer membrane of the envelope interact with the external surface of the inner envelope membrane and trigger the formation of pinocytotic vesicles. In support of this suggestion, the inner membrane of the envelope is presumably the site of synthesis of galactolipids (Douce, 1974), the major lipid components of thylakofd membranes (Benson, 1971). Subsequently, the polypeptides would become positioned in the internal half of thylakoid membranes as a consequence of the fusion of vesicles derived from the inner envelope membrane (Fig. 39). Ultrastructural studies have shown that some chloroplast ribosomes are associated with the chloroplast thylakoids in arrangements suggestive of polyribosomes (Falk, 1969; Chua et at., 1973; Margulies and Michaels, 1974). In isolated spinach and pea chloroplasts some ribosomes remain associated with green thylakofd membranes during repeated washings in hypotonic media (Tao and Jagendorf, 1973). These thylakofd-bound ribosomes could be involved in the vectorial discharge of new hydrophobic nascent polypeptides to the external half of the membrane (Margulies et at., 1975). Some of these polypeptides may bind with other polypeptides already present in the internal half of the thylakofd membrane. This scheme which involves formation of vesicles from

80

R. DOUCE AND J. JOYARD

Fig. 40. Proplastid from etiolated Hordeum leaf. Arrows indicate invaginations of the inner membrane of the plastid envelope. Micrograph provided by T. Guillot-Salomon (unpublished data).

the inner envelope membrane readily explains, for example, the association of the chlorophyll-protein complex apoproteins with thylako'id membranes in green algae and higher plants (Hoober, 1976). It is also possible, according to a recent suggestion of Ellis (1977), that besides enzyme and structural proteins some regulatory proteins must also cross the envelope in order to trigger nucleic acid and protein synthesis in the stroma of the chloroplasts and to ensure a balanced synthesis of the subunits made outside and inside the chloroplasts. It is conceivable that polypeptides synthesized in the cytoplasm which control the synthesis of subunits in the organelle are the same subunits which combine with them to form the complete protein. In support of this theory, a recent article reported that protein synthesis in isolated yeast mitochondria is strongly enhanced by addition of cytoplasmic proteins (Poyton and Kavanagh, 1976). Cobb and Wellburn (1974) have also shown that permeability changes of the plastid envelope are an important feature of the regulation of plastid

STRUCTURE AND FUNCTION OF THE PLASTlD ENVELOPE

81

development. However, this result must be interpreted with caution because the total protein per lo4 plastids they give is surprisingly high (28.3 pg total protein per lo4 plastids in 11-day old light grown Avena). Jennings and Ohad (1972) have also suggested that some chloroplast membrane proteins in Chfamydomonas reinhardtii are translated on chloroplast ribosomes, but the messenger molecules coding for them are cytoplasmic in origin. This novel suggestion seems to indicate that some mRNA molecules are able to pass through the envelope membranes. Finally, reports have suggested that pores may exist in the chloroplast envelope and that these may be identified by electron microscopy. For example, Nedukha (1976) has described complex pores of 35 nm diameter in chloroplast envelopes of the bryophyte Funaria hygrometrica, although no evidence for this proposal was found by Nurit (1979) using the same material. Scanning electron microscope studies undertaken by Mohapatra and Johnson (1976) led them to suggest that much larger pores (greater than 1 pm!) could exist in Nicotiana tabacum chloroplasts. If these pores really do exist the large molecules (proteins, RNA) would easily pass through the chloroplast envelope without the intervention of a special mechanism. However, it seems most unlikely that a finely controlled passage of metabolites across the envelope (Heldt, 1976a) could occur in the presence of large apertures of this sort. It is obvious that in spite of the great advances made recently there is still considerable work to be done to provide an understanding of the whole process of the synthesis, transport and integration into the chloroplast structure of chloroplast proteins encoded for in the cytoplasm. It is more than certain that both envelope membranes, the permanent structure of all plastids (Figs I, 9-14, 17, 40), play a fundamental role in this process. At this level, the outer envelope membrane does not constitute an “unspecific permeable barrier” in contrast with what was said previously for the transport of anions. It is likely to contain the information necessary for specific nucleic acid and/or protein binding. For example, it has been found that the chloroplast outer membrane of the chloroplast envelope is the site of TYMV-RNA replication (Laflkche et al., 1972). Finally, if envelope proteins responsible for recognition or transport of cytoplasmically synthesized proteins are manufactured by the chloroplast, the organelle may exert a greater control over its own composition than has hitherto been realized. C. LIPID SYNTHESIS BY THE CHLOROPLAST ENVELOPE MEMBRANES

The major chloroplast membrane acyl lipids consist of three glycolipids (monogalactosyldiacylglycerol, digalactosyldiacylglycerol and sulphoquinovosyldiacylglycerol) and two phospholipids (phosphatidylglycerol and phosphatidylcholine). Whether small quantities of phosphatidylethanolamine also occur in chloroplasts is a problem which is, as yet, unresolved and is

82

R . DOUCE AND J. JOYARD

certainly a reflection of the difficulties involved in the isolation of chloroplast membranes from other cellular constituents. The fatty acids of leaf chloroplasts are unusual in their high degree of unsaturation : trienoic acids, particularly a-linolenic acid (all-cis-9, 12, 15, octadecatrienoic acid), predominate (Table 111). This fatty acid can account for over 90% of the total chloroplast fatty acids (Hitchcock and Nichols, 1971) and contribute to the stability of photosynthetic membranes (Guarnierri and Johnson, 1970). The exact function of the polar lipids is uncertain. It is clear that the galactolipids which represent 80 % of the polar lipids in the chloroplasts are the principal constituents of the fluid lipid bilayer acting both as a permeability barrier to polar molecules and as a flexible framework capable of accommodating a variety of proteins. In contrast, the acidic polar lipids may be arranged as an annulus around certain specific proteins firmly embedded within the membrane and perhaps extending fully across it. These acidic lipids probably create a favourable environment for enzymatic activities. For example, Anderson (1975b) and Leech and Murphy (1976) indicated that phosphatidylglycerol molecules represent boundary lipid in association with the protein-chlorophyll complex which spans the entire thylakold membrane. A further possibility is that phosphatidylglycerol may be involved in the stacking of granal membranes and there is recent indirect evidence to support such a role (Tuquet et al., 1977). 1. Origin of Chloroplast Phospholipids The biosynthesis of phosphatidylcholine by the nucleotide pathway has been conclusively demonstrated in spinach leaves (Devor and Mudd, 1971; Marshall and Kates, 1974): CDP-choline

+ sn- 1,Zdiacylglycerol -+

phosphatidylcholine

+ CMP

and occurs almost exclusively in the “microsomal fraction”. Unfortunately, the “microsomal fraction’ is a 100000 g pellet collected from a cell-free homogenate after removal of the larger cell organelles such as chloroplasts (4000 g pellet) and mitochondria (10 000 g pellet). As well as fragments of lamellar and cristae membranes, the microsomal fraction contains numerous heterogeneous vesicles derived from the microbodies, dictyosomes and endoplasmic reticulum. It also contains chloroplast envelope vesicles. Complete chloroplast envelopes sediment at 30 000 g but the membranes also fragment into smaller vesicles which sediment at 100 000 g or more. Therefore, much of the phosphorylcholine transferase activity previously believed to be associated with the “microsomal fraction” could in fact be derived from the chloroplast envelope vesicles. However, Joyard and Douce (1976b) have demonstrated that the purified chloroplast envelope from spinach leaves although containing large amounts of sn-l,2-diacylglycerol (Joyard and Douce, 1976c), are devoid of phosphorylcholine transferase activity (Fig. 41) indicating probably

83

STRUCTURE AND FUNCTION OF THE PLASTID ENVELOPE

800

ENVELOPE

W

2 ’

gil 8‘

’p1

500 400

300

0 c 200

s5

100

0

0

20

40

60 TIME ,mln.

120

0

20

40 60

120

TIME,min.

Fig. 41. Galactolipids are synthesized by envelope membranes and not by “microsomes”. On the contrary, envelope membranes are unable to catalyse the transfer of choline from CDP-choline to phosphatidylcholine, while “microsomes” can do it easily.

that phosphatidylcholine is synthesized outside the chloroplasts. These results are essentially in agreement with the finding of Devor and Mudd (1971). It is possible that phosphatidylcholine could be synthesized in the envelope by methylation of phosphatidylethanolamine. Indeed the methylation of phosphatidylethanolamine as a route for biosynthesis of phosphatidylcholine in spinach leaves has been established in vivo as well as direct methylation of phosphatidyl-N-methylethanolamineto phosphatidyl-N,Ndimethylethanolamine and of the latter to phosphatidylcholine by S-adenosylmethionine (Marshall and Kates, 1974). This observation may explain the absence of phosphatidylethanolamine in envelope membranes. Similarly, Marshall and Kates (1972) provided evidence that the biosynthesis of phosphatidylglycerol which involves the intermediate formation of phosphatidylglycerophosphate from CDP-diacylglycerol and sn-glycerol 3phosphate occurs in spinach leaves: sn-glycerol 3-phosphate + CDP-diacylglycerol~3-sn-phosphatidyl-l’-sn-glycerol-3-phosphate + CMP 3-sn-phosphatidyl-1’-sn-glycerol 3-phosphate+3-sn-phosphatidyl-1 ’-sn-glycerol + phosphate

Subcellular localization studies of phosphatidylglycerol synthetase revealed that 3 % of the total activity is associated with the chloroplast fraction. Activity is also found in 15 000 g and 90 000 g pellets and in the 90 000 g supernatant. The fraction with the highest specific activity and largest proportion of the total activity is the 40 000 g microsomal pellet which is virtually free of mitochondria and chloroplast fragments. Although phosphatidylglycerol synthetase in green eaves is called a “microsomal” enzyme (Marshall and Kates, 1972) it is quite possible that it originates elsewhere, for example, in the chloroplast envelope. J. Joyard and R. Douce (unpublished data) have conclusively demonstrated the absence of a CDP-diacylglycerol synthesizing enzyme in the envelope membranes and thylakoi‘ds isolated from spinach

84

R. DOUCE AND J. JOYARD

chloroplasts. Such a result indicates that probably the chloroplast membranes and particularly the envelope membranes do not contain the enzymes involved in phosphatidylglycerol synthesis. This problem should be further investigated in order to determine whether or not the envelope is able to catalyse the final steps of phosphatidylglycerol synthesis. All these results strongly suggest, but do not prove, that the chloroplast membrane phospholipids may be synthesized by the endoplasmic reticulum system and then incorporated into chloroplast envelope membranes. The same is probably true for the major plant mitochondria1 phospholipids. In this case, in spite of many attempts (Douce et a/., 1968) definitive evidence for the synthesis of phosphatidylethanolamine and phosphatidylcholine by intact plant mitochondria in vitro has yet to be obtained even though these organelles are rich in these phospholipids (Douce et al., 1968; Mazliak et al., 1968; Mazliak, 1973). If the major phospholipids of the chloroplasts are really synthesized at the level of the endoplasmic reticulum, we must imagine a direct transfer of phospholipids between the reticulum and the chloroplast envelope (Fig. 42). I n animal (Dawson, 1973; Wirtz, 1974) and plant cells (Kader, 1977) exchange of phospholipids has been demonstrated between various intracellular membranes including rough and /or smooth endoplasmic reticulum, whole mitochondria, inner and outer membranes of the mitochondria and plasma membranes. This exchange is catalysed by one or several specific phospholipid exchange proteins located in the cytoplasm (Kader, 1977) and operating as phospholipid carriers. Unfortunately, evidence for phospholipid exchange between chloroplasts and other cellular fractions is lacking (Kader, 1977) although cooperation between plastids and microsomes was suggested for the biosynthesis of a-linolenic acid in young pea leaves (Trtmolikres and Mazliak, 1974). Membrane lipids are distributed asymmetrically between the two halves of the bilayer. For example, in the erythrocyte membrane, phosphatidylcholine predominates in the outer layer and amino phospholipids in the inner layer (Bretscher, 1972,1973). Recent work suggests that the unequal distribution of polar lipids may be widespread in other mammalian membranes (Nilsson and Dallner, 1977; Chap et al., 1977), bacterial membranes (Rothman and Kennedy, 1977a) and plant cell membranes (Cheesbrough and Moore, 1977). The most reliable methods for these studies appear to be either the use of phospholipase A2 digestion to form lyso derivatives of exposed phospholipids or the use of particular non-penetrating dyes or coupling reagents to form fluorescent or radioactive derivatives of phospholipids in the outer leaflet (Bergelson and Barsukov, 1977; De Pierre and Ernster, 1977; Rothman and Lenard, 1977). Transfer of phospholipids across the bilayer (“flip-flop” movement; Kornberg and McConnell, 1971) is a rather slow process. Experiments based on erythrocyte membranes have suggested a half-time of trans-

Fig. 42. Summary of different hypotheses related to phospholipid transfer from the endoplasmic reticulum to chloroplast membranes. A given phospholipid is synthesized at the level of the endoplasmic reticulum outer layer (1). It can diffuse very rapidly on the same layer of the membrane by lateral diffusion (2). Transfer to the outer envelope membrane can occur either by means of a phospholipid exchange protein (3) or fusion (4) between the endoplasmic reticulum and the outer envelope membrane. Phospholipids can be transferred to the inner layer of the membrane by a slow flip-flop process ( 5 ) or by means of an enzyme (6). Fusion (4) or contact between outer and inner envelope membranes facilitate the transfer of lipids to the inner envelope membrane. Transfer of phospholipids to the thylakolds can occur via either a specific exchange protein (3) or by formation of vesicles that are derived from the inner envelope membrane (7). Such communications between different membranes may be either continuous or intermittent.

86

R. DOUCE AND J. JOYARD

position for lipid molecules of the order of a few hours at physiological temperatures. It is possible that a specialized mechanism may enhance the rate of transfer of newly synthesized lipid across the bilayer especially during membrane assembly (Rothman and Kennedy, 1977b). Lipid molecules may also undergo rapid lateral diffusion within each monolayer (Scandella et al., 1972; Devaux and McConnell, 1972). Lateral neighbour exchange occurs some 1O1O times more rapidly than neighbour exchange across the bilayer. Although the fluidity of the bilayer varies critically with the lipid species, the chain length and degree of unsaturation of their component fatty acids and the temperature, in envelope membranes fluidity may be modulated by the presence of small quantities of sterols (Hartman-Bouillon and Joyard, unpublished data) and carotenoids (Douce et al., 1973b). Consequently, phospholipids released by the exchange protein into the outer layer of the outer envelope membrane could be transferred slowly to the inner layer of the outer envelope membrane and then, at the point where the two membranes are in contact, undergo lateral fluid translocation into either the outer or inner membrane depending on the phospholipid and its ultimate functional location (Fig. 42). Such a mechanism has been already proposed for the transfer of cytoplasmically synthesized phospholipids between both mitochondria1 membranes (Ruigrok et al., 1972). Finally, the very interesting and attractive proposal concerning possible structural continuity between various cellular membranes (MorrC et al., 1971 ; MorrC and Mollenhauer, 1974, 1976) and particularly between the endoplasmic reticulum and the outer envelope membrane (Fig. 15), if correct, would mean that phospholipids synthesized in the reticulum could diffuse laterally within a continuous membrane network to the outer envelope membrane. In other words, the endoplasmic reticulum behaves as a “generating element”, the “end product” being the envelope membranes and subsequently the thylakoids (Fig. 42). 2. Origin of the Chloroplast Galactolipids Early reports had suggested that the chloroplast is the main site of galacto. lipid synthesis in leaves (Neufeld and Hall, 1964; Ongun and Mudd, 1968). A later report showed that the highest specific activities of UDP-galactose incorporation into galactolipids is associated with the 40 000 g and 100 000 g pellets and the conclusion was drawn that the activity is associated with the “microsomal fraction” (Van Hummel, 1974). It has subsequently been unequivocally demonstrated that the chloroplast envelope is the only site of UDP-galactose incorporation into both monogalactosyldiacylglycerol and digalactosyldiacylglycerol in leaf cells (Douce, 1974; Douce and Benson, 1974). As previously noted the reason for the confusion of the earlier reports is the tendency for chloroplast envelopes to lyse during the isolation procedure. In addition, the digalactosyldiacylglycerol synthetase is relatively

STRUCTURE AND FUNCTION OF THE PLASTID ENVELOPE

87

easily removed from the envelope membranes and hence much of its activity was found in the cytoplasmic fraction of the 100 000 g supernatant (Mudd et al., 1969; Siebertz and Heinz, 1977a). The ability of the isolated chloroplast envelope to synthesize galactolipids was confirmed by other workers using leaves (Van Hummel et al., 1975; Van Hummel and Wintermans, 1975; Liedvogel and Kleinig, 1976; Williams et al., 1976b) and cells of Euglena gracilis (Blee and Shantz, 1978). Furthermore, Liedvogel and Kleinig (1976, 1977) have clearly demonstrated that the non-photosynthetic chromoplast inner membranes from the corona of Narcissus pseudonarcissus are also found to contain galactolipid synthesizing activities. The same is probably true for the envelope of amyloplasts from etiolated corn coleoptiles (HartmanBouillon and Benveniste, personal communication). In spinach leaves, the specific activity of the galactolipid synthesizing enzyme is extremely high (45 nmol mg-l protein min-l) in carefully prepared envelope preparations (Joyard and Douce, 1976c; Van Besouw and Wintermans, 1978) and exceeds the corresponding figures for total cell proteins by a factor of a t least 100. Moreover, the microsomal fraction, practically devoid of envelope membrane vesicles, is unable to synthesize the major galactolipids (Joyard and Douce, 1976b), (Fig. 41). All these results demonstrate definitively that in plant cells the plastid envelope, and probably the inner membrane (Liedvogel and Kleinig, 1976), specifically catalyses the final steps in galactolipid biosynthesis. It has also been demonstrated that two distinct enzymes responsible for the synthesis of monogalactosyldiacylglycerol and digalactosyldiacylglycerol are associated with the chloroplast envelope membranes (Joyard and Douce, 1976a). The first enzyme or UDP-galactose : diacylglycerol galactosyltransferase catalyses the synthesis of monogalactosyldiacylglycerol (Fig. 43). In this case, UDP-galactose is the galactosyl donor for galactosylation of endogenous diacylglycerol (Joyard and Douce, 1976a). The second enzyme is either a UDP-galactose: monogalactosyldiacylglycerol galactosyltransferase (Ferrari and Benson, 1961; Ongun and Mudd, 1968; Williams et a/., 1975; Heinz, 1977), (Fig. 43): MGDG

+ UDP-galactose

-

DGDG

+ UDP

or a galactolipid : galactolipid galactosyltransferase (Van Besouw and Wintermans, 1978), (Fig. 44): MGDG

+ MGDG -

+

DGDG

+ diacylglycerol

In fact we suggest that both reactions occur on the spinach chloroplast envelope (J. Joyard and R. Douce, unpublished data). The first galactosylation. enzyme has its pH optimum above pH 7.5. In contrast, the second galactosylation enzyme has its maximum activity around pH 6.5 (Joyard and Douce, 1976a). Triton X-100 (0.9% by volume) is a strong inhibitor of the second

Fig. 43. Mechanism which explains the galactosylation of diacylglycerol into MGDG and further galactosylation into DGDG. E6 and E, are two UDP-gal : galactosyltransferases (see Fig. 47) which are bound to the chloroplast envelope.

Fig. 44. Mechanism which explains the formation of DGDG and diacylglycerol by transgalactosidationfrom two molecules of MGDG.

STRUCTURE AND FUNCTION OF THE PLASTID ENVELOPE

89

galactosylation enzyme but is almost without effect on the first galactosylation enzyme (Heinz et a f . , 1978). Intact purified chloroplasts from spinach contain two acyltransferases. One is a soluble enzyme and catalyses the acylation of sn-glycerol3-phosphate with oleoyl-CoA and palmitoyl-CoA (Bertrams and Heinz, 1976; J. Joyard and R. Douce, 1977). The final product formed is lysophosphatidic acid: sn-glycerol3-phosphate

+ acyl-CoA

--t

lysophosphatidic acid

+ CoASH

This enzyme is probably loosely bound to the inner surface of the inner envelope membrane and becomes detached during the course of the envelope preparation. With palmitoyl-CoA, stearoyl-CoA or oleoyl-CoA in the incubation medium, sn-glycerol 3-phosphate is acylated in position 1 (Chuzel, Joyard and Douce, unpublished data). The second acylase in chloroplasts is an acyl CoA : monoacylglycerolphosphate acyltransferase which is firmly and specifically bound to the inner membrane of the envelope and forms phosphatidic acid (Joyard and Douce, 1977), (Fig. 45). Neither thylakoi'd membranes nor stroma contain this enzyme (Joyard and Douce, 1977): Lysophosphatidic acid

+ acyl CoA

-+

phosphatidic acid

+ CoASH.

Unfortunately fatty acid specificities of chloroplast glycerol phosphate acyltransferases have not been investigated, and it is not known whether the two transferases from chloroplasts can also use acyl-acyl carrier protein as acyl donors. However, in contrast to observations on Euglena (Renkonen and Bloch, 1969), Shine et al. (1976) have shown that in spinach chloroplasts the acyl-acyl carrier protein thioesters do not function as acyl donors. The envelope membranes contain a specific alkaline phosphatidate phosphatase which hydrolyses fairly rapidly phosphatidic acid formed from lysophosphatidic acid leading to the accumulation of diacylglycerol (Joyard and Douce, 1979), (Fig. 45): Phosphatidic acid + diacylglycerol

+ phosphate

In the presence of acyl-CoA and ~n-['~C]-glycerol3-phosphate, the envelope membranes slowly accumulate labelled diacylglycerol (Joyard and Douce, 1977).Thus the addition of UDP-galactose to the incubation medium induces a rapid decrease in the radioactivity incorporated into diacylglycerol which is accompanied simultaneously by rapid synthesis of monogalactosyldiacylglycerol (Fig. 46). These results together show that the chloroplast envelope may catalyse the transfer of fatty acid to monogalactosyldiacylglycerol establishing that chloroplasts, as a result of the enzymic complement of their envelope membranes, are autonomous for galactolipid synthesis. The enzymes involved operate in a multienzyme sequence. However, there remains the question of the synthesis of sn-glycerol 3-phosphate, UDP-galactose and unsaturated fatty acids particularly a-linolenic acid which is the major fatty acid encountered in chloroplast galactolipids.

Fig. 45. Mechanism which explains the acylation of sn-glycerol 3-phosphate in the envelope membranes. El : Acyl-CoA synthetase; E, : Acyl-CoA: sn-glycerol 3-phosphate acyltransferase; E, : Acyl-CoA: acyl-sn-glycerol 3-phosphate acyltransferase; E,: phosphatidic acid phosphatase. El, E, and E, are firmly bound to the chloroplast envelope. E, is probably loosely bound to the envelope and is released into the stroma during the isolation procedure.

STRUCTURE AND FUNCTION OF THE PLASTID ENVELOPE

91

Fig. 46. [14C]-glycerolfrom sn-glycerol 3-phosphate is incorporated into various lipids by isolated chloroplast envelopes (see Fig. 45). Labelled lipids are lyso PA: lysophosphatidic acid, PA: phosphatidic acid, M G : monoacylglycerol and DG : diacylglycerol. When UDP-galactose is added, MGDG is synthesized. After extraction, the different envelope lipids are separated by thin layer chromatography in the following solvent system: chloroform/methanol/water (65 :25 :4, v/v) and are revealed by autoradiography.

According to Konigs and Heinz (1974), Hippmann and Heinz (1976), Leech and Murphy (1976) and Heinz (1977), UDP-galactose and sn-glycerol 3-phosphate are synthesized in the cytoplasm of the cell. In marked contrast, numerous investigations have revealed that isolated chloroplasts from various leaves (Smirnov, 1960; Mudd and McManus, 1962; Appelqvist et a/., 1968; Hawke et a/., 1974; Nakamura and Yamada, 1975; Stumpf, 1976; Roughan et a/., 1976) possess the complete machinery for the biosynthesis of the hydrocarbon chains of fatty acids. The same is true for the chromoplasts from the corona of Narcissus pseudonarcissus (Kleinig and Liedvogel, 1978), the plastids from avocado mesocarp and cauliflower buds and proplastids from developing castor bean endosperm (Weaire and Kekwick, 1975; Zilkey and Canvin, 1972). The unique site of fatty acid synthesis in the post-germination castor bean endosperm is the proplastid (Vick and Beevers, 1977). Fatty acid synthetase

92

R. DOUCE AND J. JOYARD

is localized in the stroma phase (Brooks and Stumpf, 1965; Kannangara et al., 1973; Mazliak, 1973). This enzyme complex centred around a small protein named acyl carrier protein (ACP) converts acetyl-CoA and malonylCoA (which is derived from acetyl CoA) into palmityl-ACP. Palmityl-ACP is the preferred substrate for the elongation enzyme which converts it to stearyl-ACP (Jaworski et al., 1974). Stearyl-ACP is then desaturated to oleyl-ACP (Jacobson et al., 1974);the desaturase, which is probably localized in the stroma phase, requires two electrons for the activation of molecular oxygen and this pair of electrons is obtained from reduced ferredoxin loosely bound to the outer surface of the thylako'id membranes. Consequently, in the stroma, all of the chloroplast fatty acids (C16, C,,, C18:1 and C16:,) are attached to ACP during their de novo biosynthesis and during their elongation and desaturation to oleyl-ACP. There is now considerable evidence to suggest that the acyl-ACP is rapidly converted to acyl-CoA by a switching system involving acyl-ACP thioesterase and onelor several long chain acyl-CoA synthetases (Stumpf, 1976). The latter enzyme which catalyses the following reaction : Fatty acid

+ ATP + CoASH

--t

acyl-CoA

+ AMP + P

N

Pi

is specifically associated with the chloroplast envelope (Joyard and Douce, 1977; Roughan and Slack, 1977). According to Roughan and Slack (1977) the oleyl-CoA synthetase activity of the chloroplast envelope is twice that required to cope with maximum rates of oleic acid synthesis by chloroplasts in vitro. It is doubtful whether chloroplasts are capable of further desaturation of oleyl-CoA and /or oleyl-ACP although some blue-green algae, the progenitors of plastids, contain polyunsaturated fatty acids (e.g. a-linolenic acid). The mechanisms by which these high levels of unsaturated fatty acids are synthesized are key problems in understanding lipid metabolism in higher plants and algae. James (1963), Harris and James (1965) TrCmolikres and Mazliak (1974) and ChCriff et al. (1975) suggested that oleic acid in leaves is readily desaturated to linoleic and linolenic acids. In contrast, Jacobson et al. (1973) using broken spinach chloroplasts have demonstrated synthesis of linolenic acid by the addition of acetate to cis 7, 10, 13-hexadecatrienoicacid (C,6: 3). Several mechanisms have been proposed by which higher plants may attain high levels of unsaturated fatty acids in their galactolipid chloroplast membranes. Gurr et al. (1969) in studies on Chlorella and Roughan (1970) in pumpkin leaves have found that phosphatidylcholine may act as a carrier molecule involved in the desaturation. Recent work by Roughan and Slack (1977) suggests that oleic acid (C18:,) synthesized within chloroplasts is converted to an acyl-CoA form in the envelope and released into the cytosol where it is desaturated and incorporated into microsomal phosphatidylcholine. The latter may eventually be returned to the chloroplast and serve as

STRUCTURE AND FUNCTION OF THE PLASTID ENVELOPE

93

precursor for the linolenate of chloroplast monogalactosyldiacylglycerol. This shuttling of long chain fatty acids between cellular compartments is said to be essential for polyunsaturated fatty acid biosynthesis (TrCmoli6res and Mazliak, 1974). On the other hand, Safford and Nichols (1970), Siebertz (1976), Heinz (1977), Siebertz and Heinz (1977b) and Bolton and Harwood (1978) proposed that desaturation occurs in vivo after the formation of monogalactosyldiacylglycerol and that there is no need to derive C18:3 in monogalactosyldiacylglycerol from phosphatidylcholine. The third possible mechanism for obtaining high levels of unsaturated fatty acids in galactolipids is by galactosylation of specific diacylglycerol containing unsaturated fatty acids (Mudd et af.,1969; Williams et af., 1976a). In support of this mechanism, we have found recently that monogalactosyldiacylglycerol synthesis occurs in the envelope of spinach chloroplasts by galactosylation of a large endogenous pool of diacylglycerol containing polyunsaturated fatty acids (Joyard and Douce, 1976b). The large amount of diacylglycerol present in the envelope fraction most certainly represents the pool of diacylglycerol (from which the galactolipids are synthesized) postulated by Mudd et al. (1969) and Williams et al. (1975). Joyard and Douce (1979) have demonstrated that in the presence of snglycerol 3-phosphate and UDP-galactose in the incubation medium the labelled saturated fatty acids synthesized de n o w in the stroma of the chloroplast from 14C-acetate can be rapidly incorporated into diacylglycerol and monogalactosyldiacylglycerol of the envelope membranes. Since these lipids are rich in polyunsaturated fatty acids (excluding digalactosyl diacylglycerol which has a high proportion of a unique C,,:, C18:3combination absent in the other galactolipids) we propose that monogalactosyldiacylglycerol and/or diacylglycerol molecules, containing saturated fatty acids, specifically synthesized in the envelope constitute a direct substrate for one or several specific desaturases. We believe also that the desaturation which is a very slow process is confined to envelope membranes. It is likely that the quinones found in the envelope membranes (H. K. Lichtenthaler, J. Joyard and R. Douce, unpublished data) may play an important role in the mechanism of the desaturation. The question concerning the specific positioning of unsaturated fatty acids in galactolipids (Heinz, 1977) remains unanswered and further work has to be done in order to resolve this problem. It is certain that both the desaturase(s) and acyltransferases involved in the acylation of sn-glycerol3-phosphate may play an important role in fatty acyl chain location. The specific acyl galactosyl diacylglycerol-forming activity found in chloroplast envelopes (Heinz et al., 1978) and chromoplast envelopes (Liedvogel et al., 1976) which catalyses the transfer of acyl groups from the glycerol moiety of polar lipids to the galactose region of monogalactosyldiacylglycerol may also play an important role in intermixing all the polar lipid fatty acids in the envelope. It is well established that the main localization of the polar lipid synthesiz-

94

R. DOUCE AND J. JOYARD

ing enzymes in plant and animal cells is the endoplasmic reticulum and it is possible that no cellular membrane is independent of the endoplasmic reticulum in its biogenesis (MorrC, 1975). For example, in an elegant study, Gonzalez and Beevers (1976) have clearly demonstrated that enzymic constituents of the glyoxysomes, as well as membrane phospholipids, are derived almost directly from the endoplasmic reticulum. However, the data we present here show, for the first time, that the endoplasmic reticulum is not directly involved in the synthesis of an organelle's major structural lipids. It is clear that the envelope, and probably the inner membrane, contains the complete array of galactolipid biosynthetic machinery (Fig. 47). In these circumstances, during thylakoi'd biogenesis massive transport of galactolipids

Fig. 47. The chloroplast envelope plays a predominant role in the assembly of the three parts of the galactolipid molecules (galactose-glycerol-fatty acids). Saturated fatty acids are synthesized in the stroma by a multienzyme complex (fatty acid synthetase). Then the different steps occur on the envelope (probably at the level of the inner membrane). In these conditions, massive transport of galactolipids should occur very rapidly between envelope and thylakolds (dotted arrows).

should occur between the inner layer of the inner envelope membrane and the thylakoids. A possible solution to this problem may lie either with galactolipid exchange proteins localized in the stroma space or with membrane flow as discussed above. It will be interesting to know whether or not this lipid exchange occurs simultaneously with specific proteins and/or pigments such

STRUCTURE AND FUNCTION OF THE PLASTID ENVELOPE

95

as carotenoids since formation of chlorophyll and other constituents of the chloroplasts and differentiation of photosynthetic thylakoids occur simultaneously (Kirk and Tilney-Bassett, 1967). Details of the morphogenesis of photosynthetic lamellae during normal greening are lacking (Morrk, 1975). VII. ORIGIN OF THE CHLOROPLAST ENVELOPE MEMBRANES Schimper (1 885) and Mereschkowsky (1 905) were the first to postulate that the plastids in eukaryotic cells originated as free-living prokaryotes (see Stanier and Van Niel, 1962) which found shelter within primitive eukaryote cells and then became permanent symbiotic elements within them. Initially, this theory was poorly received and soon fell into disrepute. The general concept of chloroplast semi-autonomy has prompted several authors to reconsider the possibility that the Cyanophyceae (blue-green algae) are the progenitors of chloroplasts in the cells of higher plants and algae (Ris, 1961; Ris and Plaut, 1962). There is much evidence favouring the evolutionary relationships of chloroplasts (or plastids) and blue-green algae (Margulis, 1970; Raven, 1970; Taylor, 1970; Schnepf and Brown, 1971; Flavell, 1972; Cohen, 1973; Bogorad, 1975). Firstly, independent studies on the fine structure of several different blue-green algae (Hall and Claus, 1962; Lang and Whitton, 1973) and the chloroplasts of algae and higher plants (Gibbs, 1962b; Kirk and Tilney-Bassett, 1967) have shown that both are surrounded by a membrane (the plasma membrane or the inner membrane of the envelope) which serves to contain the photosynthetic apparatus (thylakoids). Secondly, chloroplast ribosomes from higher plants resemble the ribosomes of blue-green algae in their sedimentation behaviour and the size of their RNA components (Lyttleton, 1962; Jacobson et al., 1963; Boardman et al., 1966) and the fact that their ability to incorporate amino acids into proteins is inhibited by chloramphenicol and other inhibitors specific to prokaryotic cells. Thirdly, in blue-green algae, bacteria and chloroplasts, the DNA is histone free (see Raven, 1970) and bound to membranes(Herrmann et al., 1974). Fourthly, the polar lipid composition of the blue-green algae is identical to that of higher plant and algal chloroplasts (Benson, 1971; Nichols, 1973) and finally, many filamentous blue-green algae, in contrast to bacteria, have been shown to contain polyunsaturated fatty acids (Kenyon and Stanier, 1970) a characteristic chemical property of the eukaryotic chloroplasts. This elegant and attractive hypothesis is strengthened by the fact that several workers have reported the occurrence of endosymbiotic chloroplasts in the cells of numerous invertebrate species (Trench et al., 1969; Taylor, 1973). Furthermore, some blue-green algal symbionts (endocyanelles) are capable of living in association with non-photosynthetic hosts (Taylor, 1970). In this case, the cyanelle has taken on the characteristics of a functioning chloroplast and very often, its behaviour is regulated largely by the activity of

96

R. DOUCE AND 1. JOYARD

the host. This demonstrates the ease with which symbiotic blue-green algae may have been incorporated in the cytoplasm of their host’s cells and provided a wealth of information in the evolutionary potential of such a symbiotic system. Nevertheless, as indicated by Cavalier-Smith (1975), the strongest criticism of the symbiosis theory is that it fails to explain how the eukaryote condition itself (that is the “cell with nucleus”) evolved. Furthermore, this theory does not explain why nuclear genes control the numerous steps of plastid growth and development, as well as the synthesis of the enzymes and pigments involved in photosynthesis (Kirk and Tilney-Bassett, 1967). It is highly likely that in the course of evolution the initial symbiotic blue-green algae became adapted to a symbiotic existence and lost various functions which have been partly taken over by the host nucleus. The shifting of genes from one genome to another within a eukaryotic cell by a gene transfer mechanism seems a reasonable possibility (Bogorad, 1975). It is probably the case with the small subunit of ribulose- 1,5-bisphosphate carboxylase. Thus, Takabe et al. (1976), in a very attractive paper, have demonstrated that the structural make-up of the ribulose-l,5-bisphosphatecarboxylase from two blue-green algae (Plectonema boryanum and Anabaena variabilis) is of the plant type (the molecular weights of the two subunits are 5.4 x lo4 and 1.3 x lo4 respectively). The proliferation of the blue-green alga became synchronized with the cell division of its host and its limiting membrane, the prokaryotic plasma membrane, acquired the appropriate carriers (phosphate translocator, dicarboxylate translocator) necessary for its metabolism to be integrated with the metabolism of its host. It is possible that the specific carriers are the only entirely new components necessary in the evolutionary transition from the plasma membrane of blue-green algae to the inner membrane of the chloroplast envelope. Lastly, while the symbiosis theory readily explains the origin of the inner membrane of the plastid envelope (this membrane derives from the plasma membrane of the wall-free blue-green algal ancestor) it does not explain the origin of the outer membrane of the plastid envelope. Envelope membranes may have existed in their present form at least since upper Silurian times (Uzzell and Spolsky, 1974). It is possible that in the course of evolution this membrane may have arisen either from the inner membrane of the envelope or from the endoplasmic reticulum. It is also possible that the outer envelope membrane could represent a boundary membrane of an endocytotic vacuole formed by a plastid-free protoeukaryote (Schnepf, 1966). This membrane, in contrast to the outer mitochondria1 membrane does not resemble, either in its chemical composition or in its enzymic activities, the “microsomal membrane” of the eukaryotic cell (Joyard and Douce, 1976a). It is clear, but not absolutely certain, that the original function of this enveloping membrane was either to isolate the “invading” prokaryote or to protect the prokaryote

STRUCTURE AND FUNCTION OF THE PLASTID ENVELOPE

97

against the cell environment. In contrast, the present day outer envelope membrane is freely permeable, allowing all molecules either charged or uncharged up to a molecular weight of 10 000 to gain free access to the inner envelope membrane from the cytoplasm. The protein carriers if they exist are in this case the only entirely new component necessary in the evolutionary transition from the initial enveloping membrane to the outer membrane of the chloroplast envelope. It is clear that a great deal of research is needed before the notion that chloroplasts are derived from true symbiotic organelles can be asserted with any degree of confidence. Research done on primitive eukaryotic algae (red and brown algae) may help in understanding the origin of both envelope membranes typical of all plastids examined so far. The present lack of knowledge regarding the differences which may exist between the two chloroplast envelope membranes does not help the situation. VIII. SUMMARY The chloroplast envelope of higher plants is a permanent structure and consists of two morphologically and topologically distinct membranes separated by a region about 10-20 nm thick which appears electron translucent. The structure of both envelope membranes is consistent with the lipidglobular protein mosaic model of membrane structure proposed by Singer and Nicolson i.e. the polar and ionic heads of the lipid molecules together with all of the charged side chains of the amphipathic globular proteins (integral proteins) are on the surfaces of the membrane exposed to the aqueous phase. The inner membrane of chloroplast from plants having only the BensonCalvin pathway of photosynthesis is rarely completely smooth but possesses frequent discrete folds which invaginate into the plastid stroma or evaginate into the periplastidal space. In contrast, the inner envelope membrane of mesophyll cell chloroplasts from plants having the C , dicarboxylic acid pathway of photosynthesis possesses numerous folds which form an extensive system of anastomosing tubules called the peripheral reticulum. It is suggested that the sac-like infoldings of the inner membrane of the plastid envelope facilitate the passage of low molecular weight metabolites and also proteins into the developing organelle. Electron micrographs of the purified envelope fraction obtained from mature chloroplasts show relatively large vesicles or elongated profiles bordered by a single or a double membrane. In contrast with the outer mitochondrial and microsomal membranes, the chloroplast envelope membranes carefully prepared are devoid of NADH : cytochrome c oxidoreductase activity, b type cytochrome and phosphatidylethanolamine. This precludes a

98

R. DOUCE AND J. JOYARD

structural relationship between the envelope membranes and the other cell membranes. Analyses of the envelope membranes reveal that about 58% of the dry weight of the envelope membranes can be accounted for by lipid. The protein/lipid ratio of these membranes is 1 : 1.2 (1 : 0.5 for the thylakoids). In the case of plant mitochondria the protein/lipid ratio is 1 : 0.35 for the inner and I : 0.8 for the outer membrane. Thus the chloroplast envelope membranes of higher plants contain relatively less protein compared to lipid than do the thylakoid and mitochondria1 membranes. It is suggested that such a low protein/lipid ratio could explain the large scale rupture of the two membranes of the envelope during higher plant chloroplast isolation. It is also suggested that the envelope membranes, especially the outer membrane, insulate the electron carrier of the thylakoids from those of the other cell membrane systems (mitochondria, endoplasmic reticulum). Qualitatively, the polar lipids of both types of chloroplast membranes (thylakoid and envelope membranes) are identical but the proportions in which they are present are different. In the envelope membranes, monogalactosyldiacylglycerol and digalactosyldiacylglycerol are present with a ratio of 0.3-0.8 : 1 (2 : 1 for the thylakolds) and phosphatidylcholine and phosphatidylglycerol are present with a ratio of 3 : 1 (1 : 3 for the thylakoids). The significance of these differences has yet to be resolved although they almost certainly reflect the differences in function of each of the membranous systems. The fatty acids are more saturated in the envelope membranes than in the thylakolds. It is suggested that the increased degree of saturation in the envelope membranes may explain their higher thermostability when compared to the thylakofd membranes. The chloroplast envelope membranes which are unable to synthesize their own phospholipids, contain the complete array of galactolipid biosynthetic machinery. All the enzymes involved are probably located in the inner membrane of the envelope and operate in a multienzyme sequence. This indicates for the first time that the endoplasmic reticulum is not directly involved in the synthesis of an organelle’s major structural lipids. The chloroplast envelope membranes are devoid of chlorophylls but contain carotenoids. Qualitatively the carotenoids of both types of chloroplast membranes (thylakold and envelope membranes) are identical but the proportions in which they are present are different. /3-carotene accounts for a higher proportion of the thylakoldal carotenoid content when compared with the envelope fraction. On the other hand, violaxanthin accounts for a higher proportion of the envelope carotenoid content when compared with the thylakoids. These differences may be characterized by the xanthophyll to carotene ratio which is much higher ( 2 6 ) in the envelope fraction than in the thylakoids ( ~ 3 ) .

STRUCTURE AND FUNCTION OF THE PLASTID ENVELOPE

99

Envelope membranes prepared from dark-treated leaves or isolated intact chloroplasts have a violaxanthin content up to 3.5 times the lutein plus zeaxanthin content, whereas in chloroplast envelopes from illuminated leaves or intact chloroplasts this ratio is only 0.75. It is suggested that the lightinduced pigment changes in envelopes are caused by a violaxanthin exchange between envelope and thylakoids. The molecular weight distribution of the chloroplast envelope polypeptides is markedly different from that of the thylako'id polypeptides and is characterized by a series of high molecular weight polypeptides (above 70 000 daltons) and two predominant polypeptides of approximately 52 000 and 29 000 molecular weight. It is suggested that the synthesis of the thylakoids is not merely an extension of the synthesis of the inner membrane of the envelope but involves a major change in the types of proteins inserted into the growing membrane. Only one or two chloroplast envelope polypeptides are synthesized by chloroplast ribosomes. By inference, the remainder are synthesized on cytoplasmic ribosomes. It is suggested, however, that different results could perhaps be obtained if dividing plastids are used. The chloroplast envelope maintains the soluble enzymes involved in the Benson-Calvin cycle in close contact with the thylako'id network. The outer membrane is found to be unspecifically permeable to sucrose and other molecules either charged or uncharged up to a molecular weight of about 10 000. In contrast, the inner envelope membrane is impermeable to sucrose, protons and cations; is selectively permeable to a limited number of anions due to specific translocators and is freely permeable to uncharged low molecular weight molecules such as O, CO, and N0,H. So far, four specific translocators have been well characterized in the inner envelope membrane of the chloroplasts isolated from plants having only the Benson-Calvin pathway of photosynthesis (C, plants) : the phosphate translocator, the dicarboxylate translocator, the adenylate translocator and the sulphate translocator. It is suggested that the phosphate translocator plays an important role in the control of sucrose and starch synthesis whereas the dicarboxylate translocator plays a role in the indirect transfer of reducing equivalents from the chloroplast to the cytoplasm and vice-versa. A special class of proteins exists in the envelope outer membrane and/or at the point where both membranes are in contact which catalyse the unidirectional influx of all those proteins (coiled or uncoiled) made on cytoplasmic ribosomes but which are destined to function in the chloroplast. It is suggested that the proteins destined to cross the envelope membranes are synthesized with an additional sequence which contains the information necessary for specific binding to the envelope carrier protein. The inner membrane of the chloroplast envelope probably derives from the plasma membrane of a wall-free blue-green algal ancestor, whereas the outer

100

R. DOUCE AND J. JOYARD

membrane could represent a boundary membrane of an endocytotic vacuole formed by a plastid-free protoeukaryote. ACKNOWLEDGEMENTS The authors are indebted to Professor Andrew A. Benson for his kind interest. Our grateful thanks are owed to Drs D. A. Walker, U. Heber, E. Heinz and D. Harms-Siefermann for helpful discussions and comments. We would also like t o thank Dr. S. Loiseaux, Mrs. M. F. Joyard, M. Block and Mr. P. Noel for their help in preparing the manuscript. This article is largely a summary of recent work on the structure and function of the plastid envelope supported as a continuing programme by the “Commissariat h 1’Energie Atomique” and the “Centre National de la Recherche Scientifique”. REFERENCES Abreu, I. and Santos, A. (1977). J. Submicr. Cytol. 9, 239-246. Akerlund, H. E., Anderson, B., Westrin, H. and Albertsson, P. A. (1975). Abstract No. 1028, 10th meeting Fed. Eur. Biochem. SOC.,Paris. Amelunxen, F. and Gronau, G. Z. (1969). Z. Pflphys. 60, 156-168. Anderson, J. M. (1975a). Nature 253, 536537. Anderson, J. M. (1975b). Biochim. biophys. Actu 416, 191-235. Appelqvist, L. A., Stumpf, P. K. and Von Wettstein, D. (1968). PI. Physiof. 43, 163-187. Arnon, D. I. (1955). Science 122, 9-16. Amon, D. I., Allen, M. B. and Whatley, F. R. (1954). Nature 174, 394-396. Arntzen, C. J. and Briantais, J. M. (1975). In “Bioenergetics of Photosynthesis” (Govindjee, Ed.), pp. 51-113. Academic Press, New York, San Francisco and London. Asahi, T. (1964). Biochim. biophys. Actu 82, 58-66. Avrameas, S. and Ternynck, T. (1971). Zmrnunochemistry 8, 1175-1 179. Bahl, J. (1977). Plunta 136, 21-24. Bahl, J. and Moneger, R. (1975). C.r. hebd. Siunc. Acad. Sci., Puris281,1713-1716. Bahl, J., Francke, B. and Moneger, R. (1976). Pluntu 129, 193-201. Baldry, C. W., Cockburn, W. and Walker, D. A. (1968). Biochim. biophys. Actu 153, 476483. Bamberger, E. S. and Gibbs, M. (1965). PI. Physiol. 40,919-926. Bamberger, E. S.,Ehrlich, B. A. and Gibbs, M. (1975). PI. Physiol. 55, 1023-1030. Barber, J., Telfer, A. and Nicolson, J. (1974). Biochim. biophys. Actu 357, 161-165. Barton, R. (1966). Pfuntu 71, 314-325. Bassham, J. A. (1964). A. Rev. PI. Physiol. 15, 101-120. Bassham, J. A., Kirk, M. and Jensen, R. G. (1968). Biochim. biophys. Actu 153, 211-218. Bassham, J. A., Levine, G . and Forger, I11 J. (1974). Plant Sci. Left. 2, 15-21. Benson, A. A. and Calvin, M. (1947). Science 105, 648-649. Benson, A. A. (1971). In “Structure and Function of Chloroplasts” (M. Gibbs, Ed.), pp. 129-148. Springer Verlag, Berlin, Heidelberg and New York. Benson, A. A. and Jokela, A. T. (1976). In “Plant Biochemistry” 3rd Edition

STRUCTURE AND FUNCTION OF THE PLASTID ENVELOPE

101

(J. Bonner and J. E. Varner, Eds), pp. 65-89. Academic Press, New York, San Francisco and London. Bergelson, L. D. and Barsukov, L. I. (1977). Science 197, 224-230. Bertrams, M. and Heinz, E. (1976). Planta 132, 161-168. Bickel, H.and Schultz, G. (1976). Phytochemistry 15, 1253-1255. Bill’ K. Ya, Belobrodskaya, L. K. and Karpilov, Y. S. (1976). Dokl. Akad. Nauk. SSR 226, 1229-1231. Billecocq, A. (1974). Biochim. biophys. Acta 352,245-251. Billecocq, A. (1975). Ann. Immunol. (Inst. Pasteur) 126 C, 337-352. Billecocq, A., Douce, R. and Faure, M. (1972). C.r. hebd. Skanc. Acad. Sci. Paris 275, 1135-1137. Bird, I. F., Cornelius, M. J., Keys, A. J. and Whittingham, C. P. (1974). Phytochemistry 13. 59-64. Bisalputra, T. (1974). In “Algal Phsyiology and Biochemistry” (W. D. Stewart, Ed.), pp. 124-160. Blackwell, Oxford. Bisalputra, T. and Bailey, A. (1973). Protoplasma 76,443-454. Bisalputra, T., Downton, W. J. S. and Tregunna, E. B. (1969). Can. J. Bot. 47, 15-21. Blair, E. G. and Ellis, R. J. (1973). Biochim. biophys. Acta 319, 223-234. Blank, M. and Roughton, F. J. W. (1960). Trans Faraday SOC.56, 1832-1841. Blee, E. and Schantz, R. (1978). Plant Sci.Lett. 13, 247-255. Blobel, G.and Sabatini, D. (1971). In “Biomembranes” (L. A. Manson, Ed.) ~01.12, pp. 193-195. Plenum, New York. Boardman, N. K., Franki, R. I. B. and Wildman, S. G. (1966). J. mol. Biol. 17, 470-489. Boardman, N. K. (1977). I n “Photosynthesis I. Photosynthetic electron transport and photophosphorylation” Encyclopedia of Plant physiology New Series vol. No. 5 (A. Trebst and M. Avron, Eds), pp. 583-600. Springer Verlag, Berlin, Heidelberg, New York. Boasson, R., Laetsch, W. M. and Price, I. (1972). Am. J. Bot. 59, 217-223. Bogorad, L. (1975). Science 188, 891-898. Boisard, J., Marme, D. and Briggs, W. R. (1974). PI. Physiol. 54, 272-276. Bolton, P. and Harwood, J. L. (1978). Planta 138,223-228. Bomsel, J. L. and Pradet, A. (1968). Biochim. biophys. Acta 162, 230-242. Bomsel, J. L. and Sellami, A. (1975). I n “Proc. 3rd Int. Cong. of Photosynthesis”, vol. 2 (M. Avron, Ed.), pp. 1363-1367. Elsevier, Amsterdam. Bottomley, W., Spencer, D. and Whitfield, P. R. (1974). Arch. Biochem. Biophys. 164, 106-117. Bouck, G. B. (1965). J. Cell. Biol. 26, 523-537. Bourdu, R. (1975). I n “Photosynthbe et Production VegCtale” (C. Costes, Ed.), pp. 1-32. Gauthier-Villars, Paris. Bowes, G., Ogren, W. L. and Hageman, R. H. (1975). PI.Physiol. 56, 630-633. Bradbeer, J. W. (1973). I n “Biosynthesis and its Regulation in Plants” (B. V. Milborrow, Ed.), pp. 279-302. Pergamon, London. Bretscher, M. S. (1972). J . mol. Biol. 71, 523-528. Bretscher, M. S. (1973). Science 181, 622-629. Briggs, W. R. and Rice, H. V. (1972). A. Rev. PI. Physiol. 23, 293-334. Brooks, J. L. and Stumpf, P. K. (1965). Biochim. biophys. Acta 98, 213-216. Browning, G. and Saunders, P. F. (1977). Nature 265, 375-377. Bucke, C., Walker, D. A. and Baldry, C. W. (1966). Biochem. J. 101,636-641. Burton, A. C. (1970). In “Permeability and Function of Biological Membranes”

102

R. DOUCE AND 3. JOYARD

(L. Bolis, A. Katchalsky, R. D. Keynes, W. R. Loewenstein and B. A. Pethica, Eds), pp. 1-19. North-Holland, Amsterdam and London. Butler, R. D. and Simon, E. W. (1971). Adv. gerontol. Res. 3, 73-129. Buvat, R. (1958). Ann. Sci. Nut. Bot. 19, 203-261. Calvin, M. and Benson, A. A. (1948). Science 107, 476-480. Camefort, H. (1970). C.r. hebd. SPunc. Acud. Sci. Paris 271, 49-52. Campbell, P. N. and Blobel, G. (1976). FEBS Lett. 72, 215-226. Carde, J. P. (1973). J. Microscopie 9, 65-88. Catesson, A. M. (1970). J. Microscopie 15, 949-974. Cavalier-Smith, T. (1 975). Nature 256, 463468. Champigny, M. L. and Bismuth, A. (1977). Plant Cell Physiol. (Special issue), 365-375. Champigny, M. L. and Joyard, J. (1978). C.r. hebd. SPunc. Acad. Sci.Paris 286, 1791-1 794. Chap, H. J., Zwaal, R. F. A. and Van Deenen, L. L. M. (1977). Biochim. biophys. Actu 467, 146-164. Chapman, E. A., Bain, J. M. and Cove, D. W. (1975). Austr. J. PI. Physiol. 2, 207-223. Charriere-Ladreix, Y . (1977). Physiol. V&. 15, 619-640. Cheesbrough, T. M. and Moore; T. S. (1977). PI. Physiol. 59 (suppl), Abstract No. 439. D. 79. Chen, J.’L. and Wildman, S. G. (1970). Biochim. biophys. Actu 209, 207-219. Chen-She, S. H., Lewis, D. H. and Walker, D. A. (1975). New Phytol. 74, 383-392. Cheriff, A., Dubacq, J. P., Mache, R., Oursel, A. and Tremolitres, A. (1975). Phytochemistry 14, 703-706. Chevallier, D. (1974). These de Doctorat d’Etat, University of Grenoble., France. Chollet, R. (1977). Trends in Biochem. Sci.2, 155-159. Chollet, R. and Paolillo, D. J. (1972). Z. Pflphys. 68, 3 W . Chollet, R. and Ogren, W. L. (1975). Bot. Rev. 41, 137-179. Chua, N., Blobel, G., Siekevitz, P. and Palade, G. E. (1973). Proc. nutn. Acud. Sci. USA 70, 1554-1558. Cobb, A. H. (1975). Ph.D. Thesis, University of Lancaster, England. Cobb, A. H. and Wellburn, A. R. (1974). Pluntu 121,273-282. Cobb, A. H. and Wellburn, A. R. (1976). Pluntu 129, 127-131. Cockburn, W., Baldry, C. W. and Walker, D. A. (1967a). Biochim. biophys. Actu 131, 594-596. Cockburn, W., Baldry, C. W. and Walker, D. A. (1967b). Biochim. biophys. Actu 143, 606-613. Cockburn, W., Baldry, C. W. and Walker, D. A. (1967~).Biochim. biophys. Actu 143, 614-624. Coen, D. M., Bedbrook, J. R., Bogorad, L. and Rich, A. (1977). Proc. nutn. Acud. Sci. USA 74, 5487-5491. Cohen, S . S . (1973). Am. Scientist 61, 437-445. Cook, R. J. and Kendrick, R. E. (1976). Pluntu 131, 303-307. Coombs, J. and Baldry, C. W. (1971). PI. Physiol. 48, 379-381. Coombs, J. and Greenwood, A. D. (1976). In “The Intact Chloroplast” (J. Barber, Ed.), pp. 1-51. Elsevier, Amsterdam, New York and Oxford. Costes, C., Burghoffer, C., Carrayol, E., Ducet, G. and Diano, M. (1976). Plant Sci.Lett. 6, 253-260. Costes, C., Burghoffer, C., Joyard, J., Block M., and Douce, R. (1979). FEBS Lett. 103. 17-21. Cran, D: G. and Dyer, A. F. (1973). Protoplusma 76, 103-108. Cran, D. G. and Possingham, J. V. (1973). Ann. Bot. 37, 993-997.

STRUCTURE AND FUNCTION OF THE PLASTID ENVELOPE

103

Cran, D. G. and Possingham, J. V. (1974). Ann. Bot. 38, 843-847. Crotty, W. J. and Ledbetter, M. C. (1973). Science 182, 839-841. Dalling, M . J., Tolbert, N. E. and Hageman, R. H. (1972). Biochim. Biophys. Actu 283, 505-5 12. Damsz, B. and Mikulska, E. (1976). Biochem. Physiol. Pflunz. 169, 257-263. Danielli, J. F. and Davson, H. (1935). J. Cell Physiol. 5, 495-508. Dawson, R. M . C. (1973). Sub. Cell Biochem. 2, 69-89. Day, D. A. and Wiskich, J. T. (1975). Arch. Biochem. Biophys. 171, 117-123. De Duve, C. (1971). J. Cell. Biol. 50, 20 D-50 D. De Pierre, J. W. and Ernster, L. (1977). A . Rev. Biochem. 46, 201-262. Devaux, P. and McConnell, H. M. (1972). J . Am. Chem. SOC.94,4475-4481. Devor, K. A. and Mudd, J. B. (1971). J. Lipid Res. 12,403411. Dilley, R. A. and Vernon, L. P. (1965). Arch. Biochem. Biophys. 111, 365-375. Dobberstein, B., Blobel, G. and Chua, N. H. (1977). Proc. nutn. Acud. Sci. USA 74, 1082-1085 Dolzmann, P. and Ullrich, H. (1966). Z. Pj’phys. 15, 165-180. Donaldson, R. P., Tolbert, N. E. and Schnarrenberger, C. (1972). Arch. Biochem. Biophys. 152, 199-2 1 5. Douce, R. (1974). Science 183, 852-853. Douce, R. and Benson, A. A. (1974). Portug. Actu Biol. 14, 45-64. Douce, R., Guillot-Salomon, T., Lance, C. and Signol, M. (1968). Bull. SOC.Fr. Physiol. Vigkt. 14, 351-373. Douce, R., Christensen, E. L. and Bonner, W. D. (1972). Biochim. biophys. Actu275, 148-159. Douce, R., Mannella, C. A. and Bonner, W. D. (1973a) Biochim. biophys. Actu 292, 105-116. Douce, R., Holtz, R. B. and Benson, A. A. (1973b). J. Biol. Chem. 248,7215-7222. Douce, R. and Joyard, J. (1977). La Recherche 79, 527-537. Dubacq, J. P. and Kader, J. C. (1978). PI. Physiol. 61, 465-468. Dumas, C. (1974). Le Botuniste 56, 81-102. Eaglesham, A. R. and Ellis, R. J. (1974). Biochim. biophys. Actu 335, 3 9 W 7 . Echlin, P. (1964). Protoplusmu 58, 439457. Edwards, G. E. and Huber, S. C. (1978). In “Photosynthesis 1977” (D. 0. Hall, J. Coombs and T. W. Goodwin, Eds), pp. 95-106. The Biochemical Society, London and Colchester. Ellis, R. J. (1975). In “Membrane Biogenesis: Mitochondria, Chloroplasts and Bacteria” (A. Tzagoloff, Ed.), pp. 247-278. Plenum, New York and London. Ellis, R. J. (1976). In “The Intact Chloroplast” (J. Barber, Ed.), pp. 335-364. Elsevier, Amsterdam. Ellis, R. J. (1977). Biochim. biophys. Actu 463, 185-215. Ellis, R . J., Blair, G. E. and Hartley, M. R. (1973). Biochem. SOC.Symp. 38,137-162. Ellis, R. J., Highfield, P. E. and Silverthorne, J. (1978). In “Photosynthesis 77” (D. 0. Hall, J. Coombs and T. W. Goodwin, Eds), pp. 497-506. Plenum, New York and London. Esau, K. (1972). Z. Pflphys. 67, 244-254. Evans, A. (1975). In “Light and Plant Development” (H. Smith, Ed.), Butterworth, London. Evans, A. and Smith, H. (1976a). Nature 259, 323-325. Evans, A. and Smith, H. (1976b). Proc. nutn. Acud. Sci. USA 73, 138-142. Everson, R. G. and Slack, C. R. (1968). Phytochemistry 7, 581-584. Everson, R. G. (1970). Phytochemistry 9, 25-32. Falk, H. (1969). J . Cell Biol. 42, 582-587. Feierabend, J. and Schrader-Reichhardt, Ute. (1976). Plunta 129, 133-145.

104

R. DOUCE A N D J. JOYARD

Ferrari, R. A. and Benson, A. A. (1961). Arch. Biochem. Biophys. 93, 185-192. Fineran, B. A. (1975). Phytomorphology 25, 398-415. Flavell, R. (1972). Biochemical Genetics 6, 275-291. Fliege, R., Fliigge, U. I., Werdan, K. and Heldt, H. W. (1978). Biochim. biophys. Acta 502, 232-247. Fliigge, U. I. and Heldt, H. W. (1976). FEBS Lett. 68, 259-262. Flugge, U. I. and Heldt, H. W. (1977). FEBS Lett. 82, 29-33. Forde, J. and Steer, M. W. (1976). J. exp. Bot. 27, 1137-1141. Frederick, S. E. and Newcomb, E. H. (1969a). Science 163, 1353-1355. Frederick, S. E. and Newcomb, E. H. (1969b). J. Cell Biol. 43, 343-353. Frederick, S. E. and Newcomb, E. H. (1971). Plantu 96, 152-174. Frederick, S. E., Gruber, P. J. and Newcomb, E. H. (1975). Protoplasm 84, 129. Galatis, B., Apostolakos, P. and Hatzopoulou, C. (1974). Biol. Gallo-Hell. 5, 259-283. Gibbs, S. P. (1962a). J. Cell. Biol. 14, 433-444. Gibbs, S. P. (1962b). J. ultrastruct. Res. 7,418-435. Gibbs, S . P. (1970). Ann. N. Y. Acad. Sci. 175,454-473. Giles, K. L. and Sarafis, V. (1974). Nature 248, 512-513. Gimmler, H., Schafer, G., Kraminer, H. and Heber, U. (1974). Plantu 120, 47-61. Gimmler, H., Schafer, G. and Heber, U. (1975). In “3rd Int. Cong. Photosynthesis” Vol. 2 (M. Avron, Ed.), pp. 1381-1392. Elsevier, Amsterdam. Gonzalez, E. and Beevers, H. (1976). PI. Physiol. 57, 406409. Goodwin, T. W. (1958). Biochem. J. 68, 503-511. Goodwin, T. W. (1977). In “Lipids and Lipid Polymers in Higher Plants” (M. Tevini and H. K. Lichtenthaler, Eds), pp. 29-47. Springer-Verlag, Berlin, Heidelberg and New York. Granick, S. (1949). In “Photosynthesis in Plants” (J. Franck and W. E. Loomis, Eds), pp. 113-132. Iowa State College Press, Ames. Grant, B. R., Atkins, C. A. and Canvin, D. T. (1970). Planta 94, 60-72. Grant, B. R., Howard, R. J. and Gayler, K. R. (1976). Austr. J. PI. Physiol. 3, 639-65 1 . Grant, B. R., Gayler, K. R., Howard, R. J. and Wright, S. J. (1977). In “4th Int. Cong. Photosynthesis”. Abstract p. 137. UK Science Committee, London. Gray, J. C. and Kekwick, R. G. (1974). Eur. J. Biochem. 44, 491-500. Greenwood, A. D., Leech, R. M. and Williams, J. P. (1963). Biochim. biophys. Acta 78, 148-162. Gregory, R. P. F. (1977). “Biochemistry of Photosynthesis” 2nd Edition. John Wiley, London. Gronegress, J. (1974). J. Microsc. 19, 183-192. Gruber, P. J., Trelease, R. N., Becker, W. M. and Newcomb, E. H. (1970). Planta 93, 269-288. Guarnierri, M. and Johnson, R. M. (1970). Adv. Lipid Res. 8, 115-174. Guidotti, G. (1976). Trends Biochem. Sci. 1, 11-13. Guillermond, A. (1941). In “The Cytoplasm of the Plant Cell” Chronica Botan. Waltam, Massachussets. Guillot-Salomon, T. (1972). C.r. hebd. Sganc. Acad. Sci.Paris 274, 869-872. Gunning, B. E. S. and Steer, M. W. (1975). “Ultrastructure and the Biology of Plant Cells”. Edward Arnold, London. Gurr, M. I., Robinson, M. P. and James, A. T. (1969). Eur. J. Biochem. 9, 70-78. Hackenbrock, C. R. (1968). Proc. natn. Acad. Sci. USA 61, 598-605. Hackenbrock, C. R. and Miller, K. J. (1975). J. Cell Biol. 65, 615-630.

STRUCTURE AND FUNCTION OF THE PLASTID ENVELOPE

105

Hall, D . 0. (1972). Nature 235, 125-126. Hall, D. 0. (1976). In “The Intact Chloroplast” (J. Barber, Ed.), pp. 135-170. Elsevier, Amsterdam, New York and Oxford. Hall, W. T. and Claus, G. (1962). Protoplusma 54, 355-362. Hampp, R. and Ziegler, I. (1977). Planta 137, 309-312. Harris, R. V. and James, A. T. (1965). Biochim. biophys. Acta 106, 456-464. Harris, E. H., Preston, J. F. and Eisenstadt, J. M. (1973). Biochemistry 12, 12271234. Hashimoto, H. and Murakami, M. (1975). Plant CeN Physiol. 16, 895-902. Hauser, H., Levine, B. A. and Williams, R. J. P. (1976). Trends Biochem. Sci. 1, 278-281. Haussman, K. (1977). Naturwissenschaften 64, 95-96. Haverkate, F. and Van Deenen, L. L. M. (1964). Biochim. biophys. Acta 84, 106-108. Havir, E. A. and Gibbs, M. (1963). J . Biol. Chem. 238, 3183-3187. Hawke, J. C., Rumsby, M. G. and Leech, R. M. (1974). PI. Physiol. 53, 555-561. Heber, U. (1960). Z. Nuturforsch. 15b, 95-99 and 100-109. Heber, U. (1973). Biochim. biophys. Acta 305, 140-152. Heber, U. (1974). A . Rev. PI. Physiol. 25, 393-421. Heber, U. and Santarius, K. A. (1965). Biochim. biophys. Acta 109, 3 9 W 8 . Heber, U. and Santarius, K. A. (1970). Z. Nuturforsch. 25b, 718-728. Heber, U. and Krause, G. H. (1971). In “Photosynthesis and Photorespiration” (M. D. Hatch, C. B. Osmond and R. 0. Slatyer, Eds), pp. 218-225. WileyInterscience, New York and London. Heber, U. and Purczeld, P. (1978). I n “Photosynthesis 77” (D. 0. Hall, J. Coombs and T. W. Goodwin, Eds), pp. 107-118. The Biochemical Society, London. Heber, U., Hallier, U. W. and Hudson, M. A. (1967). Z. Nuturforsch. 22b, 1 2 W 1215. Heinrich, G. (1970). Protoplasma 69, 15-36. Heinz, E. (1977). I n “Lipids and Lipid Polymers in Higher Plants” (M. Tevini and H. K. Lichtenthaler, Eds), pp. 102-120. Springer-Verlag, Berlin, Heidelberg and New York. Heinz, E., Bertrams, M., Joyard, J. and Douce, R. (1978). Z. PfJphys. 87, 325-331. Heitz, E. (1936). Plantu 26, 134163. Heitz, E. (1957). Z. Naturforsch. 12b, 811-812. Heldt, H. W. (1969). FEBS Lett. 5, 11-14. Heldt, H. W. (1976a). I n “The Intact Chloroplast” (J. Barber, Ed.), pp. 215-234. Elsevier, Amsterdam, New York and Oxford. Heldt, H. W. (1976b). Horiz. Biochem. Biophys. 2, 199-229. Heldt, H. W. and Rapley, L. (1970). FEBS Lett. 10, 143-148. Heldt, H. W. and Sauer, F. (1971). Biochim. biophys. Acta 234, 83-91. Heldt, H. W., Sauer, F. and Rapley, L. (1972). In “Proc. 2nd Int. Cong. Photosynthesis” (G. Forti, M. Avron and A. Melandri, Eds), pp. 1345-1355. Junk, The Hague. Heldt, H. W., Werdan, K., Milovancev, M. and Geller, G. (1973). Biochim. biophys. Acta 314, 224-241. Heldt, H. W., Fliege, R., Lehner, K., Milovancev, M. and Werdan, K. (1975). I n “Proc. 3rd Int. Cong. Photosynthesis” (M. Avron, Ed.), pp. 1369-1379. Elsevier, Amsterdam. Heldt, H. W., Chon, C. J., Maronde, D., Herold, A., Stankovic, Z. S., Walker, D. A., Kraminer, A., Kirk, M. R. and Heber, U. (1977). PI. Physiol. 59, 1146-1 155.

106

R. DOUCE AND J. JOYARD

Herrmann, R. G., Kowallik, K. V. and Bohnert, H. J. (1974). Port. Acta Biol. 14,91-110. Herold, A., Lewis, D. H. and Walker, D. A. (1976). New Phytol. 76, 397-407. Herold, A. and Walker, D. A. (1977). Zn “Handbook on Transport across Membranes” (G. Giebisch, D. C. Tosteson and H. H. Ussing, Eds), SpringerVerlag, Heidelberg and New York. Highfield, P. E. and Ellis, R. J. (1978). Nature 271, 42G424. Hill, R. (1937). Nature 139, 881-882. Hiller, R. G. and Bassham, J. A. (1965). Biochim. biophys. Acta 109, 607-610. Hilliard, J. H. and West, S. H. (1971). Plunta, 99, 352-356. Hind, G., Nakatani, H. Y. and Izawa, S. (1974). Proc. natn. Acad. Sci. USA 71, 1484-1488. Hippmann, H. and Heinz, E. (1976).2. Pflphys. 79,408-418. Hitchcock, C. and Nichols, B. W. (1971). “Plant Lipid Biochemistry”. Academic Press, London and New York. Honda, S. I., Hongladarum, T. and Wildman, S. G. (1964). Zn “Primitive Motile Systems in Cell Biology” (R. D. Allen and N. Kamiya, Eds.), pp. 485-502. Academic Press, New York and London. Honda, S. I., Hongladarum, T. and Laties, G. G. (1 966). J . exp. Bot. 17,460-472. Hoober, J. K. (1976). I n “Genetics and Biogenesis of Chloroplasts and Mitochondria” (Th. Biicher, W. Neupert, W. Sebald and S. Wemer, Eds), pp. 87-94. North Holland, Amsterdam. Hoshina, G. and Nishida, K. (1970). Experienfia 26, 1275-1276. Huber, C. T. and Morrison, M. (1973). Biochemistry 12, 4274-4282. Huber, S. C. and Edwards, G. E. (1975a). Physiol. Plant. 35, 203-209. Huber, S. C. and Edwards, G. E. (1975b).PI. Physiof. 56, 324-331. Huber, S. C. and Edwards, G. E. (1976). Biochim. biophys. Acta 440, 675-687. Huber, S. C. and Edwards, G. E. (1977). Biochim. biophys. Acra 462, 583-602 and 603-6 1 2. Huber, D. J. and Newman, D. W. (1976). J. exp. Bot 27,49&511. Hurkman, W. J., MorrB, D. J. and Bracker, C. E. (1976). P . Physiof. 57, Abstract No. 53, p. 10. Israel, H. W. and Steward, F. C. (1967). Ann. Bot. 31, 1-18. Jacobson, A. B., Swift, H. and Bogorad, L. (1963). J. Cell Biol. 17, 557-570. Jacobsen, B. S., Kannangara, C. G. and Stumpf, P. K. (1973). Biochem. biophys. Res. Comm. 51, 487493. Jacobson, B. S., Jaworski, J. G. and Stumpf, P. K. (1974). PI. Physiol. 54,484-486. Jagendorf, A. T. (1975). In “Bioenergetics of Photosynthesis” (Govindjee, Ed.), pp. 413-492. Academic Press, New York. James, A. T. (1963). Biochim. biophys. Acta 70, 9-19. James, A. T. and Nichols, B. W. (1966). Nature 210, 372-375. Jaworski, J. G., Goldschmidt, E. E. and Stumpf, P. K. (1974). Arch. Biochem. Biophys. 163, 769-776. Jeffrey, S. W., Douce, R. and Benson, A. A. (1974). Proc. natn. Acad. Sci. USA 71, 807-8 10. Jennings, R. C. and Ohad, I. (1972). Arch. Biochem. Biophys. 153, 79-87. Jensen, R. G. and Bassham, J. A. (1966). Proc. natn. Acad. Sci. USA 56,1095-1101. Joy, K. W. and Ellis, R. J. (1975). Biochim. biophys. Acta 378, 143-151. Joyard, J. and Douce, R. (1975). FEBS Lett. 51, 335-340. Joyard, J. and Douce, R. (1976a). Physiol. Vkg. 14, 3 1 4 8 . Joyard, J. and Douce, R. (1976b). C.r. hebd. SPanc. Acad. Sci. Paris 282,1515-1518. Joyard, J. and Douce, R. (1976~).Biochim. biophys. Acta 424, 125-131.

STRUCTURE AND FUNCTION OF THE PLASTID ENVELOPE

107

Joyard, J. and Douce, R. (1977). Biochim. biophys. Acta 486, 273-285. Joyard, J. and Douce, R. (1979). Biochim. biophys. Acta, in press. Kader, J. C. (1972). These Doctorat d’Etat, University of Paris, France. Kader, J. C. (1977). In “Dynamic Aspects of Cell Surface Organization” (G. Poste and G. L. Nicolson, Eds.), pp. 127-204. North-Holland, Amsterdam. Kagan-Zur, V., Friedlander, M. and Lips, S. H. (1977). In “4th Int. Cong. on Photosynthesis” Abstract p. 185. UK Science Committee, London. Kanai, R. and Edwards, G. E. (1973a). PI. Physiol. 51, 1133-1137. Kanai, R.and Edwards, G. E. (1973b). PI. Physiol. 52, 484-490. Kandler, 0. (1954). Z. Naturforsch. 9b, 625-644. Kandler, 0. and Gibbs, M. (1959). Z. Naturforsch. 14b, 8-13. Kandler, O., Luginger, C. and Schilling, N. (1977). In “4th Int. Cong. on Photosynthesis” Abstract p. 186. UK Science Committee, London. Kannangara, C. G., Jacobson, B. S. and Stumpf, P. K. (1973). P1. Physiol. 52, 156-161. Kasemir, H., Bergfeld, R. and Mohr, H. (1975). Photochem. PhotobioI. 21,111-120. Kawashima, N. and Wildman, S. G. (1970). A. Rev. PI. Physiol. 21, 325-358. Kellems, R. E. and Butow, R. A. (1972). J. BioI. Chem. 247, 8043-8050. Kelly, G. J. and Gibbs, M. (1973). PI. Physiol. 52, 674-676. Kenyon, C. N. and Stanier, R. Y. (1970). Nature 227, 1164-1 166. Keyani, E. (1973). J. CeII BioI. 58, 480484. Kirk, J. T. 0. and They-Bassett, R. A. E. (1967). “The Plastids”. Freeman, San Francisco. Kirk, J. T. 0. (1972). SubcelI. Biochem. 1, 333-361. Kirk, P. R. and Leech, R. M. (1972). PI. Physiol. 50, 228-234. Kleinig, H. and Liedvogel, B. (1978). Eur. J. Biochem. 83, 499-505. Konigs, B. and Heinz, E. (1974). PIantu 118, 159-169. Kornberg, R. G. and McConnell, H. M. (1971). Biochemistry 10, 1111-1120. Kotyk, A. and Janacek, K. (1977). “Membrane Transport: An Interdisciplinary Approach” (Biomembranes 9), Plenum, New York and London. Krause, G. H. (1974). Biochim. biophys. Acta 333, 301-313. Krause, G. H. (1977). Biochim. biophys. Acta 460, 500-510. Krause, G. H. and Santarius, K. A. (1975). PIantu 127, 285-299. Krause, G. H. and Heber, U. (1976). In “The Intact Chloroplast” (J. Barber, Ed.), pp. 171-214. Elsevier, Amsterdam and Oxford. Krinsky, N. I. (1971). I n “Carotenofds” (0.Isler, Ed.), pp. 669-716. Birk-Hauser Verlag, Basel. Kursanov, A. L. and Paramonova, N. V. (1976). Fiziol. Rast. 23, 286-291. Laborde, J. A. and Spurr, A. R. (1973). Am. J. Bot. 60, 736-744. Laetsch, W. M. (1968). Am. J. Bot. 55, 875-883. Laetsch, W. M. (1974). A . Rev. PI. PhysioI. 25, 27-52. Laetsch, W. M. (1971). In “Photosynthesis and Photorespiration” (M. D. Hatch, C. B. Osmond and R. 0. Slatyer, Eds), pp. 323-349. Wiley, New York. Laetsch, W. M. and Price, J. (1969). Am. J. Bot. 56, 77-87. Laetsch, W. M. and Kortschak, H. P. (1972). PI. Physiol. 49, 1021-1023. Laetsch, W. M. and Sprey, B. (1977). In “4th Int. Cong. on Photosynthesis” Abstract p. 108. UK Science Committee, London. Laflkhe, D., BovC, C., Dupont, G . , Mouches, C., Astier, T., Gamier, M. and BovC, J. M. (1972). In “RNA Viruses/Ribosomes” vol. no. 27, pp. 43-71. North-Holland, Amsterdam. Lance, A. (1958). Ann. Sci. Nat. Bot. 19, 167-202. Lang, W. and Potrykus, I. (1971). 2. Pflphys. 65, 1-12.

108

R. DOUCE AND J. JOYARD

Lang, N. J. and Whitton, B. A. (1973). I n “The Biology of Blue-green Algae” (N. G. Carr and B. A. Whitton, Eds), pp. 66-79. Blackwell, Oxford. Larsson, C., Anderson, B. and Roos, G. (1977). Plant Sci. Left. 8, 291-298. Larsson, C., Collin, C. and Albertsson, P. A. (1971). Biochim. biophys. Acta 245, 425-438. Laulhkre, J. P. and Dorne, A. M. (1977). Plant Sci.Lett. 8, 251-256. Leech, R. M. (1964). Biochim. biophys. Acta 79, 637-639. Leech, R . M. and Murphy, D. J. (1976). In “The Intact Chloroplast” (J. Barber, Ed.), pp. 365401. Elsevier, Amsterdam. Lehner, K. and Heldt, H. W. (1978). Biochim. biophys. Acta 501, 531-544. Lerbs, S. and Wollgiehn, R. (1975). Biochem. Physiol. Pflanzen. 168, 167-174. Lichtenthaler, H. K. (1968). Endeavour 27, 144-149. Lichtenthaler, H. K. and Sprey, B. (1966). Z. Naturforsch. 21b, 690-697. Liedvogel, B. and Kleinig, H. (1976). Planta 129, 19-21. Liedvogel, B. and Kleinig, H. (1977). Planta 133, 249-253. Liedvogel, B., Sitte, P. and Falk, H. (1976). Cytobiology 12, 155-174. Lilley, R. McC., Holborow, K. and Walker, D. A. (1974). New Phytol. 73, 657662. Lilley, R. McC., Fitzgerald, M. P., Rienits, K. G. and Walker, D. A. (1975). New Phytol. 75, 1-10. Lilley, R. McC., Chon, C. J. A,, Mosbach, A. M. and Heldt, H. W. (1977). Biochim. biophys. Acta 460, 259-272. Loiseaux, S., Nurit, F. and Mache, R. (1974). “Atlas Micrographique de Cytologie Vc5gCtale”. Masson, Paris. Loomis, W. D. and Croteau, R . (1973). In “Recent advances in Phytochemistry” (V. C. Runeckles and T. J. Mabry, Eds), Academic Press, New York and London. Loveys, B. R. (1977). Physiol. Plant. 40, 6-10. Lucy, J. A. (1975). I n “Cell Membranes: Biochemistry, Cell Biology and Pathology” (G. Weissman and R. Clairborne, Eds), pp. 75-83. H.P. Publishing, New York. Lunney, C. A., Davis, G. J. and Jones, M. N. (1975).J. ultrastruct. Res. 50,293-296. Lyons, J. M. (1973). A. Rev.PI. Physiol. 24, 445466. Lyttleton, J. W. (1962). Exp. Cell Res. 26, 312-317. Lyttleton, J. W., Ballantine, J. E. M. and Forde, B. J. (1971). I n “Autonomy and Biogenesis of Mitochondria and Chloroplasts” (N. K. Boardman, A. W. Linnane and R. M. Smillie, Eds), pp. 447-452. North-Holland, Amsterdam and London. McCarty, R. E., Douce, R. and Benson, A. A. (1973). Biochim. biophys. Acta 316, 266-270. McLachlan, T. L. and Porter, H. K. (1959). Proc. R. SOC.Lond. 150,460-473. McLaren, J. S. and Barber, D. J. (1977). Planta 136, 147-151. McLean, J. D. (1956). Ph.D. Thesis, University of Sydney, Australia. Mache, R. and Loiseaux, S. (1973). J. Cell Sci. 12, 391-401. Mackender, R. 0. and Leech, R. M. (1970). Nature 228, 1347-1349. Mackender, R. 0. and Leech, R. M. (1972). I n “Proc. 2nd Int. Cong. on Photosynthesis” (G. Forti, M. Avron and A. Melandri, Eds), pp. 1431-1440. Junk, The Hague. Mackender, R. 0. (1971). Ph.D. Thesis, University of York, England. Mackender, R. 0. and Leech, R. M. (1974). PI. Physiol. 53,496-502. Mackenzie, J. M., Coleman, R. A., Briggs, W. R. and Pratt, L. H. (1975). Proc. natn. Acad. Sci. USA 72, 799-803.

STRUCTURE AND FUNCTION OF THE PLASTID ENVELOPE

109

Margulis, L. (1970). “Origin of Eukaryotic Cells”. Yale University Press, New Haven and London. Margulies, M. M. and Parenti, F. (1968). PI. Physiol. 43, 504-514. Margulies, M. M. and Michaels, A. (1974). J. Cell Biol. 60, 65-77. Margulies, M. M., Tiffany, H. L. and Michaels, A. (1975). Biochem. biophys. Res. Comm. 64, 735-739. Marmt, D., Mackenzie, J. M., Boisard, J. and Briggs, W. R. (1974). PI. Physiol. 54, 263-271. Marra, E., Doonan, S., Saccone, C. and Quagliariello, E. (1978). Eur. J. Biochem. 83,427-435. Marshall, M. 0 . and Kates, M. (1972). Biochim. biophys. Acta 260, 558-570. Marshall, M. 0. and Kates, M. (1974). Can. J. Biochem. 52, 469-482. Mason, T. L. and Schatz, G. (1973). J. Biol. Chem. 248, 1355-1360. Mathieu, Y. (1967). Photosynthetica 1, 57-63. Mayer, F. (1971). I n “Structure and Function of Chloroplasts” (M. Gibbs, Ed.), pp. 35-49. Springer-Verlag, Berlin, Heidelberg and New York. Mazliak, P., Stoll, U. and Abdelkader, A. B. (1968). Biochim. biophys. Acta 152, 414-417. Mazliak, P. (1973). A. Rev. PI. Physiol. 24, 287-310. Mazliak, P. (1977). I n “Lipids and Lipid Polymers in Higher Plants” (M. Tevini and H. K. Lichtenthaler, Eds), pp. 48-74. Springer-Verlag, Berlin, Heidelberg and New York. Mendiola-Morgenthaler, L. R. and Morgenthaler, J. J. (1974). FEBS Lett. 49, 152-155. Mendiola-Morgenthaler, L. R., Morgenthaler, J. J. and Price, C. A. (1976). FEBS Lett. 62, 96-100. Menke, W. (1960). Z. Naturforsch. 15b, 479482. Menke, W. (1961). 2. Naturforsch. 16b, 334-336. Menke, W. (1962). A. Rev. PI. Physiol. 13, 2 7 4 . Menke, W. (1964). Ber. Deut. Bot. Ges. 77, 340-354. Mereschkowsky, C. (1905). Biol. Zentr. 25, 593-604. Milborrow, B. V. (1976). In “Perspectives in Experimental Biology” 2, Botany (N. Sunderland, Ed.), pp. 111-124. Pergamon, Oxford. Miflin, B. J. (1974). Plunta 116, 187-196. Miginiac-Maslow, M. and Hoarau, A. (1977). Plant Sci. Lett. 9, 7-15. Miller, K. R. and Staehelin, L. A, (1973). Protoplasma 77, 55-78. Moor, H. and Muhlethaler, K. (1963). J. CeN Biol. 17, 609-628. Moore, F. D. and Shepard, D. C. (1977). Protoplasma 92, 167-175. Mohapatra, S. C. and Johnson, W. H. (1976). Trans. Am. Microsc. SOC.95,228-232. Montes, G. and Bradbeer, J. W. (1976). Plant Sci. Lett. 6, 35-41. Moreau, F. and Lance, C. (1972). Biochimie 54, 1335-1348. Moreau, F., Dupont, J. and Lance, C. (1974). Biochim. biophys. Acta 345, 294-304. Moreau, F. (1978). T h b e de Doctorat d’Etat, University of Paris, France. Morgenthaler, J. J., Price, C. A., Robinson, J. M. and Gibbs, M. (1974). PI. Physiol. 54,532-534. Morgenthaler, J. J., Mendiola-Morgenthaler, L. R. and Vasconcelos, A. C. (1975). Fed. Proc. 34, 510. Morgenthaler, J. J. and Mendiola-Morgenthaler, L. R. (1976). Arch. Biochem. Biophys. 172, 51-58. Morrt, D. J. (1975). A. Rev. Pl. Physiol. 26, 441-481. Morrt, D. J., Merritt, W. D. and Lembi, C. A. (1971). Protoplasma 73,4349.

110

R. DOUCE AND J. JOYAKD

Morre, D. J. and Mollenhauer, H. H. (1974). In “Dynamic Aspects of Plant Ultrastructure” (A. W. Robards, Ed.), pp. 84-1 37. McGraw-Hill, London. Morre, D. J. and Mollenhauer, H. H. (1976). I n “Transport in Plants 111” (C. R. Stocking and U. Heber, Eds.), pp. 288-344. Springer-Verlag, Berlin, Heidelberg and New York. Mourioux, G. and Douce, R. (1 978). C.r. hebd. Sianc. Acad. Sci.Paris 286,277-280. Mudd, J. B. and McManus, T. T. (1962). J. Biol. Chem. 237, 2057-2063. Mudd, J. B., Van Vliet, H. H. D. M. and Van Deenen, L. L. M. (1969). J. Lipid Res. 10, 623-630. Muhlethaler, K. and Frey-Wyssling, A. (1959). J. Biophys. Biochem. Cytol. 6, 507-5 12. Muhlethaler, K., Moor, H. and Szarkowski, J. W. (1965). Planta 67, 305-323. Muhlethaler, K. (1971). In “Structure and Function of Chloroplasts” (M. Gibbs, Ed.), pp. 7-34. Springer-Verlag, Berlin. Miihlethaler, K. (1977). In “Photosynthesis I. Photosynthetic Electron Transport and Photophosphorylation” Encyclopedia of Plant Physiology. New Series, vol. No. 5 (A. Trebst and M. Avron, Eds), pp. 503-521. Springer-Verlag, Berlin, Heidelberg and New York. Murakami, S. and Packer, L. (1971). Arch. Biochem. Biophys. 146, 337-347. Murakami, S. and Strotmann, H. (1978). Arch. Biochem. Biophys. 185, 3&38. Nakamura, Y. and Yamada, M. (1975). Plant Cell Physiol. 16, 139-149. Nakatani, H. Y. and Barber, J. (1977). Biochim. biophys. Acta 461, 510-512. Nedukha, E. M. (1976). Tsifol. Genet 10, 43-44. Neuburger, M., Joyard, J. and Douce, R. (1977). PI. Physiol. 59, 1178-1181. Neufeld, E. F. and Hall, C. W. (1964). Biochem. biophys. Res. Comm. 14, 503-508. Neumann, J. and Jagendorf, A. T. (1964). Arch. Biochem. Biophys. 107, 109-119. Neupert, W., Ludwig, G. D. and Pfaller, A. (1972). In “Biochemistry and Biophysics of Mitochondria1 Membranes” (G. F. Azzone, E. Carafoli, A. L. Lehninger, E. Quagliariello and N. Siliprandi, Eds), pp. 559-568. Academic Press, New York and London. Newcomb, E. H. and Frederick, S. E. (1971). In “Photosynthesis and Photorespiration” (M. D. Hatch, C. B. Osmond and R. 0. Slatyers, Eds), pp. 442-457. Wiley, New York. Nichols, B. W. (1973). In “The Biology of Blue-Green Algae” (N. G. Carr and B. A. Whitton, Eds), pp. 144161. Blackwell, Oxford. Nilsson, 0. S. and Dallner, G. (1977). Biochim. biophys. Acta 464, 453-458. Nishimura, M. and Akazawa, T. (1975). PI. Physiol. 55, 712-716. Nishimura, M., Graham, D. and Akazawa, T. (1976). PI. Physiol. 58, 309-314. Nobel, P. S. (1969). Biochim. biophys. Acta 172, 134-143. Nobel, P. S. and Cheung, Y. S. (1972). Nature 237, 207-208. Nobel, P. S. and Wang, C. (1970). Biochim. biophys. Acta 211, 79-87. Northcote, D. H. (1974). Phil. Trans. R. SOC.London 268, 119-128. Northcote, D. H. and Wooding, F. B. P. (1965). Proc. R. SOC.London 163,524-537. Nurit, F. (1979). These Doctorat d‘Etat, University of Grenoble, France. O’Brien, R. L. and Brierley, G. (1965). J. Biol. Chem. 240, 4527-4531. Ohad, I., Siekevitz, P. and Palade, G. E. (1967). J. Cell Biol. 35, 521-551. Ohad, I. (1975). In “Membrane Biogenesis: Mitochondria, Chloroplasts and Bacteria’’ (A. Tzagoloff, Ed.), pp. 279-350. Plenum, New York. Ohmori, M. and Yamada, M. (1974). Plant Cell Physiol. 15, 1129-1 132. Orci, L., Perrelet, A., Malaise-Lagae, F. and Vasalli, P. (1977). J. Cell Sci. 25, 157-1 61,

STRUCTURE AND FUNCTION OF THE PLASTID ENVELOPE

111

Osmond, C. B., Troughton, J. H.and Goodchild, D. J. (1969).Z. Pflphys. 61,171-178. Ongun, A. and Mudd, J. B. (1968). J . Biol. Chem. 243, 1558-1566. Pacini, E. and Cresti, M. (1976). J . Ultrastruct. Res. 57, 260-265. Pallas, E. J. and Mollenhauer, H. H. (1972). Am. J. Bot. 59, 504514. Paramonova, N. V. (1977). Dokl. Akad. Nauk. SSSR 235, 1225-1227. Park, R. B. and Sane, P. V. (1971). A . Rev. PI. Physiol. 22, 395430. Parsons, D. F., Bonner, W. D. and Verboon, J. G. (1965). Can. J. Bot. 43, 647-655. Peavey, D. G., Steup, M. and Gibbs, M. (1977). PI. Physiol. 60, 305-308. Pfaff, E., Klingenberg, M., Ritt, E. and Vogell, W. (1968). Eur. J. Biochem. 5, 222-232. Pfliiger, R. (I 973). Z. Naturforsch. 28c, 779-780. Pineau, B. and Douce, R. (1974). FEBS Lett. 47, 255-259. Platt-Aloia, K. A. and Thomson, W. W. (1977). New Phytol. 78, 599-605. Plaut, Z., Lendzian, K. and Bassham, J. A. (1977). PI. Physiol. 59, 184188. Poincelot, R. P. (1972). Biochim. biophys. Acta 258, 637-642. Poincelot, R. P. (1973). Arch. Biochem. Biophys. 159, 134-142. Poincelot, R. P. (1974). PI. Physiol. 54, 520-526. Poincelot, R. P. (1975). PI. Physiol. 55, 849-852. Poincelot, R. P. (1976). PI. Physiol. 58, 595-598. Poincelot, R. P. and Day, P. (1974). PI. Physiol. 54, 780-783. Poincelot, R. P. and Day, P. (1976). PI. Physiol. 57, 334-338. Portis, A. R. and Heldt, H. W. (1976). Biochim. biophys. Acta 449, 43-6. Portis, A. R., Chon, C. J., Mosbach, A. M. and Heldt, H. W. (1977). Biochim. biophys. Acta 461, 3 13-325. Possingham, J. V., Vesk, M. and Mercer, F. V. (1964). J. ultrastruct. Res. 11,68-83. Poyton, R. D. and Kavanagh, J. (1976). Proc. natn. Acad. Sci. USA 73, 3947-3951. Preiss, J. and Levi, C. (1978). In “Photosynthesis 77” (D. 0. Hall, J. Coombs and T. W. Goodwin, Eds), pp. 457-468. The Biochemical Society, London and Colchester. Pressmann, B. C. (1973). In “Inorganic Biochemistry” vol. No. 1 (G. L. Eichhorn, Ed.), pp. 203-226. Elsevier, Amsterdam. Price, C. A. (1974). Methods Enzymol. 31, 501-519. Priestley, D. A. (1977). Ph.D. Thesis, University of Leeds, England. Priestley, D. A. and Woolhouse, H. W. (1977). In “4th Int. Cong. on Photosynsynthesis” Abstract p. 304. UK Science Committee, London. Purczeld, P., Chon, C. J., Portis, A. R., Heldt, H. W. and Heber, U. (1978). Biochim. biophys. Acts 501, 488-498. Quail, P. H., Marme, D. and Schafer, E. (1973). Nature 245, 189-191. Quail, P. H., Gallagher, E. A. and Wellburn, A. R. (1976). Photochem. Photobiol. 24,495-498. Qureshi, A. A., Kim, M., Qureshi, N. and Porter, J. W. (1974). Arch. Biochem. Biophys. 162, 108-1 16. Racusen, D. and Poincelot, R. P. (1976). PI. Physiol. 57, 53-54. Radunz, A. (1969). Hoppe-Seylers Z. Physiol. Chem. 350, 41 1-417. Radunz, A. (1972). Z. Naturforsch. 27b, 822-826. Radunz, A. and Berzborn, R. (1970). Z. Naturforsch. 25b, 412419. Raison, J. K. and Chapman, E. A. (1976). Austr. J. PI. Physiol. 3, 291-299. Rathnam, C. K. M. and Das, V. S. R. (1974a). Z. Ppphys. 75, 360-364. Rathnam, C. K. M. and Das, V. S. R. (1974b). Can. J . Bot. 52,2599-2605. Raven, P. H. (1970). Science 169, 641-646. Renkonen, 0. and Bloch, K. (1969). J. Biol. Chem. 244,4899-4903.

112

R. DOUCE A N D J . JOYARD

Ridley, S. M. and Leech, R. M. (1970). Nature 227, 463465. Ris, H. (1961). Can. J. Genet. Cytol. 3, 95-104. Ris, H. and Plaut, W. (1962). J. Cell Biol. 13, 383-391. Ritenour, G. L., Joy, K. W., Bunning, J. and Hageman, R. H. (1967). PI. Physiol. 42,233-237. Robards, A. W. and Humpherson, P. C. (1967). Planta 76, 169-178. Robertson, J. D. (1959). Biochem. SOC.Syrnp. 16, 3-43. Robinson, S. P. and Wiskich, J. T. (1976). PI. Physiol. 58, 156-162. Robinson, S. P. and Wiskich, J. T. (1977a). FEBS Lett. 78, 203-206. Robinson, S. P. and Wiskich, J. T. (1977b). Biochim. biophys. Acta 461, 131-140. Robinson, S. P. and Wiskich, J. T. (1977~).PI. Physiol. 59, 422-427. Robinson, S . P. and Wiskich, J. T. (1977d). Arch. Biochem. Biophys. 184, 546-554. Rogers, L. J., Shah, S. P. J. and Goodwin, T. W. (1968).Photosynthetica2, 184-207. Rosado-Alberio, J. T., Weier, T. E. and Stocking, C. R. (1968). PI. Physiol. 43, 1325-1331. Rosenberg, T. and Wilbrandt, W. (1963). J. theoret. Biol. 5, 288-305. Rosinski, J. and Rosen, W. G. (1972), Q. Rev. Biol. 47, 160-191. Rosso, S. W. (1968). J. ultrastruct. Res. 25, 307-322. Rothman, J. E. and Kennedy, E. P. (1977a). J. mol. Biol. 110,603-618. Rothman, J. E. and Kennedy, E. P. (1977b). Proc. natn. Acad. Sci. USA 74, 1821-1 825. Rothman, J. E. and Lenard, J. (1977). Science 195, 743-753. Roughan, P. G. (1970). Biochem. J. 117, 1-8. Roughan, P. G., Slack, C. R. and Holland, R. (1976). Biochem. J. 158, 593-601. Roughan, P. G. and Slack, C. R. (1977). Biochem. J. 162, 457-459. Roux, S. J. and Yguerabide, J. (1973). Proc. natn. Acad. Sci. USA 70,762-764. Roy, H., Patterson, R. and Jagendorf, A. T. (1976). Arch. biophys. Biochem. 172, 64-73. Ruigrok, Th. J. C., Van Zaane, D., Wirtz, K. W. A. and Scherphof, G. L. (1972). Cytobiologie 5, 412-421. Sabnis, D. D., Gordon, M. and Galston, A. W. (1970). PI. Physiol. 45,25-32. Safford, R. and Nichols, B. W. (1970). Biochim. biophys. Acta 210, 57-64. Sager, R. (1972). “Cytoplasmic Genes and Organelles”. Academic Press, New York. Sane, P. V. (1977). In “Photosynthesis I. Photosynthetic Electron Transport and Photophosphorylation”. Encyclopedia of Plant Physiology. New Series Vol. No. 5 (A. Trebst and M. Avron, Eds), pp. 522-542. Springer-Verlag, Berlin, Heidelberg and New York. Santarius, K. A. and Stocking, C. R. (1969). 2.Naturforsch. 24b, 1170-1179. Santarius, K. A. and Heber, U. (1965). Biochim. biophys. Acta 102, 39-54. Scandella, C. J., Devaux, P. and McConnell, H. M. (1972). Proc. natn. Acad. Sci. USA 69,2056-2060. Scanu, A. M. (1972). Biochim. biophys. Acta 265, 471-508. Schafer, G., Heber, U. and Heldt, H. W. (1977). PI. Physiol. 60, 286-289. Schiff, J. and Hodson, R. C. (1973). A . Rev. PI. Physiol. 24, 381-414. Schimper, A. F. W. (1885). Jahrb. Wiss. Botun. 16, 1-247. Schnarrenberger, C. and Fock, H. (1976). In “Transport in Plants 111” (C. R. Stocking and U. Heber, Eds), pp. 185-234. Springer-Verlag, Berlin, Heidelberg and New York. Schnarrenberger, C. and Burkhard, C. (1977). Planta 134, 109-1 14. Schneider, M. M., Hampp, R. and Ziegler, H. (1977). PI. Physiol. 60, 518-520. Schnepf, E. (1966). In “Problem der Biologischen Reduplikation” (P.Sitte, Ed.), pp. 372-393. Springer-Verlag, Berlin, Heidelberg and New York.

STRUCTURE AND FUNCTION OF THE PLASTID ENVELOPE

113

Schnepf, E. (1969). Protoplasma 67, 195-203. Schnepf, E. and Brown, R. M. Jr. (1971). In “Origin and Continuity of Cell Organelles” (J. Reinert and H. Ursprung, Eds), pp. 299-322. Springer-Verlag, Berlin, Heidelberg and New York. Schnepf, E. and Klasova, S. (1972). Ber. Dt. Bot. Ges. 85, 249-259. Scholte, H . R. (1973). Biochim. biophys. Acra 330, 283-293. Schotz, F. and Diers, L. (1967). In “Le Chloroplaste” (C. Sironval, Ed.), pp. 21-29. Masson, Paris. Schotz, F. and Diers, L. (1975). Planta 124, 277-285. Schwenn, J. D. and Trebst, A. (1976). In “The Intact Chloroplast” (J. Barber, Ed.), pp. 315-334. Elsevier, Amsterdam. Schwertner, H. A. and Biale, J. B. (1973). J. Lipid Res. 14, 235-242. Sellami, A. (1976). Biochim biophys. Acta 423, 524539. Shine, W. E., Mancha, M. and Stumpf, P. K. (1976). Arch. Biochem. Biophys. 172, 110-116. Shumway, L. K. and Weier, T. (1967). Am. J. Bot. 54, 773-780. Siddell, S. G. and Ellis, R. G. (1975). Biochem. J. 146, 675-685. Siebertz, H. P. (1976). Ph.D. Thesis, University of Koln, Germany. Siebertz, H.P. and Heinz, E. (1977a). Hoppe-Seylers Z. Physiol. Chem. 358, 27-34. Siebertz, H. P. and Heinz, E. (1977b). 2.Naturforsch. 32c, 193-205. Siefermann-Harms, D. (1977). In “Lipids and Lipid Polymers in Higher Plants” (M. Tevini and H. K. Lichtenthaler, Eds), pp. 218-230. Springer-Verlag, Berlin, Heidelberg and New York. Siefermann-Harms, Do., Joyard, J. and Douce, R. (1978). PI. Physiol. 61, 530-533. Siekevitz, P. (1972). A. Rev.PI. Physiol. 34, 117-140. Silayeva, A. M. (1969). Dokl. Akad. Nauk. SSSR 184, 486-488. Silayeva, A. M. and Shiryayev, A. I. (1966). In “Electron Microscopy 2”. (R. Uyeda, Ed.), pp. 375-376. Maruzen, Tokyo. Singer, S. J. (1974). A. Rev. Biochem. 43, 805-833. Singer, S. J. and Nicolson, G. L. (1972). Science 175, 720-731. Sitte, P. (1962). Port. Acta Biol. Ser. A 6, 269-277. Sitte, P. (1974). Z. Pjlphys. 73, 243-265. Sitte, P. (1977). In “Lipids and Lipid Polymers in Higher Plants” (M. Tevini and H. K. Lichtenthaler, Eds), pp , 1-28. Springer-Verlag, Berlin, Heidelberg and New York. Sjolund, R. D. and Weier, T. E. (1971). Am. J. Bot58, 172-181. Smirnov, B. P. (1960). Biokhimija 25, 545-555. Smith, H. (1975). “Phytochrome and Photomorphogenesis”. McGraw-Hill, Maidenhead. Spencer, D. and Unt, H. (1965). Aust. J. Biol. Sci. 18, 197-210. Sprey, B. and Laetsch, W. M. (1975). Z . Pjlphys. 75, 38-52. Sprey, B. and Laetsch, W. M. (1976a). Z. Pfiphys. 78, 146-163. Sprey, B. and Laetsch, W. M. (1976b). Z. Pflphys. 78, 360-371. Sprey, B. and Laetsch, W. M. (1978). Z. PJJphys.87, 37-54. Staehelin, L. A. and Probine, M. C. (1970). In “Advances in Botanical Research” (H. W. Woolhouse, Ed.), Vol. 3, pp. 1-52. Academic Press, London and New York. Stanier, R. Y. and Van Niel, C. B. (1962). Archs. Mikrobiol. 42, 17-35. Stanier, R. Y.(1970). In “Organization and Control in Prokaryotic and Eukaryotic Cells” (M. P. Charles and B. C. Knight, Eds), pp. 1-38. Cambridge University Press, Cambridge.

114

R. DOUCE AND J. JOYARD

Stankovic, Z. C. and Walker, D. A. (1977). PI. Physiol. 59, 428-432. Stetler, D. A. and Laetsch, W. M. (1969). Am. J. Bot. 56, 260-270. Steup, M., Peavey, D. G. and Gibbs, M. (1976). Biochem. biophys. Res. Comm. 72, 1554-1561. Stocking, C. R. and Larson, S. (1969). Biochem. biophys. Res. Comm. 37,278-282. Stocking, C. R., Nivison, H. and Lin, C. H. (1977). I n “4th Int. Cong. on Photosynthesis’’ Abstract p. 366. UK Science Committee, London. Stoekenius, W. (1970) I n “Membranes of Mitochondria and Chloroplasts” (E. Racker, Ed.), pp. 53-90. Van Nostrand, New York. Stokes, D. M. and Walker, D. A. (1971). In “Photosynthesis and Photorespiration” (M. D. Hatch, C. B. Osmond and R. 0. Slatyer, Eds), pp. 226-231. Wiley, New York. Strotmann, H. and Berger, S. (1969). Biochem. biophys. Res. Comm. 35,20-26. Strotmann, H. and Murakami, S. (1976). I n “Transport in Plants I11 Intracellular Interactions and Transport Processes” (C. R. Stocking and U. Heber, Eds), pp. 398416. Springer-Verlag, Berlin, Heidelberg and New York. Strugger, S. (1951). Ber. Dt. Bot. Ges. 64, 69-83. Stumpf, P. K. (1976). In “Plant Biochemistry” 3rd edition (J. Bonner and J. E. Varner, Eds), pp. 427-461. Academic Press, New York. Surzycki, S. J., Goodenough, U. W., Levine, R. P. and Armstrong, J. J. (1970). I n “Control of Organelle Development” (P. L. Miller, Ed.), pp. 13-37. Cambridge University Press, Cambridge. Swader, J. A. and Stocking, C. R. (1971). PI. Physiol. 47, 189-191. Takabe, T., Nishimura, M. and Akazawa, T. (1976). Biochem. biophys. Res. Comm. 68, 537-544. Taylor, D. L. (1970). Int. Rev. Cytol. 27, 29-64. Taylor, D. L. (1973). A. Rev. Microbiol. 27, 171-187. Taylor, A. 0. and Craig, A. S. (1971). PI. Physiol. 47, 719-725. Tao, K. L. J. and Jagendorf, A. T. (1973). Biochim. biophys. Acta 324, 518-532. Thomson, W. W., Lewis, L. N. and Coggins, C. W. (1967). Cytologia 32, 117124. Thomson, W. W. (1974). Zn “Dynamic Aspects of Plant Ultrastructure” (A. W. Robards, Ed.), pp. 138-177. McGraw-Hill, London and New York. Tolbert, N. E. (1971). I n “Photosynthesis and Photorespiration” (M. D. Hatch, C. B. Osmond and R. 0. Slatyer, Eds), pp. 458471. Wiley, New York. TrCcul, A. (1858). Ann. Sci. Nat. Bot. 10, 127-163. Trench, R. K., Greene, R. W. and Bystrom, B. G. (1969). J. Cell Biol. 42, 404-417. Trench, R. K., Boyle, J. E. and Smith, D. C. (1973). Proc. R. SOC.Biol.184, 51-61. TrCmoliBres, A. and Mazliak, P. (1974). Plant Sci. Lett. 2, 193-201. Tuquet, C. (1972). C.r. hebd. SPanc. Acad. Sci. Paris 274, 1473-1476. Tuquet, C. and De Lubac, M. (1975). C.r. hebd. SPanc. Acad. Sci. Paris 281, 1313-131 6. Tuquet, C., Guillot-Salomon, T., De Lubac, M. and Signol, M. (1977). Plant Sci. Lett. 8, 59-64. Tzagoloff, A. (1971). J. Biol. Chem. 246, 3050-3056. Uzzel, T. and Spolsky, C. (1974). Am. Scientist 62, 334-343. Valanne, N. (1975). Can. J. Bot. 53, 1072-1077. Van Hummel, H. C. (1974). Z. Ppphys. 71, 228-241. Van Hummel, H. C. and Wintermans, J. F. G. M. (1975). In “Proc. 3rd Int. Cong. on Photosynthesis” vol. 2, pp. 2029-2037. Elsevier, Amsterdam, Oxford and New York.

STRUCTURE AND FUNCTION OF THE PLASTID ENVELOPE

115

Van Hummel, H. C., Huselbos, T. J. M. and Wintermans, J. F. G. M. (1975). Biochim. biophys. Actu 380, 219-226. Van Besouw, A. and Wintermans, J. F. G. M. (1978). Biochim. biophys. Actu. 529, 4453.

Vasconcelos, A. C. (1976). PI. Physiol. 58, 719-721. Vasconcelos, A. C., Mendiola-Morgenthaler, L. R., Floyd, G. L. and Salisbury, J. (1976). P1. Physiol. 529, 44-53. Vick, B. and Beevers, H. (1977). PI. Physiol. 59 (suppl.), Abstract No. 169, p. 31. Vidaver, G. A. (1966). J. theoret. Biol. 10, 301-306. Vigil, E. L. (1973). Sub. Cell Biochem. 2, 237-285. Vignais, P. V. (1976). Biochim. biophys. Acta 456, 1-38. Von Wettstein, D. (1958). Brookhuven Symp. Biol. 11, 138-157. Von Wettstein, D. (1959). Brookhuven Symp Biol. 16, 123-160. Walker, D. A. (1964). Biochem. J. 92, 22c-23c. Walker, D. A. (1965a). I n “Biochemistry of the Chloroplast” (T. W. Goodwin, Ed.), vol. 2, pp. 53-69. Academic Press, New York. Walker, D. A. (1965b). PI.Physio/. 40, 1157-1161. Walker, D. A. (1974). In “MIP Int. Rev. of Sci. Biochemistry Series 1, vol. 11; Plant Biochemistry” (D. H. Northcote, Vol. Ed.), pp. 1-49. Butterworths and University Park Press, London and Baltimore. Walker, D. A. (1971). Methods Enzymol. 23, 211-220. Walker, D. A. (1976a). I n “The Intact Chloroplast” (J. Barber, Ed.), pp. 235-278. Elsevier, Amsterdam and Oxford. Walker, D. A. (1976b). In “Transport in Plants I11 Intracellular Interactions and Transport Processes” (C. R. Stocking and U. Heber, Eds), pp. 85-136. Springer-Verlag, Berlin, Heidelberg and New York. Walker, D. A. and Herold, A. (1977). Plant CellPhysiol. (special issue), 295-310. Walker, D. A. and Hill, R. (1967). Biochim. biophys. Actu 131, 330-338. Walker, D. A. and Crofts, A. R. (1970). A . Rev. Biochem. 39, 389-428. Walker, D. A. and Lilley, R. Mc.C. (1974). PI. Physiol. 54, 950-952. Walker, D. A., Slabas, A. R. and Fitzgerald, M. P. (1976). Biochem. biophys. Actu 440, 147-162.

Wang, C. T. and Nobel, P. S. (1971). Biochim. biophys. Actu 241,200-212. Weaire, P. J. and Kekwick, R. G. 0. (1975). Biochem. J. 146, 425-437. Weier, T. E. (1938). Bot. Rev. 4, 497-530. Weier, T. E. (1961). Am. J. Bot. 48, 615-630. Weier, T. E. (1966). J. ultrustruct. Res. 15, 38-56. Weier, T. E. and Stocking, C. R. (1952). Bot. Rev. 18, 14-15. Weier, T. E. and Thomson, W. W. (1962). Am. J. Bot. 49, 807-820. Weier, T. E., Engelbrecht, A. H. P., Harrison, A. and Risley, E. B. (1965). J. ultrustruct. Res. 13, 92-1 11. Weier, T. E., Stocking, C. R. and Shumway, L. K. (1967). Brookhuven Symp. Biol. 19, 353-374.

Wellburn, A. R. and Hampp, R. (1976a). Pluntu 131, 95-96. Wellburn, A. R. and Hampp, R. (1976b). Biochem. J. 158, 231-233. Wellburn, A. R. and Wellburn, F. A. M. (1971). J. exp. Bot. 22,972-979. Werdan, K. and Heldt, H. W. (1972). I n “Proc. 2nd Int. Cong. on Photosynthesis” (G. Forti, M. Avron and A. Melandri, Eds), pp. 1337-1344. Junk, The Hague. Werdan, K., Heldt, H. W. and Geller, G. (1972). Biochim. biophys. Act0 283, 430-441.

Werdan, K., Heldt, H. W. and Milovancev, M. (1976). Biochim. biophys. Actu 396, 276292.

116

R. DOUCE AND J. JOYARD

Westrin, H., Albertsson, P. and Johansson, G. (1976). Biochim. biophys. Actu 436, 696-706. Whatley, J. M. (1974). New Phytol. 73, 1097-1110. Whatley, J. M. (1977). New Phytol. 78, 407-420. Whatley, J. M., Allen, M. B., Rosenberg, L. L. Capindale, J. B. and Arnon, D. I. (1956). Biochim. biophys. Acta 20, 462468. Wieler, A. (1936). Protoplusma 26, 295-311. Whitton, B. A., Carr, N. G. and Craig, I. W. (1971). Protoplusmu 22, 325-357. Wildman, S. G. (1967). In “Biochemistry of Chloroplasts” (T. W. Goodwin, Ed.), vol. 2, pp. 295-319. Academic Press, New York and London. Wildman, S. G. (1971). In “Autonomy and Biogenesis of Mitochondria and Chloroplasts” (N. K. Boardman, A. W.Linnane and R. M. Smillie, Eds), pp. 402-412. North-Holland, Amsterdam and New York. Wildman, S. G., Hongladarum, T. and Honda, S. I. (1962). Science 138, 434-436. Wildman, S. G., Hongladarum, T. and Honda, S. I. (1966). “Organelles in Living Plant Cells”, 16 mm Sound Film (Educational Film Sales and Rentals, University Extension, University of California, Berkeley). Wildman, S. G., Jope, C. and Atchison, B. A. (1974). PI. Physiol. 54,231-237. Williams, J. P., Watson, G. R., Khan, M. U. and Leung, S. P. K. (1975). PI. Physiol. 55,1038-1042. Williams, J. P., Watson, G. R. and Leung, S. P. K. (1976a).PI. Physiol. 57,179-184. Williams, J. P., Simpson, E. E. and Leung, S. P. K. (1976b). In “Lipids and Lipid Polymers in Higher Plants” Abstract book, pp. 32-33. Karlsruhe, Germany. Wirtz, K. W. A. (1974). Biochim. biophys. Actu 344, 95-1 17. Wojtczak, L. and Zaluska, H. (1969). Biochim. biophys. Actu 193, 64-72. Wooding, F. B. P. and Northcote, D. H. (1965). Am. J. Bot. 52, 526531. Woolhouse, H. W. (1978). Endeavour, N.S. 2, 35-46. Zilkey, B. F. and Canvin, D. T. (1972). Can. J. Bot. 50, 323-326.

Sodium as an Essential Micronutrient Element for Plants and its Possible Role in Metabolism

P. F. BROWNELL Department of Botany. James Cook University of North Queensland. Tounsville. Queensland. Australia

I. Introduction . . . . . A. Scope . . . . . B. Historical Perspective-I

.

. . . . . . . . . . . . . . . . . . . . . . . . 860 to 1977

. . . . . . . .

118 118 119

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

126 126 127 132

I11. Responses to Sodium at Low Concentrations . . . . . . . A. Lower Plants . . . . . . . . . . . . . . . B. Atriplex vesicaria Heward ex Benth . . . . . . . . C. Other Species having the C, Dicarboxylic Photosynthetic Pathway . . . . . . . . . . . . . . . . D. Response by a Species having Crassulacean Acid Metabolism . E. Discussion . . . . . . . . . . . . . . .

144 144 149

157 159 168

IV . Metabolic and Physiological Effects of Sodium at Low Concentrations General Strategies . . . . . . . . . . . . . A. B. Anabaena cylindrica . . . . . . . . . . . . . C. Other Lower Plants . . . . . . . . . . . . . D. C, and CAM Plants . . . . . . . . . . . . .

172 172 172 184 186

I1

Low-sodium Culture Conditions A. Definition . . . . B. Determination of Sodium C. Culture Techniques . .

. . . . . . . . . . .

118

P. F. BROWNELL

V. Tentative Schemes for the Role of Sodium in C4 and CAM Plants and in Blue-green Algae . . . . . . . . . . . . . . . 207 A. C4 and CAM Plants. . . . . . . . . . . . . 207 B. Anabaena cylindrica . . . . . . . . . . . . . 212 VI. Summary and Conclusions .

. . . . . . . . . . . .

21 5

Acknowledgements . . . . . . . . . . . . . . . 219 References . . . . . . . . . . . . . . . . . 219

I. INTRODUCTION A. SCOPE

This article is concerned almost entirely with the effects of sodium on plants including microorganisms for which it can be considered essential according to the criteria of Arnon and Stout (1939). These plants include those angiosperms having the C, dicarboxylic photosynthetic pathway (Brownell and Crossland, 1972) and at least some members of the Cyanophyta (Allen and Arnon, 1955; Kratz and Myers, 1955). Concentrations of sodium of about 0.1 mol m-3 (2.3 ppm) and below, support optimum growth in these species and its effects at these concentrations on growth and metabolism are discussed. References are also made to certain bacteria and fungi, largely of marine origin, for which sodium has been shown to be essential. Sodium is generally required at relatively high concentrations by these organisms but reference to them is considered to be justified due to their specific requirement for sodium. A recent review by Jennings (1976) has dealt with the effects of sodium at considerably higher concentrations i.e. 50 mol m-3 (1 150 ppm) and above on plants having the C, as well as on plants with the C, pathway. Although sodium at these concentrations may stimulate growth in certain species and affects the plants in other ways such as by increasing their succulence (Jennings, 1968), it is apparently not essential for them. Plants with C, photosynthesis have been found to develop and grow normally with an almost complete absence of sodium viz. less than 0.256 mmol m-3 (0.006 ppm) with tomatoes (Woolley, 1957), less than 0.17 mmol m-3 (0-004ppm) for cotton (Pleunneke and Joham, 1972) and less than 0-07mmol m-3 (0.0016 ppm) for more than 15 species with C, photosynthesis (Brownell, 1968). It is difficult to discuss sodium as a nutrient without making some reference to its effects at high concentration on C , plants but the main emphasis in this article is on the consideration of the role of sodium as a nutrient in plants for which it is essential. Reviews dealing with effects of sodium at high concentrations on plants include those by Wybenga (1957), Hewitt (1963), Rains (1972), Marschner

SODIUM AND PLANT METABOLISM

119

(1971), Waisel (1972), Marschner (1975), Flowers (1975) and Flowers et al. (1977). Comprehensive descriptions of the effects of sodium salts on Salicornia herbacea and Aster tripolium were made by van Eijk (1939) and on sugar beet by Tullin (1954) and reviews have been published by Lehr (1953, 1957). This article comprises a brief history of the understanding of sodium as a nutrient element for plants. The methods for obtaining sodium-free conditions are discussed in the second section and in a further section, the evidence for sodium being an essential element for various plants is given and its known physiological effects are described. Finally, tentative schemes are proposed to explain its possible involvement in plant metabolism. B. HISTORICAL PERSPECTIVE - 1860 TO 1977 Between 1860 and 1890, it was indicated from experimental work by Knop, Nobbe and Pfeffer that ten elements were absolutely essential for plant life. They were carbon, hydrogen, oxygen, phosphorus, potassium, nitrogen, magnesium, calcium, sulphur and iron. The other elements found in plants were considered of little or no physiological importance and their presence was thought to be due to them entering the cell passively. If large amounts were present, it was attributed to a specific property of the plant to accumulate them. It was not until early in the twentieth century that the essentiality of further elements was demonstrated. Until then it was generally assumed that the essential elements were all needed in relatively large amounts and little attempt had been made to eliminate impurities from salts used in nutrient solutions. Due to the general abundance of sodium, it was thought that if it were essential for plants it would be needed in relatively large amounts. There was a tendency to regard sodium which resembled potassium in its chemical properties as possibly having a similar role to that of potassium in the nutrition of plants. Birner and Lucanus (1866) had shown potassium to be essential for oats in water culture but Nobbe, Schroeder and Erdmann (1871) (cited by Hartt, 1934) concluded from their work on potassium nutrition of plants that sodium was not needed by plants. Hellriegel and Willfarth (1898) (cited by Dorph-Petersen and Steenbjerg, 1950) from their observations of growth increases under conditions of potassium deficiency suggested that sodium should be regarded as a beneficial element. During the next decade, Osterhout (1909, 1912) found sodium to be necessary for certain marine algae. The replacement of sodium by ammonium, calcium, magnesium, potassium, barium, strontium, caesium, rubidium or lithium was “distinctly injurious” to the plants. Much of the research on sodium as a possible nutrient element for higher

120

P. F. BROWNELL

plants was conducted on a field scale from the late nineteenth century until the mid twentieth century. In many instances, growth responses have been reported but the interpretation of the results of these experiments is often difficult. When massive amounts of sodium salts were applied, the responses could have been due to indirect effects of the treatment and not necessarily to the direct involvement of sodium in the metabolism of the plants. In experiments with complicated media such as sand, vermiculite, ion exchangers or soils, the treatment application could affect differentially a wide array of conditions including the availability of other nutrient elements, the pH, or the growth of microorganisms which could, in turn, affect the growth of plants indirectly. In experiments conducted before the discoveries of the requirements for micronutrients, increases in yield could have been brought about by the supply of a limiting micronutrient associated with the sodium treatment salt as an impurity. The designs of some experiments can be criticized for the lack of control treatments. When only one salt of sodium was supplied, the effects observed could have been due to the anion of the salt and not necessarily to the sodium. In other experiments, a salt of sodium e.g. sodium nitrate has been substituted for the salt of the same anion and a different cation e.g. calcium nitrate. Any differences between such treatments could be ascribed to the associated change in the ratio of monovalent to divalent cations in the medium and not specifically to the sodium. However, despite these possible limitations in experimental procedures, it appears that sodium at high concentrations (10 mol m-3 and above) increases growth in many instances. Perhaps the most puzzling feature of reported responses is their apparent inconsistency. Some workers have obtained no response to sodium even when potassium was limiting. Hartt (1934) found that sodium at 2-25 mol m-3 (51.75 ppm) was unable to substitute for potassium in the nutrition of sugarcane. No special measures were taken to grow the plants in a low-sodium medium in this work. Similarly, Montasir et al. (1966) obtained no growth response to sodium in sesame when potassium was limiting. There are many cases of responses to sodium when potassium is limiting. Early publications from the Rhode Island Agricultural Research Station by Hartwell and Pember (1908) and Hartwell and Damon (1919) showed that sodium could act as a partial substitute for potassium for the growth of wheat in water culture. In barley, sodium has been shown to replace potassium to some extent (Mullison and Mullison, 1942; Lehr and Wybenga, 1958; Montasir et al. 1966). Many other species have also responded to sodium in a deficiency of potassium including cotton (Joham, 1955; Joham and Amin, 1965) and oats (Truog et al., 1953). Ulrich and Ohki (1956) observed growth responses with sugar beet only under low potassium conditions.

SODIUM AND PLANT METABOLISM

121

It has also been shown that sodium a t 10 mol m-3 (230 ppm) has restored the translocation of carbohydrates in calcium-deficient cotton plants to the levels obtained with non-limiting levels of calcium present (Joham and Gossett, 1974; Joham and Johanson, 1973; Whitenberg and Joham, 1974). Partial substitution of sodium for calcium in the growth of excised cotton roots was also demonstrated (Johanson and Joham, 1971). There have been numerous reports of growth responses to sodium at high concentrations (10 mol m-3 and above) even when other nutrient elements are present at optimal levels. These responses have been obtained generally (but not always) in species of Chenopodiaceae, including sugar beet, Swiss Chard, spinach, Atriplex hastata and mangolds (Harmer and Benne, 1945; Harmer et al., 1953; Kushizaki and Yasuda, 1964; El-Sheikh er al., 1967; El-Sheikh and Ulrich, 1970; Montasir et al., 1966), Salicornia herbacea and the Composite species, Aster tripolium (Van Eijk, 1939; Baumeister and Schmidt, 1962). Connor (1969) showed that sodium chloride at 256mol m-3 (5897 ppm) sodium added to a complete culture solution gave a 69% increase in yield, whereas all levels of potassium chloride or calcium chloride suppressed the yield of Avicennia marina (grey mangrove) by 12% and 78 %, respectively. No visual signs of deficiency were noticeable in the plants not receiving sodium. Hewitt (1963) concluded that the response to sodium is dependent upon the species and the composition of the culture solution. This is shown clearly in the experiments of Montasir et a/. (1966) who examined a number of plants for the effect of sodium with different levels of potassium in the culture solutions. Atriplex hastata, alfalfa, barley, spinach and flax all responded by increased dry weight production on receiving sodium chloride when potassium was limiting, but sesame did not respond to sodium at any levels of potassium. Alfalfa and flax showed medium responses with adequate potassium levels while Atriplex and spinach showed large responses with ample potassium present. It was suggested that sodium has a specific function in the nutrition of Atriplex and spinach as it increased growth even at the highest level of potassium while in sesame it was detrimental. The calcium concentration on a dry weight basis decreased with decreasing sodium chloride in the solution. The treatments were applied in such a manner that the sums of concentrations (on an equivalence basis) of sodium and potassium were the same in all treatments. Harmer and Benne (1945) and Harmer et al. (1 953) surveyed investigations on the effects of sodium on plant growth and placed crops into one of two classes each of which was subdivided into two tentative groups with regard to their response to sodium: A. Benefited by sodium in a deficiency of potassium Group 1 . None to slight benefit e.g. buckwheat, corn, lettuce.

122

P. F. BROWNELL

Group 2. Slight to medium benefit e.g. asparagus, barley, flax, wheat. B. Benefited by sodium in sufficiency of potassium Group 3. Slight to medium benefit e.g. cabbage, mustard, radish, rape. Group 4. Large benefit e.g. celery, mangold, sugar beet, turnip. Lehr (1953) in a similar scheme classified crops according to the ability of sodium to replace their potassium requirements and to the independent effect of sodium on its yield. The discussion to this point has been concerned with the effects of sodium at high concentrations in increasing dry weight yields. Sodium nutrition also appears to affect the plant qualitatively. There have been many observations of increased sugar concentration in sugar beet fertilized with sodium (ElSheikh and Ulrich, 1970; Palladina and Bershtein, 1974). The quality of fibre crops including flax and cotton has been influenced by sodium treatment. Flax has been found to have optimal fibre quality when 6 % of the exchangeable ions were sodium (Lehr and Wybenga, 1955; Wybenga and Treggi, 1958; Treggi, 1961). Szymanek (1952) found the straw yield was unaffected but that there were more, stronger fibres in flax receiving nitrogen as sodium nitrate than as calcium nitrate. Moscolov and Aleksandrovskaya (1962) obtained increased yields of flax and fat content of seeds with increased synthesis of sucrose in the leaves in response to supplying sodium chloride instead of potassium. With low substrate levels of calcium or potassium, both the dry weight production and the bolls : leaves plus stem ratio of cotton were reduced. Sodium at 10mol m-3 (230ppm) partly restored these values in calciumdeficient plants and fully restored them in potassium-deficient plants (Joham, 1955). Little is known about the way in which sodium is involved in increasing yields and bringing about qualitative changes in response to high concentrations of sodium in certain plants. Sodium may act as a monovalent cation activator of enzymes. It appears that although K+ is specifically required by some enzymes and appears to be the most effective activator of the monovalent cations in many other enzymes there are no known instances where a higher plant enzyme is specifically activated by Na+ (Evans and Sorger, 1966). Evans and Sorger (1966) considered that the effectiveness of K+, NH,+, Rb+, Na+ and Li+ as activators of many enzymes was consistent with their effectiveness to stimulate growth. In many of the 46 enzymes they listed from animals, higher plants and microorganisms which require monovalent ions for maximal activity their effectiveness as activators decreases in the following order: K+, Rb+, NH,+, Na+, Li+. Na+, for example, is only about 20% as effective as Rb+ as a monovalent cation activator of pyruvate kinase and K+

SODIUM AND PLANT METABOLISM

123

and both NH,+ and Rb+ were more effective than Na+ (Miller and Evans, 1957; McCollum et al., 1958). Particulate starch synthetase has been shown to have an absolute requirement for K+ but Rb+, Cs+ and NH,+ were 80% and Li+ 21 % and Naf only 8 % as effective as K+ as monovalent cation activators (Nitsos and Evans, 1969). Hawker et al. (1974) studied the effects of sodium and potassium on starch synthesis in leaves of Spinacia oleracea and Beta vulgaris. Growth was decreased with low levels of potassium but the addition of sodium increased the growth to that of plants with adequate potassium. By comparison Phaseolus vulgaris grew poorly when part of the potassium was replaced by sodium. They found that whereas sugar beet plants grown with adequate levels of potassium contained nine times more starch than leaves of plants grown on high sodium low potassium culture solutions, the starch content in spinach was not affected. Potassium, but not sodium stimulated, by about loo%, the activity of starch granule bound ADP-glucose starch synthetase from sugar beet, bean and Atriplex nummularia but not from spinach leaves. Neither potassium nor sodium caused marked stimulation in other enzymes associated with carbohydrate metabolism. Oji and Izawa (1969) obtained evidence that potassium ions stimulated the formation of nitrate reductase and nitrite reductase in rice seedlings. Sodium could partially substitute for potassium in the enzyme formation. Potassium and the other monovalent cations are required to be present in very high concentrations as activators of enzymes compared to those required for other mineral cations (with the exception of magnesium) which act as physiological activators of plant enzymes. For maximal activity often concentrations of 50 to 100 mol m-3are required. Epstein (1972) suggests that such loose binding between potassium and the enzymes activated by it may not have been a limiting factor in the early marine phases of evolution when high concentrations of potassium were readily available in the sea water. The other monovalent cations are also required at high concentrations for their optimum activity. Evans and Wildes (1971) suggest the following mechanisms to explain the activation by univalent cations of certain enzymes for which they function as activators: (i) subunit structure of some enzymes may depend on univalent cations; (ii) capacity of some enzymes to bind a particular coenzyme is ion dependent; (iii) univalent cations may act as alloteric effectors; (iv) univalent cations may influence the conformation of some enzymes without causing gross changes in physical structure; (v) univalent ions may stabilize reaction intermediates during enzyme catalysis by mechanisms not yet fully understood. The finding that sodium was an essential micronutrient element for certain, but not all plants, came as a surprise. Here was a ubiquitous element, closely resembling the macronutrient element potasssium in many of its chemical

124

P. F. BROWNELL

properties being needed in only very small quantities for the plants requiring it. The first suggestion that sodium might be needed as a micronutrient element was probably due to Pfeffer (1899) who stated:

“. . . it is not always easy to decide with certainty whether an element is essential or not. Moreover, if but a little substance is required, the presence of the merest trace as impurities in the water or the salts employed, or dissolved from the walls of the glass vessel containing the culture fluid, may produce a marked effect especially in the fungi, & although the traces may be so small that the tests employed failed to reveal them. Nor has a flowering plant ever been developed in the complete absence of silicon or sodium, a condition which, however, could only be secured by a cultivation in a fluid which was not in contact with the glass. Both seeds and spores always contain a certain amount of the essential elements . . .”

Little attention was given to the suggestion of Pfeffer (1899) that sodium could have a role as an essential micronutrient until the early 1950s. The late Professor J. G. Wood, of the University of Adelaide suggested the possibility that sodium and/or chlorine could be essential for plants in very small amounts since at that time no growth experiments had been reported in which these elements had been carefully excluded from the plants’ environment. He suggested a Ph.D. project to the author to examine these elements as possible micronutrients for plants. Two developments greatly assisted the investigation. The first was the introduction of the emission flame photometer (followed soon after by the invention of the atomic absorption spectrophotometer (Walsh, 1955)), which enabled sodium to be determined rapidly at low concentrations. The second was the use of apparatus of plastic materials virtually free of sodium which had recently become available. Professor Wood suggested that we work with Atriplex vesicaria (bladder salt bush) as one of the species to be investigated. He had shown that it had a propensity to accumulate large quantities of sodium and chlorine (Wood, 1925) and hence had a possible need for larger amounts of these elements than other plants. He was also interested in its possession of the bundle sheath in its leaf anatomy which he had described previously (Wood, 1925). This he thought might have some special physiological significance! Dry weight responses to sodium chloride were obtained in preliminary water culture experiments in this species as observed by Ashby and Beadle (1957) with A triplex infata and Atriplex nummularia. During the course of this work Broyer et al. (1954) demonstrated the essentiality of chlorine in tomatoes and subsequently the requirement for chlorine was shown for ten further species by Johnson et al. (1957). There now seems little doubt that chlorine is an essential micronutrient for all higher plants. Although the requirement for chlorine was considerably higher than for the other micronutrients, its essentiality was not readily demonstrated until the chlorine (probably partly in dust and HCl vapour) was filtered from

SODIUM AND PLANT METABOLISM

125

the air surrounding the plants. The first clear evidence of sodium being essential for plant life was produced by Allen and Arnon (1955) who demonstrated a specific requirement for the blue-green alga Anabaena cylindrica. Only 0.22 mol m-3 (5 ppm) sodium were needed and no other monovalent cation would substitute for sodium. Two years later, Atriplex vesicaria (bladder salt bush) was also shown to have a specific requirement for sodium (Brownell and Wood, 1957). Plants not receiving sodium made only little growth and showed signs of sodium-deficiency including chlorosis and necrosis of leaves. Optimal growth was restored on the addition of 0.1 mol m-3 (2.3 ppm) of sodium to the culture solution. From the pattern observed with the other essential elements, it was expected that sodium would be required by all plants. However, the range of species for which sodium was shown to be essential was surprisingly restricted to Australian species of Atriplex. None of the other species examined could be shown to have a requirement for sodium (Brownell, 1968). Even when great care was taken to eliminate sodium from the environment of the plants, the other 23 species examined including other species of Atriplex grew normally without addition of sodium salts to their culture (Brownell, 1968). It was concluded that these species either had an extremely small requirement for sodium compared to that of the Australian species of Arriplex or that they did not require it at all. Since no distinguishing features were then known between the Australian and other species of Atriplex which could explain the difference in their responses, it still seemed possible that all higher plants might require sodium but those plants which had grown normally without added sodium might require only extremely small amounts compared to the Australian species. When great care was taken to further eliminate sodium from the salts of the culture solution, the concentration of sodium in the basal culture solution was reduced to 0.0685 mmol m-3 (0.0016 ppm). However, all plants examined with the exception of the Australian species of Atriplex, still grew normally without added sodium even though the Australian species of Atriplex needed approximately 0.1 mol mp3 (2.3 ppm) sodium for maximum growth. This is 1460 times the concentration of sodium in the purified basal solution which supported normal growth of the other species. With the discovery of the C , dicarboxylic acid photosynthetic pathway (Hatch and Slack, 1970), the correlation between the requirement for sodium and the possession of the C , pathway was established (Brownell and Crossland, 1972). It is, of course, still possible that C , plants have a requirement for sodium but it would have to be less than about 0.0685 mmol rn-, (0.0016 ppm). Responses to small amounts of sodium were obtained in Bryophyllum tubzjlorunl, a Crassulacean acid metabolizing plant when grown under shortday conditions with a large diurnal variation in temperature but not under

126

P. F. BROWNELL

long-day conditions with small diurnal temperature variation (Brownell and Crossland, 1974; Boag, 1976). At the present time, little is known of the physiological role of sodium in plants. The respiration rate of leaves of C, Atriplex species is depressed under low sodium conditions but was rapidly restored on the addition of sodium but no other monovalent cation (Brownell and Jackman, 1966). In the bluegreen alga, Anabaena cylindrica, sodium appears to control the activity of nitrate reductase; its activity increases up to twenty times in sodium-deficient compared to normal nitrate-grown cells. The nitrogen-fixing activity is decreased in sodium-deficient compared to normal cells (Brownell and Nicholas, 1967). A similar response of nitrate reductase to sodium has not been observed in a higher plant with the C4 decarboxylic acid photosynthetic pathway. The similarity of the C4 decarboxylic acid systems in C, and CAM plants and its absence in C, plants immediately suggests that this could be the area of metabolism in which sodium is involved. Some recent work, however, suggests that sodium may affect a part of metabolism not yet defined which occurs in C , and CAM but not in C, plants. Present studies indicate that sodium may have a role in preventing a “leakage” of carbon compounds from the metabolic system of plants for which it is essential. A scheme is presented in this article to explain this hypothesis.

11. LOW-SODIUM CULTURE CONDITIONS A. DEFINITION

It is not possible to specify an exact concentration of sodium at which growth is limited in plants for which it is essential. In Atriplex vesicaria (bladder salt bush), concentrations of less than 0-05 mol rn-, (1.25 ppm) sodium may limit growth but the concentration would be expected to vary according to the species and conditions of the experiment (Brownell and Wood, 1957). In Anabaena cylindrica, a concentration of 0.22 mol m-, (5 ppm) Na was needed for optimum growth (Allen and Arnon, 1955). Although these concentrations are high compared to those of some other elements required for optimal growth e.g. Mn, 0.01 mM (0.55 ppm); Cu, 0.001 mM (0.0635 ppm); Mo, 0.0005 mM (0.048 ppm) (Hewitt and Smith, 1975) special methods are needed to obtain culture media with concentrations of sodium low enough to produce lesions resulting from sodium-deficiency in plants, consistently, due to the ubiquity of sodium. Untreated salts even of analytical reagent grade, often contain large, variable amounts of sodium. Brownell (1 965) prepared a culture solution composed of unpurified analytical reagent salts containing 4-63 mmol m-, (0.106 ppm) sodium compared to 0-069 mmol m-3 (0.0016 ppm) sodium in the solution made up from salts especially prepared to have only low sodium concentrations. The culture solution of

SODIUM AND PLANT METABOLISM

127

Williams (1960) containing reagent grade chemicals and deionized water had concentrations of sodium between 52.2 and 78.3 mmol m-3 (1.2 to 1.8 ppm). Plants of Halogeton glomeratus showed a marked decline in yield when grown on this culture solution compared to those grown on the same solution to which 1 mol m-3 (23 ppm) sodium was supplied. Woolley (1957) prepared a culture solution containing 0.256 mmol m-3 (0.006 ppm) sodium in experiments designed to determine if sodium were essential for tomatoes. In experiments in which sodium was shown to be essential for Anabaena cylindrica, Allen and Arnon (1955) prepared culture solutions by twice recrystallizing the macronutrient salts but did not state the concentration of sodium in their purified medium. The addition of sodium chloride to give the solution a sodium concentration of 217 mmol m-3 (5 ppm) was sufficient for optimal growth of the alga. In a study of the effects of sodium on nitrogen metabolism in the same organism, the concentration of sodium was reduced to 0.174 mmol m-3 (0.004 ppm) (Brownell and Nicholas, 1967) by methods described by Brownell (1965). Ward and Wetzel(l975) estimated a concentration of 10.26 mmol m-3 (0.236ppm) sodium in their basal solution in a study on the effects of sodium on blue-green algal growth. They used unpurified salts. The plants for which sodium has been shown to be essential, including certain blue-green algae, C , and CAM plants appear to require less than about 217 mmol m-3 (5.00 ppm) of sodium for optimum growth. Therefore, it is necessary to purify the culture solution salts to free them from sodium and thus obtain decisive signs of sodium-deficiency in these plants. It is obvious that these levels of sodium are very much less than those of about 10 mol m-3 (230 ppm) and above needed to elicit growth responses in certain other species including sugar beet (Harmer and Benne, 1945). B. DETERMINATION OF SODIUM

Early work on the effects of sodium in plant nutrition was hampered by the lack of suitable methods for determining sodium. Indirect methods employed were too inaccurate and insensitive to enable quantitative determination of small amounts of sodium in plants and their culture solutions. Bertrand (1929) studied the sodium contents of plant material by the uranyl acetate precipitation method. This was satisfactory for determinations of tissues with high concentrations of sodium but was not sufficiently sensitive for estimating sodium at the lower concentrations. The method was timeconsuming and tedious making it difficult to carry out determinations on many samples. From 1940-1950 the use of the emission flame photometer became general. Jones (1960) describes an early version of flame photometer constructed by Klemperer in 1910. In this instrument, spray-charged oxyacetylene flames carrying unknown and standard solutions (which were varied) were com-

128

P. F. BROWNELL

pared visually in a divided spectroscope eye-piece. Lundegardh in the 1930s used a monochromator, vacuum photocell and amplifier with an air-acetylene flame. During the 1930-1950 period, monochromators were replaced by optical filters transmitting only narrow ranges of the optical spectrum. This became the basis for the modern emission photometer which provided a rapid, sensitive method for determining the sodium concentrations in water and extracts of plant tissues. Vogel(l961) has described the method and some of the main types of instruments employed. Wallace et al. (1948) used the emission flame photometer to determine the sodium contents of 300 samples from over 100 native and cultivated plants growing in New Jersey. They ranged from 0.00 to 3.00% on a dry weight basis. Brownell (1965) found the flame photometer to be suitable for the determination of sodium in distilled water by carefully evaporating it down to one-thousandth of its original volume in a silica beaker and determining the sodium it contained with the flame photometer but large interferences occurred when determining sodium at low concentrations in solutions of nutrient salts and ashed plant material. To overcome these interferences Woolley (1957), using a flame spectrophotometer, made readings at wavelengths of 586 nm and 594 nm as well as at 589.5 nm. Interpolation between the wavelengths of 586 nm and 594 nm gave values representing sodium-free luminescence which was subtracted from the value obtained at the sodium emission line at 589.5 nm. The method of Pleunneke and Joham (1972) was similar except that they estimated the background radiation at 586 nm and 592 nm. They used a 1 % (v/v) purified HCl solution as a reagent blank in plant tissue analysis. When estimating sodium with a filter flame photometer in solutions of nutrient salts, particularly in those of calcium and potassium, the relative errors due to the interferences of other ions increased as the salts were progressively purified (Brownell, 1965). This is shown in Fig. 1 in which the estimated concentration of sodium in 2.4 M KNO, as determined by the flame photometer decreased after three successive recrystallizations to a constant level. Further recrystallizations did not reduce the apparent concentration of sodium. This apparent level of sodium could have been due either to a small quantity of sodium which was not eliminated by repeated recrystallization or to interference by potassium. Estimations of sodium in the same solutions with an atomic absorption instrument (Box and Walsh, 1959) were virtually free of interference (Fig. 1). However, when estimating sodium in highly concentrated solutions of nutrient salts and digests of plant material, the response of the atomic absorption instrument to sodium was reduced. Under these conditions, it was necessary to prepare calibration curves showing the response of the atomic absorption instrument to known concentrations of sodium in solutions similar to those in which sodium was to be estimated. Calibration curves

129

SODIUM AND PLANT METABOLISM

were prepared for determining sodium in solutions containing potassium and calcium (Fig. 2).

B 3-

Eel Flame photometer

\--

C

0 c

__.

0

0 -

e

c

E

s 5

0

2-

1-

Atomic absorption photometer 0-0-d-

0-

L

I

1

I

1

I

1

I

I

I

I

1

0

1

2

3

4

5

6

7

8

9

10

Number of times recrystallized

Fig. 1 . The apparent concentration of sodium in 2.4 mol I-' KNO, determined with the Eel flame photometer and by an atomic absorption photometer (Box and Walsh, 1959) (P. F. Brownell, unpublished work).

The estimates of sodium concentrations in solutions of purified salts used for making up culture solutions when made with the Eel filter flame photometer, the Bechman DU flame spectrophotometer and the atomic absorption instrument showed poor agreement, particularly in the calcium nitrate data (Table I, Brownell, 1959). Emission methods were generally unsatisfactory due to the interference from certain ions. Furthermore, results obtained with the filter flame photometer varied according to the filter used in the instrument (Fig. 3). Neutron activation analysis has been used to determine sodium in plant material (Singh and Dieckert, 1973). These determinations are subject to interference from magnesium if the sample contains relatively more magnesium than sodium and the thermal neutron used has an appreciable fast neutron component. In peanut flour containing 3.3 pg 8-I dry weight Na and

130

P. F. BROWNELL

3.7 mg g-1 Mg the proportion of apparent sodium contributed by magnesium was 42 % but this was an unusually critical test for the method as the tissue contained an abundance of magnesium and only traces of sodium.

water

"3

water

0 2M

500 5M 1 OM 1 5M

-.z .

0 5M

40-

..

c

0

1 OM

-

5 300

.30-

.

2 4

.

-

20 -

10-

10-

20

0-

0 1020

06

50

1 00

Concentrotion of sodium ( ppm 1

0 1020

06

50

10 0

Concentralionof sodium (ppm1

Fig. 2. Calibration curves showing the absorption by sodium in solutions of potassium nitrate and calcium nitrate at varying concentrations (P. F. Brownell, unpublishedwork).

Special care is required for the determination of sodium in plant tissues at low concentrations. Woolley (1957), at harvest, placed material from tomato plants in perforated polyethylene bags which were in turn placed in paper bags and left in an 80°C blower-oven for several days. The dried material was ground in a Wiley mill to pass a 40-mesh sieve. A polythene plate was substituted for the glass plate normally supplied as a part of the Wiley mill. Samples were stored in new, but unwashed polyethylene bags. They were ashed at 700°C in platinum crucibles and after cooling made up in 0-16 M HNO,. With the sample sizes and dilutions used, it was possible to determine sodium down to concentrations of 0.005 pmol g-1 dry weight (0.12 ppm (dry basis)). Pleunneke and Joham (1972) dry-ashed 2 g samples of dried plant material from cotton plants in nickel crucibles for 4 h at 225°C and then overnight at 550°C. The ash was dissolved in purified dilute HC1 and the crucibles were rinsed with four 20ml aliquots of warm water. The dissolved sample was

131

SODIUM AND PLANT METABOLISM

TABLE I Concentrations of Sodium in the Culture Solution

Salt

Concentration of salt in culture solution

Estimated concentration of sodium in culture solution (ppm) Eel flame photometer

Ca(N0,)2

0.004 mol 1-l

0.10

KNO, KH,PO,

0.006 mol 1 -l 0.001 mol I-' 0.001 mol I-' 0.006 mmol 1 -l 0.46pmol 1 0.182 pmol 1

0.007

MgS04 FeCI, HzBO,

MnSO,

-

Beckman D.U. and spectrophotometer

0.0024 O~ooo02 0.002 0.00032 OW028 ~~~

-

0.03 0.0054

0.0013

0.0017 O~ooo02 O~oooo1

OW025 no response

O~ooo08

O~ooo01 0.000005

0~0002 ~~

SI-RO-SPEC and atomic absorption OW075 no response -

~~

From Brownell (1959).

collected in a polyethylene beaker and filtered and brought to volume for the sodium determination. Concentrations of sodium as low as 0.1 pmol g-1 dry weight (2.3 ppm (dry basis)) were determined by this method. Both Woolley (1957) and Pleunneke and Joham (1972) determined the sodium in the extracts by emission flame photometry described above. Brownell (1965) determined sodium in acid-digested material by atomic absorption spectroscopy in the following method : From finely ground plant material two representative fractions were taken of less than 0.5 g. These were dried at 95" until they had reached constant weight and then placed in a desiccator. When at room temperature, they were reweighed and placed in quartz Kjeldahl tubes of approximately 15 ml capacity and digested with 1 cm3 H,SO/ (S.G. 1+36), 1 cm3 HCI04 (S.G. 1.70) and 5 cm? HNO, (S.G. 1.42) (which had been redistilled in silica). When the digestion was complete, the digest was made up to a suitable volume with distilled water, and the concentration of sodium determined with the atomic absorption instrument. Digests of material containing only very low concentrations of sodium were made up to small volumes so that the sodium concentration would be high enough to be determined with the atomic absorption instrument. It was found that the response of the instrument to known concentrations of sodium in these highly concentrated solutions was less than in water. The depression in the response to sodium in the solution appeared to be due to the sulphuric acid used in the digestion of the plant material and calibration curves were prepared using similar amounts of sulphuric acid (Fig. 4). It is possible to determine concentrations of sodium as low as 0.043 pmol g-1 dry weight (1 ppm (dry basis)) by this method. Longitudinal profiles of sodium and potassium ions have been made in roots of Hordeum distichon L. and Atriplex hortensis by flameless atomic absorption spectroscopy by Jeschke and Stelter (1976). Concentrations of

132

P. F. BROWNELL r16

-15 b

/

6-

1-

/

/

/

/

/'

-14

-13 -12

-2

b//

d/

-1

Fig. 3. Comparison of responses of an Eel flame photometer fitted with different filters and the atomic absorption photometer (Box and Walsh, 1959) to sodium in a 0.02 moll-' calcium nitrate solution. Response by the flame photometer withfilterA O--o,with filter B +o and the atomic absorption photometer (Box and Walsh, 1959) &----A (P. F. Brownell, unpublished work).

sodium as low as 0.1 pmol g-l (2.3 ppm) on a fresh weight basis were determined in 0.5 mm sections of single roots equilibrated or grown in potassium-free, 1 mM sodium solution. The roots were injected into a flameless ionization chamber on the tip of a pipette. C. CULTURE TECHNIQUES

The techniques for obtaining low sodium culture conditions are critical in these studies as the sodium levels must be extremely low to obtain clear signs of its deficiency in experimental plants. Sodium is very abundant in water, most reagents and in the atmosphere as dust or sea spray and although the requirements by plants for sodium appear to be much greater than for some of the other micronutrients, it can be as difficult to demonstrate the essential nature of sodium as for some of the other micronutrients. The known sources of sodium to plants include water, the air surrounding the plants and that used for aeration of cultures, seeds, the salts of the culture solution and the culture apparatus.

SODIUM AND PLANT METABOLISM

133

9

0

2.0

5.0

10.0

Concentration of sodium ( ppm 1

Fig. 4. Responses to known concentrations of sodium in water 0--------0 , the acid digest of plant material 0-0 and 3.6 N sulphuric acid (approximately the concentration of sulphuric acid in the acid digest O----O (P. F. Brownell, unpublished work).

I . Water Sodium must be carefully removed from water for it to be suitable for growth experiments in which sodium is studied as a micronutrient. Water distilled in pyrex glass contains too much sodium for decisive signs of sodium deficiency to be obtained in plants. The use of deionizers does not appear to be completely satisfactory either. In experiments carried out in Adelaide,

134

P. F. BROWNELL

South Australia rainwater containing approximately 87 mmol m-3 (2 ppm) sodium was passed through a commercial deionizer consisting of columns of cation and anion exchange resins arranged in series. The treated water contained 3.5 mmol m-3 (0.08 ppm) sodium (Brownell 1965). However, Williams (1960) obtained evidence of a sodium requirement by Halogeton glomeratus using resin deionized water. The sodium content was 52 to 78 mmol m-3 (1.2 to 1.8 ppm) and unpurified reagent grade nutrients were used. Pleunneke and Joham (1972) in their study of the effects of sodium upon the free amino acid content of cotton leaves passed distilled water through two Barnstead mixed-bed demineralizer cartridges. It was then stored in polyethylene vessels and contained less than 0.078 pmol (0.0018 ppm) sodium. Woolley (1957) using a polyethylene condenser for his study on the sodium and silicon requirements of plants obtained water containing 0.026 mmol m-3 (0.0006ppm) sodium. Hewitt (1966) suggests that some of this sodium could have been derived from the polyethylene which is manufactured by a process involving alkali. Brownell and Wood (1957) redistilled water three times in tinned metal stills and stored it in polyethylene containers. It contained 0.02 18-0.0348 mmol m-3 (0.0005-0.0008 ppm) sodium. Redistilling demineralized rainwater from a silica still reduced its sodium content from 3.5 to 0.0087 mmol m-3 (0-08-0.0002 ppm) (Brownell, 1965). Storage in large polyethylene containers appears to be satisfactory. 2. Air Purification Relatively large accessions of sodium occur particularly in coastal areas. Hutton (1953) and Turton (1953) presented data showing that approximately 20 lb/acre/year (22 kg ha-' year-l) of sodium are deposited in rain over areas including Yorke Peninsula and Keith in South Australia and at Coolup in Western Australia. Brownell (1965) estimated accessions of sodium in a conventional glasshouse by placing filter papers horizontally in different sites within the glasshouse and determining the increase in sodium of each paper at the end of each week of exposure when the papers were replaced by another set. The mean amount of sodium collected per cmz per week was 0-02 pmol. This is of the same order as the accession of chlorine (0.022pmol cm-2) collected by Johnson et al. (1957) at Berkeley, California. It was calculated that this rate of accession of sodium (approximately 210 pmol sodium per 780cm2 of plant cover in 93 days) was almost sufficient to account for the increase (294pmol) observed in the amount of sodium recovered in the culture solution and the plant material above that originally present as an impurity of the culture solution, seed and water at the beginning of the experiment (Table 11). In subsequent experiments, Brownell and Wood (1957) and Brownell (1965) grew plants in a small greenhouse designed to prevent contamination of plants and their cultures by sodium from the atmosphere (Fig. 5). A slightly positive pressure was maintained

TABLE I1 Sodium Supplied: Sodium Recovered Amount supplied (pmol) Conditions of experiment

Seeds

Solution

Watefl

43c

80

Amount recovered (pmol)

Total

Leaves

126

153

Stems and petioles

Roots

Remaining in culture solution

Total

~~

In conventional

3 (10 seeds)

glasshousea In pressurized cabinetb

1.07 (4 seeds)

5C

0.04

6.11

1.52

33

1-64

117 1 *66

117C 1*13c

420 5.95

a Ten plants of Atriplex vesicuria were grown in 4.5 litres of basal culture solution to which no sodium had been intentionally added. Experiment was of 93 days duration. b Four plants of Atriplex vesicuria were grown in 2 litres of basal culture solution to which no sodium had been intentionally added. The experiment was of 48 days duration. C These data were obtained using emission flame photometry. Due to positive interferencefrom ions in the culture solution, these values are generally higher than they should be. d Water used in the experiment in the glasshouse contained 4 mmol Na m3 and water used in the pressurized cabinet contained 0.010 mmol Na ma. Amounts of sodium were calculated on the total volume of water supplied to the culture during the experiment. From Brownell (1 965).

136

P. F. BROWNELL

Fig. 5. The pressurized greenhouse. Note the housing of the air compressor on the left, which pumped air through a series of cotton-wool filters to a T-piece at this end of the greenhouse from which part of the air passed directly into the greenhouse and part was used for aeration of cultures. Windows were of polyvinylchloride and access was through a screw-clamped door on the left. The polythene tube which entered the greenhouse on the left carried wiring for electrical heating and the thermostat controlling it. The clothcovered disc on the immediate right of this tube was the outlet for surplus air. From Brownell and Wood (1957).

within the greenhouse by a compressor which supplied air continuously to both cultures for aeration and to the greenhouse itself through Whatman No. 1 filter papers and washed absorbent cotton wool contained in metal cylinders. The amount of sodium known to have been added intentionally was approximately the amount of sodium recovered in the culture solution and plant organs at the end of the experiment in this greenhouse (Table 11). No increase in the amount of sodium could be detected after the cultures had remained in the experimental glasshouse for 48 days. All air entering the compressor was drawn through Whatman No. 1 filter papers which were changed at 24-hourly intervals and the sodium they had trapped determined. The amounts of sodium trapped per day (from about lo5 litres of air) rose and fell periodically (Fig. 6). It was found that the amounts were greatest when strong winds blew from the west (the seaward side). Under these conditions opening of the cabinet and manipulation of cultures was avoided. Woolley (1957) in experiments designed to determine if sodium and silicon were essential for tomatoes had a room in a greenhouse supplied with air at a

SODIUM AND PLANT METABOLISM

137

Fig. 6. Amounts of sodium trapped per day from air drawn through a Whatman No. 1 filter paper; about lo6 litres of air passed through this filter per day. From Brownell (1965).

slight positive pressure which had been passed through four filters in the following order : 1. A charcoal and limestone filter to trap smog and acid gases. 2. A fibreglass filter to trap macroscopic dust, including particles of carbon from the smog filter. 3. A Mine Safety Appliance paper filter designed to stop particulate matter as small as 0.5 prn diameter. 4. An excelsior pad, continuously wet with distilled water. It was found in the sodium supply-recovery balance that the “minus” treatment plants received significant amounts of sodium from unknown sources and the recovery of sodium from the plants and that remaining in the culture solution was 3.15 pmol which is almost twice that known to have been supplied (2.9 pmol) in the culture solution salts, water, seeds, dacron, polyethylene bag and polyethylene container (Table 111). It is extremely unlikely that all the sodium ( 2 pmol) in the polyethylene container would have been supplied to the system so that the contamination of sodium from unknown sources must have been large. Pleunneke and Joham (1972) grew cotton plants in a clear polyvinyl chloride-covered box located in a greenhouse. All inner wood surfaces were impregnated with plastic spray and coated with mineral oil. Air circulation was provided by a fan mounted in a metal plenum chamber attached to one end of the growth chamber. Air entering the chamber passed through a distilled, deionized wash and a series of three filters separated by mineral oil coated compartments. Most of the sodium present in the harvested plants could be accounted for by that supplied in the seed and the culture solution. Singh and Diekert (1975) described a simple system for the production of low-sodium peanut seedlings to seven days of age. During this period they picked up only 2 pg of extraneous sodium per seedling. Brownell (1965) could not detect any increase in the concentration of

138

P. F. BROWNELL

TABLE 111 Sodium Supply-Recovery

Balance

Av Na in plants Treatment

Known Na supplied

Shoots pg

Minus Na Na, Si, A1 Elkhorn sand

2.9a 2000 2000 6.6a

Roots

atoms/culture 0.88

2.0 740 770 2.1

640 640 9.2

Av Na remaining in solutions 0.27 510 470 0.29

Na recovered 3.15 1890 1880 11.6

a These values are probably excessive, since they include the entire 2pg atoms that could possibly have been supplied by the polyethylene container. From Woolley (1957).

sodium in water contained in open vessels after a fortnight in the greenhouse pressurized with filtered air. However, the concentration of sodium in water in a culture vessel after a fortnight of continuous aeration even when covered, increased by 0.4mmol m-3 (0.01 ppm). At this rate of contamination, the amount of sodium in the most highly purified solution (0.069 mmol m-3) (0.00158 ppm) would be increased many times in even a short experiment. In subsequent experiments the air for aeration of cultures was effectively freed from sodium by passing it through distilled water contained in a train of plastic vessels. Pleunneke and Joham (1972) filtered air for aeration of nutrient solutions through two 12-5 cm plugs of acid-washed cotton and four distilled, doubly deionized water washes (changed daily). Woolley (1957) inserted a Dacron plug at the top of each aeration tube to act as filter.

3. Salts of the Culture Solution Responses to sodium have been reported in experiments in which unpurified salts were used. Ashby and Beadle (1957) obtained significant increases in yield in Atriplex injlata and Atriplex nummularia grown on Solution 1 of Hoagland and Arnon (1938) prepared from A.R. salts and distilled water when sodium was added to 50 mol m-3 as either NaCl or Na,SO,. Tomatoes did not respond to this sodium treatment in the same experiment. Black (1960) observed similar significant growth responses in Atriplex vesicaria receiving up to 20 mol m-3 NaCI. Further increase in concentration of NaCl decreased growth. Williams (1960) obtained significant responses to 1 rnol m-3 NaCl, NaNO, and 4 mol m-3 Na,SO, applied to Halogeton glomeratus grown in the solution of Hoagland and Arnon (1938) using unpurified salts. In establishing the essentiality for sodium by the blue-green alga, Anabaena cylindrica, Allen and Arnon (1 955) twice recrystallized the macronutrient salts but used untreated A.R. salts for the micronutrients. In attempting to determine if tomatoes had a sodium or silicon require-

SODIUM AND PLANT METABOLISM

139

ment, Woolley (1957) prepared all the nutrient salts by two recrystallizations except molybdic acid. The recrystallizations consisted of two operations; a rapid crystallization from a hot supersaturated solution, and a slow crystallization as the supernatant from the first operation gradually cooled. It was suggested that most impurities would be associated with the first small crystals rapidly formed at high temperatures by adsorption or occlusion. These crystals were discarded and the larger crystals slowly formed at lower temperatures were used in the preparation of nutrient solutions. The molybdic acid was prepared by the prolonged heating a t 70" of a 10% solution of ammonium molybdate. The sodium concentration of the final culture solution was 0.511 mmol m-3 (0.012 ppm). I n their study of the effect of low sodium levels upon the free amino acid content of cotton leaves, Pleunneke and Joham (1972) purified the macronutrient salts by recrystallization in stainless steel and polyethylene vessels. Three recrystallization cycles were employed with potassium salts and four cycles were used for the purification of MgS0,.7H20 and Ca(N0J2.4H,0. Micronutrients apart from FeS04-7H20were not recrystallized. The EDTA used in the preparation of iron chelate was dissolved in polyethylene distilled NH,OH and precipitated with purified HCI. The process was repeated three times and the EDTA then reacted with recrystallized FeS04.7H20.The final culture solution contained less than 0.17 mmol Na m-3 (0.004 ppm). Brownell (1965) prepared a culture solution containing less than 0.069 mmol m-3 (0.0016 ppm) sodium. The concentrations of sodium in solutions of purified salts and untreated salts are compared (Table IV). The following methods were used to obtain salts with low sodium content. Potassium nitrate, potassium dihydrogen phosphate, magnesium sulphate and manganese sulphate were recrystallized in silica vessels up to six times. Alternatively, diammonium phosphate was prepared by first distilling phosphorous oxychloride from silica, cautiously hydrolysing it in water, boiling off hydrochloric acid and adding two equivalents of redistilled ammonium hydroxide. Calcium nitrate was obtained by preparing calcium salicylate from calcium carbonate and salicylic acid, recrystallizing it several times and then ashing. The oxide was dissolved in nitric acid redistilled from silica. Sodium contamination of the resulting purified salt was d a of that of the analytical reagent salt. Boric acid was freed from sodium by making a saturated solution of boric acid in ethanol in the boiling flask of a silica still. The boron was volatilized as the ethyl ester of boron and the distillate collected in a platinum vessel. After drying the distillate in a waterbath, boric acid remained. This was placed in a desiccator until its weight was constant then made up into a stock solution 10OOO times as concentrated as it was in the final solution. The concentration of sodium was reduced from 435 mmol kg-l(l0 000 ppm)

TABLE IV Sodium Contributed to the Culture Solution as Impurities of Component Salts Before and After Purification

salt Calcium nitrate Potassium nitrate Potassium dihydrogen phosphate Diammonium sulphate Magnesium sulphate Boric acid Manganese phosphate Copper sulphate Zinc sulphate Ammonium molybdate Ferric ammonium ethylene tetra acetate Ammonium chloride Total sodium in culture solution due to sodium impurities of all component salts From Brownell (1965).

Sodium contributed to culture solution by component salts (pmol. 21-9 Conc. of salt in Untreated analytical culture solution ( p ~ ) reagent salts Prepared salts 4Ooo

5000 1000 1000 1000 46 9.1 0.3 1 0.76 0.10

90 350

435 0.52 2.18 052 0-26 0.0026 0.0252 0-00026 0.0065 0.00022 1.39 0.00569

0.0174 0.0109 0-0174 0.0347 0.00174 0.00087 0-01320 0-00026 0.000435 0-000218 0.0347 0.00565

9-26

0.137

SODIUM AND PLANT METABOLISM

141

in boric acid (to which sodium had been intentionally added) to less than 218 pmol kg-' (5 ppm) in boric acid purified in this way. Solutions of copper sulphate, zinc sulphate and ammonium molybdate were made up from A.R. grade salts without purification as the amounts of sodium they contributed to the culture solutions were extremely small. Iron was supplied to cultures in a single addition of ferric ammonium ethylene tetra acetic acid (EDTA). This was prepared by a method similar to that of Jacobson (1951) except that ammonium hydroxide was used instead of potassium hydroxide in equivalent amount. Potassium hydroxide, which contained much sodium as an impurity would have been difficult to purify, whereas the ammonium hydroxide redistilled in silica contained an amount of sodium too small to be detected. Ferrous sulphate was recrystallized six times from solutions acidified by small quantities of sulphuric acid, and the resulting crystals were dried in an oven at 50". EDTA was dissolved in 2 mol I-' ammonium hydroxide (redistilled in silica) and then precipitated by the addition of 2 mol 1-1 HC1 (redistilled in silica). This procedure was repeated four times and the resulting precipitate was washed in several changes of distilled water and dried in an oven at 50". Ammonium chloride was formed by the addition of ammonium hydroxide to an equivalent amount of hydrochloric acid (both redistilled in silica). The resulting solution was concentrated by boiling, cooled to room temperature, and placed in a refrigerator overnight. The ammonium chloride crystals formed were dried in a desiccator to constant weight and made up in a stock solution 20 000 times the concentration required in the full concentration culture solution.

4. Culture Apparatus To obtain clear-cut signs of sodium-deficiency, the choice of culture apparatus is critical. The use of glassware even borosilicate glass should be avoided and in its place materials including platinum, silica-ware, stainlesssteel and selected plastics have been found to give minimum contamination. Ashby and Beadle (1957) obtained significant responses to sodium in Atriplex nummularia and Atriplex inflata seedlings grown in vermiculite which were transferred to cultures in which the plants were supported by cotton-wool plugs in holes of masonite tops previously impregnated with paraffin wax supported over four-litre enamelled pots of nutrient solution. Woolley (1957) germinated tomato seeds on moist acid-washed cheesecloth in polyethylene containers. The plants were cultured in two-litre polyethylene freezer containers with holes drilled in the lids, and were supported by Dacron wool which had previously been washed with 10% HNO, and 10% HF, followed by repeated rinsing with water until the pH of the rinse water remained constant. Aeration of the culture solutions was by a continuous flow of air bubbling from a polyethylene tube fitted with a Dacron plug to

142

P. F. BROWNELL

act as a filter into each culture vessel. The total amount of sodium per container was estimated to be 2.0pmol (46pg). Probably very little of this sodium would have been available. Hewitt (1966) considered that this high sodium figure for the polyethylene could have been due to the use of alkali in manufacture. Pleunneke and Joham (1972), in studying the influence of sodium nutrition upon the free amino acid content of cotton leaves, germinated the seeds on cheesecloth platforms in sterile polypropylene beakers covered with Saran wrap. Seedlings were transferred to eight-litre polyethylene nutrient solution containers containing three plants each. The outside of the containers was painted black and coated with aluminium foil to prevent light penetration. Seedlings were initially supported by rinsed cotton plugs and later by plugs on ring stands. Plastic tubing was attached to the base of each solution container passing through the growth chamber wall thus providing a means by which the solutions were changed without opening the chamber. Brownell (1965) germinated seeds on nylon gauze sewn into a circle of polyethylene tubing to form a flat disc supported by polystyrene legs in a circular polythene vessel. Water or culture solution in this vessel was aerated through a fine-bore silica tube. Culture vessels of two-litre capacity were made from half-gallon polythene containers by cutting off their tops. The vessels had covers of grey Perspex which held four evenly spaced plants, secured by white terylene (equivalent to Dacron) fibres washed in many changes of silica distilled water, clamped between the longitudinally split halves of polythene tubing. Cultures were aerated continuously, with air filtered through cottonwool and bubbled through frequently changed distilled water and filter papers, by means of centrally placed silica tubes dipping to the bottom of culture vessels. Paper, black on one side and white on the other was wrapped around the culture vessels to exclude light from the culture solution and the roots of plants. Samples of all materials associated with the cultures were boiled in small amounts of concentrated HNO, (redistilled in silica) ; the amounts of sodium extracted by this treatment were small in all cases. 5 . Seeds The amounts of sodium per seed vary widely. A seed of Beta vulgaris (sugar beet) contained 13.04 pmol (300 pg) sodium whereas a lettuce seed 0.0043 pmol (0.1 pg) sodium (Brownell, 1968). Many other species have seeds with amounts of sodium between these limits. Lycopersicum esculentum tomato cv. “Marglobe” contained 0.008 pmol (0.18 pg) sodium per unwashed seed (Woolley, 1957). Cabbage contained 0.0109 pmol (0.25 pg) per seed and Atriplex vesicaria (bladder salt bush) had a sodium content of 0.27 pmol(6.21 pg) per seed after washing (Brownell, 1968).

SODIUM AND PLANT METABOLISM

143

A two-litre culture solution prepared from purified salts and silicadistilled water contains only approximately 0.16 pmol(3.68 pg) sodium as an impurity (Brownell, 1965). The contribution of sodium by the seeds could be a relatively important source under these conditions. Methods of reducing the amounts of sodium in the seed include harvesting the seeds from plants grown under conditions of low sodium. By this method, Woolley (1957) reduced the sodium per seed in tomatoes cv. “Marglobe” to 0.002 pmol (0.046 pg) which was approximately one quarter of the amount present in the original seed. Another method of reducing the sodium per seed is by washing. Large amounts of sodium were found in the water which had been used to wash seeds of sugar beet, Atriplex vesicaria and Kochia childsii (Brownell, 1958, 1968). The sodium content per seed of tomato cv. “Grosse Lisse” was reduced from 0-004pmol (0.095 pg) to 0.001 pmol (0.031 pg) by washing in four changes of distilled water over a period of 33 hours (Brownell, 1968). The amounts of sodium in the water which had been used to wash the seeds of other species including barley, cabbage, white clover, tomato, Chenopodium capitatum, Exomis axyrioides and Aster tripolium were low in all cases (Brownell, 1958). Johnson et al. (1957) could only show losses of chlorine from carrot seeds when the seeds of lettuce, tomato, cabbage, carrot, sugar beet, barley, alfalfa, buckwheat, corn, beans and squash were washed ; the washed carrot seed contained five times less chlorine in the unwashed seeds. Vigorous washing may damage the seed by removing other mineral nutrients or organic substances essential for the normal development of the plant. Although the mineral nutrients could be replaced by soaking the seed in a nutrient solution, the organic substances would not be replaced. By growing plants to a large size, the sodium present is distributed over a large amount of metabolizing tissue and the relative contribution of sodium from the seed becomes less. However, long-term experiments are open to objections arising from the depletion of other nutrient elements from the culture solution, the possible infection of the culture solution by microorganisms, or the extra risk of contamination from manipulation and the greater area of leaves capable of absorbing sodium from the atmosphere. As a general procedure, Brownell (1965) washed seeds in many changes of distilled water until the amount of sodium in the wash water could not be detected with the flame photometer adjusted to its maximum sensitivity. 6. Composition of Culture Solution The culture solution must provide all the elements known to be essential for plant growth with the exception of sodium at concentrations which support vigorous growth when sodium is added to the solution. It is an advantage

144

P. F. BROWNELL

to select a culture solution composed of salts which are amenable to purification procedures. The basal culture solution used by Brownell and Wood (1957) and Brownell (1965) was similar to that of Broyer et al. (1954) and had the following composition expressed in pmol 1-l: KNOB, 5000; Ca(NO,),, 4000; MgSO,, 1000; (NH,), HPO,, 1000; KH,PO,, 1000; &.B03,46; MnSO,. 7H20, 9.1 ; CuSO4.5H2O,0.31 ; ZnS0,.7H20, 0.76; (NH,), M070,,.4H20, 0.1 ; NH,Cl, 350. Iron was supplied as the ferric ammonium EDTA at 90 pmol 1-1 in the basal culture solution by a method similar to that of Jacobson (1951). For the growth of the blue-green alga, Anabaena cylindrica, Brownell and Nicholas (1967) used culture media similar to that described by Allen and Arnon (1955) with the following composition (mol m-3): MgSO,, 1;KH,PO,/ K,HPO, (pH 7.2), 2; CaCl,, 0.5; and KC1, 10 (in the combined nitrogen-free medium), or KNO,, 10 (in the nitrate-containing medium). Micronutrients were supplied (pg c m 3 as follows: Fe (as the EDTA complex) 5, Mn (as MnS0,.4H20) 0.5, Mo (as (NH,), M0~0,~.4H,0)0.1, Zn (as ZnS0,.7H20 4H,O) 0.05, Cu (as CuS04.5H,0) 0.02, B (as H,BO,) 0.05, and Co (as Co (NO3)& 0.01. 111. RESPONSES TO SODIUM AT LOW CONCENTRATIONS A. LOWER PLANTS

1. Blue-green Algae There have been several suggestions in the early literature of blue-green algae having a requirement for sodium. In 1898, Benecke (cited by Allen and Arnon, 1955) described a species of Oscillatoria, which grew in a medium in which all potassium salts had been replaced by sodium salts. Although it is likely that some potassium would still have been present in the medium as an impurity, the work suggested that sodium was acting as an independent nutrient element. Emerson and Lewis (1942) found Chroococcus supplied with 0.026 mol m-3 (1 ppm) potassium made only “very poor growth” without the addition of sodium. Allen (1952) grew 23 out of 30 cultures of Myxophyceae without added potassium salts but found they required sodium for growth in the presence of potassium. Some potassium probably would have been present in the cultures as an impurity of the ordinary C.P. grade chemicals which were used. Kratz and Myers (1955) obtained a decisive growth response to small additions of sodium in Anacystis nidulans even though no attempt had been made to purify the salts to eliminate their sodium contamination. The growth of Anabaenaflos aquae A37 was shown to be severely limited by the absence of either Na or K from the culture medium (Bostwick et al., 1968). Optimal growth was achieved with 4-81 mol m-3 (188 ppm) potassium and 1.76 mol m-3 (40.5 ppm) sodium. To obtain optimum

SODIUM A N D PLANT METABOLISM

145

growth of Nostoc muscorum Eyster (1970) found it necessary to increase the sodium concentration of the medium with increasing concentration of potassium in the medium. These observations suggested a possible nutrient effect of sodium in small amounts. Allen and Arnon (1955) critically examined sodium as a nutrient for a blue-green alga, Anibaena cylindrica. They carefully obtained low sodium conditions in their cultures. Glass was washed with 3 mol 1-' HCl followed by thorough rinsing with glass distilled water. The C.P. grade salts of the culture solution, MgSO,, KCl, NaCl, KH,PO, and Na,HPO, were purified by three successive recrystallizations from hot water. CaSO, was prepared by the addition of H,SO, to a solution of calcium chloride, followed by thorough washing of the CaSO, precipitate. When these precautions were taken to exclude sodium contamination, normal development of the alga did not occur in the absence of sodium but was restored by the addition of sodium. Five parts per million of sodium or higher was required for optimal development of the alga. Potassium at 8.0 mol m-3 was supplied to all cultures to ensure that the response to sodium occurred in the presence of adequate potassium. It was also found that potassium was essential in the presence of adequate sodium. The elements lithium, rubidium and caesium a t 1 or 10ppm in the culture were unable to replace the requirement for sodium in these cultures. Sodium-deficient cultures were shown to have less phycocyanin than those with adequate sodium but their chlorophyll content was not affected. Brownell and Nicholas (1967) found a marked difference in ratio (1 : 62) between the amounts of phycocyanin in sodiumdeficient cultures containing less than 0.17 mmol m-3 (0.004ppm) sodium as an impurity and normal cultures supplied with 4 mol m-3 (92 ppm) sodium in cultures containing nitrate. On the other hand, the cultures containing no combined nitrogen showed relatively small differences (1 : 2.3) in the ratio of phycocyanin between sodium-deficient and normal cultures. The ratio between the amounts of chlorophyll in sodium-deficient and normal cultures were (1 : 6.7) for nitrate containing cultures and (1 : 3.07) for cultures containing no combined nitrogen. The lack of agreement between the chlorophyll observations made by Allen and Arnon (1955) and Brownell and Nicholas (1967) could be due to the differences in the amounts of sodium in the cultures. Furthermore, the effects of sodium nutrition on the amounts of phycocyanin and chlorophyll in cultures of blue-green algae may well be conditioned by other properties of the cultures including the age of cells, light intensity and presence or absence of combined nitrogen. Brownell and Nicholas (1967) found that dry weight yields increased with sodium supply when the cells were grown with nitrate nitrogen or without combined nitrogen (Table I). It is noteworthy that cells grown with nitrate, at suboptimal levels of sodium, contained much less chlorophyll and phycocyanin than those grown without combined nitrogen (Table V; Fig. 7).

146

P. F. BROWNELL

TABLE V Effect of Sodium on Dry Weight, Chlorophyll and Phycocyanin Contents of Anabaena cylindrica Grown With Nitrate Nitrogen and Without Combined Nitrogen

Dry weight (gj100 cm3culture)

Chlorophyll

Amount of No NaCl in 10 No 10 combined mol m-3 combined mol rn-, culture KNO, nitrogen KNO, (mol m-,) nitrogen None 0.10 2.00 440

0.058 0.066 0.092 0.097

0.041 0.063 0.118 0.094

Phycocyanin

(pg/100cm3culture) a(units/100cm3culture)

6.59 8.55 20.50 20.20

3.07 2.66 24.00 20.50

No 10 combined mol rn-, nitrogen KNO, 0.268 0.315 0.600 0.670

0.008 0.032 0.645 0.496

~~

Cultures harvested 12 days after inoculating. a Arbitrary Unit = E$$ of phycocyanin in 100 ml. From Brownell and Nicholas (1967).

Fig. 7. Effect of sodium on Anabaena cylindrica grown without combined nitrogen (KCl) or with nitrate (KNO,). Note darker algae (more chlorophyll and phycocyanin) without combined nitrogen (KCl) to the paler ones in nitrate (KNO,) grown in cultures at equivalent rates of sodium below 0.4 mol m-3 (9.2 ppm). From Brownell and Nicholas (1967).

2. Other Algae There appears to be little firm evidence for the essentiality of sodium in algae other than in members of the Cyanophyta. There have been reports (Osterhout, 1909, 1912) that sodium was necessary for the growth of the green alga, Ulva and red algae in the genera Gigartina, Ptilota, Iridaea and Prionitis. The evidence was based on experiments in which little or no growth was obtained when the sodium of sea water was replaced by other cations including ammonium, calcium, magnesium, potassium, barium, strontium, caesium, rubidium or lithium. The best growth occurred with magnesium, calcium and potassium as the substitutes for sodium. It is not surprising that the cations substituting for the sodium were unable to support normal growth. The substitution of an element normally present at a concentration

SODIUM AND PLANT METABOLISM

147

of 500 mol m-3 by another element would be expected to have severe effects on the algae in addition to any possible effect caused by the lack of sodium. The range of tolerance to the concentration of sodium for some euryhaline species including Hemiselmis virescens, Monochrysis lutheri, Phaeodactylon tricornutum, Nannochloris oculata and Skeletonema costatum was studied by Droop (1 958). The organisms generally tolerated concentrations of sodium between 13 mol m-3 (300 ppm) and 522 mol m-3 (12 000 ppm) with maximum growth occurring between 87 mol m-3 (2000 ppm) and 261 mol m-3 (6000 ppm). The results suggest that sodium has an independent role as a nutrient element for these species and is generally required in high concentrations. McLachlan (1960) showed that sodium chloride at 80 mol m-3 (1840 ppm) supported optimal growth in another euryhaline algae, Dunaliella tertiolecta. He found the minimum requirement for sodium to be much greater than for any other element and it was not possible to substitute other monovalent cations for the minimum requirements. The alga could tolerate high concentrations of sodium chloride. Treatments of sodium chloride have been shown to stimulate the uptake of phosphate by the unicellular green alga, Ankistrodesmus braunii (Simonis and Urbach, 1963). The enhancement of phosphate uptake by sodium was much greater than by potassium or calcium. The stimulation of phosphate binding in the organisms by sodium was rapid occurring after five seconds (UllrichEberius, 1973). There does not appear to be any evidence that sodium is essential for the growth of Ankistrodesmus braunii. 3. Bacteria The literature describing the salt requirements of marine bacteria was reviewed recently by Pratt (1974). Sodium has been shown to be required specifically for the growth of certain marine bacteria and MacLeod (1965) suggested that the possession of a sodium requirement distinguishes marine bacteria from most non-marine species. Most marine bacteria which have been examined require 0.2-0-3 mol 1-1 sodium for optimal growth. Such organisms require the addition of sodium salts or sea water even to the complex laboratory media commonly used for isolation, though such media are usually contaminated with appreciable amounts of sodium. Lithium, rubidium and caesium were not able to replace the requirement of sodium for the growth of the organisms. The requirement of marine bacteria for sodium at high concentrations apparently can be lost when cultured on low-sodium media. MacLeod and Onofrey (1 963) trained a marine pseudomonad to grow on a medium prepared without added sodium salts by streaking cultures serially on to the surface of plates of the medium containing progressively lower concentrations of sodium, down to 0.028 mol 1-1 sodium. The organism still required sodium for growth when the adapted culture was tested in a chemically defined

148

P. F. BROWNELL

medium containing less than 6.5 x mol 1-1 sodium. The optimal requirements of sodium for the marine bacteria so far examined arein the range of 0.005 to 0.2 mol 1-1 depending upon the species. Later observations have shown some non-marine bacteria also to require sodium. MacLeod (1968) found two strains of the non-halophilic Rhodopseudomonas to have small but specific requirements for sodium. Kodama and Taniguchi (1976) demonstrated the growth rate of Pseudomonas stutzeri, a non-halophytic bacteria, to be a sigmoidal function of sodium concentration from 2 to 50 mol m-3 (46-1 150 ppm). The growth rate was half-maximal at 0.5 mol m-s (1 1.5 ppm). It was found that five strains (including isolate S85) of an important cellulolytic species in the bovine rumen, Bacteriodes succinogenes, required 84 mol m-3 (1932 ppm) sodium when potassium was present at concentrations between 3 mol m-3 (1 17 ppm) and 50 mol m-s (1955 ppm) in the medium for maximal growth (Bryant et al., 1959). With lower levels of sodium, higher levels of potassium were needed and the reverse was also true suggesting that the role of these ions was to act as an osmoticum; little growth occurred when the calculated total ion concentration was outside the range of 0.6 to 1.2% solution of sodium chloride. However, sodium appeared also to have a specific role in the nutrition of the organism as only poor growth occurred if sodium was omitted regardless of the potassium concentration in the medium. A non-marine photosynthetic bacterium, Rhodopseudomonas spheroides has a specific requirement for sodium. Maximum growth was obtained with 1.74 mol m-3 (40 ppm) sodium (Sistrom, 1960). 4. Fungi There have been few studies on the sodium requirement of fungi. Siegenthaler et al. (1967) showed sodium to be specifically required for the uptake of phosphate in Thraustochytriurn roseurn, a lower marine Phycomycete; its growth rate was greatest with 200-400 mol m-3 (4600-9200 ppm) sodium. Vishniac (1960) also obtained increased phosphorus and oxygen uptake in Thraustochytriumglobosum when receiving 200 mol m-3 (4600 ppm) sodium as sodium chloride. Smaller or negligible increases of phosphorus or oxygen uptake were obtained when the chlorides of lithium, potassium and rubidium and sucrose at the same molarity were supplied. This suggested the specific involvement of sodium in the processes of phosphorus and oxygen uptake. The marine yeast, Candida marina when grown in a carefully purified nutrient medium to remove sodium responded to sodium a t 1% level of significance (W. A. Shipton and K. Watson, unpublished data), (Table VI). Many more lower plants may be shown to have a requirement for sodium at low concentrations when they are examined. There are extra technical difficulties encountered when preparing the organic media needed to support these heterotrophic organisms free of sodium.

149

SODIUM AND PLANT METABOLISM

TABLE VI

Eflect of Sodium and Potassium Salts on the Growth of Candida marina. The Medium Contained 4.8 x lo-? mol I - I Na+ and mol l-1 K+ Before Addition of Specific Ions

Yield Treatment Nil NaCl KCIb KCI + Na,SO, (K+:Na+ = 1:1)

Cations added

(g dry weight/50cm3

(moll-l)

culture medium)

0 6.7 x 1 0 - 3 6.7 x 10-3 13.4 x 10-3

0.047 ba 0.108 a 0.069 c 0.106 a

same letter are not significantly different (P < 0.05). Level of Na+ contamination of KCI was 5 x 10-7 mol I-'. (From unpublished data of W. A. Shipton and K. Watson)

a Figures followed by the b

B.

Atriplex vesicuriu HEWARD EX BENTH.

Before 1954, no serious attempts were known to have been made to remove sodium and chlorine from the environment of plants before adding the treatment salts containing sodium and/or chlorine. It was therefore possible that plants could have a requirement for these elements so small that it had always been satisfied under the conditions in which plants had been grown. The late Professor J. G. Wood suggested using the Australian bladder salt bush (Atriplex vesicuriu) as one species in a trial as it had interesting features including the possession of a bundle-sheath in leaves (Wood, 1925; Black, 1954) and the ability to accumulate large amounts of sodium and chlorine compared to other species growing in similar habitats. During the course of this work, chlorine was demonstrated to be essential for tomatoes by Broyer et ul. (1954) and it was subsequently shown to be essential for a wide range of species (Johnson et ul., 1957; Ozanne et al., 1957). This led to the acceptance of chlorine as an essential micronutrient element for higher plants. During the same period Allen and Arnon (1955) discovered the requirement for sodium as a specific micronutrient by the blue-green alga. In preliminary experiments in an ordinary glasshouse, plants of Atriplex vesicuriu growing in the basal culture solution showed significant increases in dry weight when sodium salts (but not when additional potassium salts in equivalent amounts) were supplied at levels of 0.35, 1.75 and 7.0 mol m-3 of sodium. When the plants were relatively large and had acquired tertiary branches, the leaves of all plants not receiving sodium appeared paler green than those of plants with added sodium. Analysis of plant organs and culture solutions at the end of the experiment showed an increaae of 300pmol sodium above the amount supplied in the culture solutions and seeds. This

150

P. F. BROWNELL

suggested that plants had received sodium, presumably as cyclic salt, from the atmosphere (Brownell and Wood, 1957) (see also Section II.C, this chapter). To preclude this possibility, a cabinet was constructed which was maintained at a slightly positive pressure to prevent entry of dust. Other precautions to prevent contamination are described in Section 1I.C. Bracteoles were removed from seeds, which were then washed in distilled water and germinated on nylon gauze over silica-distilled water; after emergence of the radicle, the distilled water was replaced by basal culture solution of one-fifth full concentration. When 11 to 14 days old, the seedlings which had acquired cotyledons and apical buds were selected for uniformity and transferred to the polythene culture vessels containing two litres of culture solution. Differential treatments were supplied at this stage and the following experiments performed (Brownell, 1965) : I . Efect of Small Graduated Amounts of Sodium on Growth Different treatments were applied to cultures in each of four blocks (Table VIII). The concentration of sodium in the full concentration culture solution due to the sodium contributed by potassium sulphate was reduced from 7.1 to 0.039 mmol m-3 by recrystallizing the potassium sulphate five times. The cultures within each block were placed in random positions at the beginning of each experiment. By the twenty-fifth day after germination plants which had not received sodium sulphate could be distinguished from those which had by their yellow colour and fewer leaves each of smaller area. White necrotic areas appeared along tips and margins of the cotyledons and older leaves on the thirtieth day (Fig. 8). Some plants died by the thirty-fourth day. A plant which showed symptoms just described and another which had died, were examined by plant pathologists at the Waite Agricultural Research Institute for the presence of pathogenic organisms. None were found in these plants. When harvested on the forty-eighth day, plants which had received sodium sulphate appeared markedly different from those which had not, having many more leaves of darker green colour which showed no necrosis (Fig. 9). It should be noted that these symptoms were easily distinguished from those of chlorine-deficiency in the same species. Chlorine-deficiency was manifested by sudden wilting in leaf-tips at a relatively late stage of growth in plants which otherwise resembled normal plants. There was no sign of chlorosis which occurs at an early stage of growth in sodium-deficient plants. The difference between the root systems of plants grown with and without the addition of sodium sulphate was observable at an early stage and was pronounced by the forty-eighth day (Fig. 10). From the results in Table VII obtained when the plants were harvested on the forty-eighth day, the yield is seen to have increased asymptotically with increasing sodium sulphate. As plants which received 0.60mol m-3 of potassium

TABLE VII Yields of Atriplex vesicaria Following the Application of the Sulphates of Sodium and Potassium

Fr wt/vessel (g) Treatment

I No addition I1 0.02 mol m-3 Na2S04 111 0.10 rnol ma Na,SO, IV 0.60 mol m-3 Na,S04 V 0.60mol m-3 K,SO,

Leafblades

Stems and petioles

Roots

0.301 2.101 2.926 2.940 0.436

0.021 0.144 0.224 0.228 0.019

0-221 1.441 2.148 2.940 0.326

Total

Leafblades

Dry wt/vessel (g) Stems and petioles

Roots

Total

0.543 3.686 5.298 6.108 0.781

0-0324 0.2354 0.3468 0.3475 0.0433

0.0042 0.021 2 0.0324 0.0355 Oa03 1

0.0153 0.0913 0.1357 0.1286 0.0206

0.0519 0.3479 0.5149 0.5116 0.0670

All values are the means of yields from four 2-litre cultures of four plants each. The statistical treatment of total dry weight data W E S as follows: I1 > I at 1% level of significance; I11 > I at 0.1 % level of significance; I11 > I1 at 5 % level of significance. From Brownell (1965).

TABLE VIII Fresh and Dry Weight Changes After Recovery of Sodium Deficient Plants of Atriplex vesicaria Following the Application of Sodium Sulphate Treatments

Treatment

I N o sodium sulphate I1 0.10 mol m3 Na2S04 applied on day 16 111 0.10 mol m4 Na2S04 applied on day 31

Dry wt/vessel (g) Stems and petioles Roots

Leafblades

Fr wt/vessel (g) Stems and petioles

Roots

Total

Leafblades

0877 8.139

0.078 1.007

0.671 6.923

1.626 16.069

0-0894 1.0236

00124 0.1619

0.0481 0.4810

0.1499 1.6665

3.713

0.334

2.800

6.847

0.4713

0.0544

0.2133

0.7390

Total

All values are the means of yields of two vessels of four plants each. The statistical treatment of total dry weight data was as follows: 111 > I at 1% level of significance; I1 > I at 0.1 % level of significance; I1 > 111 at 5 % level of significance. From Brownell (1965).

SODIUM A N D PLANT METABOLISM

153

Fig. 8. A plant of Arriplex vesicaria grown in a basal culture solution from which sodium was eliminated. Note the few yellow-green leaves of small area possessing, in some cases, white necrotic patches at their tips and the absence of secondary shoots. This plant had a height of 1.0 inch and was photographed on the 48th day. From Brownell and Wood (1957).

sulphate, a concentration equivalent to the highest concentration of sodium sulphate treatment in their cultures, could not be distinguished from the plants grown in the “no addition” cultures, it was evident that the increase in yield with increasing sodium sulphate was not due to the sulphate but to the sodium of the sodium sulphate treatment. This also showed that the part played by sodium in the nutrition of Atriplex vesicaria could not be performed by additional potassium when supplied in an amount equivalent to the highest sodium sulphate treatment. The lowest sodium sulphate treatment for maximum dry weight production was about 0.2 mmol/Zlitre culture and

154

P. F. BROWNELL

Fig. 9. Comparison between the growth of tops of plants of Atriplex vesicaria growing in the basal culture solution with addition of 0.60 mol m-3 potassium sulphate (left), with no addition (centre) and with 0.02 mol m-3 sodium sulphate (right). The plants had a height of approximately 2.5 cm (left), 2.5 cm (centre) and 5.1 cm (right). From Brownell (1965).

the leaf material contained about 295 pmol/g (6785 ppm) (dry basis). Although these data would be expected to vary markedly according to the conditions of the experiment, the sodium requirements of Atriplex vesicuriu were high in comparison with the requirements of micronutrients by plants of other species (W. R. Meager, see Hewitt (1966) pp. 163-165).

Fig. 10. Comparison between the tops and root growth of plants of Atriplex vesicaria grown in the basal solution which received no addition (left), and 0.60 mol m-s sodium sulphate (right). The heights of the tops of the plants were approximately 2.5 cm (left) and 6.4 cm (right). The photograph was taken on the 48th day. From Brownell (1965).

SODIUM AND PLANT METABOLISM

155

2. Recovery of Sodium-deJicient Plants Following the Application of Sodium Plants, 14 days old, were transferred to culture vessels containing the basal culture solution without added sodium. On the sixteenth day sodium sulphate (0.1 mol m-3) was added to one set of culture vessels and seven days later plants growing in these solutions could be distinguished from the controls by their darker green colour. Signs of deficiency, similar to those described, again appeared in cultures that had not received sodium, and by the thirtyfirst day, when a second set of deficient cultures received a treatment of 0.1 mol m-3 of sodium sulphate, symptoms were severe. Four days after receiving this delayed sodium treatment, plants showed signs of recovery by a progressive change of colour in older leaves (and in some cases cotyledons) from yellow to green; greening commenced at tips and around midribs and gradually spread over the lamina. Plants growing in the set of cultures which received no sodium treatment throughout the experiment became progressively more chlorotic making little further growth. On the other hand, vigorous growth occurred in both sets of cultures which received added sodium. When harvested on the forty-ninth day, the results in Table VIII were obtained. The complete recovery of plants growing in cultures which received a small addition of sodium sulphate (even though they were adequately supplied with sulphate), is convincing evidence for sodium being an essential nutrient element for Atriplex vesicaria.

3. Efects of Lithium, Sodium, Potassium, Rubidium and Caesium on Sodiumdejcient Plants Seedlings were grown by the techniques described above in the previous experiments and transferred to culture vessels on the fifteenth day after germination. On the twenty-second day when the symptoms of sodium deficiency were clearly recognizable, four cultures each containing four plants were harvested. The mean dry weight per culture was 0.0187 f 0.0012 g. On the same day differential treatments shown in Table IX were applied to the cultures of each of four blocks. The concentration of sodium in the culture solution due to the sodium associated with the treatment application is also given and was not greater than 0.035 mmol m-3 (0.0008 ppm) in any of the treatments other than that of sodium sulphate. On the twenty-seventh day, plants which had received the sodium sulphate treatment showed signs of recovery. Plants growing in cultures which received no sodium sulphate became progressively more chlorotic, making little further growth. On the other hand marked growth occurred in the set of cultures which had received sodium. Plants in untreated, and in lithium, potassium and rubidium sulphate treated cultures were indistinguishable. By the thirty-third day, some plants had died in the cultures which had not received sodium. The mean dry

TABLE IX Effects of Equivalent Amounts of the Sulphates of Lithium, Sodium, Potassium or Rubidium When Applied to Cultures of Sodium Deficient Plants of Atriplex vesicaria

Treatment

Sodium/Z-litre culture (CLmol) As impurity of Due to basal culture treatment salts solution

I Control I1 Li,S04 0.10 mol ma

0.14 0.14

I11 Na2S0, 0.10 mol m3

0.14

IV KISOI0.10 mol m-3

0.14

V Rb2S040.10 mol m3

0-14

(No addition) 0.052 (Impurity) 200 (Treatment) 0.017 (Impurity) 0.069 (Impurity)

Total

Yield (Each value is the mean of 4 replications) Mean dry wt/culture (g) LeafStems and Roots petioles blades

Total

0.179 0.163

0.030 0.027

0.061 0.051

0.270 0.241

0.761

0.183

0.288

1-232

0.157

0.169

0.024

0.049

0.242

0.209

0.220

0.045

0.071

0.336

0.14 0.192 200.14

-

Plants were harvested on the44th day. The statistical treatment of total dry weight data was as follows: 111 > I, 11, IV, V at 0-1% level of significance; I, 11, IV, V indistinguishable. From Brownell (1965).

SODIUM AND PLANT METABOLISM

157

weights per culture obtained for each treatment on the forty-fourth day are shown in Table IX. These experiments showed sodium to be an essential nutrient element for Atriplex vesicaria according to the criteria of Arnon and Stout (1939). Plants, protected from atmospheric contamination of sodium, grown in culture solutions containing only small amounts of sodium showed characteristic deficiency symptoms by the yellowing of their leaves and development of white, necrotic patches on their tips and margins. Plants developed few or no secondary shoots and in some cases died at an early stage; no pathogenic organisms could be found in their tissues. The first criterion of Arnon and Stout (1939) that “a deficiency of it makes it impossible for the plant to complete its life cycle”, was satisfied. It was also shown that only sodium of the group 1 elements, lithium, potassium and rubidium brought about the recovery of sodium-deficient plants. It therefore appeared that the second criterion that “such deficiency is specific to the element in question and can only be prevented by supplying this element”, has almost certainly been satisfied. The fulfilment of the third of the criteria, that “the element is directly involved in the nutrition of the plant quite apart from its possible effects in correcting some unfavourable microbiological or chemical condition of the soil medium” is difficult to achieve. However, plants in these experiments were grown in solution culture so that the sodium supplied in the treatment was more likely to have affected the nutrition of the plant directly than if more complicated media had been used. Even so the possibility still exists that the sodium corrected an unfavourable chemical or microbiological condition of the culture solution, and this possibility cannot be dismissed until a specific role of sodium in the metabolism of the plant has been demonstrated. These experiments were of short duration so that the possible complicating effects due to the depletion of nutrients in the culture solution were avoided, and the risk of heavy infection of the cultures by organisms such as algae, fungi and bacteria was minimized. C. OTHER SPECIES HAVING THE C4 DICARBOXYLIC PHOTOSYNTHETIC PATHWAY

The finding that sodium is essential for Atriplex vesicaria suggested that sodium might be an essential micronutrient element for other higher plants. Trials were carried out on a diverse selection of species to find out if the requirement for sodium was general for higher plants (Brownell, 1968). Species were chosen for these trials after consideration of the following: suggestions in the literature of responses to sodium; suitability for culturing; low sodium content per seed ;taxonomic or ecological relationship with Atriplex vesicuriu or unusual features in common with Atriplex vesicariu such as vesicles arising from the leaf epidermis. The techniques used to grow plants in a low sodium environment are

158

P. F. BROWNELL

described in Section 1I.C. The culture solution had a sodium concentration of about 0.08 mmol m-, (0.0018 ppm) derived from impurities of the salts and silica-distilled water. Seeds were thoroughly washed in distilled water before germination. The dry weight yields of species grown with and without the addition of sodium to their cultures and the significance of their responses to sodium are shown in Table X. The most striking feature of the results of these experiments was the extremely restricted group of species for which sodium is apparently essential. Even within the genus Atriplex, there was remarkable specialization in the requirement of sodium. Only the Australian species of Atriplex developed symptoms due to sodium deficiency. One other species of Atriplex, Atriplex hortensis significantly increased its yield on receiving sodium but it did not develop sodium-deficiency symptoms. Species of other genera of Chenopodiaceae viz. Beta, Kochia, Exomis and Chenopodium did not appear to require sodium. Even Beta vulgaris (sugar beet) which has been reported to increase its yield when supplied with sodium salts (Adams, 1961 ; Harmer and Benne, 1945) showed no significant responses in these trials. Possibly, very much larger amounts of sodium are needed to induce yield responses in sugar beet than in Atriplex, suggesting that the roles of sodium in the nutrition of sugar beet and species of Atriplex responding to sodium differ. Hordeum vulgare cv. “Pallidium” which showed a small significant response to sodium is a salt-tolerant variety of barley. Significant responses to sodium by Lycopersicum esculentum cv. “Grosse Lisse” were not obtained in these trials whereas Woolley (1957) reported a 12% increase in the mean dry weight of tomatoes (P < 0.01) on the addition of 1 mol m-, NaCl to the basal solution which contained 0.25 mmol rn-, (0.0058 ppm) sodium as an impurity. However, no symptoms due to a deficiency of sodium were observed. Other species which did not respond to sodium either have no requirement for sodium, or they require it in amounts so small that the plants obtained adequate sodium under the conditions of the experiment, i.e. less than approximately 0.08 mmol rn-, (0.00184 ppm) sodium in their cultures. Subsequently, with the discovery of the C , dicarboxylic photosynthetic pathway (Hatch and Slack, 1966; Hatch and Slack, 1970; Osmond, 1969a; Osmond, 1969b; Slack and Hatch, 1967), it was observed that the species of Atriplex which required sodium had characteristics of plants with the C, pathway whereas those not requiring sodium had features of the C , photosynthetic pathway. This suggested that all species with the C, pathway might require sodium as a micronutrient and that all species with C, photosynthesis might have no requirement. In addition to the ten Australian species of Atriplex which have been shown to need sodium (Table X), the requirement for sodium has now been demonstrated in a further nine C, species belonging to five families (Gramineae, Cyperaceae, Amaranthaceae, Chenopodiaceae and Portulacaceae) but no re-

SODIUM AND PLANT METABOLISM

159

sponses to sodium were detected in two C, species of Gramineae, Poa pratensis and Panicum milioides (Table XI). The latter species has certain features of a C, plant including the possession of Kranz anatomy and reduced photorespiration (Quebedeaux and Chollet, 1977). In Table XII, the responses of 37 species to 0.10 mol m-, of sodium chloride or sodium sulphate are summarized. Only certain species of Echinochloa, Cynodon, Chloris, Panicum, Eleusine (Gramineae), Kyllinga (Cyperaceae), Amaranthus (Amaranthaceae), Atriplex, Kochia, Halogeton (Chenopodiaceae) and Portulaca (Portulacaceae) were shown to respond to sodium by marked increases in dry weight. Plants of these species grown in “sodiumfree” cultures had the characteristic leaf lesions of chlorosis and necrosis observed previously in sodium-deficient Atriplex vesicaria. These species have characteristics of plants with the C, dicarboxylic pathway which include the “Kranz type” specialized leaf anatomy (Moser, 1934; Black, 1954; Hatch and Slack, 1970; Smith and Brown 1973), a high activity of phosphoenol pyruvate carboxylase (Hatch and Slack, 1970), a low CO, compensation value and a reduced 13C discrimination (Smith and Epstein, 1971). The known C, pathway characteristics for each species are shown in Table XII. No data relating to the C, pathway were available for certain species, including Atriplex semilunalaris, A . lindleyi, A. angustifolia, A . albicans, Exomis axyrioides and Aster tripolium as their responses to sodium had been determined before the discovery of the C, dicarboxylic pathway and the material has not been available since. Three other species, barley, Atriplex hortensis (Brownell, 1968) and tomato (Woolley, 1957) responded to sodium but only marginally. The plants of these species in “sodium-free’’ cultures did not exhibit sodium-deficiency lesions. Within Atriplex and Kochia which include both C, and C, species, only the C, species have been shown to respond to sodium; no decisive response was obtained in the C, species. It appears from these data that species having characteristics of the C, photosynthetic pathway generally have a requirement for sodium. D. RESPONSE BY A SPECIES HAVING CRASSULACEAN ACID METABOLISM

The finding that species having the C, photosynthetic pathway require sodium as a micronutrient (Brownell and Crossland, 1972) suggested that CAM species might also require small amounts of sodium due to the similarities of their mechanisms for CO, fixation. In preliminary experiments conducted by Brownell and Crossland (1974) in a greenhouse under long-day conditions, Bryophyllum tubijlorum failed to respond to small amounts of sodium added to the cultures from which sodium had been carefully eliminated. The absence of a response to sodium was attributed to the possibility that the plant already contained sufficient sodium for normal growth. However, under these conditions it appeared

TABLE X Yield Responses by Various Higher Plants to Sodium

Specie9 Gramineae Hordeum vulgare L. cv. “Pallidium” (1) (barley) (2) Chenopodiaceae Chenopodiumcapitatum (L.) Aschers Beta vulgaris L. (sugar beet) Atriplex nummularia Lindl. (Old man, Giant salt bush) Atriplex paludosa -R.Br. (Marsh salt bush) Atriprex quinii Fv.M. Atriplex semibaccata R.Br. (Berry salt bush) Atriplex injlata Fv.M. (1) (2) Atriplex leptocarpa F.v.M. Atriplex lindleyi Moq. Atriplex spongiosa Fv.M. (Pop salt bush) Atriplex semilunalaris (Aellen) Atriplex hortensis L.var. atrosanguinea Garden orache

Yield (dry weight per plant, g) Sodium No 0.2 mmol supplied Harvested addition Na,SO, (days after (days after to per 2-litre germination) germination) culture culture 11 10

26 45

0.77 3.82

4.01

17 11 15

48 61 43

12.25 3.86 0.166

14.37 5.07 0.830

10 10 15

56 51 42

0.215 0.116 0.104 0.202 0.036 0.050 0093 0.570 0.098 2-849

2.789 0.815 0.653 9.745 1.865 1.329 0.560 12.472 0.526 3.677

-

-

8 14 14 18

65 55 54 41

-

-

12

39

-

Significance of differenceb

Remarks

* N.S. N.S.

***

* * ** *** *** ** *

*** *** *

75 pg Na seed-’

Atriplex angustiflia Sm. Atriplex glabriuscula Edmondston Atriplex albicans Ait.

(1) (2)

Kochia pyramidata Benth. Exomis axyrioides Fenzl. ex Moq. Cruciferae Brassica oleracea L. cv. “Savoy” (Cabbage)

Leguminosae Triflium repens L. cv. “Palestine” (White Clover) Solanaceae Lycopersicum esculentum Mill. cv. “Grosse Lisse” (Tomato) Compositae Lactuca sativa L. cv. “Great Lakes” (Lettuce) Aster tripolium L.

(1)

(2) (1) (2) (3)

14 19 23 17

-

-

22

54

0.531 24.804 8.818 9.237 35.390 0.864

17

44

14.244

17.766

N.S.

19

43

2-299

2-897

N.S.

20 16

47 42

4.84 6.93

4.76) 7.83

N.S.

17 14 9 17

46 32 37 50

3.871 0.822 2-230 0.282

45 43 49 49

a Names given in this table are those under which the seeds were received. Levels of significance: Not significant-N.S.; 5 % * ; 1%**; 0.1 %***. From Brownell (1968).

0.377 24-100 9.401 33.990 0.925

6.1 13 1-228

N.S. N.S. N.S. N.S. N.S.

N.S.

2444 0-458

N.S.

1 seed contained 0.25 pg Na

1 seed contained 0.03 pg Na after washing 1 seed contained 0-1pg Na

TABLE X I Responses of VariousPlants to Sodium

Age at harvest (days)

Gramineae Poa pratensis L. (Kentucky blue grass)c Echinochloa utilis Ohwi et Yabuna (Japanese mil1et)C Cynodon dactylon L. (Bermuda grass)c Chloris barbata Swartz (Purple top Ch1oris)d Chlorisgayana Kumpth (Rhodes grass) Panicum maximum Jacq. (Guinea grass) Panicum milioides Nees ex. Trin. Eleusine indica(L.) Gaertn. (Crowsfoot grass)d Cyperaceae Kyllinga brevifolia Rottb.c Amaranthaceae Amaranthus tricolor L. cv. “Early Sp1endour”c Chenopodiaceae Kochia childsii H0rt.c Portulacaceae Portulaca grandijlora Hook (Rose moss)c

a b

Lesions in plants not receiving sodium

No addition

0.1 molm3 NaCl

Significance of difference

Yield

52

None

0.0236

0.0206

NSb

22

Chlorosis and necrosis

0.404

0.713

0.1

47

Chlorosis

0.178

0.337

1

23

Chlorosis

0.071

0.239

0.1

21

Chlorosis

0.018

0.271

0.1

17

Chlorosis

0.335

0.427

50

None

0539

0533

NSb

21

Chlorosis

0.221

0.435

0.1

57

Chlorosis

0.628

1*245

1

40

Chlorosis

0.884

2-099

0.1

21

Chlorosis

0.125

0.442

1

29

Chlorosis, necrosis and failure to set flower

0.242

0-789

1

Names given in this table are those under which the seeds were received. Not significant. Data from Brownell and Crossland (1972). Data from unpublished work of T.S. Boag.

TABLE XI1 Sodium Requirement in Relation to C, Pathway Characteristics Yield

Species Gramineae Hordeum vulgare L. cv. “Pallidium” (barley) Poa pratensis L. (Kentucky blue grass) Echinochloa utilis L. Ohwi et Yabuno (Japanese millet) Cynodon dactylon L. (Bermuda grass) Chloris barbata Swartz (purple top chloris) Chloris gayana Kumpth (Rhodes grass) Panicum maximum Jaw. (Guinea grass) Panicum milioides Nees ex. Trin. Eleusine indica (L.) Gaertn. (Crowsfoot grass) Cyperaceae Kyllinga brevifolia Rottb.

Lesions in plants not receiving sodium None

None

0.1 mol m-3 Significance of No NaCl addition or Na$04 difference Reference C4 pathway Reference Probable (g dry Wplant) O/O / characteristics‘ pathway

0.77 3-82 0-024

Cs

N.S.

N, €3 N

f

0.021

C

c 3

K

C

c 4

C

c 4

5

Chlorisis and necrosis Chlorosis

0.404

0-713

0.1

0.178

0.337

1

d

K, H14C4

Chlorosis

0.071

0.239

01

a

K

Chlorosis

0.108

0271

0.1

a

K, P. a r b

Chlorosis

0.335

0.427

N.S.

a

P. a r b

None Chlorosis

0.539 0.221

0.533 0.435

N.S. 01

a a

L, L”C4 K

Chlorosis

0.628

1.1245

1

d

K

continued

TABLE XI-continued Yield

Species Amaranthaceae Amaranthw tricolor L. cv. “Early Splendour” Chenopodiaceae Chetwpodium capitatum L. Aschers Beta vulgaris L. (sugar beet) Atriplex numularia Lindl. (oldman, giant saltbush) Atriplex paludosa R. Br. (marsh saltbush) Atriplex quinii Fv. M. Atriplex semibaccata R.Br. (berry salt bush) Atriplex inflata Fv.M. Atriplex Ieptocarpa Fv.M.

Lesions in plants not receiving sodium Chlorosis and necrosis

0.1 mol m“ Significance No NaCl of C4pathway Probable addition or Na2S04 difference ts dry wt/plant) % Reference characteristicsl Reference pathway 0.884

2.099

0.1

d

K

C

N.S.

C

N

n

None

12-25

None

3.86 0.166

5.07 0.830

N.S.

C

0.1

C

0.215

2.789

5

C

N, H, L14C4 K, L, Hl4C4, L13C K

0.116

0.815

5

C

K

0.104

0-653

1

C

K, L W

1, n

0.202

9.745

01

C

K, H1*C4

j, m

0.050

1.329

1

C

K

Chlorosis Chlorosis and necrosis Chlorosis and necrosis Chlorosis and necrosis Chlorosis and necrosis Chlorosis and necrosis

14.37

f j, 1, m

Atriplex spongiosa Fv.M. (pop salt bush) Atriplex semilunalaris Aellen Atriplex lindleyi Moq. Atriplex vesicaria Heward ex

Benth. (bladder salt bush)

Chlorosis and necrosis Chlorosis and necrosis Chlorosis and necrosis Chlorosis and necrosis None

Atriplex hortensis L. var. atrosanguineae(garden orache) Atriplex angustifolia Sm. None Atriplex glabriuscula Edmonton None None Atriplex albicans Ait. Kochia pyramidata Benth. Kochia childsii Hort.

None Chlorosis and necrosis Exomis axyrioides Fenzl ex Moq. None Halogeton glomeratus (Bieb) Smaller, curved Meyer leaves, tendency to wilting Cruciferae Brassica oleracea L. cv. “Savoy” None (cabbage)

0.570

12.472

0.1

C

K, L, P. w b

0.098

0.526

0.1

C

Unknown

0.093

0.560

5

C

Unknown

0.013

0-129

0.1

b

K, H14C4,P. carb, L13C

2.849

3.677

5

C

0531 24.804 8.818 9.237 35.390 0.125

0.377 24 100

N.S. N.S.

N, H, L14C4, H13C Unknown N, H, H W

9.401 33.990 0.442

N.S.

Unknown

N.S. 1

N, L14C4 K, L, L13C

0.864

0.925 0-800

N.S. 1

C

0.285

n

Unknown K

14.244

17.766

N.S.

C

N

j, 1

c 4

n

c 4

...

G continued

TABLE MI-continued

Lesions in plants not receiving sodium

Species

Yield 0.1 mol m3 Significance No NaCl of C, pathway Probable addition or NazS04 difference Reference characteristics1 Reference pathway (I3 dry W P l d %

LegUminOSae

Trifolium repens L.cv. “Palestine” (white clover) Solanaceae Lycopersicum esculentum Mill. cv. “Grosse Lisse” cv. “Marglobe” Compositae Lactuca sativa L.cv. “Great

None None None None

Lakes” Aster tripolium L.

None

2.299 4.84 6-93 13.76

2-897

N.S.

C

N

4*76} 7.83

N.S.

C

N

1

P

N

N.S.

a

N

N.S.

a

Unknown

15*402

}

3.871 0.822 2.230 0.282

6.113 1*228 2.444 0.458

0.242

0.789

Portulacaceae Portulaca grand’ora (rose moss)

Hook

Chlorosis no flowers

1

K

Leaf anatomy: N, normal (bifacial); K, Kranz. CO, compensation: L, low: H, high. 14Cin C, compounds (malate, aspartate): Ll4C4,low; H14C4 high. Phosphoenol pyruvate carboxylase activity: P. carb, high. 13Cdiscrimination: LW, low; H T , high. * Cultures contained 1 mol ma NaCI. g Guttierrez et al. (1974) 1 Smith and Epstein (1971) a T. S. Boag, unpublished work m Tregunna et af.(1970) b Brownell and Wood (1957); Brownell (1965) h Hatch et af.(1975) c Brownell (1968) i Osmond (1969a) n Welkie and Caldwell(l970) o Williams (1960) j Osmond (1969b) d Brownell and Crossland (1972) p Woolley (1957) k Quebedeaux and Chollet (1977) e Chen et al. (1971) f Downton and Tregunna (1968)

Fig. 11. Comparison between plants of Echinochloa utilis (left), Amaranthus tricolor (centre) and Kochia childsii (right) which received no addition (-Na) and 0.10 rnol m3 sodium chloride (+Na). From Brownell and Crossland (1972).

168

P. F. BROWNELL

possible that the plants would have assimilated COz almost entirely by the C, option with little involvement of the CAM option. Ting (1970), reviewing previous literature, concluded that CAM activity is most pronounced when night temperatures are low and day temperatures are high and that short-day conditions were also conducive to CAM. Brulfert el al. (1973) found the activities of the enzymes specifically involved with CAM metabolism in Kalanchoe blossfeldiana, a short-day plant, to be phytochrome-controlled. In short days there was a progressively rapid increase in the activity of all the enzymes of the CAM pathway whereas in long days or in short days with nights interrupted with red light, the pathway was not operative, presumably due to low activity of PEP carboxylase. In several experiments, Brownell and Crossland (1974) found that growth of Bryophyllurn tubijlorum was greatest under conditions of long days and small diurnal temperature variation. Under these conditions there was no growth response to sodium. When plants were grown under conditions of short day length and large diurnal temperature variation, significant responses to sodium (0.1 mol m-3 (2.3 ppm)) were obtained although the overall growth was still less than that of the former. The results of these experiments (Table XIII, Fig. 12) suggested that sodium is not required by plants grown under conditions thought to be conducive to the C, option of photosynthesis, i.e. under long day periods. Plants grew more actively under these conditions but did not respond to sodium. Plants taken from the same population, however, when grown under conditions conducive to CAM photosynthesis, i.e. under conditions of short days and large diurnal temperature variation, responded significantly to small amounts of sodium. Under these conditions of growth, metabolic processes common to those operating in C4 dicarboxylic photosynthesis are active and this observation would suggest that sodium is involved in this area of metabolism both in species having the C, dicarboxylic photosynthetic system and in members of CAM carbon fixation. This supported the hypothesis that sodium may be required for the primary dicarboxylic C0,-fixation system characteristics of C, and CAM plants. E. DISCUSSION

There is little doubt that sodium is specifically required in small amounts for the blue-green algae, Anabaena cylindrica and Anacystis nidulans. It may be generally essential for blue-green algae but this is still open to conjecture as few other blue-green algae have been tested critically for their sodium requirement. Sodium, at similar levels, is also specifically required by Atriplex vesicaria according to the criteria of Arnon and Stout (1939) (Brownell, 1965) and evidence from further experiments (Brownell and Crossland, 1972) now makes it likely that all plants with the C4dicarboxylicphotosynthetic pathway

169

SODIUM AND PLANT METABOLISM

TABLE XI11 Responses by Bryophyllum tubiflorum to NaCl Grown Under Different Conditions Mean dry wt per plant Conditions of growth - .~

No addition

0.1 mol m-3

NaCl

-~

Short days Long days-short days Long days

0.123

g

Significance of difference

% <1

0.175

0.176 0.218

<1

0.548

0.521

N.S.

-.

Plants were transferred on day 24 from seedling cultures to culture vessels and received differential NaCl additions under the following conditions. Short days: in artificially illuminated growth cabinet with an 8 h light period at approximately 2800 ft-c and 16 h dark period. Temperature during the light period was 33°C and in the dark period 13°C. Long days-short days: in a naturally illuminated cabinet with the normal day length of 11 h extended to 16 h per day by a 100 w incandescent lamp giving an intensity of approximately 100 ft-c. The overall temperature range was 15" to 38°C. On day 65, cultures were transferred to short day conditions in the artificially illuminated growth cabinet. Long days: in the naturally illuminated cabinet with the normal day length of 11 h extended to 16 h per day. Plants were harvested on day 100. From Brownell and Crossland (1974).

have a requirement for sodium although it has been shown to be essential for species in only five families of C4plants to the present time. Responses to sodium at low levels were also obtained in a CAM species, Bryophyllum tubgorum only when grown under certain conditions i.e. with short days and high diurnal variation in temperature but not with long days with low diurnal temperature variation (Brownell and Crossland, 1974). The question of whether or not sodium is essential in small amounts for species with the C, photosynthetic pathway has not been resolved. It can be stated, however, that if the C , species examined by Brownell (1968) do require sodium, it would be at a concentration less than 0-08 mmol m-3 (0.0018 ppm) since the plants grew normally in this purified culture solution. On the other hand, plants with C4photosynthesis required a concentration of oodium of about 0.1 mol rn-, (2.3 ppm) for optimal growth-about 1250 times the concentration of that remaining as an impurity in the basal culture solution which supported optimal growth of the C , species. The possibility that sodium is essential for all species, but in extremely small amounts, cannot be disproved until the amount of sodium available to the plant is reduced to a minute quantity ;Steinberg (1937) estimated tentatively (without divulging the basis of the estimation) that if an element is present in greater quantities than 1 part per billion (1 ppb (U.S.) = 0401 ppm), one cannot be reasonably certain that it is not essential. Hewitt and Smith (1975) consider that the maximum concentration of molybdenum required by Scenedesmus obliguns with ammonia or urea as nitrogen sources could be as low as O~OOOOOO1ppm. On this basis, since the concentration of sodium used in this

170

P. F. BROWNELL

Fig. 12. Comparison between plants of Bryophyllum tubiflorum receiving the following treatments. Left to right: no addition, 0.1 rnol m3 NaCl (16-h light period; overall temperature range, 15"to 38"); no addition, 0.1 mol m-3 NaCl (8-h light period; light ternperature 33", dark temperature 13"). From Brownell and Crossland (1974).

study was approximately 0.08 mmol m-3 (0.0018 ppm) and the possibility existed of sodium also reaching the plant from other sources, it would have been necessary to further reduce the level of sodium in the culture solution before sodium could have been considered non-essential for the species examined in this work. The discoveries that plants with C , photosynthesis require sodium and that CAM plants (employing under certain conditions photosynthetic machinery resembling that of C4 plants) may respond to sodium suggest a common metabolic basis not shared by C , species to explain their sodium requirement. It, therefore, seems unlikely that C3 species have a requirement for sodium even at a minute concentration. It is not possible to explain the sodium requirement by the blue-green algae in similar terms as there is little evidence of them having the C , dicarboxylic carbon fixation pathway except under specialized conditions of high temperature and CO, tension which do not normally occur (Dohler 1974). Sodium-deficient cells of Anabaena cylindrica growing in nitratecontaining solutions have much higher nitrate reductase activity than that of normal cells (Brownell and Nicholas, 1967). No such effect of sodium deficiency on nitrate reductase activity was observed in Echinochloa utilis, a C , plant (P. F. Brownell, unpublished data). Thus, no common basis has been discovered to explain the sodium requirement of the blue-green algae on the

SODIUM AND PLANT METABOLISM

171

one hand and the C4 and CAM species on the other. They may require sodium for entirely different functions. In conclusion, it appears that sodium is almost certainly only essential for a t least certain blue-green algae, C, and CAM species but not for C, species of higher plants. This would make sodium a unique essential element in that it would only be needed by certain species. Reports have been made of sodium chloride treatments having changed the photosynthetic options in certain species. It was found (Winter and von Willert, 1972) that Mesernbryanthenum crystallinum, normally a C, species, grown in the presence of 350 mol rn-, sodium chloride, showed typical gas exchange reactions of CAM plants, exhibiting a clear CO, uptake in the dark accompanied by an increase in malate content of leaves. Neither CO, uptake in the dark nor increase in malate content was found in plants grown in the absence of sodium chloride. It appeared to undergo typical C, photosynthesis under these conditions. The leaves of sodium chloride treated plants also underwent cytological changes by the formation of new vacuole-like spaces under the chloroplasts between the plasmalemma and cell wall (von Willert and Kramer, 1972). These changes in metabolism and leaf cytology occurred following the application of an extremely high concentration of sodium chloride (350 mol m-,)). This concentration would cause water-stress, one of the factors proposed by Osmond (1975, 1976) and Ting (1970, 1976) that induces the CAM option. It is probable that a similar water stress from droughting treatments would have given a similar response. It seems most unlikely that this is a specific response to sodium. Sodium chloride (50 mol rn-, (1 150 ppm Na)) was claimed to have affected the balance between C, and C, pathways of carbon fixation in young leaves of a grass, Aeluropus litoralis Par1 (Shomer-Ilan and Waisel, 1973). This grass normally fixes CO, by the C, photosynthetic pathway, having no detectable PEP carboxylase activity, but when exposed to 50 mol rn-, sodium chloride solution, a high level of PEP carboxylase activity was observed and further evidence for the fixation of CO, by the C, pathway was the I4C labelling of aspartate in leaves fixing 14C02in the light. The C, species, Zea mays and Chloris gayana were not consistently affected by the 100 mM sodium chloride treatment. It seems likely that the high levels of sodium chloride involved in this response were also due to water stress imposed on the plants and not to a specific effect of the sodium ion. Kennedy (1977) observed similar effects in the C, plants, Zea mays and Portulaca oleracea, from NaC1-, polyethylene glycol- or naturally ind uced water stress. With increased water stress, irrespective of its cause, the percentage of C, acids decreased and the C, acid percentage increased. The CO, compensation point increased and the photosynthetic rates were reduced. Clearly the effects of the NaCl treatment could not be ascribed to the sodium it supplied.

172

P. F. BROWNELL

IV. METABOLIC AND PHYSIOLOGICAL EFFECTS O F SODIUM AT LOW CONCENTRATIONS A. GENERAL STRATEGIES

The almost complete lack of information on the involvement of sodium at low concentrations in any biological system makes it difficult to find effective approaches to defining its role in plants for which it is essential. One approach has been to look for an early response by sodium-deficient plants to sodium, a response that precedes the obvious signs of recovery shown by the greening of chlorotic leaves and increased growth. A metabolic response to sodium detected near the beginning of this period may represent a primary step in recovery and hence contribute to an understanding of the function of sodium in the nutrition of the plant. The cascade of measurable responses which follow in the longer-term ranging from increased production of chlorophyll, changes in the soluble and insoluble nitrogen fractions and increases in carbohydrate content are probably indirect effects of the sodium treatment. An early response detected in the C, species of Atriplex was an increased rate of respiration which occurred within a few hours of applying sodium whereas dry weight responses were observed only after six days (Brownell and Jackman, 1966). This response was studied in detail and is described later in this section. A further approach was to study metabolic responses to sodium in the blue-green alga, Anabaena cylindrica (Brownell and Nicholas, 1967). This alga has a specific requirement for sodium and it seemed likely that the function of sodium in the alga could be similar to that in Atriplex. The alga has advantages over higher plants for metabolic studies in having a short generation time, being easy to sample and manipulate in metabolic experiments. Following the discoveries of the correlation between the possession of the C, pathway and the requirement for sodium (Brownell and Crossland, 1972) and the response to sodium by a plant undergoing CAM (Brownell and Crossland, 1974), it appeared likely that sodium might play a role within the C, dicarboxylic acid system common to C, and CAM but not to C, plants. This area of metabolism is being investigated currently in this laboratory. B.

Anabaena cylindrica

1. Nitrogen Metabolism

It was found that Anabaena cylindrica had a higher requirement for sodium when grown with nitrate than without combined nitrogen (Fig. 7), (Brownell and Nicholas, 1967). When nitrate was supplied, nitrite increased markedly in sodium-deficient cultures as shown in Fig. 13. The addition of sodium to deficient cultures, after 45 h growth, depressed nitrite production after a further 40 h period, compared with the deficient cultures.

SODIUM AND PLANT METABOLISM

0

20

40

60

173

80 100 120 140 160 180 200 220 240 260 Time (hours 1

Fig. 13. Effect of sodium on nitrite production by Anabaena cylindrica grown in culture solution containing 10 r n M KNO,. No addition, o--O--o; 4.0 mol m-3 NaCL at onse tof experiment, @--to;4.0rnol m-3 Na CI added to deficient cultures 45 h after inoculation, O.... ....0. From Brownell and Nicholas (1967).

Fig. 14. Eighteen day old cultures containing nitrate. From left to right: dark culture which received 4 mol m3 NaCl at inoculation; dark culture which received 4 rnol m-3 NaCl 7 days after inoculation after washing and resuspending in a fresh medium; pale culture which received 4 mol m-3 NaCl without washing; pale sodium-deficient culture. From Brownell and Nicholas (1967).

The effect of washing deficient cells in water, to remove nitrite, on their subsequent recovery on adding sodium is shown in Fig. 14. Unless nitrite was removed, the addition of sodium did not reconstitute chlorophyll and phycocyanin. Graded amounts of nitrite when added to normal cultures resulted in a chlorosis of the cells within a few days, as shown in Table XIV.

174

P. F. BROWNELL

TABLE XIV Eflect of Nitrite on Chlorophyll Content of Anabaena cylindrica Final concentration of nitrite in the culture (mmol ma)

Treatmenta Normal culture Normal culture + 0-2mol m3 KNO, Normal culture + 0.4 mol ma KNOa Normal culture + 1.0 mol m-3 KNO, Normal culture + 3.0 mol m-3 KNO, Sodium-deficient culture Harvest 15 days after inoculation.

Chlorophyll/culture (mg)

235 910 1050

120

2750

500

5100

370

1700

500

a Nitriteadded to cultures beforeinoculation. Normalculturescontained 4mol m-s NaCI.

From Brownell and Nicholas (1967).

TABLE XV Effect of Sodium on Nitrate Reductase Activity in Cell Extracts of Anabaena cylindrica Grown in Culture Solution Containing Nitrate Amounts of NaCl (mol m3 culture solution) None 0.004 0.4 4.0

Nitrate reductase activity Enzyme activity Dry wt Specific activity (total unitsal (g/100 cm3 (unitsalg culture) dry weight) 100 cm*culture) 173 139 51 36

0.040 0.067 0.053 0.093

4300 2080 960 385

Cultures were harvested 13 days after inoculation. Cells broken by ultrasonication for enzyme assay. a Enzyme unit (nmol NO,- formed h-l). From Brownell and Nicholas (1967).

Extracts of normal cells and of those given suboptimal amounts of sodium were assayed for nitrate reductase and the results are given in Table XV. The enzyme had markedly increased activity when the element was limiting. This accounted for the accumulation of free nitrite in deficient cultures which in turn resulted in a chlorosis of the cells. The effect of adding sodium aseptically in vivo to sodium-deficient cells is shown in Table XVI. It is clear that the addition of sodium reduced the enzyme activity to normal levels. When chloramphenicol was added at the same time as sodium to sodium-deficient cells, nitrate reductase was not repressed to normal levels (Fig. 15). This

175

SODIUM AND PLANT METABOLISM

TABLE XVI Nitrate Reductase Activity of Extracts of Sodium Deficient and Normal Cells of Anabaena cylindrica and those Recovering from the Deficiency ~~

~

~~-

Enzyme activity Total protein Specific activity (mg/100 cm3 of enzyme (total unitsal culture) (unitsa/mgprotein) 100 cm3culture)

Cultural treatment

~-

-

~~~

~~~~~

~

Complete Omit sodium Sodium added to deficient cultures, 50 h after inoculation and enzyme assayed 135 h later

540 10 800 1240

~~

23.6 12.6 20.0

11.5 8 60 31

~-

a Enzyme unit (nmol NO,- formed h-l). P. F. Brownell and D. J. D. Nicholas (unpublished

data).

suggests that a newly-formed protein factor(s) dependent on sodium may exert a control on the enzyme. Not only was the reduction of nitrate to nitrite enhanced by sodium deficiency but also the rates of incorporation of nitrogen-1 5 , labelled nitrateammonia and 14C-glutamate into protein were greatly increased as shown in Table XVII. Thus a sodium-deficiency appears to decontrol the assimilation of nitrate via nitrite, ammonia and glutamate into cell-protein. This was also found for the incorporation of 15NH, and 14C-glutamate by sodium-deficient cells grown without combined nitrogen and therefore relying solely on nitrogen gas as a nitrogen source. Smith (1977) by observing the rates of nitrate disappearance also found the rate of nitrate reduction by Anabaena cylindrica to be greater in sodiumdeficient than in sodium-sufficient cultures both on a protein and on a culture basis (Fig. 16). The sodium-deficient cultures with nitrate supplied (Fig. 17) or without combined nitrogen (Fig. 18) had less cellular nitrogen but increased extracellular organic nitrogen per culture compared to normal cultures. There was a greatly increased amount of extracellular organic nitrogen on a protein basis in sodium-deficient compared to normal cultures. Although the findings of Smith (1977) and Brownell and Nicholas (1967) agree on the enhanced rate of nitrate utilization in the sodium-deficient compared to normal cells, Smith (1977) did not observe the accumulation of nitrite in sodium-deficient cultures reported by Brownell and Nicholas (1 967). This anomaly could be explained by the system operating differently due to differences between the growth conditions of these studies or a possible physiological difference between the strains of algae used. However, it is apparent that nitrate reduction was accelerated under conditions of sodiumdeficiency in both studies.

140C

120c

l0OC

-

lo

E -

-=? E

80C

0

=

r

C

e

60C

e

d

s 400

200

I

0-

0

I

I

I

I

I

50

70

100

150

200

Time from inoculation (hours)

Fig. 15. Effects of sodium and chloramphenicol on nitrite production by Anabaena cylindrica grown with 10 mM KNOB. 4 mol m4 NaClNo sodium supplied Chloramphenicol (0.005 mg ~ m - (A) ~) From Brownell and Nicholas (1967).

-----

SODIUM AND PLANT METABOLISM

177

Time (days)

Fig. 16. Nitrate reduction by Anabaena cylindrica grown with 10 mol m-3 KNO,. N o added sodium (0) per culture; (A) per mg protein. 0.26 mol m-3 NaCl (0)per culture; ( A ) per mg protein. From Smith (1977).

Brownell and Nicholas (1967) found that 15N2fixation into cell protein was reduced in sodium-deficient cells compared with the normal ones (Table XVII). Ward and Wetzel (1975) observed decreased rates of acetylene reduction in cells grown in cultures to which no sodium was added. Smith (1977) also found the rate of acetylene reduction to be lower in sodiumdeficient than normal cultures (P < 0.01) but on a cellular basis there was no significant difference. Cultures with added nitrate had decreased rates of acetylene reduction (Fig. 19). These observations suggest that sodium is required for the reduction of nitrogen gas to ammonia assuming the latter to be a key intermediate in nitrogen-fixation. The addition of either ammonia, glutamine, asparagine, citrulline or ornithine a t low levels (0.1 mM) to sodium-deficient cultures reduced nitrite content within 20 h. Ammonia, arginine, citrulline and ornithine were the most effective as shown in Table XVIII. Similar effects of these compounds on nitrate reductase have been reported in isolated tobacco cells (Filner, 1966). No similar effect of sodium-deficiency in decontrolling nitrate reductase in Echinochloa utifis (a C, species) could be detected.

1 8

"1

6

"

1-4

13-

12I

11-

-E 100

i

m

2

2 09-

-b 2 - 0 8(Y

~ e07-

w

0 403-

1

-. 0

I

2

4

6

8

1

0

1

2

Time (doys)

Fig. 17. Cellular and extracellular nitrogen of Anabuenu cylindricu grown with 10 mM KNO,. Cellular nitrogen of 12-day old culture without added sodium (open bar). Cellular nitrogen of 12-day old culture with 0.26 mol m4 NaCl (closed bar). Extracellular nitrogen : No added sodium (0)per culture; (A) per mg protein. 0.26 mol m3 NaCl (0)per culture; (A) per mg protein. From Smith (1977).

16

14

17

1.2

1-0

-

-*

0 C

-8- 0.8 ?

C

L

0 3 -

-e w

0-6

x

0.4

0.2

0 2 Time (days

Fig. 18. Cellular and extracellular nitrogen of Anubuenu cylindricu grown without combined nitrogen. Cellular nitrogen of 12-day old cultures without added sodium (open bar). Cellular nitrogen of 12-day old cultures with 0.26 mol m-3 NaCl (closed bar). Extracellular nitrogen: No added sodium (0) per culture; (A) per mg protein. 0.26 mol m-3 NaCl ( 0 )per culture; (A) per mg protein. From Smith (1977).

TABLE XVII Incorporation of 16N, into Cellular Material and l6NOS,l6NO2,lSNH4and U14C-glutamateinto Protein

From 16N,a

From l6NOS

Specific enrichment From 16N02

From 16NH4

From U14G glutamate

Pg N/mg cell N/h N o combined nitrogen in culture solution Complete (4 rnol ma NaCI) Omit sodium Nitrate nitrogen in culture solution (10 rnol ma KN03) Complete Omit sodium

Pg Nlmg protein/h

cpm/mg protein

6.05

9.10

20.7

4-13

17-30

72.0

9-24 13-56

43-0 68.5

4.21 2-60

Pg Nlmg proteinlh

2.96 8-70

Pg Nlmg proteinlh

4-92 12.00

Cells collected after 6 days growth. a Cells suspended in 5 cm3nitrogen and sodium-free culture solution incubated at 30" for 2 h in a Warburg flask with continuous agitation. Pardee buffer in sidearm provided 0.2% CO,. Gas phase: 0-2 atm. N, enriched with 31.4% 16N,, 0-20atm. 0,and 0.60 atm. He. Suspensions were illuminated

at an intensity of approximately 400 footcandles. Cells incubated in 5 cmanitrogen and sodium-free culture solution at 30" as follows: 10 mol ma ISNOS(31 atom % excess); 10 mol m4 lSNO2(27.5 atom % excess) for 1 h or 0.1 mol ma l6NH4NO3(95-65 atom % excess) for 10 min. lV3-glutamate (63 OOO cpm) added to cells suspended in 10 cms of the nitrogen and sodium-free culture solution for 10 min. Cells illuminated at 400 footcandles. From Brownell and Nicholas (1967).

181

SODIUM AND PLANT METABOLISM

i

0.3

0.2

0.1

-

Ti

.-E 0 c

g

I

0 0.4

CII

;

2

C

0 0

v

0.3

g

C

f

W

0.2

0.1

0 Time (days 1

Fig. 19. Effect of sodium on acetylene reduction by Anubuena cylindrica grown with 10 mM KNO, (A); without combined nitrogen (B). No added sodium (0) per cm3;(A) per rng protein. 0.26 rnol m-3 NaCl (0)per cm3; (A) per mg protein. From Smith (1977).

2. Carbon Metabolism It has been found that cells of Anabaena cylindrica grown in low-sodium media (0.0103 rnol m-3 (0.236 ppm)) have decreased 14Cassimilation and released extracellularly a higher proportion of previously fixed carbon as organic carbon (Ward and Wetzel, 1975). The carbon uptake (as lac)in cultures without added sodium was about one-third that of cultures grown at concentrations of sodium of 0.22 mol m-3 (5 ppm) to 2.2 mol m-3 (50 ppm). There

I82

P. F. BROWNELL

TABLE XVIII Effect of Sodium and Nitrogenous Compounds on Nitrite Production by the Alga Grown With Nitrate Nitrogen (Nitrite mmol ~ n - ~ )

Time after inoculation (h)

20

48

91

Normal culture Deficient culture Deficient culture, 0.1 mol m-3 NH4Cl Deficient culture, 0.1 mol m3arginine Deficient culture, 0.1 mol m-3 citrulline Deficient culture, 0.1 mol m-3 ornithine Deficient culture, 0.1 mol m-3 glutamic acid Deficient culture, 0.1 mol m-3 glutamine Deficient culture, 0.1 mol m-3 asparagine Deficient culture, 0.1 mol IT-^ proline

12 16 1 7 9 12 16 5 3 9

44 80 1 15 18 15 70 50 34 58

50 148 5 75 50 69 125 95 100 108

From Brownell and Nicholas (1967).

was no significant difference between cultures with these higher levels of sodium. Rates of excretion of organic carbon in cultures with no added sodium were about half those of cultures receiving the sodium treatments but a greater proportion of carbon assimilated (as inorganic lac)was excreted as organic carbon in the cultures with no added sodium. The actual percentages of excreted carbon were small (2-3 %) but they suggest reduced efficiency in carbon utilization in low sodium cultures. These increased losses of organic carbon occurred even though the concentration of sodium in the lower sodium culture solutions was 0.0103 rnol m-3 (0.236 ppm) which is relatively high compared to the level needed to produce the definite signs of sodium deficiency obtained when only about 0.174 mmol m-3 (0.004 ppm) of sodium are present as an impurity (Brownell and Nicholas, 1967). It is possible that a much greater loss of carbon would occur in cells subjected to an extreme sodium deficiency. The quantity of organic matter present in cultures receiving from 0.22 mol m-3 (5 ppm) to 2.2 rnol m-3 (50 ppm) was twice that present in cultures not receiving sodium. As the rates of carbon assimilation were over three times greater in cultures receiving sodium than in sodium-deficient cultures, more carbon appears to .be lost after assimilation in cultures with added sodium than can be accounted for by excreted organic carbon. Ward and Wetzel (1975) found that the particulate carbon and rates of 14Cassimilation did not vary with increases in nitrate concentration. Particulate organic carbon in cultures receiving 0.22 mol m-3 (5 ppm) sodium was approximately twice that of cultures not receiving sodium. The proportion of 14Cassimilated and then released as organic carbon increased by approximately 25 % and 58 % with increasing nitrate concentrations of 0.0323 mol m-3 (2 ppm) and 0.323 rnol m-3 (20 ppm) in sodium-deficient cultures.

183

SODIUM AND PLANT METABOLISM

0

I

0

2

4

6 Time ( doys )

I

I

I

8

10

12

Fig. 20. Effect of sodium on glycolate release by Anubuenu cylindricu grown without combined nitrogen. No added sodium (0) per culture; (A) per mg protein. 0.26 mol m-3 NaCl (0)per culture; ( A ) per mg protein. From Smith (1977).

Smith (1977) also showed that sodium-deficient cultures of Anabaena cylindrica released greater proportions of previously fixed 14C as organic carbon than sodium-sufficient cells. Significantly more glycolate was released extracellularly in sodium-deficient cultures not containing added combined nitrogen in the media (Fig. 20). Glycolate determinations could not be made on nitrate-containing cultures due to nitrate interference.

184

P. F. BROWNELL

C. OTHER LOWER PLANTS

I . Bacteria The function of sodium in the metabolism of bacteria and fungi is considered here even though the concentrations of sodium involved are generally much higher than those required for the growth of blue-green algae and C, plants. The role of sodium is of interest in these organisms as its effects were shown to be specific and decisive. Furthermore, there is evidence that at least some of these organisms, after a number of stages of subculturing on media with progressively lower concentrations of sodium, have a greatly decreased but an absolute requirement for sodium (MacLeod and Onofrey, 1963). MacLeod (1965) suggested that the role of sodium is in the transport of substrates into the bacterial cell. Drapeau and MacLeod (1963) obtained evidence for the function by dissociating the uptake of the substrates from their subsequent metabolism with the use of non-metabolizable analogues of metabolizable substrates. When washed cells of a marine pseudomonad B-16 were incubated with 14C-a-aminoisobutyricacid, the analogue of one of the naturally occurring amino acids, it accumulated inside the cells but could not be metabolized. Its uptake required the presence of sodium in the suspending medium. As the uptake took place without a lag period from an incubation medium containing chloramphenicol, it seemed unlikely that the accumulation was due to the preliminary induction of a penetration mechanism. Lithium, potassium, rubidium, ammonium or sucrose could not substitute for sodium in the transport process (Table XIX) and equivalent amounts of the sulphate or chloride of sodium were equally effective. The process was an active one as the substrate was accumulated at about 3000 times its concentration in the medium and the uptake was stimulated by the presence of an oxidizable substrate (e.g. galactose). MacLeod (1965) concluded that since galactose needed less sodium for the maximum rate of TABLE XIX Specixcity of the Requirement for the Marine Pseudomonad B-I6 for N a f for the Uptake of a-Aminoisobutyric Acid (From Drapeau and MacLeod, 1963) ~~

-

~~

Addition to suspending mediuma ~

~ .~ _

0

_

14Cactivity of cells (cpm) 66

NaCl 6467 KCl 4 RbCl 22 NHICl 27 LiCl 29 Sucrose 44 a At concentration of 200 mol m-3. Incubation time 45 minutes.

SODIUM AND PLANT METABOLISM

185

oxidation than was needed for the optimal rate of uptake of the amino acid analogue, sodium had a role in the uptake process in addition to any possible role in oxidative metabolism. A non-metabolizable analogue of galactose, D-fUCOSe also required sodium for uptake suggesting that the requirement for sodium for galactose oxidation actually represented a requirement for its transport. The uptake of the amino acid analogue, a-aminoisobutyric acid by cells of the marine luminous bacterium Achromobacter (Photobacterium) jscheri has also been found to be a sodium-dependent process. MacLeod (1 965) using washed-cell preparations of two marine pseudomonads found different concentrations of sodium to be required for the maximum rate of oxidation of different substrates. For the maximal rate of oxidation of acetate, butyrate, propionate or an oxidizable sugar 50 mol m-3 ( I I50 ppm) sodium was required. For malate, citrate and succinate 150-200 mol m-3 (3450-4600 ppm) sodium was necessary. These differences could be explained by assuming the presence of a number of permeases in the cell membrane with quantitatively different requirements for sodium. Kahane et a/. (1975) demonstrated sodium-dependent glutamate transport in membrane vesicles of Escherichia coli K-12. Mutants of E. coli K-12 are able to utilize glutamate as a major carbon source and transport glutamate more effectively than wild type strains which are unable to grow on this amino acid. A marked similarity was observed between the effects of sodium on glutamate transport in membrane vesicles and in whole cells of E. coli K-12. On the basis of the data, it was concluded that the glutamate “carrier” of E. coli is in the plasma membrane and that it is possible that a glutamatebinding protein is involved in transport. Kodama and Taniguchi (1977) found in the presence of potassium, that sodium rapidly enhanced respiration and activated cellular motility and transport of potassium, amino acids and phosphate in the non-halophilic bacterium, Pseudomonas stutzeri. Inorganic phosphate was taken up rapidly and esterified to nucleoside triphosphates and diphosphates. Respiration and phosphate transport responded to sodium which increased with temperature. Respiration was more sensitive to polymyxin B in the presence of sodium. Since dinitrophenol stimulated respiration when either potassium or sodium was limiting, sodium appears to be required for the cytoplasmic membrane of Pseudomonas stutzeri to perform energy-linked functions coupled to respiration. It appears that sodium exerts its primary effect on the membrane so that energy stored as the energized state of the membrane can be used to drive ATP synthesis, active transport, cellular motility and other energy-linked processes. Approximately 1 mol m-3 (2-3 ppm) of sodium was needed for full activity. 2. Fungi Vishniac (1960) showed 49 isolates of non-filamentous phycomycetes of

186

P. F. BROWNELL

marine origin to require sodium chloride for optimum growth. In one species, Thraustochytrium globosum, it was found that the optimum concentration of sodium chloride was unchanged after five serial cultures using large inocula. In Thraustochytrium roseum uptake of phosphate and oxygen was much greater in the presence of 0.2 mol 1-1 sodium chloride than in the presence of 0.2 moll-' of the chlorides of lithium, potassium, rubidium or magnesium or sucrose. Siegenthaler et al. (1967) observed that phosphate uptake and transport in Thraustochytrium roseum was maximally stimulated by 200-400 mol m-3 sodium chloride. They suggested that the effectiveness of sodium chloride in phosphate transport was not related to its osmotic pressure but that the increase in respiration obtained with the sodium chloride treatment was due to its osmotic pressure as sucrose at the same concentration also increased the respiration rate without affecting the ability of the cells to take up phosphate. D. C4 AND CAM PLANTS

1. Changes Following the Supply of Sodium in Sodium-de$cient C4 Atriplex Plants A study was made of the changes in leaf respiration rate, chlorophyll content, soluble and insoluble nitrogen fractions, and sugar and starch contents upon the addition of sodium to sodium-deficient Atriplex plants (Brownell and Jackman, 1966). The sequence of changes that follow the supply of sodium to sodium-deficient plants of Atriplex nummularia is shown in Fig. 21. For at least five days, the growth of plants which received sodium (expressed as fresh weight of shoots) was similar to that of plants to which no sodium was added (Fig. 21A). After seven days, however, the growth of the sodiumfed plants began to increase rapidly compared with that of untreated controls. This delay in growth response occurred despite a rapid uptake of sodium into the shoots from the time of application (Fig. 21B). By the fifth day the plants had taken up most of the sodium supplied in the culture solution, thus the concentration in the leaves remained steady until the seventh day when it showed a slight decrease presumably due to the onset of rapid growth. The respiration rate per unit fresh weight of shoots (Fig. 21C) increased rapidly (for about three days) after receiving sodium, reaching about twice that of the sodium-deficient plants. Thereafter, there was little further change in rate. The concentration of chlorophyll in the leaves increased rapidly to almost double the initial concentration by the third day after the addition of sodium (Fig. 21D) and continued to increasethroughout theten daysof theexperiment. The concentrations of both sugars and starches in sodium-deficient plants were low (Fig. 22). During recovery, however, these levels gradually rose to several times those found in deficient plants. Even though, for at least five days, the growth of plants which received

0

I

2

3

4

5

6

7

8

9

10

Time after opplying differential Ireolmenls ( d a i s )

Fig. 21. Changes in fresh weight of shoots (A), sodium concentration (B), rate of oxygen uptake by leaves (C), and chlorophyll concentration (D) following the addition of sodium sulphate (0.6 mol m-3), to 32-day old sodium-deficient Atriplex nummuluria (old man salt bush) plants. All harvests were made at 9 a.m. Each point is derived from two cultures of eight plants of which six plants were used for chlorophyll determinations, and the remainder for measurements of sodium concentration and respiration rate. Total fresh weights of all 16 plants are recorded. No addition, 0----0; 0.60 mol m-3 Na,SO,, 0-0. From Brownell and Jackman (1966).

188

P. F. BROWNELL

I

0

1

2

3

4

5

6

7

8

I

I

I

9

10

11

I

I

I

121314

Time after applying differenlial trealmentd d a y s )

Fig. 22. Changes in concentration of sugars (A) and starch (B) following the addition of sodium sulphate (0.6 mol m-3) to 45-day old sodium-deficient Atriplex numrnuluriu (old man salt bush) plants. Points are the means of determinations of six samples each containing the leaves of two plants. N o addition, 0----0; 0.60 mol m4 Na,SO,, 0-0. From Brownell and Jackman (1966).

sodium (expressed as fresh weight of shoots) was similar to that of plants to which no sodium was added, there is evidence of an increase in the production of photosynthate during this period. Sugar concentrations increased steadily in the plants which received sodium compared to those not receiving sodium. The levels of starch also increased in the sodium-treated plants compared to those not receiving sodium after a lag period of about eight days (Fig. 22). 2. Uptake and Distribution of Sodium and Potassium Numerous studies have been made of the uptake and transport of sodium in plants but these generally have been with C , plants for which sodium does not appear to be essential or with C , plants growing in cultures with concentrations of sodium too high to limit growth. It seems that it is more relevant in terms of attempting to understand the function of sodium as a nutrient to work with C , plants with the amount of sodium available ranging from deficiency to optimum levels. A few studies have been made with C , plants under these conditions. Brownell (1965) supplied graded amounts of sodium sulphate to plants of Atriplex vesicaria. The dry weight production and the

TABLE XX Effects of Treatments of Sodium Sulphate on Dry Weight Production and Concentrations of Sodium and Potassium in Leaves, Stems and Roots of Atriplex vesicaria

Treatment

Dry wt (€9 All values are the mean of 5 cultures of 4 plants each

Conc. of sodium and potassium (mmol/kg) All values are the means of duplicate samples taken from 5 replicated cultures of each treatment Sodium Potassium

~

Leaves Stems Roots Total Leaves Stems Roots Leaves Stems Roots 0.008 0.022 0,086 100 7.1 I No sodium sulphate 0.0560 7.1 2834 1913 1547 0.398 47.8 6.5 6.5 0-043 0.098 4450 I1 0.01 rnol m3 Na,SO, 0.257 2583 1442 78.3 11.7 0.066 0.138 0.581 7.0 2504 I11 0.02 mol ma Na,S04 0.377 2197 1563 0.722 213.0 20.2 0.088 0.173 11.7 2476 2169 1540 IV 0.06 mol ma Na,S04 0.461 0.193 0.771 295.7 51-0 29.1 0.489 0.089 2225 2205 1683 V 0.10 mol ma Na,SO, 1129 0-267 1.101 338.7 257.8 0-149 1688 VI 0.60mol ma Na,SO, 0.685 1934 1445 The statistical treatment of total dry weight data was as follows: VI, V, IV, 111, I1 > I at 0.1 % level of significance: VI > V at 5 % level of significance: 11,111; 111, N ;IV, V indistinguishable. Seedlings were transferred to cultures 9 days after germination, the different treatments applied on day 10, and the plants harvested on day 51. From Brownell (1965).

190

P. F. BROWNELL

concentrations of sodium and potassium in leaves, stems and roots of Atriplex Ivesicaria obtained are shown in Table XX. The dry weight production increased asymptotically with increasing applications of sodium sulphate. The concentrations of sodium increased strikingly in all fractions, especially in the leaves where the increase was more than 100-fold when 0.60 mol m-3 (13.80 ppm) of sodium sulphate was supplied although their dry weight production increased only 12 times. The concentrations of potassium in leaves, stems and roots increased when 0.01 mol m-3 (0.23 ppm) sodium sulphate was supplied, but decreased when the amounts of sodium sulphate were further increased to 0.60 mol m-3. Similar results were obtained by Ashby and Beadle (1957) in which they showed significant increases in dry weight yield in the C , species, Atriplex inflata and Atriplex nummularia following the addition of either sodium chloride, sodium nitrate or sodium sulphate at 10 mol m-3 (230 ppm Na) to the culture solution. The concentration of sodium in the leaves increased from 93- to 227-fold when 10 mol m-3 (230 ppm) Na was supplied. Williams (1960) using another C, species, Halogeton glomeratus, also obtained significant dry weight increases on the supply of 1.0 mol m-3 (23 ppm) sodium as either the chloride, nitrate or sulphate. The concentration of sodium in the leaves increased from 110 mmol kg-l(2.530 ppm) to 2800 rnol kg-' (64 400 ppm) on the addition of 1 mol m-3 of sodium chloride to the culture solution. In both studies, potassium concentrations in leaves were markedly depressed by the sodium treatments. The uptake of sodium by Atriplex nummularia, a C , species, was found to be rapid. Brownell and Jackman (1966) calculated that after five days little sodium remained in culture solutions initially containing 0.6 mol m-3 sodium sulphate. The uptake was calculated by the increase in the amounts of sodium determined in the leaves. It appears that the species of Atriplex used in these studies and Halogeton glomeratus may not be typical in their ability to accumulate sodium in their tissues at high concentrations in relation to the concentration of sodium in the cultures. Contrasting data have been obtained in the C, species, tomato (Woolley, 1957) and cotton (Pleunneke and Joham, 1972) and in the C, species Echinochloa utilis and Kochia childsii (Table XXI) (D. W. Mill and P. F. Brownell, unpublished work). It is obvious that the ability to accumulate sodium at high concentrations in the tissues from low substrate sodium levels is not general for all C4species. The high concentration of ions in the leaves of some Atriplex species has been explained by the active transport of certain ions including chlorine and sodium into epidermal vesicles or bladders which they possess (Osmond et al., 1969). Foliar applications of sodium have brought about localized recovery in sodium-deficient plants (P. F. Brownell and D. W. Mill, unpublished work).

191

SODIUM AND PLANT METABOLISM

~.~

TABLE XXI Concentration of Sodium in Leaves (Dry Wt Basis) ~-

-

~~~

._

~

No addition ppm pmollkg -~

~

~

.~

~

.~

Atriplex vesicariaa 230e loo00 (0.08mmol m-3sodium) Kochia childsiib 10.17f 442 (0.08 mmol

2.3g

~

~

~~

--

6801

295 700

0.1 mol m-3

233

10 110

0.1 mol m-3

100

4.14

0.0047 mol m-3

180

sodium)

Lycopersicum esculentumd cv. “Marglobe” (0.35 mmol m-3sodium) -

~

Added N a concn

sodium)

Gossypium hirsutumc cv. “Stonefield” 2B-S9 (0.17 mmol

Added sodium ppm pmollkg

~

~

~~

2.0Sh .~

-~

89 --

1.0mol m-3

29000

667 ~. ~

~~

.

Brownell (1965). D. W. Mill and P. F. Brownell, unpublished work. Pleunneke and Joham (1972). Woolley (1957). Culture solution contained 0.08 mmol m-3 sodium as an impurity. f Culture solution contained 0.08 mmol m-3 sodium as an impurity. g Culture solution contained 0.17 mmol m-3 sodium as an impurity. Culture solution contained 0.35 mmol m-3 sodium as an impurity.

a

One microlitre of a solution containing 5 pg Na as sodium chloride and 40 % ethanol to facilitate infiltration was applied to the upper epidermis of one leaf of a sodium-deficient Kochia childsii plant. Within two or three days the leaf receiving the sodium solution greened-up and made rapid growth. Even after two weeks the other leaves of the same plant remained small and chlorotic. Similar results have been obtained with Atriplex vesicaria. Mill (1977) applied 22NaCl to leaves of sodium-deficient and sodium-treated Kochia childsii plants. Even after seven days of applying the treatment over 80% of the radioactivity was retained in the young treated leaves and over 60% in the old treated leaves (Table XXII). It therefore appears that small amounts of sodium applied to the leaves are relatively immobile and are able to bring about the observed localized recovery of a single leaf while the other leaves still show signs of the deficiency. 3. Respiration Since it was one of the first detected, the respiratory response by Atriplex inj?ata and A . numrnularia was studied more fully by examining the effects of different salts of sodium, other univalent cations, varying concentrations of sodium, feeding sodium to cut shoots, feeding sucrose, and the effects of sodium on anaerobic CO, production (Brownell and Jackman, 1966). Whether supplied as chloride or sulphate, equivalent amounts of sodium brought about similar increases in the rate of 0, uptake (Fig. 23). It appeared, ‘therefore, that these increases depended upon the sodium ion and

192

P. F. BROWNELL

TABLE XXII Distribution of Sodium in Plants of Kochia childsii .

~

__

~

~

Radioactivity (percentage) Sodium-deficient plantsa Sodium-treated plantsb Young leaf Old leaf Young leaf Old leaf application application application application

Part of plants Leaf receiving sodium treatment Leaves above treated leaf Leaves below treated leaf Stems and roots

80.6 f 0.65

85.1 f 0.66

63.1 k 0.57

10.2 k 0.19

19.5

0.28

8.3

0.21

12.4 & 0.23

3.0 f 0.15

8.3

0.22

4.1

0.19

5.1 i 0.17

9.1 1.7 % 0.10 ~ ~ _ _ _ _ _ _ _

i 0.23

7.0 i 0.17

2.3 f 0.15

80.3 f 0.68

a Plants were grown under low-sodium culture conditions and harvested 27 days after

germination. Plants received 0.1 mol m-3 NaCl in their cultures 7 days after germination. 5 p g Na as NaCl with *%Naactivity of 0.05pCi in 40% ethanol was applied in a total volume of 1 pl to the upper epidermis 21 days after germination. Data are the means of values from four plants. From Mill (1977).

were not due to excess cation absorption (Jacobson, 1955; Jacobson and Ordin, 1954), as with the latter phenomenon, different rates of oxygen uptake would be expected as a result of the different mobilities of chloride and sulphate ions. The rates of 0, uptake by leaves of sodium-deficient plants 48 hours after receiving sulphates of univalent cations (0.1 mol m-3) in culture solutions are shown in Fig. 24. Sodium sulphate increased the rate of oxygen uptake by 73 % above that of the control whereas the sulphates of lithium, potassium, rubidium and caesium did not increase the rate. It seems unlikely that this is salt-stimulated or anionic respiration (Lundegardh, 1955; Robertson and Turner, 1945) for two reasons; first, the salt-stimulated response generally occurs only in tissues initially containing low concentrations of ions, whereas the concentrations of ions in leaf tissues of Atriplex are high compared with the low concentrations of sodium required to elicit the response; second, saltstimulated respiration may be brought about by various ions, unlike this respiratory response in Atriplex which requires sodium, specifically. Both respiration rate (0, uptake) and growth (fresh weight) tend towards a maximum value with similar concentrations of sodium in the culture solution (Fig. 25). From these results it appeared that lack of sodium limited the rate of respiration of Atriplex leaf tissues. If this brought about a reduction in phosphorylation, growth could conceivably be restricted by the decreased amounts of high energy phosphate compounds available for cellular work and syntheses.

193

SODIUM AND PLANT METABOLISM

1

0

10

1

,

1

1

20 30 40 50

,

,

,

1

1

60 70 80 90 100

Time (minutes)

Fig. 23. Effect of sodium either as chloride or sulphate on the rate of 0, uptake by leaves of sodium-deficient Atriplex nummulariu (old man salt bush). Sodium chloride or sodium sulphate (0.22 mol m-3) was added to the culture solution 48 hours before harvesting. Points are the means of 0, uptake of four Warburg flasks each containing the leaves of two plants. No addition, 0----0; NaCI, A-A ; Na,SO,, 0-0. From Brownell and Jackman (1966).

When sodium was introduced to the tissue through cut stems, the rate of

0,uptake by the leaves was 20 % greater than that of untreated leaves (Table XXIII) and analyses showed that there had been an increase in the sodium concentration. This respiration response, only 13 h after the application of sodium, was detected in a considerably shorter time than in other experiments when sodium was added to the culture solution. It is difficult to determine precisely the time required for this response to sodium, first because it is not known when sodium arrives a t the actual site of action, and second because it would not be expected to reach all cells in a tissue at the same time. Since the concentrations of sugars and starch were low in leaves of sodiumdeficient plants (Fig. 22), it seemed possible that lack of respiratory substrates could have limited their respiration rates. Hence cut leaves were supplied with sucrose and their respiration rates measured. The results

194

P. F. BROWNELL

no addition

mmol m-3 ~ 1 2 ~ 0 4

0

100

200

300

400

600

500

Mean roles of oxygen uploke ( ~ g-'1 FW h-'

700

1

Fig. 24. The rates of 0, uptake by sodium-deficient plants of Afriplex nummularia (old man salt bush) which received 0.1 mol m of the sulphate of lithium, sodium, potassium, rubidium or caesium 48 hours before harvesting compared with the control plants which received no addition. Points are the means of 0, uptake o f four Warburg flasks each containing the leaves of two plants. From Brownell and Jackman (1966).

(Table XXIV) showed that addition of sucrose increased the respiration rates of leaves from both sodium-deficient and normal plants suggesting that the small amounts of endogenous substrate were limiting the rate of respiration. The supply of sodium to the culture solution 43 hours before harvesting also caused an increase in the respiration of leaves which received no additional sucrose. This could have been due to either a direct effect of sodium on the respiratory system or alternatively to an indirect effect, for example by increasing the amount of substrate derived from photosynthesis during the period after addition of sodium. When sucrose was fed in an attempt to remove the substrate limitation, there was still a response to sodium, suggesting that its effect is a direct one on some part of the respiration process. If however, the substrate limitation was not entirely removed by the exogenous sucrose, the apparent effect of sodium on respiration rate could be through its effect on photosynthesis. In all experiments, sodium was supplied in the light when the high rates of transpiration needed for the rapid uptake of the treatment solutions would be favoured. Even so, it is unlikely that in the short-term experiments (13 h) there would have been sufficient time for differences in the photosynthate produced between treatments to be great enough to affect the results of the subsequent respiration experiment. Thus, it seems that the effect of the sodium treatment on respiration in these experiments was direct and not due to its possible effect on photosynthesis. In an initial attempt to locate the part of the respiratory system in which

195

SODIUM AND PLANT METABOLISM

70

6.0

5.0

m I

i

4.0

g

S

I

I

30

I I I I

20

? LL

I I

1.0

I

I I I

00.2

I

0.1

I

D

0.6

Concentrationofsodium (mol m-3)

Fig. 25. Effect of concentration of sodium in culture solution on fresh weight (*-----a) and rate of oxygen uptake in Atriplex nummularia (old man salt bush). Fresh weight data are means of five cultures of four plants each, sodium being supplied 16 days after germination and the plants harvested 32 days later. The rates of O2 uptake were determined 28 hours O........... and 76 hours (0-0) after supplying sodium to 46-day old plants. From Brownell and Jackman (1966).

sodium might be involved, rates of CO, output of sodium-deficient and sodium-fed plants were compared. The results obtained (Table XXV) showed that the addition of sodium to sodium-deficient plants stimulated the rate of anaerobic CO, production and that the rate of 0, uptake when leaves were returned to air was unaffected by the anaerobic treatment. Hence, the increase in rate of CO, production under anaerobic conditions, suggests that sodium acts in the glycolytic stages of respiration. However, an earlier effect of sodium on some other system leading to this respiratory response is still possible. This effect of sodium in increasing the respiratory rate in leaves does not appear to be general for all C, species. Mill (1977) was not able to obtain similar results when sodium was supplied to sodium-deficient plants of Kochia

TABLE XXIII Effect of Feeding Sodium to Cut Shoots on Rate of 0, Uptake and Sodium Concentration of Leaves

No sodium added Added sodiuma

0,uptakeb (p1g-l fr. wt h-l)

Sodium conc.c (mmol Na per kg dry wt)

21 1 256

12.1 18.5

a Cut shoots were placed in sodium sulphate (2.2 mol m-3) for

14 h in the light.

* Values are mean rates of uptake of six Warburg flasks each containing the leaves of

2 to 3 plants. Values are the means of two determinations on the leaves from six Warburg flasks. From Brownell and Jackman (1966).

TABLE XXIV Effect of Feeding Sucrose to Cut Leaves on Rate of 0,Uptake ~

0, uptake (pI g-' fr. wt h-l)

No sodium added Sodium addeda Percentage increase due t o sodium

Without sucrose

With sucroseb

244 302 24

322 400 24

~~

Percentage increase due to sucrose

32 32

_ _

Data are the mean rates of 0, uptake of three Warburg flasks each containing the leaves of 2 or 3 plants. a 0.60 mol m-3 sodium sulphate added to culture solution 43 hours before harvesting. b Cut petioles were immersed in 0.1mol I-' sucrose for 13 h in light. From Brownell and Jackman (1966). TABLE XXV Effect of Sodium on Rate of Anaerobic CO, Production of Sodium Deficient Atriplex nummularia Leaves

No sodium added Added sodiuma

Anaerobic phase co, output

Aerobic phase 0, uptake

85 105

316 426

Data are the mean rates of gas exchange (PI g-' fr. wt h-l) of sixteen Warburg flasks each containing the leaves of 2 to 3 plants. a 0.1 mol m-3 sodium sulphate was supplied to the culture solution 24 hours before harvesting. From Brownell and Jackman (1966).

SODIUM AND PLANT METABOLISM

197

childsii. At present it is not possible to explain this difference in the respiratory responses of these species. As already pointed out, there is a large difference between these C, plants in their ability to accumulate sodium from a lowersodium substrate and it is possible that the response described for the Atriplex species could be attributable in some way to their propensity for accumulating sodium from low substrate concentrations. The respiratory response may not necessarily be a sequential step in the events taking place from when the sodium treatment was applied to when the signs of recovery appear. 4. Nitrogen Fractions Brownell and Jackman (1966) found marked differences in the ratios of 80 % ethanol-soluble nitrogen to total nitrogen between the leaves of sodiumdeficient and normal plants of A triplex b y 20 days after supplying sodium but no difference was detectable at two days after the addition of sodium (Table XXVI) although by this time there was a definite respiratory increase. Hence, the apparent long-term effect of sodium in decreasing the ratio of soluble nitrogen to total nitrogen is probably indirect and only one of the many changes likely to take place in the later stages of recovery.

5. Photosynthetic Responses The finding that plants having the C , photosynthetic pathway require sodium as a micronutrient (Brownell and Crossland, 1972) and that a CAM plant, Bryophyllum tubiporum, responds to sodium only when grown under certain conditions (Brownell and Crossland, 1974) suggested that sodium is involved in the first carboxylation and the decarboxylation of the resulting C, dicarboxylic acid. This appeared likely since this metabolic system operates only in C, and CAM plants but not in C, plants. Three types of C4 plants have been described on the basis of their C , acid decarboxylating systems and ultrastructural features (Gutierrez et al., 1974; Hatch et al., 1975). They comprise: 1. Malate type NA DP-malic enzyme-type 2. Aspartate type NAD-malic enzyme-type 3. Aspartate type Phosphoenol pyruvate carboxykinase type. Species for which sodium has been shown to be essential fall into all these categories. Echinochlou utilis which has relatively high NADP malic enzyme activity (P. F. Brownell, unpublished work), Kochiu childsii and Portulaca grandijlora (rose moss) (Gutierrez et al., 1974) would be in type 1. Type 2 includes Amaranthus tricolor (Gutierrez et al., 1974), Atriplex spongiosa (pop salt bush) (Hatch et al., 1975) and Cynodon ductylon (Bermuda grass) (Chen et al., 1971). Chloris gayuna (Hatch et al., 1975) and Eleusine indica (Gutierrez

TABLE XXVI Concentrations of Soluble and Insoluble Nitrogen in Leaves of Atriplex

A friplex nummularia a two days after receiving sodium Nitrogen content Fr. wt (yofr. wt) per plant (g) ethanol extract Residue No sodium added Sodium added C

Sol N Total N

Atriplex inflata twenty days after receiving sodium Nitrogen content Fr. wt (yofr. wt) per plant (g) ethanol extract Residue

Sol N Total N

1.33

0.29

0-42

0.41

0.1 1

0.32

0.39

0.45

1.76

0.25

0.39

0.39

0.50

0.17

0.46

0-27

~

Data are the means of determinations on five individual plants. Data are the means of determinations on four samples of two plants when no sodium was added and one plant when sodium was supplied. 0.60 and 0.10 mol m-3 sodium sulphate to Arriplex nummularia and Atriplex inpafa, respectively. From Brownell and Jackman (1966).

4

SODIUM AND PLANT METABOLISM

199

et al., 1974) are in type 3 and have been shown to have a sodium requirement (T. S . Boag, unpublished work). If sodium is needed for the system operating in C, and CAM plants comprising the first carboxylation and the subsequent decarboxylation, it would be expected to affect a part of the system common to types 1,2 and 3 species. Of the enzymes highly active in the three types of C, species but not in C , species, only the four enzymes, phosphoenol pyruvate carboxylase, pyruvate orthophosphate di kinase, adenylate kinase and pyrophosphatase have similar activities in all types of C, plants (Hatch, 1976). These enzymes also appear to be active in the group of CAM species defined by Dittrich et al. (1973) as having high activities of NADP malic enzyme but not detectable phosphoenol pyruvate carboxykinase activity. This group includes Bryopliyllum tubiflorum, the species which responded by increased dry weight yield to sodium when grown under conditions of short day and large diurnal temperature fluctuations (Brownell and Crossland, 1974; Boag, 1976). The possible requirement of sodium for the formation or activation of the other enzymes indicated by Hatch (1976) to be present only in the individual types of C4 species would not explain the general requirement of sodium for all the three types of C, species. The other CAM species described by Dittrich et a/. (1973) contained high activities of phosphoenol pyruvate carboxykinase but low activities of NADP malic enzyme. The only enzyme occurring in plants of this group common to the C , and the other CAM group of species but not active in C, species would be phosphoenol pyruvate carboxylase. No CAM species in this group has yet been examined for a sodium requirement. If sodium is involved in the primary carboxylation (phosphoenol pyruvate carboxylase) and the subsequent decarboxylation steps of photosynthesis in C , and CAM plants, it could be affecting the activity of phosphoenol pyruvate carboxylase or the formation of phosphoenol pyruvate, the carbon dioxide acceptor. In both C, and the CAM species, Bryophyllum tubijlorum, the enzyme, pyruvate orthophosphate dikinase, is involved in the conversion of pyruvate (derived from the decarboxylation) to phosphoenol pyruvate. This is the CO, acceptor in C, plants but is converted by reverse glycolysis to sugars and starches in Bryophyllum tubiyorum. These compounds are the sources of the PEP, the CO, acceptor in the dark. The other enzymes implicated in the supply of PEP would be adenylate kinase required in the following reaction :

ATP

+ AMP

-

2 ADP

and pyrophosphatase required i n the reaction: PPi

--f

2 Pi.

It seemed possible that sodium could be involved in these reactions. It appears that sodium nutrition in C, plants does not affect the formation of phosphoenol pyruvate carboxylase. Echinochloa utilis plants were harvested at four times in an experiment to determine the effect of sodium on the enzyme

200

P. F. BROWNELL

(P. F. Brownell, unpublished work). Although there was a marked growth response to the sodium addition, there was no significant difference between the specific activity of PEPC extracted from sodium-deficient and normal plants at any harvest (Fig. 26). These results indicate that sodium does not affect the formation of PEPC but it is still possible that sodium may be necessary for the activation of the enzyme in vivo. This would not have been revealed in this experiment as no measures were taken to obtain sodium-free assay conditions. If the enzyme requires sodium for its activation, there would probably have been sufficient sodium present in the assay medium not to have limited its activity in the in vitro assay. Holtum (1975) detected no difference in PEPC activity on a fresh weight basis in leaves of sodium-deficient and normal plants of Kochia childsii but the assay was not carried out under sodium-free conditions. No activation effect of sodium was observed on pyruvate orthophosphate dikinase, phosphoenol pyruvate carboxylase or adenylate kinase in assay media containing about 0.2 mol m-3 (4.6 ppm) sodium (M.D. Hatch, personal communication). It is still possible that sodium is needed for activation of these enzymes in lower concentrations. Evidence for the lack of an effect of sodium on the activity of PEPC in vivo was obtained from preliminary 14C labelling experiments with Kochia childsii exposed to 14C0, in the light (Holtum, 1975; Webb, 1977). No consistent differences were observed between the labelling patterns of sodiumdeficient and normal plants which could suggest an effect of sodium nutrition on the activity of phosphoenol pyruvate carboxylase. Further indirect evidence that phosphoenol pyruvate carboxylase activity is not influenced in vivo is presented by Boag (1976) who followed the patterns of carboxylation and decarboxylation (determined by changes in the titratable acidity of leaves) in the CAM plant, Bryophyllum tubzjlorum. It is possible to dissect out the processes of carboxylation and decarboxylation which are separated in time and to observe the possible effects of sodium nutrition on them. It was found that the rate of carboxylation in the dark in plants grown under both short- and long-day regimes was not affected by the sodium treatment (Fig. 27A and B). This suggests that sodium had not affected the in vivo activity of PEPC, the enzyme responsible for the increase of acid (as malic acid in the vacuoles). Similarly, there was no evidence for differences in the rates of decarboxylation in the light between sodiumdeficient and normal plants. It thus appears that this system which collects carbon in the dark in CAM plants to be converted eventually to photosynthate by the Calvin cycle in the light is not affected by sodium nutrition even though there are significant increases in dry weight in response to sodium under certain conditions e.g. short days and large diurnal temperature variations. It is probably justified to extrapolate this finding to C , plants which have similar systems for the primary carboxylation and to postulate

Fig. 26. Changes in fresh weight (upper) and specific activity of phosphoenol pyruvate carboxylase (lower) of shoots, (- - -) and roots (---) of Echinochloa utilis L. following the addition of 0.1 mol m-3 of sodium chloride to eleven-day old sodium-deficient plants. Fresh weight data are means of 24 plants in the first and 12 plants in the subsequent harvests. Activities of phosphoenol pyruvate carboxylase are the means of values from four determinations in extracts of tops and roots. The extraction of the enzyme was by the method of Slack and Hatch (1967) and the assay according to Osmond (1969a). No addi0.1 mol m-a NaCI, ( 0 )(P.F. Brownell, unpublished work). tion, (0);

202

P. F. BROWNELL

carbohydrate (pmol g-'

FW)

Titratableacid (pmol g- F W 1 Carbohydrate Ipmol g-'FW) 0

5 8 g 8 88ao 5 8 g

8 8

8 8 m

Fig. 27. Diurnal changes in titratable acidity (-)

and acid-soluble carbohydrates (0.1 mol m-3 sodium chloride treated) ( 0 )BryophyNum tubiflorum grown under (A) a sixteen-hour light period, temperature 33"-36"C/eight-hour dark period, temperature 24"-26"C. Illumination was 183pE m-z sec-l (9480 lux) and (B) under an eight-hour light period, temperature 33"C/sixteen-hour dark period, temperature 13°C. Illumination was 255 pE m-2 s-' (10 200 lux). The plants were 72 days old. From Boag (1976).

(..........) in leaves of sodium-deficient (0) and normal

that sodium is not acting in this system but in a later stage possibly in the photosynthetic carbon reduction cycle. There is ample evidence for sodium increasing the rate of photosynthesis when supplied to sodium-deficient C, plants and the CAM plant, Bryophylfum tubifZorum grown under conditions of short days and large diurnal temperature variation. The simplest evidence is from marked increases in growth as fresh or dry weight yields which are now well established (Brownell and

203

SODIUM AND PLANT METABOLISM

Wood, 1957; Williams, 1960; Brownell and Crossland, 1972; Brownell and Crossland, 1974). Evidence for short-term differences is suggested from calculation of data from Brownell and Jackman (1966) in which the respiration of leaves of sodium-deficient and normal plants of Atriplex nummuluriu (old man salt bush) was expressed as glucan equivalents respired per day. These quantities, added to the glucan equivalents in the glucose and starch (Fig. 21) produced by sodium-deficient and normal plants, respectively, give an estimate of their relative photosynthetic performance during the early period of recovery prior to a growth response (Fig. 28). By calculation, it can be shown that the combined glucan equivalents of glucose plus starch production and respiration per day would be about lo%, of the dry weight for sodium-deficient plants compared to 20 for normal plants. The 14C-labellingof amino acid, organic acid and sugar fractions extracted from leaves of Atriplex nummularia (old man salt bush) when exposed to

H

g

10000-

i Y

z

1 0

1

2

3

4

5

6

7

8

9

10

Time after applying differential treatments (days)

Fig. 28. Increase in “photosynthate” in leaves of Afriplex nurnrnularia (old man salt bush) during the recovery from sodium-deficiency following the addition of 0.60mol m-3 sodium sulphate to their cultures. “Photosynthate” was calculated indirectly as glucans from glucose and starch data added to glucan equivalents respired per day on the assumption that the respiratory quotient = 1 and that 0 . 7 5 ~ 10%(STP) = 1 p g glucan. No addition, 0----0; 0.60 mol m-3 sodium sulphate, 0-0. From Brownell and Jackman (1966).

204

P. F. BROWNELL

14C02 illuminated at approximately 1000 ft-c intensity for periods of two, five and ten minutes are shown in Fig. 29 (P. F. Brownell, unpublished work).

0

1

2

3

4

5

6

7

8

9

10

Time ( minutes)

Fig. 29. The effect of sodium on the 14Clabelling of amino acid, ...........;organic acid, fractions from leaves of Atriplex vesicaria. One set of leaves (approximately 0.5 g fresh weight) from sodium-deficient, (O), and the other from plants which had received 0.60 mol m-3 sodium sulphate in their cultures four days previously, (a), were simultaneously exposed for each period (2, 5 and 10 minutes) in a 200ml chamber to lOpCi W O , at a light intensity of 2000 ft-c. Extractions and separations of carbon fractions were carried out according to the method of Canvin and Beevers (1961) (P. F. Brownell and S. E. Knowles, unpublished work).

__-- ;and sugar, -,

The radioactivity of organic acids and amino acids from leaves of sodiumtreated plants was greater than from sodium-deficient plants after two minutes, but the labelling of amino acids after fifteen minutes was similar for both treatments. On the other hand, the labelling of the sugar fraction from sodium-deficient and normal leaves was similar after two minutes but the difference between the labelling of sodium-deficient and normal leaves became greater with time. After fifteen minutes of exposure, the radioactivity of sugars extracted from leaves of normal plants was almost four times greater than that from sodium-deficient plants. The greater amount of 14Cfrom the 14C02 being incorporated into the sugar fraction in normal than sodiumdeficient plants indicates that they were photosynthesizing more rapidly.

SODIUM A N D PLANT METABOLISM

205

6. Transport

There have been numerous reports of adenosine triphosphatase (ATPase) systems using ATP as the energy source for the active transport of ions in plants. An active mechanism for the extrusion of Naf in Scenedesmus was described by Kylin (1966) similar to the phenomenon observed in tissues and cells of animals (Skou, 1964). ATPase activity has been demonstrated in homogenates from higher plant tissues including roots of sugar beet (Hansson and Kylin, 1969), oats (Fisher and Hodges, 1969; Leonard et al., 1973), oats, wheat, barley and maize (Fisher et al., 1970), sugar beet cotyledons (Karlsson and Kylin, 1974) and in leaves of mangroves (Kylin and Gee, 1970). Hodges et al. (1972) purified an ion-stimulated ATPase from oat roots by discontinuous sucrose density gradient centrifugation and showed it to be substrate specific and associated with the plasma membranes. The plasma membrane-associated ATPase of oat roots was activated by divalent cations Mg2+ = Mn2f > Zn2f > Fe2+ > Ca2f and further stimulated by KCl and other monovalent salts both inorganic and organic (Leonard and Hodges, 1973). Fisher et al. (1970) established correlations between ion fluxes and ionstimulated ATPase activity in roots of barley, oats, wheat and maize. According to Jennings (1976) only roots of sugar beet (Hansson and Kylin, 1969; Karlsson and Kylin, 1974) and leaves of the mangrove, Avicinniu nitida have been shown to have (Na+ K+)-activated ATPase with an activity specifically related to the ratio of Na+: Kf in the assay medium as shown in the animal system (Skou, 1964). The ATPase prepared from homogenates of sugar beet roots had highest activity with the ratio of Na+: K+ at 1 : 1 in homogenates at constant ionic strength (Hansson and Kylin, 1969). In homogenates from leaves of mangroves with varying ratios of Na+ : K+, Kylin and Gee (1970) demonstrated salt-stimulation at three peaks. Attempts have been made to correlate activities of ATPases with physiological functions. Kylin and Gee (1970) suggest that the occurrence of (Na+ K+) activated ATPases might be correlated with salt tolerance in certain species including Avicennia marina in which optimum absorption of Kf is a function of the concentration of Na+ in the medium. They interpret this dependence on Naf as a coupling between uptake of potassium and extrusion of sodium. Palladina and Bershtein (1974) postulate that both Nafand K+-activated ATPases exist in sugar beet petioles. The ATPase has its highest activities at certain Na+: Kf ratios. They suggest that the highest dry weight yields and sugar contents occurred when the K+: Naf ratio of the tissues was near these ratios. The demonstration of the activity of (Naf K+) ATPase in certain plants could suggest that they may require sodium as an essential element. However, it would be necessary to have evidence that the (Naf Kf) ATPase is functional in such plants and that it is absolutely dependent on sodium for its activity. From data presented by Palladina and Bershtein (1974), Kylin and

+

+

+

+

206

P. F. BROWNELL

Gee (1970) and Leonard and Hodges (1973) it appears that K+ in the absence of Na+ supports high activities of the ATPases. It therefore follows that the ATPases are not absolutely dependent upon sodium for their function. Furthermore, the ATPases dependent upon (Na+ K+) for their activity, occur in plants such as sugar beet and mangroves for which sodium does not seem to be essential. The concentrations of monovalent ions required for maximum activation of ATPases are high, approximately 50 mol m-3 K+ in oat roots (Leonard and Hodges, 1973), 100 mol m-3 (Na+ K+) in mangroves (Kylin and Gee, 1970) compared to the concentrations of sodium in the tissues of C4 plants receiving sufficient sodium for maximum growth (maximum growth was obtained in Kochiu chifdsii when its leaves contained only 10.1100 mmol kg-l sodium on a dry basis). It thus appears unlikely that sodium is required solely for the activation of ATPases in a C, species. It is possible that sodium might be involved in some way with the transport of metabolites at intra- and extracellular levels between structures having specialized functions in plants for which sodium is essential. In blue-green algae with heterocysts, there is good evidence that the heterocyst is the site of nitrogen-fixation and the vegetative cells the site of carbon assimilation (Fay el af., 1968; Stewart er al., 1969). The heterocysts supply combined nitrogen to the vegetative cells which are the source of assimilated carbon compounds for the heterocysts (Wolk, 1968). Similarly, in plants with the C, photosynthetic pathway, there is rapid shuttling of metabolites between organelles and cytoplasm and between mesophyll and bundle-sheath cells (Osmond and Smith, 1976).Although concentration gradients are considered to be sufficient to explain observed rates of transport without invoking an active process (Hatch and Osmond, 1976), it is still possible that there may be situations where active processes are involved. From work with leaf discs of Amarunthus panicufutus, an aspartate NADmalic enzyme type of C, plant, Raghavendra and Das (1977) suggest a possible light-dependent active transport of amino acids into the bundlesheath, mediated by Na+-dependent ATPase activity. They reported alanine production by illuminated discs when infiltrated with asparate which was accelerated by additions of a-oxoglutarate and glutamate. Alanine production was also supported by oxaloacetate but was only stimulated by glutamate. The aspartate-dependent alanine production was increased in the light or by the addition of ATP and sodium ions. They interpret these results as indicating a light-dependent transport of alanine into the bundle-sheath involving an Na+-dependent ATPase. It would be of interest to know if this effect is specific to sodium ions and if a similar phenomenon occurs in the PEP carboxykinase aspartate type of C4 plants. The concentration of sodium needed to give maximum rates of alanine production was high (approximately 100 mol m-3) compared to the concentrations in leaves of plants

+

+

SODIUM AND PLANT METABOLISM

207

receiving sufficient sodium for maximum growth. However, the actual concentration of sodium in the leaf discs may have been considerably lower than those of the incubating solutions. In certain bacterial cells, sodium has been shown to play a role in the transport of substrates (MacLeod, 1965; Kahane et al., 1975), (see Section IV. C. 1.). There seems to be little information on the mechanisms involved in the transport of metabolites in these organisms. If these organisms, blue-green algae, C , plants and sodium-requiring bacteria have active transport systems for metabolites, it is possible that sodium may be involved in their functions.

V. TENTATIVE SCHEMES FOR THE ROLE OF SODIUM IN C, AND CAM PLANTS AND I N BLUE-GREEN ALGAE Two schemes are suggested to summarize our present ideas on the role of sodium in plants for which it is essential. The first scheme applies to C , and CAM plants and the second to Anabaena cylindrica and possibly to other members of Cyanophyta. Both schemes must be regarded as extremely speculative due to the present lack of information but they appear to be consistent with the data available. At this time, it appears unlikely that sodium has a similar role in the two groups of plants. A.

C4

AND CAM PLANTS

As sodium is probably essential for all C , plants and increases growth of CAM plants (under conditions where they are largely dependent upon darkfixation of carbon dioxide) but does not appear to be essential for C , plants, it seemed likely that sodium might function within the C , dicarboxylic acid system operating in C , and CAM plants but not in C , plants. However, it has not been possible to obtain evidence for sodium having a direct role in the C, dicarboxylic acid system of either C , or CAM plants. Neither has sodium been found to affect the enzymes specifically associated with this system nor the operation of the overall system. The enzymes involved in the C4 dicarboxylic acid system include phosphoenol pyruvate carboxylase, pyruvate orthophosphate dikinase, adenylate kinase and pyrophosphatase (Hatch, 1970; Dittrich et al., 1973). Sodium nutrition does not appear to affect the formation of phosphoenol pyruvate carboxylase extracted from leaves of C, plants (Holtum, 1975; P. F. Brownell, unpublished work). Furthermore, sodium did not increase the activity of phosphoenol pyruvate carboxylase, pyruvate orthophosphate dikinase, or adenylate kinase in low-sodium assay media (M. D. Hatch, personal communication). The overall C, dicarboxylic acid system does not appear to be specifically dependent upon sodium nutrition. The pattern of 14Clabelling of C , acids in

208

P. F. BROWNELL

leaves of the C, species, Kochia childsii, exposed to 14C0, in the light was similar in sodium-deficient and normal plants (Holtum, 1975; Webb, 1977). In the CAM species, Bryophyllum tubiflorum, the rates of carboxylation in the dark and decarboxylation in the light were not affected by sodium nutrition (Boag, 1976). Despite the lack of evidence for any effects of sodium on this C, dicarboxylic system which supplies CO, to the photosynthetic carbon reduction cycle with the ultimate production of photosynthate, the sodium-deficient C, and shortday CAM plants have lower rates of photosynthesis than normal plants as shown in 14C-labellingand growth experiments (see section 1V.D). Therefore, it appears sodium is in some way essential for efficient photosynthetic assimilation of endogenous CO, produced by decarboxylation of malic or aspartic acids in the light. It would follow that sodium is involved in an area of metabolism (possibly not yet defined) associated with the PCR/PCO cycles and not with the C, dicarboxylic acid system common to C, and CAM plants. The following scheme proposed by Boag (1976) (Fig. 30) summarizes our present ideas on the role of sodium in C, and CAM plants primarily based on work with Bryophyllum tubijorum. The assumption is made that sodium has the same role in the nutrition of C, and CAM plants since they have similar primary carboxylation and decarboxylation systems. There are differences in the origin of the phosphoenol pyruvate and in the separation of the carboxylation and decarboxylation processes being spatial in C, plants and temporal in CAM plants. Plants with C, photosynthesis appear to have only one photosynthetic option whereas CAM plants are able to utilize a dark C0,-fixation system or direct CO, assimilation by the PCR cycle in the light in varying proportions depending upon the conditions of growth. In Fig. 30, the effects of sodium nutrition are compared in Bryophyllum tubijorum under short- and long-day regimes. The processes occurring in the dark, the primary carboxylation of phosphoenol pyruvate to form oxaloacetate and its reduction to malate are shown in the stippled portion of the diagrams. The processes occurring in the light, the decarboxylation of malate and the reduction of the CO, to photosynthate by the PCR cycle are shown in the unstippled areas. The CO, enters the PCR cycle (represented as a cube with two halves). Two sources of CO, are shown; one from the decarboxylation of malate generated in the dark-fixation process (supplying the left-hand half of the cube) and the other exogenous atmospheric CO, (supplying the right-hand half of the cube) where it is assimilated directly by RuDP carboxylase via the C,-like option. It is postulated, that under conditions of sodium-deficiency (Fig. 30a and b, lower diagrams) there is a loss of carbon derived from the C, dicarboxylic CO, fixation system in the PCR/PCO systems under both short- and long-day conditions. Evidence for this loss is that compared to normal plants sodiumdeficient plants produce less photosynthate despite the apparently similar

Fig. 30. Proposed scheme to explain the responses to sodium by plants of B. tubiforurn maintained under different growth regimes. This scheme indicates the possible pathways of carbon assimilation by +Na (i.e. plants supplied with 0.1 mmol NaCl I-l) and -Na (i.e. no addition) plants grown under; (a) long-days (16 h; 34"C)/short-warm-night (8 h, 24°C) regime. (b) short-days (8 h; 33"C)/long-cool-night (16 h; 13°C) regime. Thickness of line is a quantitative indication of significance of alternate pathways, CO, produced endogenously is shown enclosed by a box and exogenous or atmospheric CO, is not enclosed by a box. 1. PEP-carboxylase; 2. MDH; 3. Malic Enzyme; 4. RuDP-carboxylase; 5. RuDPoxygenase. From Boag (1976).

210

P. F. BROWNELL

supply of CO, from dark CO, fixation. This carbon is represented as being lost or “escaping” (through a “hole” in this part of the cube). Sodium is indicated preventing loss of carbon (as a “plug” filling this hole). Depending upon the growth regime of CAM plants, this loss of carbon would be expected to have a variable effect upon overall plant growth. If plants are grown under long-day conditions (Fig. 30a), a large proportion of the CO, assimilated by the plant will be derived directly from the exogenous atmospheric source to be assimilated by RuDP carboxylase: only a small part will be supplied by the decarboxylation of malate, produced in the dark-CO, fixation process. Under these conditions the decreased conversion of endogenous CO, to phosphate will be relatively unimportant compared to the overall rate of photosynthesis and the effect of sodium nutrition on the growth of the plants will not be readily detected. However, the situation in CAM plants under a short-day regime is very different (Fig. 30b). Little atmospheric carbon dioxide will be fixed directly by RuDP carboxylase and the plant will be thus largely dependent upon the photosynthetic assimilation of CO, produced by the decarboxylation of malate from dark C0,-fixation. Under these conditions, this decreased conversion of endogenous COz to photosynthate in sodium-deficient plants will be manifested by reduced growth and in severe cases, chlorosis and necrosis of the leaves. On the other hand, plants receiving sodium will not suffer this loss of carbon and will make normal, vigorous growth. Fig. 31. Proposed scheme to explain the role of sodium in the blue-green alga, Anabaena cylindrica grown in nitrate-containing cultures.

The following suggestions are made: A. Sodium-deficient cells: 1. The permeability of the membrane increases but the cell remains intact. 2. Nitrate diffuses rapidly into the cells inducing nitrate reductase activity which leads

to the more rapid reduction of nitrate. 3. The subsequent steps of nitrate assimilation leading to the formation of protein are accelerated (indicated by the relative thickness of lines). 4. When 15N-labellednitrite, nitrate or ammonia is supplied to sodium-deficient cells, the protein becomes more rapidly labelled than in normal cells. However, the increase in the amount of protein is less rapid in deficient than in normal cells suggesting the loss of nitrogenous compounds to the medium. Therefore, growth is retarded in deficient cultures. 5. A greater proportion of the carbon assimilated is also lost into the medium in deficient cultures in compounds including glycolate and certain nitrogenous compounds. B. Normal cells: 1. The membrane integrity is maintained permitting the uptake of nitrate at the usual lower rate. 2. Nitrate assimilation proceeds at a normal rate. 3. Larger pools of protein accumulate with only a low rate of loss of nitrogenous compounds, thus more material is available for growth. 4. Loss of carbon through normal release processes is moderate. 5. The rate of nitrogen-fixation on a protein basis is similar for both sodium-deficient and normal cells. From Smith (1977).

*I

cycle

hexose

I I

-

p glycolate Glycolate

I

General

II

Loss

I I I I

a - ketoglutarate

N2

I Nitrogenous

I

Material

I

I

growth I

TCF --

- ---

------hexose

cycle

+;;;ico2 Glycolysis p-glycolate /

I

+

General

- - - - --

I-"----

cycle

j

a-ketoglutarate

amino

acid and proteir

NH4

Carbon

3Loss

-A

+

Nitrogenous Materia I

growth Fig. 31

-3

212

P. F. BROWNELL

The way in which carbon could be lost from the PCR/PCO system is not known. However, it is possible that it is lost as CO,; perhaps sodium nutrition in some way affects the mode of action of RuDP carboxylase/oxygenase. B.

Anabaena cylindrica

The following tentative schemes (Figs 31 and 32) proposed by Smith (1977) summarize our present view of the involvement of sodium nutrition in the carbon and nitrogen metabolism of Anabaena cylindrica and possibly other members of Cyanophyta for which sodium is an essential element. Two schemes are presented, one for algae supplied with nitrate (Fig. 31) and the other for algae not supplied with combined nitrogen and thus dependent upon nitrogen-fixation for their source of nitrogen (Fig. 32). (In each figure, A represents sodium-deficient cells and B normal cells.) Sodium-deficient cells incorporate nitrogen from nitrate more rapidly than normal cells (indicated by the relative thickness of arrows in Fig. 31A and B). The evidence for this is the increased rates of 15Nenrichment of protein of cells fed K16N0, in their cultures, the enhanced specific activity of nitrate reductase (Brownell and Nicholas, 1967) and the greater rate of nitrate uptake in sodium-deficient compared to normal cultures (Smith, 1977). Smith (1977) also observed a more rapid release of nitrogenous compounds by sodium-deficient algae (Fig. 17). This was 33% of the total cellular nitrogen in sodium-deficient cultures compared to only 6 % for normal cultures supplied with nitrate. While the same trend occurred in cultures not supplied with combined nitrogen (Figs 18,32), the overall release of combined nitrogen was less, being only 17% of the total cellular nitrogen in sodium-

Fig. 32. Proposed scheme to explain the role of sodium in the blue-green alga, Anabaena cylindrica grown without combined nitrogen. The following suggestions are made: A. Sodium-deficient cells: 1. Nitrate assimilation does not occur, therefore sodium nutrition is not involved in this system. 2. Assimilation of nitrogen occurs exclusively via nitrogen fixation at overall rates less than for cells receiving nitrate. There is a smaller pool of protein and release of extracellular nitrogen compounds compared to those in nitrate-grown cultures. 3. Nitrogenous compounds needed for growth are lost more rapidly into the medium and growth is therefore reduced. 4. Glycolate and other carbon compounds are released in higher amounts by deficient cells. 5. It is suggested that the membranes are affected in a similar manner as in nitrategrown cells. B. Normal cells: 1. Similar to nitrate-grown cultures but the nitrate assimilation system is not operative. 2. Nitrogen fixation on a protein basis is similar in sodium-deficient and normal cells. 3. Release of nitrogen and carbon-containing compounds is less by normal than sodium deficient cells. From Smith (1977).

I N O i k >

a - ketoglutarate

NOS

43 > -

amino I acid and protein pools

NHt

'

N2

I

growth

a-ketoglutamte

acid and protein pools

growth Fig. 32

Material

214

P. F. BROWNELL

deficient cultures compared to 7 % in normal cultures. The production of extracellular nitrogenous compounds by blue-green algae has been recorded by Fogg (1952) who found increased production of mainly polypeptide and lesser amounts of amide in cultures deficient in iron and molybdenum. The amounts also increased with age of culture. Jones and Stewart (I 969) also found greater losses of extracellular nitrogen in Calothrix scopulorum under suboptimal conditions. They suggested that the peptides and traces of amino acids found may have possibly originated from the mucilaginous sheath surrounding the alga. The increased release of nitrogenous compounds may be a general effect of suboptimal growth conditions and not a specific effect of sodium deficiency. However, the greater rate of uptake and reduction of nitrate appears likely to be a specific effect of sodium deficiency. Sodium-deficient cultures of A . cylindrica released increased organic carbon, expressed as a percentage of the total carbon assimilated (Ward and Wetzel, 1975; Smith, 1977). Sodium-deficient cells in cultures not supplied with combined nitrogen released significantly more glycolate into the media than normal cells (Smith, 1977). Glycolate is one of the more common substances excreted by algae (Fogg, 1952; Cheng et al., 1972; Hellebust, 1965). The production and release of glycolate is of interest in that its possible source could be the PCO cycle (Cheng and Colman, 1974; Ingle and Colman, 1976) which would indicate a higher rate of photorespiration in sodiumdeficient cells. The loss of carbon would be expected to be closely associated with the loss of nitrogen as much of the nitrogen compounds would contain carbon. Brownell and Nicholas (1967) suggested that sodium may exert its control specifically on nitrate reductase through the production of a protein factor. They based their suggestion on the increased levels of nitrite in sodiumdeficient cultures and the failure of sodium to control the activity of nitrate reductase when supplied simultaneously with chloramphenico, an inhibitor of protein synthesis. However, evidence from 15Nand 14Clabelling experiments indicated that the overall activity of the nitrate assimilation system increased with a deficiency of sodium. Smith (1977) also observed greater rates of nitrate reduction but they were not accompanied by the marked increases in nitrite levels in sodium-deficient compared to normal cultures that had been found by Brownell and Nicholas (1967). To explain the increased rates of nitrate assimilation, it was initially suggested that sodium deficiency may have the multiple effect of increasing the activities of all the enzymes concerned in this process. However, Smith (1977) considered it more likely that sodium could be needed to maintain the integrity of membranes. In a deficiency of sodium, the membranes would offer little resistance to the uptake of nitrate which would be available for the induction of the inducible enzyme, nitrate reductase. Ohmori and Hattori (1970) found that nitrate reductase is induced in Anabaena cylindrica by its

SODIUM AND PLANT METABOLISM

215

substrate, nitrate, and its activity depends upon the rate of entry of nitrate into the cells. This report is consistent with the model proposed by Butz and Jackson (1977) for the transport and reduction of nitrate by a membranelocated nitrate reductase enzyme. With the increased availability of nitrate at the sites of nitrate reductase induction, both the nitrate reductase activity and the rate of nitrate reduction would increase in sodium-deficient cells. The rate of the subsequent steps of nitrate assimilation would also be expected to increase in response to the higher substrate levels in the sodium-deficient compared to normal cells. The leaky membranes in sodium-deficient cells could also explain the observed losses of nitrogenous compounds including polypeptides, amides and amino acids and the carbon compounds including glycolate. Under these conditions, compounds of nitrogen and carbon, which in normal cells would contribute to growth would be lost to the surrounding medium. In cells grown without nitrate, there could also be losses of carbon- and nitrogen-containing compounds due to the defective membranes (Fig. 32). Loss of compounds containing carbon and nitrogen would reduce the amount of metabolites available for growth and this would account for the poorer growth in sodium-deficient cultures. Losses of these compounds from cells receiving sodium would be minimal and normal growth would result. With the data available, it is not possible to ascertain at what stage of assimilation the losses occur. Assimilated carbon could be released from the PCRjPCO system as glycolate or it may be released with nitrogen in amino acids, amides, peptides or proteins. Carbon and nitrogen could be released at steps prior to the formation of proteins or they may be derived from the breakdown of protein. This scheme for the involvement of sodium in the nutrition of blue-green algae requires firm evidence for sodium being needed for the integrity of the cell membranes. The evidence for sodium having this role is based only on physiological data ; there is no histological evidence for this possible effect of sodium on membrane integrity. No obvious differences between membranes of sodium-deficient and normal cells have been revealed by light microscopy. If sodium is involved in the maintenance of membrane structure, this may become apparent following studies of ultrastructure by electron microscopy.

VI. SUMMARY AND CONCLUSIONS This article concentrates on sodium as an essential micronutrient element for certain higher plants and microorganisms. 2. A clear distinction is drawn between the roles of sodium as an essential micronutrient element for some species and as a beneficial element (at much higher concentrations) for some others. 3. Results of many experiments to determine the effect of sodium on 1.

216

4.

5.

6. 7.

8.

9.

10.

11.

P. F. BROWNELL

growth have been difficult to interpret due to the lack of control treatments, the use of complicated media and to the omission of micronutrients in the culture media. Sodium, at high levels above 10mol rn-, (230ppm) has stimulated growth in certain species (mainly members of Chenopodiaceae and other species including Aster tripolium and some mangroves) even when other essential elements were supplied at optimum concentrations. Other species responded to high levels of sodium by increased growth only in a deficiency of a nutrient element. Wheat, barley, oats and cotton have been shown to respond to sodium in a deficiency of potassium and cotton in a deficiency of calcium. Some species do not respond to high levels of sodium even when other nutrient elements including potassium are limiting. Sodium, at high levels, has been shown to affect some plants qualitatively. Increases in the concentration of sugar have been recorded in sugar beet and the quality of fibre crops improved by application of salts of sodium. Following the suggestion that sodium may be required in very small amounts as an essential element, experiments were conducted in which sodium was carefully removed from the environment of the plants. Critical techniques for the determination of sodium in low concentrations, sometimes in the presence of high concentrations of other substances were needed to assess the relative sources of sodium to the plant. Initially, emission flame photometry was used but this method was largely replaced by the atomic absorption spectroscopy method in which interferences were minimal. Potential sources of sodium to the cultures were the culture solution salts, the water, the culture apparatus, the seeds, air for aeration of cultures and the air surrounding the cultures. Methods are described which greatly reduce the magnitude of these sources. Atriplex vesicaria was the first higher plant for which sodium was shown to be essential as a micronutrient. Following this discovery, it was suggested that sodium might be required by other species which are halophytes or which accumulate sodium at high concentrations in their tissues. However, it must be stressed that there is no clear correlation between the possession of halophytic or sodium-accumulating properties and the requirement of sodium as a micronutrient. Sodium has been shown to be essential for some blue-green algae including Anabaena cylindrica, and for C, species in the families, Gramineae, Cyperaceae, Amaranthaceae, Chenopodiaceae and Portulacaceae. In two genera of Chenopodiaceae, Atriplex and Kochia, which contain both C, and C, species, sodium was shown to be essential only for the C4 species. The C, species apparently had no requirement for sodium.

SODIUM AND PLANT METABOLISM

217

12. The minimum concentrations of sodium required for maximum growth of Atriplex vesicaria were about 0.05 mol m-3 (1.25 ppm) and for Anabaena cylindrica 0.22 mol m-3 (5 ppm). No other monovalent cation would support growth in the absence of sodium. 13. The Crassulacean acid metabolism species, Bryophyllum tubiflorum responded to low concentrations of sodium when grown under conditions of short-days with large diurnal temperature variation. Under these conditions, the bulk of the CO, assimilated was taken up in the dark involving the C, dicarboxylic system. 14. The signs of sodium deficiency are similar in all species for which it is essential. Leaves become chlorotic and in severe cases the margins and tips become necrotic. Full recovery can be obtained from chlorosis by the addition of a sodium salt at about 0.1 mol m-3 (2.3 ppm) to the culture solution. If 5 pg Na in a solution of NaCl or Na,S04 is applied to a single leaf of a sodium-deficient plant of Kochia childsii, it will green-up and expand rapidly whereas even after two weeks the remainder of the plant retains its signs of sodium deficiency. When 22NaC1in the same quantity is applied to leaves a large proportion of the radioactivity (63 % to 85 %) is retained by the leaf after one week. 15. Sodium has been reported to be essential (generally at concentrations of 50 mol m-3 and higher) for some marine bacteria and fungi and nonhalophilic bacteria. 16. Little is known of the function of sodium in organisms for which it is essential. Sodium-deficient compared to normal Anabaena cylindrica produces less dry weight and protein and has lower nitrogen-fixation rates. 17. In nitrate-containing cultures, sodium-deficient compared to normal Anabaena cylindrica takes up nitrate nitrogen and incorporates it into protein more rapidly both on a culture and protein basis. Nitrate reductase extracted from sodium-deficient cells has a specific activity many times greater than from normal cells. No similar effect on nitrate reduction was detected in C, plants. 18. Sodium-deficient cells, with or without combined nitrogen release nitrogenous compounds more rapidly and a greater proportion of assimilated carbon including glycolate than normal cells. 19. An early response by sodium-deficient leaves of C, Atriplex to sodium was an increase in the rate of respiration as oxygen uptake or CO, output. The concentration of sodium to elicit this response was of the same order as that required for growth responses. Sodium was specifically required; no response was obtained to other monovalent cations. This does not appear to be a general response for C , plants as no clear respiratory response to sodium was obtained in Kochia chifdsii. Chlorophyll content increased rapidly following the supply of sodium.

218

P. F. BROWNELL

20. It seemed likely that sodium was required for the functioning of the C, dicarboxylic C0,-fixing system operating in C, and CAM species (under certain conditions) but was not required in C, species which lack this system. However, there is little evidence to support this. Neither the whole system nor its key enzymes appear to depend on sodium for their operation. 21. A tentative scheme is suggested to explain the role of sodium in C , and CAM plants. A deficiency of sodium apparently does not affect the rate of CO, supply via the C , dicarboxylic system for reduction in the PCR/PCO system yet the overall photosynthetic rates are reduced. This suggests that carbon, in some form, is being lost from the system. It is postulated that sodium prevents this loss in a normal plant. One possibility is that sodium may, in some way, control the ribulose diphosphate carboxylase/oxygenase option, allosterically. 22. The role of sodium in the nutrition of Anabaena cylindrica and possibly in other blue-green algae is shown in a separate scheme. Sodium is postulated as being needed to maintain membrane integrity. In a deficiency of sodium, the membranes of cells growing in nitrate permit the rapid entry of nitrate resulting in the induction of high nitrate reductase activity. The subsequent steps of nitrate assimilation are accelerated due to high concentrations of substrates. It is suggested that increased losses of compounds containing carbon and nitrogen may be the result of increased permeability of membranes in a deficiency of sodium. There is no electron micrograph evidence for an effect of sodium on the ultrastructure of membranes, at present. 23. It has been suggested that sodium may act as a monovalent cation activator for certain enzymes in plants for which it is essential. This does not seem to be a likely function for sodium as no plant enzymes are known that are specifically activated by sodium. Furthermore, enzymes which are activated by monovalent cations require very high concentrations of the ions for maximum activity. These concentrations are generally higher than the concentrations of sodium in tissues of plants receiving enough sodium for maximal growth. 24. The proposal that sodium is required for the activation of ATPase systems is difficult to support. ATPases stimulated by Na+ and Kf have been found in some species but these are not species for which sodium has been shown to be essential. The (Na+ K+) ATPases appear to be at least partially activated by K+, alone. The levels of Na+ and K+ required for maximal activity are high compared to the tissue concentrations of sodium necessary for maximal growth. It is possible that sodium has a role in the transport of metabolites in plants for which it is essential but no mechanisms have been suggested for this.

+

SODIUM AND PLANT METABOLISM

219

ACKNOWLEDGEMENTS

I thank Dr C. J. Crossland for his helpful comments and constructive criticisms of parts of the manuscript. I am also grateful t o Messrs T. S. Boag, D. W. Mill, M. K. Smith and M. J. Webb for making available unpublished material and for their encouragement. I acknowledge Mr H. Lamont for photography of plants and Messrs J. Ngai and P. J. Maccarone for making diagrams. I am especially grateful to Mrs D. McNamara and Miss C. McPherson for their careful preparation of the manuscript. Finally, I wish to thank Mrs N. Incoll for carrying out the major task of converting the data in this chapter to SI units. Part of the research by the author was supported by the Chilean Nitrate Corporation and the Australian Research Grants Committee. REFERENCES Adams, S. N. (1961). Rep XXIV Winter Congr. Int. Sugar beet Research. Allen, M. B. (1952). Arch Mikrobiol. 17, 34-53. Allen, M. B. and Arnon, D. I. (1955). Physiologia PI. 8, 653-660. Arnon, D. I. and Stout, P. R. (1939). Pl. Physiol., 14, 371-375. Ashby, W. C. and Beadle, N. C. W. (1957). Ecology 38, 344-352. Baumeister, W. and Schmidt, L. (1962). Flora, Jena 152, 24-56. Benecke, W. (1898). Bot. Ztg. 56, 83-96. Bertrand, G. (1929). Annls. Sci. agron. ,fr. 46, 1-8. Birner, H. and Lucanus, B. (1866). Landwirtsch. Versuchsstat. 8, 128-177. Black, R. F. (1954). Aust. J . Bot. 2, 269-286. Black, R. F. (1960). Ausr. J . biol. Sci. 13, 249-266. Boag, T. S. (1976). “Responses to sodium by the Crassulacean Acid Metabolism Plant Bryophyllum tubiflorum Harvey grown under different environmental conditions”. Honours Thesis, James Cook University of North Queensland. Bostwick, C. D., Brown, L. R. and Tischer, R. G. (1968). Physiologia PI. 21, 466469. Box, G. F. and Walsh, A. (1959). C.S.I.R.O. Ind. Res. News No. 17. Brownell, P. F. (1958). “Sodium as an essential micronutrient element for higher plants”. Ph.D. thesis, University of Adelaide. Brownell, P. F. (1959). Report on Second Australian Spectroscopy Conference. Nature, Lond. 184, 1195-1 197. Brownell, P. F. (1965). PI. Physiol. 40, 460-468. Brownell, P. F. (1968). PI. Soil 28, 161-164. Brownell, P. F. and Crossland, C. J. (1972). PI. Physiol. 49, 794-797. Brownell, P. F. and Crossland, C. J. (1974). PI. Physiol. 54, 416417. Brownell, P. F. and Jackman, M. E. (1966). PI. Physiol. 41, 617-622. Brownell, P. F. and Nicholas, D. J. D. (1967). PI. Physiol. 42, 915-921. Brownell, P. F. and Wood, J. G. (1957). Nature, Lond. 179, 653-656. Broyer, T. C., Carlton, A. B., Johnson, C. M. and Stout, P. R. (1954). PI. Physiol. 29, 526-532. Brulfert, J., Guerrier, D. and Queiroz, 0. (1973). PI. Physiol. 51, 220-222. Bryant, M. P., Robinson, 1. M. and Chu, H. (1959). J. Diary Sci. 42, 18311847.

220

P. F. BROWNELL

Butz, R. G. and Jackson, W. A. (1977). Phytochemistry 16, 409-417. Canvin, D. T. and Beevers, H. (1961). J. bid. Chem. 236, 988-995. Chen, T. M., Brown, R. H. and Black, C. C. (1971). PI. Physiol. 47, 199-203. Cheng, K. H. and Colman, B. (1974). PIanta 115, 207-212. Cheng, K. H., Miller, A. G. and Colman, B. (1972). Planta 103, 110-116. Connor, D. J. (1969). Biotropica 1, 36-40. Dittrich, P., Campbell, W. H. and Black, C. C. (1973). PI. Physiol. 52, 357-361. Dohler, G. (1974). PIanta 118, 259-269. Dorph-Petersen, K. and Steenbjerg, F. (1950). PI. Soil 2, 283-300. Downton, W. J. S. and Tregunna, E. B. (1968). Can. J. Bot. 46,207-215. Drapeau, G. R. and Macleod, R. A, (1963). Biochem. biophys. Res. Commun. 12, 11 1-115. Droop, M. R. (1958). Verhandf.int. Verein. theor. angew. Limnol. 13, 722-730. El-Sheikh, A. M. and Ulrich, A. (1970). PI. Physiol. 46, 645-649. El-Sheikh, A. M., Ulrich, A. and Broyer, T. C. (1967). PI. Physiol. 42, 1202-1208. Emerson, R. and Lewis, C. M. (1942). J. gen. Physiol. 25, 579-595. Epstein, E. (1972). “Mineral Nutrition of Plants: Principles and Perspectives”, pp. 294-295. John Wiley and Sons Inc., New York. Evans, H. J. and Sorger, G. J. (1966). A . Rev. PI. Physiol., 17, 47-76. Evans, H. J. and Wildes, R. A. (1971). I n “Proceedings of 8th Colloq. lnt. Potash Inst., Uppsala”, pp. 13-39. (Int. Potash Inst. Berne.) Eyster, C. (1970). I n “Taxonomy and Biology of Blue Green Algae”. (T. V. Desikachary, ed.), pp. 508-520. Univ. Madras, India. Fay, P., Stewart, W. D. P., Walsby, A. E. and Fogg, G. E. (1968). Nature. Lond. 220, 810-812. Filner, P. (1966). Biochim. biophys. Acta 118, 299-310. Fisher, J. D. and Hodges, R. K. (1969). PI. Physiol. 44,385-395. Fisher, J. D., Hansen, D. and Hodges, R. K. (1970). PI. Physiol. 46, 812-814. Flowers, T. J. (1975). I n “Ion Transport in Plant Cells and Tissues’’ (D. A. Baker and J. L. Hall, Eds), pp. 309-334. North-Holland Publishing Company, Amsterdam, Oxford American Elsevier Publishing Company, Inc., New York. Flowers, T. J., Troke, P. F. and Yeo, A. R. (1977). A . Rev. PI. Physiol. 28, 89-121. Fogg, G. E. (1952). Proc. R . SOC.B. 139, 372-397. Gutierrez, M., Gracen, V. E. and Edwards, G. E. (1974). Planta 119, 279-300. Hansson, G. and Kylin, A. (1969). Z. Pfphys. 60, 270-275. Harmer, P. M. and Benne, E. J. (1945). Soil Sci. 60, 137-148. Harmer, P. M., Benne, E. J., Laughlin, W. M. and Key, C. (1953). SoilSci. 76, 1-17. Hartt, C. E. (1934). PI. Physiol. 9, 399-452. Hartwell, B. L. and Pember, F. R. (1908). Univ. Rhode Isl. Agric. Exp. Sta. Ann. Rept. 21, 243-285. Hartwell, B. L. and Damon, S. C. (1919). Univ. Rhode Isl. Agric. Exp. Sta. Bull. 177, 4-32. Hatch, M. D. (1970). I n “Photosynthesis and Photorespiration” (M. D. Hatch, C. B. Osmond and R. 0. Slatyer, Eds), pp. 139-152. Wiley-Interscience, New York. Hatch, M. D. (1976). I n “Plant Biochemistry” (J. Bonner and J. E. Varner, Eds) 3rd edn. pp. 797-884. Academic Press, New York. Hatch, M. D. and Osmond, C. B. (1976). I n “Encyclopedia of Plant Physiology” New Series, Volume III (C. R. Stocking and U. Heber, Eds), pp. 144-184. Springer-Verlag, Berlin, Heidelberg, New York. Hatch, M. D. and Slack, C. R. (1966). Biochem. J. 101, 103-111.

SODIUM AND PLANT METABOLISM

22 1

Hatch, M. D. and Slack, C. R. (1970). A . Rev. PI. Physiol. 21,141-162. Hatch, M. D., Kagawa, T. and Craig, S. (1975). Aust. J. PI. PhysioI. 2, 111-128. Hawker, J. S., Marschner, H. and Downton, W. J. S. (1974). Aust. J . PI. Physiol. 1, 491-501. Hellebust, J. A. (1965). Limnol. Oceanogr. 10, 192-206. Hellriegel, H. and Willfarth, H. (1898). Arb. dt. Ges. 34, Cited by Dorph-Petersen and Steenbjerg (1950). Hewitt, E. J. (1963). In “Plant Physiology” (F. C. Steward, Ed.) Vol. 111, pp. 137360, Academic Press, New York and London. Hewitt, E. J. (1966). “Sand and Waterculture Methods Used in the Study of Plant Nutrition”. Technical Communication No. 22 2nd ed. Commonwealth Bureau of Horticulture and Plantation Crops, East Malling, Maidstone, Kent, Commonwealth Agricultural Bureaux, Farnham Royal, Bucks, England. Hewitt, E. J. and Smith, T. A. (1975). “Plant Mineral Nutrition”. p. 31. The English Universities Press, Ltd., London. Hoagland, D. R. and Arnon, D. I. (1938). Circ. Calif. Agric. Exp. Sin. No. 347, pp. 1-9. Hodges, T. K., Leonard, R. T., Bracker, C. E. and Keenan, T. W. (1972). Proc. natn. Acad. Sci. USA 69, 3307-331 1 . Hokum, J. A. M. (1975). “The effect of sodium on carbon assimilation in plants possessing the C, dicarboxylic acid pathway of carbon fixation”. Honours Thesis, James Cook University of North Queensland. Hutton, J. T. (1953). “Atmospheric Accessions”, Aust. Conf. Soil Sci., Adelaide, 1953, Paper No. 6, 1 1 . 1 . Ingle, R. K. and Colman, B. (1976). Planta 128, 217-223. Jacobson, L. (1951). PI. Physiol. 26,411-413. Jacobson, L. (1955). PI. Physiol. 30,264-269. Jacobson, L. and Ordin, L. (1954). PI. Physiol. 29, 70-75. Jennings, D. H. (1968). New Phytol. 67, 899-911. Jennings, D. H. (1976). Biol. Rev. 51, 453486. Jeschke, W. D. and Stelter, W. (1976). Planta 128, 107-112. Joham, H . E. (1955). PI. Physiol. 30, 4-10. Joham, H. E. and Amin, J. V. (1965). Soil Sci. 99, 220-226. Joham, H. E. and Gossett, D. (1974). “Proceedings Beltwide Cotton Production, Research Conferences”. p. 35. Joham, H. E. and Johanson, L. (1973). Physiologia PI. 28, 121-126. Johanson, L. and Joham, H. E. (1971). PI. Soil 35, 323-336. Johnson, C. M., Stout, P. R., Broyer, T. C. and Carlton, A. B. (1957). PI. Soil8, 337-353. Jones, A. G. (1960). “Analytical Chemistry some New Techniques”. pp. 1-46. Butterworths Scientific Publications, London. Jones, L. W. and Stewart, W. D. P. (1969). J. Mar. Biol.Ass. UK 49, 475488. Kahane, M. M., Barash, H. and Halpern, Y. S. (1975). FEBS Lett. 56, 235-239. Karlsson, J. and Kylin, A. (1974). Physiologia Pl. 32, 136-142. Kennedy, R. A. (1977). Z. Pflphys. 83, 11-24. Kodama, T. and Taniguchi, S. (1976). J. gen. Microbiol. 96, 17-24. Kodama, T. and Taniguchi, S. (1977). J. gen. Microbiol. 98, 503-510. Kratz, W. A. and Myers, J. (1955). Am. J. Bot. 42, 282-287. Kushizaki, M. and Yasuda, T. (1964). Research Bulletin No. 84, Hokkaido National Agricultural Experiment Station, Japan. Kylin, A. (1966). PI.Physiol. 41, 579-584.

222

P. F. BROWNELL

Kylin, A. and Gee, R. (1970). PI. Physiol. 45, 169-172. Lehr, J. J. (1953). J. Sci. Food Agric. 4, 460-471. Lehr, J. J. (1957). Chilean Nitrate Agricultural Service No. 30. Lehr, J. J. and Wybenga, J. M. (1955). PI. Soil6, 251-261. Lehr, J. J. and Wybenga, J. M. (1958). PI. Soil9, 237-253. Leonard, R. T. and Hodges, R. K. (1973). PI. Physiof. 52, 6-12. Leonard, R. T., Hansen, D. and Hodges, T. K. (1973). PI. Physiof. 51, 749-754. Lundegardh, H. (1955). A. Rev. PI. Physiol. 6, 1-24. McCollum, R. E., Hageman, R. H. and Tyner, E. H. (1958). Soil Sci. 86, 324331. McLachlan, J. (1960). Can. J. Microbiof. 6, 367-369. MacLeod, R. A. (1965). Bacteriof. Rev. 29, 9-23. MacLeod, R. A. (1968). Adv. Microbiol. Sea 1, 95-126. MacLeod, R. A. and Onofrey, E. (1963). I n “Symposium on Marine Microbiology” (C. H. Oppenheimer, Ed.), pp. 481-489, Charles C. Thomas, Springfield, Illinois. Marschner, H. (1971). I n Proceedings of 8th Colloq. Int. Potash Inst., Uppsala, pp. 50-63 (Int. Potash Inst. Berne). Marschner, H. (1975). Prog. Bot. 37, 98-106. Mill, D. W. (1977). “The transport and distribution of sodium in C,plants”. Honours Thesis, James Cook University of North Queensland. Miller, G. W. and Evans, H. J. (1957). PI. Physiol. 32, 346-354. Montasir, A. H., Sharoubeem, H. H. and Sidrak, G. H. (1966). PI. Soil25, 181-194. Moscolov, I. V. and Aleksandrovskaya, V. A. (1962). Chem. Abstr. 56(2), Abstr. 178431. Moser, H. (1934). Beih. bot. Zbf. 52, 378-388. Mullison, W. R. and Mullison, E. (1942). PI. Physiof. 17, 632-644. Nitsos, R. E. and Evans, H. J. (1969). PI. Physiol. 44, 1260-1266. Nobbe, F., Schroeder, J. and Erdmann, R. (1871). Landw. Vers-Sta. 13, 321-399; 401423. Cited by Hartt (1934). Ohmori, K. and Hattori, A. (1970). PI. Celf Physiol., Tokyo 11, 873-878. Oji, Y.and Izawa, G. (1969). P1. Celf Physiol., Tokyo 10, 665-674. Osmond, C. B. (1969a). Biochim. biophys. Acta 172, 144-149. Osmond, C. B. (196913). Z . Pflphys. 62, 129-132. Osmond, C. B. (1975). I n “Environmental and Biological Control of Photosynthesis” (R. Marcelle, Ed.), pp. 311-321. Dr. W. Junk, The Hague. Osmond, C. B. (1976). I n “CO2 Metabolism and Plant Productivity” (R. I. Burris and C. C. Black, Eds), pp. 217-233. University Park Press, Baltimore. Osmond, C. B. and Smith, F. A. (1976). In “Intercellular Communication in Plants: Studies on Plasmodesmata” (B. E. S. Gunning and A. W. Robards, Eds), 229-241. Springer-Verlag, Berlin, Heidelberg, New York. Osmond, C. B., Luttge, U., West, K. R., Pallaghy, C. K. and Shacher-Hill, B. (1969). Aust. J. biol. Sci. 22, 797-814. Osterhout, W. J. V. (1909). Bot. Guz. 48, 98-104. Osterhout, W. J. V. (1912). Bot. Gaz. 54, 532-536. Ozanne, P. G., Woolley, J. T. and Broyer, T. C. (1957). Aust. J. biof. Sci. 10, 66-79. Palladina, T. A. and Bershtein, B. I. (1974). Soviet PI. Physiol. 20, 664-667. Pfeffer, W. (1899). In “The Physiology of Plants” (A. J. Ewart, Translator and Ed.). Vol. I, pp. 41 1-430. Clarendon Press, Oxford. Pleunneke, R. H. and Joharn, H. E. (1972). PI. Physiol. 49, 502-505. Pratt, D. (1974). I n “Effect of the Ocean Environment on Microbial Activities”

SODIUM AND PLANT METABOLISM

223

(R. R. Colwell and R. Y. Morita, Eds), pp. 3-15. University Park Press, Baltimore, London, Tokyo. Quebedeaux, B. and Chollet, R. (1977). PI. Physiol. 59, 4 2 4 4 . Raghavendra, A. S. and Das, V. S. R. (1977). New Phytol. 79,481-487. Rains, D. W. (1972). A . Rev. PI. Physiol. 23, 367-388. Robertson, R. N. and Turner, J. S. (1945). Aust. J . exp. Biol. med. Sci. 23, 63-79. Shomer-ilan, A. and Waisel, Y. (1973). Physiologia PI. 29, 190-193. Siegenthaler, P. A., Belsky, M. M. and Goldstein, S. (1967). Science 155, 93-94. Simonis, W. and Urbach, W. (1963). Arch. Mikrobiol. 46, 265-286. Singh, J. and Dieckert, J. W. (1973). Analyt. Biochem. 53, 445-447. Singh, J. and Dieckert, J. W. (1975). Soil Sci. 119, 238-241. Sistrom, W. R. (1960). J. gen. Microbiol. 23, 778-785. Skou, J. C. (1964). Progr. Biophys. Mol. Biol. 14, 131-166. Slack, C. R. and Hatch, M. D. (1967). Plantu 103, 660-665. Smith, B. N. and Brown, W. V. (1973). Am. J . Bot. 60, 505-513. Smith, B. N. and Epstein, S. (1971). PI. Physiol. 47, 380-384. Smith, M. K. (1977). “The effect of sodium on the nitrogen and carbon metabolism of the blue-green alga, Anabaena cylindrica”. Honours Thesis, James Cook University of North Queensland. Steinberg, R. A. (1937). J . Agric. Res. 55, 891-892. Stewart, W. D. P., Haystead, A. and Pearson, H. W. (1969). Nature Lond. 224, 226-228. Szymanek, J. (1952). Bulletin de 1’Institute Textile de France, No. 86. Ting, I. P. (1970). In “Photosynthesis and Photorespiration”. (M. D. Hatch, C. B. Osmond and R. 0. Slatyer, Eds), pp. 169-185. Wiley-Interscience, New York. Ting, I. P. (1976). I n “ C 0 2 Metabolism and Plant Productivity”. (R. I. Burris and C. C. Black, Eds), pp. 251-268. University Park Press, Baltimore. Treggi, G . (1961). PI. Soil 15, 291-294. Tregunna, E. B., Smith, B. N., Berry, J. A. and Downton, W. J. S. (1970). Can. J . Bot. 48, 1209-1214. Truog, E., Berger, K. C. and Attoe, 0. J. (1953). Soil Sci. 76, 41-50. Tullin, V. (1954). Physiologia PI. 7, 810-834. Turton, A. G. (1953). “Atmospheric Accessions”. Aust. Conf. Soil Sci., Adelaide, 1953. Paper No. 6. 14.1. Ullrich-Eberius, C. I. (1973). PIunta 109, 162-176. Ulrich, A. and Ohki, K. (1956). PI. Physiol. 31, 171-181. Vishniac, H. S . (1960). Limnol. Oceanogr. 5, 362-365. Vogel, A. I. (1961). “A Text-book of Quantitative Inorganic Analysis including Elementary Instrumental Analysis”. pp. 879-889. Longmans, Green and Co. Ltd., London. van Eijk, M. (1939). R e d . Trav. bot. ne‘erl. 36, 559-659. Waisel, Y. (1972). “Biology of Halophytes”. Academic Press, New York and London. Wallace, A., Toth, S. J. and Bear, F. E. (1948). Soil Sci. 65, 245-259. Walsh, A. (1955). Spectrochim. Acta 7 , 108-1 17. Ward, A. K. and Wetzel, R. G. (1975). J. Phycol. 11, 357-363. Webb, M. J. (1977). “The effect of sodium on the pathway of photosynthetically assimilated carbon in the C, plant Kochia childsii Hort.” Honours Thesis, James Cook University of North Queensland. Welkie, G. W. and Caldwell, M. (1970). Can. J. Bot. 45, 2135-2146.

224

P. F. BROWNELL

Whitenberg, D. C. and Joham, H. E. (1974). Bot. Guz. 125, 56-62. Willert, von. D. J. and Kramer, D. (1972). Pluntu 107, 227-237. Williams, M. C. (1960). PI. Physiol. 35, 500-505. Winter, K. and Willert, von. D. J. (1972). Z. Pflphys. 67, 166-170. Wolk, C. P. (1968). J. Bucteriol. 96, 2138-2143. Wood, J. G. (1925). Aust. J. exp. Biol. med. Sci.2, 45-56. Woolley, J. T. (1957). PI. Physiol. 32, 317-321. Wybenga, J. M. (1957). In “Investigations with Radioactive Sodium”. Kantoordruk de Goede, Wageningen, Holland. Wybenga, J. M. and Treggi, G. (1958). PI. Soil 10, 205-260.

Aeration in Higher Plants

W . ARMSTRONG Department of Plant Biology. The University of Hull. England

I . Introduction

. . . . . . . . . . . . . . . . .

I1. Principles of Aeration and Aeration Modelling . A. Entry and Dispersal of Respiratory Gases B. Diffusion and the Ventilating Process . C. The Oxygen Source . . . . . . . D. The Aeration Model . . . . . .

.

111

. . . . . .

. . . . . . . . . . . .

227 227 242 260 265

. . . . . . .

272

. . . . . . . . . .

278 278 298 305

The Cylindrical Platinum Electrode Technique

IV. Aeration in the Wetland Condition A. The Wetland Plant . . . B. The Non-wetland Plant . . C. Trees . . . . . . .

. . . . . . . . . . . .

226

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

V. Root Aeration in the Unsaturated Soil Acknowledgements . . . . . .

. . . . . . . . . . . . . . . . . .

313 324

Appendix 1 The Transport of Diffusible Species in Media Moving by

. . . . . . . . . . . . .

325

. .

326

. . . . . . . . . . . . . . . . . . . . . . . . . . . .

328 332

Mass Flow

Appendix 2 Radial Diffusion into Respiring Spherical Bodies

References . . . Note Added in Proof

226

W. ARMSTRONG

I. INTRODUCTION While there is a considerable volume of literature dealing with the diffusive influx of carbon dioxide and efflux of water vapour across leaf surfaces, the gas movements necessary to sustain respiratory activities throughout the plant body attract little comment in physiological texts. Historically, this can be attributed to the relative intensities of research in the two fields: whilst Browne and Escombe had by 1900 laid the foundation of the diffusion mathematics concerned with gaseous fluxes through stomata1 surfaces, as recently as 1937 Conway was commenting that the general assumption of gas-phase continuity of intercellular spaces thoughout the wetland plant body had not yet been tested. A gas phase continuum in the cortical intercellular space system of non-wetland species was not even a general assumption ten years later. Brown (1947) concluded that longitudinal movements of nitrogen, oxygen and hydrogen which he could induce between shoot and root and vice versa in seedlings of Cucurbita pep0 were not diffusion controlled but rather were dependent upon the active translocation of gases in solution. Brown was perhaps the first to demonstrate conclusively that there was oxygen leakage from root surfaces. We now know that in wetland and nonwetland plants alike, the major movements of respiratory gases between shoot and root occur in the gas-phase and are largely controlled by diffusion. Current interest in whole plant aeration (but more particularly the aeration of below ground parts) can be traced back about 40 years, to experiments performed on wetland species. First Conway (1937), followed by Laing (1940) and the Dutch scientist Van Raalte (1941, 1944) convincingly demonstrated the continuity of the gas space of shoot and root, and the dependence of the submerged root upon the shoot for its oxygen supply. That the intercellular space system also provides for the escape of carbon dioxide was indicated by reverse-order gradients of concentration. Van Raalte (1944) also drew attention to the rhizosphere oxidizing activity of the rice root, an important property for counteracting the reduced nature of the wet soil and also dependent upon longitudinal oxygen flow from the aerial parts. The oxidizing activity of roots had been overlooked since the pioneer work of Molisch (1888), Raciborski (1905a, 1905b), Schreiner and Sullivan (1910) and Schreiner and Read (1909). Van Raalte’s findings were perhaps the greatest stimuli to further research and the oxidizing activities of root systems still continue to attract interest (Engler and Patrick, 1975; Green and Etherington 1977). Tracer studies gave the first proof of gaseous oxygen diffusion through non-wetland species (Evans and Ebert, 1960; Barber et al., 1962), and this time marked the beginning of an upsurge in research activity concerned with both soil and root aeration. The earlier work of Buckingham (1904), Penman (1940), Van Bavel (1951) and others on gas movements in the soil was taken further (Currie,

AERATION IN HIGHER PLANTS

227

1961a, b, 1965; Millington, 1959; Greenwood, 1961) and the application of diffusion mathematics to root aeration began (Lemon, 1962; Lemon and Wiegand, 1962; Kristensen and Lemon, 1962). The following account is centred chiefly around developments since 1960 which have culminated in the modelling of the oxygen movements within the plant, and a recognition that internal oxygen transport in the mesophyte can be substantial. At the same time it has become apparent that root respiration and growth might proceed unimpaired at quite low internal oxygen pressures and under anoxic conditions some extraordinary “malformations” of mitochondrial structure have been observed recently (Vartapetian et al., 1977). Since oxygen transport has been the foremost research topic during this period it is this which I have chosen to emphasize below. To this end I have attempted to collate the mathematical approaches to the aeration process and to explain the concepts of modelling in a manner which I hope may prove simple to understand. As to the future, there is a pressing need to study further the responses of the roots, and their cells and organelles to sustained anoxia and low oxygen pressure, particularly in the intact plant. Hopefully this might result in a consensus on the role of anaerobic metabolism in the wetland condition; it might eventually lead to the discovery of what it is that enables the wetland plant to develop aerenchyma, perhaps the most teasing problem of all. 11. PRINCIPLES OF AERATION AND AERATION MODELLING A. ENTRY AND DISPERSAL OF RESPIRATORY GASES

1. Introduction*

The environment exerts a considerable influence on the directional flow of the respiratory gases within the plant and the directional exchange with the atmosphere. Oxygen may enter the plant body in a variety of ways. In non-aquatic species, where cuticularization and suberization lower the gas permeabilities of aerial surfaces, the internal gas-space system of the plant connects with the external environment through the stomata and lenticels which provide paths of low resistance for the entry and exit of both oxygen and carbon dioxide. In submerged astomatal aquatics, surface permeabilities are sufficiently high to allow for the necessary gas transference across the epidermal layers by liquid-phase movement. Plants rooted in unsaturated soils may be exposed to an oxygen-rich

* Note added in proof. For a broader view of transport processes in the plant the reader is recommended to consult the review by J. A. Raven (1978): The Evolution of Vascular Land Plants in Relation to Supracellular Transport Processes ( A h . Bot. Res. 5, 153-219).

228

W. AKMSTRONG

environment over the greater part of their shoot and root surfaces. In such circumstances there may be little longitudinal movement of gases within the plant: the leaves and shoots will be aerated by simple planar and radial gas exchange with the atmosphere; the oxygen requirements of the root system may be met largely by a diffusive transfer from the soil atmosphere supplemented by transpirational flow (p. 236). In most instances gaseous oxygen must enter the liquid-phase at a point external to the root and it is thought that its passage across the cellular layer(s) of the root wall occurs chiefly in the liquid phase. A substantial proportion of this will then pass into the gas phase of the intercellular space system at the outer perimeter of the root cortex. In saturated soils the situation is very different: little or, more often, no oxygen is available for radial entry to the root system and it becomes essential for longitudinal transport to take place between shoot and root. There is potential for longitudinal gas transport in both the intercellular gas-space system and the stele and the relative merits of these paths are discussed in the following sections. Radial movements will tend to be bidirectional : from root cortex to stele and from root to soil. The latter, radial oxygen loss (ROL) may be of very great importance for the survival of plants in the wetland habitat (see p. 281). Finally, oxygen enters the plant in the combined state as water. As water it is transported from root to shoot in the xylem where a proportion is finally released into the liquid phase within the chloroplasts during the photolysis stage of photosynthesis. At light intensities above the compensation point a net transfer of oxygen from the cellular liquid phase of the leaf to the gasphase of the adjoining intercellular spaces will be ensured. The roles of light and darkness in the aeration process are considered in detail in Section II.C.2. 2. Longitudinal Transport (a) The gas-space system. In the more complex animals oxygen enters and leaves an elaborate distribution network (the blood system) by means of diffusion, but within the distribution network itself movement is essentially a mass-flow energy-dependent process. In contrast, most, if not all, of the steps in the gas flow process in the higher plant, including the major one of dispersal, are to a greater or lesser extent diffusion-dependent : gas-phase diffusion is the major mechanism for long-distance transport. Dispersal by diffusion is made possible because the higher plant has evolved as a porous structure with a labyrinth of intercellular gas-space. Diffusion in the gas-phase is extremely rapid (lo4 times more rapid than in an aqueous medium) and it is this which enables the higher plant to rely on diffusion for its gas exchange requirements to the extent it does.

AERATION IN HIGHER PLANTS

229

Intercellular gas-space is characteristic of most extrastelar ground tissues in the plant such as the mesophyll and palisade tissues of leaves, and the cortical parenchymas of stem and root, while pith within primary vascular cylinders may be similarly structured. Much ground-tissue gas-space interconnects and we find for example that herbaceous species are characterized by an uninterrupted gas-space continuum within the cortical ground tissues extending from the sub-stomatal cavities to within 20 pm of the root/root-cap junction (Plate I). The volume per cent of the gas-space within an organ at any point is referred to as the porosity, and the proportion of gas-space within individual tissues and organs varies enormously: in wetland plants the porosities of roots, stems and leaves can be as high as 60%. However, despite this, there are places within the plant structure which are not so liberally permeated; in other places adjacent gas-spaces may be separated by non-porous tissue barriers. Secondary growth can introduce areas of low or high porosity. In the primary plant body the stelar parenchymas most closely associated with the conducting tissues may have little if any significant gas-space. Likewise, the extremities of meristems are devoid of spaces. The endodermis is a tissue which is not usually traversed by gasspaces and may be a significant barrier to gas exchange between stele and cortex in roots (and perhaps also in the stems of Gymnosperms and Pteridophytes). The primary vascular cambia may also have little if any permeating gas-space and may form a significant aeration barrier to both stem and root aeration in woody species. The compacted tissues of root-shoot and rootroot junctions are potential dispersal barriers in gas exchange. The movement of gases through the individual cell and across aporous tissues is briefly discussed below under the heading “Lateral Transport”. Although there is little doubt that the longitudinal movements of respiratory gases in the intercellular space systems of plants are essentially diffusional it is as well to realize that some mass-flow movement must occur: both the diffusion-exchange system itself and environmental factors such as temperature and barometric changes will tend to produce total pressure differences. Wood and Greenwood (1971) have drawn attention to the probable induction of mass-flow in soils brought about by diffusional inequalities. The molecules of the three atmospheric gases oxygen, nitrogen and carbon dioxide have different weights and consequently different rates of random movement (Table I); also the total pressure at any point in the gas-phase is proportional to the number of molecules per unit volume irrespective of molecular species. In aerobic soils oxygen is consumed and carbon dioxide produced in approximately a 1 : 1 ratio. This induces a flow of molecules, and in general because of the different rates of oxygen and carbon dioxide diffusion, incipient differences in total pressure must arise which can only be

Plate I

AERATION IN HIGHER PLANTS

23 1

balanced by mass-flow involving not only the oxygen and carbon dioxide, but also the nitrogen. By similar reasoning one can anticipate similar mass-flow effects within the plant. Mass-flow supplementation of soil aeration in this way must be rather small but in roots the situation is more complicated because of the potential for carbon dioxide escape in the transpiration stream. (From simple solubility considerations it can be calculated that the water flow into the stele of a young root could be sufficient to accommodate much of the carbon dioxide efflux of stelar respiration.) Carbon dioxide removed in this way would further induce changes in total pressure and add to the mass-flow component in the aeration process. However, it is exceedingly difficult to estimate the likely magnitude of this effect; at this stage we can only assume it to be small. We know little of the relative respiratory rates of the different stelar zones or the diffusional characteristics of the tissues; the rate of carbon dioxide entry to the xylem will depend upon the carbon dioxide pressures generated within the stelar zones and upon the relative route resistances between the living stelar elements and the xylem elements and cortical gas spaces. Furthermore, in situations where this mechanism might assume some physiological importance, e.g. the wetland condition, it is by no means certain that there may not be a compensating inflow of carbon dioxide from soil to root in the transpiration stream. A mass-flow of somewhat greater proportions is likely in tidally-inundated plants: a total pressure deficit of c. 0.05 atm in the roots of mangroves during periods of submergence has been reported by Scholander et al. (1955). This deficit is quickly corrected by mass-flow during the ebb-tide. It is very unlikely that internal mass-flow induced by temperature or barometric fluctuations will be of any great significance in the overall aeration process. Undoubtedly mass-flow can be effected in this way: a 10°C fall in temperature can cause a mass-flow into soils of as much as 0.9 1 0, m-, (Bouyoucos 1915). For a varied range of soils Grable (1 966) calculated this as sufficient to satisfy from 11-1 50 of the daily oxygen requirements of the soil population. How valuable such a recharging of the soil atmosphere might be will depend very much on the circumstances in particular soils. It seems reasonable to assume that a temperature-induced mass-flow within the plant would be of the greatest value for supplementing root aeration in Plate I. Gas-space characteristics of wetland plants: Rice and Eriophorum angustifolium: Adventitious roots from waterlogged soils: transverse and longitudinal appearance of gas-spaces within the apex. (1)-(3) Rice: transverse sections at distances of 20pm, 120pm and 1.5 mm from the root/root cap junction (magnification x 120, x 122.5 and x 75). (4) Rice: tangential longitudinal section (magnification x 1375). (5) E. angustifolium: longitudinal section through segment of cortex at a position which in rice would be approximately halfway between sections (2) and (3) (magnification x 500). Sections (4) and (5) both show clearly the lack of tortuosity in the cortical gas-spaces.

TABLE I Diffwivities, Solubilities and Fractional Volumes in Air, of the Atmospheric Gases Nitrogen, Oxygen and Carbon Dioxide

1. Oxygen Temp. Fractional vol. in (“C) moist air at NP

Conc. in moist Solubility in pure water air at NP from moist air at NP (1W6 g ~ m - ~ )(10V g cm-*) mol 1-l) cms cm-l) a

0

3 5 10 15 20 23 25 30

0.2086 0-2083 0.208 1 0.2074 0.2064 0.2050 0-2041 0.2033 0.201 1

297.9 294-2 291.7 285.6 279.3 272.8 268.7 265.9 258-7

1463 1346 12.77 11-28 10.07 9.08 8-57 8-26 7-57

Self diffusion Diffusivity Diffusivity coefficient in air in water (cmz s-l) (cm2 s-l) cm2 s-’)

a*

a*

b

C

457 4.20 3.99 3.52 3-15 2.84 2.68 2.58 2-36

1025 9.53 9.11 8-19 7.44 6-83 6.508 6-32 5.88

0.181

0.178 0181 0.184 0.189 0.195 0.201 0.205 0.207 0.214

d

099f

1-16? 1*27? 1.54? 1.82 2.10 2.267 2-38 2*67t

NB: The fractional volume of oxygen in dry air at NJ? is given as 0.2099 in the International Critical Tables (Humphries, 1926) and as 0.20946 O-ooOo2 in Weast (1974). The fractional volumes and concentrations of oxygen in moist air are based on the former of these two values.

i

2. Carbon dioxide Temp. (“C)

Solubility in water. CO, gas source at a pressure of 760 mm g cm-8) mol 1-l) e*

e*

Self diffusion coefficient (cm2 s-l) b

Diffusivity in air (cm2 s-l) C

3. Nitrogen Fractional volume in dry air at NP: 0.7803 (Humphries, 1926) 0.7809 (Weast, 1974). Solubility in water from N, source at a pressure of 760 mm: 18-1 x 10-o g cm-s @ 23°C (extrapolated from data of Kaye and Laby, 1966).$ Self diffusion coefficient : 0-178cm2 s-l.b Other diffusivities are not comprehensively listed but closely approximate to those for oxygen.

Diffusivity in water cm2 s-l) f

0 3

747-7 0-0965 0-138 1-15? 3.29 677.3$ 0.141 1-24t 2*98$ 625$ 0.143 2.75$ 5 1.30t 2.28 10 518.8 0.148 1.46 1.94 15 440.7 0.153 1.63 1-66 377-3 0.159 1.77 20 1-52i 346i 0.162 1*86 23 25 144i 327-3$ 0.164 30 1-28 290.0 0.170 1*92t 2.08t NB: The fractional volume of carbon dioxide in dry air at NP is given as O W 0 3 (Humphries, 1926) and 0-00033 by Weast (1974). a Montgomery et al. (1964); Chapman and Cowling (1939); Boynton and Brattain (1929); d Millington (1959); e Kaye and Laby (1966); f Bruins (1929); * calculated from data given in a different form; t linear extrapolation from other data; 2 curvilinear extrapolation from other data.

I

234

W. ARMSTRONG

saturated soils. However it can be calculated that it would require many recharging operations within a 24 h period to satisfy the oxygen requirements of root systems, even in highly porous plants (see also p. 296). (b) The stele. It might be thought that the specialized long-distance transport systems for water and solute movement in the plant could also play a significant role in the dispersal of oxygen to remote parts. For example, it might be particularly advantageous in wetland species if phloem were to provide a route for the transference of oxygen from the aerial shoot system to the submerged organs subject to the anoxia of the wetland soil regime. In such circumstances the reverse flow in the xylem might provide the escape route for carbon dioxide. Conversely it has been suggested that in nonwetland conditions the inner tissues of most broadleaved tree species may rely on oxygen dissolved in the transpiration stream (Hook et al., 1972). These possibilities are neither wholly supported nor contradicted in the specialist literature on phloem and xylem transport: phloem physiologists have divergent views on the oxygen relations of phloem transport. However, with the aid of some simple calculations, and taking a biased view from the literature I find myself drawn to the view that neither the phloem nor the xylem is likely to play a major role in the oxygen dispersal process. (i) The phloem. Peel (1974) has noted that despite the difficulties encountered in making velocity measurements most workers in the field of phloem physiology would accept that solutes can be transported at speeds of up to 100 cm h-l. Some experiments have indicated values greatly in excess of this (Nelson et al., 1958), others considerably less (Canny, 1961). Using this figure of 100 cm h-l, consider a phloem bundle of length 100 cm and radius 0.005 cm. Assuming that the sieve tubes occupy approximately one-fifth of the total phloem volume (Peel, 1974) the total sieve tube volume cm3. If the oxygen partial pressure will be (10015) (O.00Sz T ) or 1-57 x at the loading site is 0.2043 atm and the oxygen solubility in the sieve tube cm3 ~ m - ~then ) the cytoplasm approximates to that in water (6.5 x maximum rate at which the phloem will conduct oxygen should be (1.57 x x (6.5 x or 10.2 x cm3h-l. Assuminga density of l.O(Canny, 1973) the total fresh weight of the phloem strand would be 7.85 x g. If the oxygen transported was freely available for phloem respiration the respiratory rate could not exceed 10.2 x 10-s/7+35 x or 1.3 x cm3 g-l fw h-l. A shorter phloem strand would naturally lead to an increase in the potential for respiratory activity: with a strand of 10 cm the value becomes 13 x cm3 g-' fw h-l. However even this contrasts markedly with the values for phloem respiration given in the literature: it is only one twentieth of the value cited by Canny (1973) as an acceptable mean level for oxygen consumption by phloem. With efficient oxygen extraction and a cm3 8-l fw h-l up to 95% of the potential respiratory rate of 230 x

AERATION IN HIGHER PLANTS

235

10 cm phloem strand could become anaerobic; in a strand of 100 cm this could rise to 99.5 On this basis it would appear that the oxygen-carrying capacity of the phloem not only fails to meet the requirements for long-distance transport but is also inadequate for the vital activities of the phloem itself. Hence, although the velocity of solute movement in phloem can accord with that of gas-phase diffusion the suggestion that phloem might play a major role in gas transport appears to falter on the phloem’s capacity in a gas-exchange role. On the experimental side, a divergence of views emerges. Oxygen transport in the phloem has not been studied directly and there is even confusion concerning the phloem’s own oxygen requirements and how these may be satisfied. Mason and Phillis (1936) found that provided they excluded oxygen from a sufficient length of stem, translocation of materials in the phloem could be reduced or stopped. However, the apparent difficulty experienced in actually excluding oxygen from the stem was taken as evidence for a special oxygen carrier in the phloem. (The evidence of later oxygen diffusion studies in plants would suggest that the difficulties in oxygen exclusion experienced by Mason and Phillis point to a rapid intra- or extra-stelar gas-phase diffusion of oxygen from more remote parts.) The more recent findings of Quereshi and Spanner (1973) are particularly interesting in this respect. The movement of applied 13’Cs and naturally assimilated 14C down the long uniform stolons of Saxifraga sarmenfosa was strongly inhibited by confining 20-30cm of stolon in an atmosphere of nitrogen but was readily reversed by re-exposure to air. The results (see Fig. 1) provide very convincing proof of the oxygen-dependence of the phloem transport phenomenon but offer little support for the oxygen carrier hypothesis of Mason and Phillis: the distance travelled within the N,-treated portions is probably an indication of the length of stolon receiving adequate aeration from the untreated portion by diffusion along the gas-filled spaces within cortical tissues (see p. 256). Ullrich (1 961) whose work preceded that of Quereshi and Spanner favoured the carrier hypothesis of Mason and Phillis that an internal hydrogen acceptor, hydrogen peroxide (H,O,) or an organic peroxide, together with the peroxidase detectable in young uncallosed sieve tubes, could provide for the respiratory needs of the phloem. Ullrich’s calculations show that c. 1.4 mg H,O, g-1 fw h-l would sustain the respiratory level of 230 x cm3 0,g-1 fw h-l. Whether these levels of H,O, are present in phloem is not known but the experiments on which Ullrich bases his hypothesis are open to alternative interpretation. The conclusions of Quereshi and Spanner seem less open to doubt and indicate the need for a constant lateral transference of molecular oxygen into the phloem along its whole length to sustain its activity. (ii) The xylem. The likelihood of a significant transpirational dispersal of

x.

236

W. ARMSTRONG

-

0

Extent of sleeve

0

VI +

.-C c 0

.-> .-

l o - " " " ' " ~ ' ~

~

"

"

"

"

"

"

"

"

"

Mid-points of segments of stolon

Fig. 1. The movement of 14Clabelled natural assimilates along the stolon of Saxifraga surmenfosu sleeved for part of its length in darkness and in atmospheres of either air or nitrogen. Dose of 14C02, 50pCi. Temp. 25-32°C. Nitrogen treatment given 5 h, and l4CO24 h, before harvesting (after Quereshi and Spanner, 1973).

dissolved oxygen in roots located in unsaturated soils is probably as unreal as dispersal in the phloem. It is as well to consider first the extent to which the transpiration flow may satisfy the respiratory needs of the roots themselves. If we accept a mean water uptake of 0.3 cm3 day-l cm-l for young roots (r = 0.05 cm) (Nobel, 1974) water will enter one centimetre of root (vol.: 7.85 x cm3) at a cm3 s-l. If we assume a respiratory rate of 120 ng 0, s-l rate of 3.5 x ~ r n fw - ~root tissue the respiration of one centimetre of root will be 0.9425 ng 0, s-l (0.029 nmol s-l). At air-saturation the oxygen in 3.5 x cm3 of g ~ m - or ~ )0.0299 ng 0,. water will amount to (3.5 x cm3) (8.57 x Thus the influx of oxygen attributable to transpiration can at best satisfy only 3.2 % of the root's requirements and in most instances the water entering the root will be below air-saturation. Clearly this cannot be the only source of oxygen for root metabolism and the major mechanism for entry and dispersal i.e., diffusion, is considered at length in subsequent sections. In unsaturated soils the diffusive flow of oxygen through the root wall can ensure high levels of oxygen throughout the root cortex (Section V). Consequently it remains possible that the water entering the stele may approach air-saturation and

"

AERATION IN HIGHER PLANTS

237

micro-electrode analysis of oxygen pressure within the protoxylem elements of intact Heliunthus roots (Bowling, 1973) supports this contention. A stelar respiration of c. 200 x cm3 0, ~ m fw - h-l ~ in maize roots has been reported by Hall et al. (1971) and if the stele occupies approximately one quarter of the root volume used in the previous example it will have a potential rate of oxygen consumption of 0.117 ng 0, s-l. This is still four times greater than the rate at which the oxygen enters the stelar strand by transpirational flow and runs contrary to a direct oxygen transport role for the xylem. However, the possibility remains that stelar parenchymas and phloem will be unable to compete effectively for this oxygen supply. Anatomical considerations must play a part here: the path from the endodermis into the nonrespiring protoxylem and metaxylem elements is usually short and relatively non-tortuous and this could perhaps preclude any substantial oxygen loss from the inflowing water. In these circumstances some of the stelar oxygen requirements of basal regions of root or stem might be met by the transpirational flow. This effect could also be additive along the root to some extent as the radial water flow into subapical regions increases the velocity of water movement in the xylem and hence the potential rate of oxygen flow. The extent to which this might occur would depend greatly on the water permeability of the root in subapical regions and the efficiency with which the living stelar tissues can extract the oxygen from the tracheary elements. Nevertheless, it remains clear that if the stele is not to receive a supplementary oxygen supply from other sources much of it must experience the suboptimal aeration predicted by Crafts and Broyer (1938). The existence of supplementary oxygen sources for stelar aeration seems beyond doubt (e.g. liquid and gas-phase diffusion and symplastic cyclosis) and will be discussed later. The transpiration stream has a vastly greater potential for carrying the carbon dioxide from respiratory sites than it has for oxygen transport: carbon mol ~ m a -t 23°C) ~ is 27 times dioxide solubility in pure water (346 x greater than that of oxygen (Table I). Carbon dioxide carried from root to shoot in the transpiration stream could be either utilized in the photosynthetic process or escape to the atmosphere through the stomata. The total volume of carbon dioxide released per centimetre of root in the previous example would be (0.029 nmol s-l) and a significant proportion of this could be accommodated in the transpiration fluid. However, although the carbon cm3 of water is (346 x lo-' mol ~ m - ~(3) x dioxide capacity of 3 x cm3) or 0.1038 nmol at a carbon dioxide partial pressure of 1 atm, this cannot be realized in aerated tissues. In the presence of oxygen the partial pressures of carbon dioxide are unlikely to exceed (or even approach) 0.2atm and will lie within the range zero to 0.2atm. Pressures will vary radially being highest in compact aporous stelar tissues and lowest in the

238

W. ARMSTRONG

TABLE I1 Surface Oxygenation in Saturated Soils (see p . 281) The depth, x, at which the solution oxygen concentration falls to zero, has been comand soil oxygen puted for various combinations of potential soil oxygen consumption, Ms, diffusivity, De (s) (see equation 34). The solution oxygen concentration at the air:soil g ~ m (20.41 - ~ %). interphase was taken as 8.57 x

a

Depth of aeration (cm) Oxygen diffusivity in the wet soil (lo+ cm2 s-l)

Soil oxygen consumption (g cm-3 s-l)

0.56a

3.54a

1O b

5.27 5.27 5.27 5.27 5.27

0.01 3 0.042 0,135 0.426 1.349

0.034 0.107 0.339 1.073 3.393

0.057 0.180 0.570 1.803 5.702

x lO-*C x lo-@ x x lo-” x

Computed from the data of Currie (1965). Greenwood and Goodman (1967). From data of Teal and Kanwisher (1961).

cortex where gas-phase diffusion is an aid to dispersal. The complexity of the system makes it exceedingly difficult to predict the rate at which the carbon dioxide may pass into the tracheary elements of the young root and if carbon dioxide pressures were to approach 0.2 atm the harmful effects of high carbon dioxide concentration could be a further complication. Hence although significant carbon dioxide removal in the stele is an attractive and indeed a likely possibility it certainly could not fully accommodate the carbon dioxide production of whole-root respiration cited above. It might more realistically accommodate much of the carbon dioxide efflux of stelar respiration but it is still extremely difficult to estimate the proportion which could reasonably be expected to be transported in this way. One might anticipate that carbon dioxide escape in the xylem will increase in importance as root aeration becomes subnormal. It is known for example that the anaerobic by-product ethanol is translocated in the xylem (Fulton and Erickson, 1964) and similar claims have been made for other products of anaerobic metabolism, e.g. malic acid (Crawford, 1972). In more normal circumstances carbon dioxide can probably escape equally or more readily from respiratory sites in roots and stems by alternative means, such as radial diffusive loss to the soil or gaseous diffusion in cortical gas spaces. In woody plants gas-filled pathways within secondary xylem may be a major route for both carbon dioxide and oxygen exchange (see p. 309).

3. Lateral Transport The intercellular space system is an aid to lateral as well as longitudinal transport although within any particular tissue lateral resistance might differ

AERATION IN HIGHER PLANTS

239

markedly from that in a longitudinal direction : the geometry of cortical intercellular space in roots is illustrative of this (see Plates I and 111). Furthermore, whilst longitudinal transport occurs chiefly in the gas-phase, in lateral transport there is almost invariably a restrictive liquid-phase component. Even in porous tissues the transfer of gases between gas-space and mitochondrion and chloroplast are usually lateral movements occurring radially at right angles to the long axis of the cell. It would seem that in non-porous tissues gas movement must be restricted entirely to the liquid phase and normally this will tend to occur at right angles to the long axis of the organ in question. For the most part this will offer the route of least resistance between source and sink. As with ion transport (Spanswick, 1976; Goodwin, 1976; Robards and Clarkson, 1976; Gunning and Robards, 1976) the manner in which gases are transported across aporous tissue blocks or even through the cytoplasm of individual cells is still far from being unequivocally resolved. There is a pressing need for a detailed analysis of the system and among other things to determine the relative contributions of cyclosis and liquid-phase diffusion in the aeration process. It is possible that gas transport might be enhanced by some as yet unidentified process. High plasmalemma resistance to electrolytes has given rise to the view that ion movement may occur preferentially through the symplast via the plasmodesmata. However, although definitive values for plasmalemma or protoplasmic permeabilities to oxygen and carbon dioxide are apparently unavailable, oxygen is known to be highly lipid soluble and it is generally agreed (Collander, 1959; Nobel, 1974) that both gases are amongst the most rapidly penetrating of molecules (plasmalemma permeabilities 2 1 0-2 cm s-I). Consequently the transfer from cell to cell in aporous tissue ought to be appreciable over the whole inter-cell interface although the partition of carbon dioxide between the dissolved gas and its ionic species adds a further dimension to the problem. It is difficult to do more than comment briefly on the relative magnitudes of transport by cyclosis and diffusion: the problem is extremely complex. Pertinent literature may be found in Tyree (1970) and Robards and Clarkson (1976). We lack a knowledge of path characteristics such as streaming rates and directions and the examples which are known are very limited: for cm s-l is often quoted. The protoplasmic cyclosis an upper limit of 2 x diffusivities of carbon dioxide and oxygen are also uncertain although there are reasons for believing them to be close to those in water (Tyree, 1970). We are unaware of the path lengths and areas available for diffusion and cyclosis; if, as seems likely, the respiratory gases readily cross the tonoplast the diffusion path in the vacuolated cell might be substantially shorter than the path of cyclosis. By adopting the most extremely simplistic view of the streaming process it

240

W. ARMSTRONG

is possible to gain some appreciation of how cyclosis and diffusion might contribute to oxygen transportation across tissue barriers. Let the cytoplasmic streams between the two opposite sides X and Y of a hypothetical transport barrier be simplified as two parallel tubes, P and Q, of equal radius, through which water flows at constant velocity V, from X to Y in P, from Y to X in Q. If the side X behaves as an oxygen source at constant concentration C,, and side Y behaves as an oxygen sink at constant, but lower, concentration C,, there will be the potential for diffusive as well as mass flow of oxygen between X and Y. Within P the mass flow will act in conjunction with diffusion, within Q the two processes will be in opposition. (NB since the concern here is simply to elucidate the relationship between the diffusion and mass flow processes no attempt will be made to model the linkages between P and Q at the two surfaces X and Y.) For movement along the x-coordinate perpendicular to the surfaces X and Y we have the boundary conditions C = C, on x = 0, and C = C, on x = 1. For the tube P the concentration C at any distance between x = 0 and x = 1 is given by the equation:

c = c, - (C,

- C,)

-1 ehx -~ (eh' - 11 (see Appendix 1 for derivation).

where h = V/D, D being the diffusion coefficient for oxygen in water. The corresponding equation for tube Q is:

For tube P the diffusive flux at x = 0 is given by the expression V(C, C,)/(eX' - l), at x = 1 it is equal to V(C, - C,) (eh'/eh' - l), whilst the total oxygen flux through P which is constant from x = 0 to x = 1, is equal to VC, + V (C, - Cl)/(eh' - 1); the corresponding expressions for tube Q are given in Appendix 1. If for C,, C,, V, and D in equation 1 we respectively substitute the values 8.57 x g ~ m - zero, ~, 2x cm s-l, and 2.267 x cm2 s-l, and for path length I, we substitute distances which are multiples of average cell thickness, 30 x cm, we obtain the series of oxygen concentration profiles 2-6 shown in Fig. 2. As the distance I increases so d o the concentration profiles deviate the more from the linear relationship found where V = 0 and transport is entirely diffusive (Fig. 2, graph 1). Concentration gradient, dC/dx, is indicative of diffusion rate (p. 244), the steeper the gradient the greater is the flux, and hence it will be seen that when streaming is in conjunction with diffusion (curves 2-6, Fig. 2) it leads to a lowering of the diffusion rate close to the oxygen source whilst near to the sink the diffusion increases. Where the path length is the equivalent of 16

- 9-

t-7

5

Y

8

7-

1

C

6-

c 0.

2 c

5-

C Q,

-gguiCPx c 3

0 v)

#

8-

0

1 234

a-

7

0 0

- --

L

6

5

-

OO

x =o

75

15

Distance along diffusion path

2 2 5 25

(lo-‘

cm 1

275 2

X=l

Fig. 2. To show how a stream of water having a constant velocity, V, numerically equal cm s-’) can modify the oxygen concentration to that of cytoplasmic streaming (2 x gradients of a simple source-sink diffusion system (plot l), where C = Co= 8.57 x g ~ m on - ~x = 0, and C = C, = 0, on x = I. The effects of streaming both in the direction of diffusion (curves 2-6) and against it (curve 7) are shown together with the changes brought about by extending the source-sink path length from 30pm (c. one cell thick, curve 2) to 480pm (c. 16 cells thick, curves 6 and 7). The oxygen flux from source to sink ( x = 0 to x = I) under the various circumstances is tabulated and the ratio, total flux with streaming: total flux without streaming, indicates how the enhancement of flux by streaming is reduced in the presence of a “return” flow (double path).

242

W. ARMSTRONG

cells in thickness (curve 6, Fig. 2) the diffusive flux at x = 0 is only 0.25 ng cm-2 s-l, whereas at x = Z it rises to 17.37 ng cm-2 s-l. It is interesting to note that this is also the value attributable to the total flux at this point. In other words at x = I, the streaming component has been totally masked by the diffusive force. Where V = 0 diffusive flux is constant from X to Y at 4.04ng cm-2 s-l. Hence it can be seen that the combined streaming and diffusive movement increases the oxygen transport along P by a factor of (17.37/4.04) i.e. 4.34. Where the transport barrier is only one cell thick and V =2 x cm2 s-l (curve 2, Fig. 2) the oxygen flux at x = Z is 73.6 ng cm-2 s-I; where V = 0 the value becomes 64.7 ng cm-2 s-l and the fractional increase due to streaming is only 1.14. When V is in opposition to the diffusive forces (equation 2) the concentration profiles obtained are as before but rotated through 180 degrees about an axis perpendicular to the plane of the graph (cf. curves 7 and 6, Fig. 2); diffusion is greatest near the oxygen source and is diminished towards the cm, the final oxygen flux at x = Zis sink. For the path length, 480 x cm, it is 56.5 ng only 0.25 ng cm-2 s-l; for the path length, 30 x cm-2 s-l. If the oxygen movements from X to Y along P and Q are combined it seems that there is little advantage to be gained from the streaming process where path length is short: if the barrier is the equivalent of one cell in thickness the ratio between the streaming and non-streaming condition is as small as 1.005 : 1. However at eight times this thickness the ratio rises to 1.35 : 1, at sixteen times it becomes 2.18 : 1. Consequently it seems unlikely that cytoplasmic streaming will enhance the transport of oxygen across narrow cellular barriers but it is conceivable that it might enhance transport in thicker zones of aporous tissue. B. DIFFUSION AND THE VENTILATING PROCESS

I . Introduction Diffusion is the process by which matter is transported from one part of a system to another as a result of random molecular movement: if a chemical species j is present at concentration C’jat some point in an isotropic medium and is present at a lower concentration C”j elsewhere within the medium there will be a net transfer of material towards C”j and this net transfer will continue until the two sites have attained the same uniform concentration. To appreciate this diffusion process the picture of random molecular motion has to be reconciled with the observed transfer of material from sites of higher to lower concentration. Consider a plane within an isotropic medium such that on opposite sides of the plane the species j is at the concentrations C’j and C”j. Although it is not possible to say in which direction any particular molecule of j will move in a given period, it can be said that, on average, be-

AERATION IN HIGHER PLANTS

243

cause of random movement, a definite fraction of molecules will cross the plane from the area of high concentration, while the same fraction of molecules from the region of lower concentration will cross the plane in the opposite direction. Hence although the fractional movement from each region is the same, the total number of molecules moving from the higher concentration must exceed that from the lower concentration. It will also be apparent that the rate of net transfer must diminish with time as the concentration difference is reduced. The velocity of the random walk process within the isotropic medium is governed by the characteristics of the medium and sometimes by the concentration of the diffusing species. These find expression in a term which quantifies diffusivity, namely the diffusion coefficient (D). Diffusion coefficients are normally accorded the units cm2 s-l; they vary with temperature and can vary with concentration (Crank, 1975), but for the diffusion of simple molecular species through gases or dilute aqueous solutions they may be considered to be relatively independent of concentration.

2. The Fundamental Difusion Equations: Fick's Laws The mathematical description of diffusive transfer originates from the researches of the Zurich Professor, A. von Fick (1855) who recognized that diffusion could be likened to the transfer of heat by conduction: the fundamental differential equation for diffusion attributed to Fick and commonly referred to as Fick's second law is a direct adaptation from the equation for heat conduction derived by Fourier (1822). In its simplest form for planar diffusion where diffusion is one-dimensional, i.e. there is a gradient of concentration only along the x-axis, the equation is:

where C is concentration, t represents time and D is the diffusion coefficient for the species in question. For diffusion to or from cylinders of radius r where diffusion is everywhere radial the equation becomes :

ac

1

a

ac

t - r-* a r ( ' D a r )

(4)

while for spherical diffusion in which movement is again confined to the radial direction we get:

Solutions of these equations can show how the concentration of a diffusate changes with position and time as a result of the diffusion process and the

244

W. ARMSTRONG

various equations for steady-state diffusion may be derived from them (see later). Both steady-state and non-steady-state solutions have found application in the study of the aeration process although the former are by far the more commonly used. The hypothesis upon which the differential equations are based is referred to as Fick’sfirst law. This states that the rate of transfer of diffusing substance through unit area of section is proportional to the concentration gradient measured normal to that section. For diffusion in one dimension this can be expressed mathematically as:

J=-D-

ac ax

where J is the rate of transfer per unit area of section, C is the concentration of diffusate, x is the space coordinate normal to the section and D is the diffusion coefficient for the diffusate in the medium concerned. The negative sign indicates that diffusion takes place in the direction away from increasing concentration ; the partial derivative indicates that all but one of the possible independent variables is held constant. In equation (6) aC can be likened to a “force” analogous in part to the potential difference in an electric circuit while the term ax/D is a measure of the resistance to diffusion. From a consideration of the numerical values of D for the two respiratory gases (see Table I) it will be apparent that diffusive resistance in water is approximately lo4 times greater than in air and that resistance to oxygen diffusion in air and water is c. 0.8 times that of carbon dioxide.

3. Planar Digusion: The Simple Case A number of stages in the aeration process in higher plants approximate in the short term to fairly simple steady-state one-dimensional diffusion systems in which there is no net lateral movement in y or z directions, and where a linear gradient of concentration is developed between a source of diffusible molecule and an adjacent or more remote sink. During darkness the entry of oxygen into leaves and the corresponding efflux of carbon dioxide probably deviates little from this diffusion pattern. Localized linear gradients may also arise in the longitudinal transport pathway under certain circumstances; they can be brought about by experimental techniques used to assess aeration parameters (Armstrong and Wright, 1975). An equation describing the steady-state linear diffusion from planar source to planar sink of equal area, through an intervening isotropic medium in which there is no lateral loss or gain of the molecular species, can be derived from the differential equation for planar diffusion previously described : consider the case of one-dimensional diffusion through a plane sheet or membrane of thickness 1and diffusion coefficient D whose surfaces, x = 0, x = I, are maintained at constant concentrations C, and C, respectively. After a

245

AERATION IN HIGHER PLANTS

time a steady state is reached in which the concentration remains constant at all points of the sheet and provided that D is constant (and ignoring such effects as gravitation) the diffusion equation (3) in one dimension reduces to: d2C =o dx2

(7)

-

On integration we get: dC - = A dx where A is a constant > 0 confirming that the concentration must change linearly through the sheet from C, to C,. Integrating a second time we obtain :

(9)

C=Ax+B where B is a second constant of integration. Applying the boundary conditions C = C, on x then C , = Band C, = A1 C, and hence

+

=0

and C

=

C, on x

=

I

In the situation under consideration equation (6) can be written: dC J=-Ddx and substituting from (10) into (6) we get

J

=

c, - Cl D ( 7 )

where J, the one-dimensional flux of diffusate is constant throughout the length of the diffusion path from x = 0 to x = 1. Graphically the plot of C against x for this system has the form illustrated in Fig. 3(a). J which has dimensions of quantity, area and time may also be written: J = -

Q

At where Q = grammes, moles or cm3; A tion (12) is often rearranged as

=

cm2 and t

=

seconds, hence equa-

Q = D co A -(Cl~ ) t

Source

'='

Rodiol distonce between sourceand sink

Distance from the source ( x

1

r=b Sink

'='

Fig. 3 (a). Linear diffusion gradient in simple planar source-sink system. Oxygen source, C,, on x = 0 and sink, C,, on x = I. (b). Showing how a linear increase in the diffusive resistance between source and sink in a simple diffusive system is accompanied by a curvilinear decrease in flux. (c). A curvilinear concentration profile characteristic of a diffusive system in which source and sink lie on concentric cylinders. The example given was computed for an oxygen source (r = 0.01 cm) at concentration C, (= 8.57 x lo-" g cm") separated from a concentric oxygen sink (r = 0.1125 cm) at concentration C, (= 0 ) by a "shell" of water in was taken as 2.267 x cm* s-' (see equation 23). which, DO~-,H~O (d) Curvilinear gradients characteristic of planar diffusion systems in which the diffusate is removed uniformly along the length of a homogeneous diffusion path (see equation 30).

247

AERATION IN HIGHER PLANTS

where the term Q/t is referred to as the diflusion rate (g s-l) for the finite system of planar sectional area A (cm2). 4. Ohm’s law and the Diffusion Analogy In discussing equation (6) it was pointed out that the term ax/D was a measure of the resistance to diffusion while aC could be considered as analogous with a force and hence the flux is the resultant of the interaction between “force” and “resistance”. In equation (12) there is no restricting area term and the resistance I/D is simply a measure of the linear resistance between planar surfaces, providing that the areas of these surfaces are equal. The units of I/D are

or s cm-l and the use of resistance in this form has found wide application in considering water vapour and carbon dioxide fluxes across leaf surfaces. Equation (14) specifies the area through which diffusion takes place and the resistance term becomes I/DA and has the units cm x -sx - ) 1 1 cm2 cm2 or s ~ m - Resistance ~. expressed in this form is of great value in considering the longitudinal transport of gases through shoot and root. There are close similarities between equations (12) and (14) and the expression of Ohm’s law for the conduction of electricity through a homogeneous conductor and it is often helpful to consider diffusion problems using such electrical analogues; it can also prove helpful to develop functional models of diffusion using electrical systems. In its expanded form Ohm’s law may be written: t

where e is the quantity of electricity (coulombs) flowing through a conductor in time t (seconds), I is the length of the conductor (cm), 0 is its sectional area (cm2), V and V represent the electrical potential (volts) at the beginning and end of the conductor, and f is the conductivity constant, the value of which depends on the quality of the conducting material and on temperature. Comparing equations (14) and (15) it will be apparent that Q/t is analogous with e/t, D with f, C , - C , with V , - V A/I with O / l , and that diffusive resistance I/DA is an analogue of electrical resistance Ilf.0. In the condensed version of Ohm’s law e/t is reduced to the termI(amperes), V - V reduces to V ,and l/f.O becomes R, the resistance of the conductor whichismeasured in ohms (Q). Ohm’s law is then written

,

,

,

248

W. ARMSTRONG

and is equivalent to a condensed form of equation (14), i.e. Q/t = dC/R, where dC represents C, - C, and R represents IIDA. At this stage it may be useful to note that just as in an electrical circuit one may calculate the voltage drop (V’) along any section of conductor by applying the relationship V‘ = IR’ where I is the current flowing through the whole conductor and R’ is the resistance in the segment, in a diffusion system one may similarly calculate a localized concentration drop. For homogeneous conductors R’ = RI’/I where 1‘ is the length of the segment, 1 the length of conductor and R its total resistance. For a number of conductors in series Ohm’s law reads:

I=

V

R’

+ R” + R ” + . . . .

V R

- -

(17)

Similarly, diffusive resistances in series become additive and, as with the flow of electricity where only as R approaches co does I approach zero, so too with diffusion: Q/t remains finite at all values of R c 00. This important principle is illustrated graphically in Fig. 3(b), where the change in diffusion rate consequent upon extending the distance between source and sink across an isotropic medium is plotted against the change in diffusive resistance. While diffusive resistance increases linearly with increasing path length the diffusion rate decreases in a curvilinear fashion. Just as conductors in series are additive in their resistance to flow in both electrical and.diffusion systems so do conductors in parallel effectively reduce the total resistance. For electrical conductors in parallel the total effective resistance is found from the equation 1

1

_ -- F R

1 + -R” + R”’ -1+ . . .

and can be directly applied in the appropriate diffusional context. In longdistance oxygen transport to submerged roots the presence of several leaves arising on a condensed submerged axis can behave as parallel resistances and similarly the distribution of stomata is akin to a parallel resistance network. 5. Pore-space Resistance and Effective DiJiusion Coeficient The simplest analogue of linear (long-distance) gas transport in roots would be that of planar diffusion along a simple tube of uniform radius and having an impermeable wall. Equation (14) would suffice to describe gas flow in this system. Although such an analogue differs in many respects from the situation pertaining in roots, it serves to illustrate one particular point: that the physical resistance to diffusion in this system is a direct function of the length and

249

AERATION IN HIGHER PLANTS

sectional area of the diffusion path, i.e. the diffusional impedance will be . the linear given by the term I/DA and will have units of s ~ m - ~Although resistance to diffusion along the whole tube would be effectively increased if the diffusate could leak away laterally through a permeable tube wall (see p. 305), nevertheless the term f/DA could be retained to represent the physical impedance of the linear path. In plants this particular feature of the gastransport path may be termed the pore-space resistance, R,. R, can be categorized as a non-metabolic diffusive resistance to distinguish it from the resistance effects which arise from the metabolic usage of diffusate along the diffusion path. Plant organs rarely if ever (see next section) fully approximate with the open-tube analogy and the gas-space volume can vary enormously (p. 290). The occlusion of potential “low resistance” pathways by cellular structures effectively increases the diffusional impedance and for any uniform segment the expression f/DA may be modified to f/eDA to accommodate this, where E is the fractional porosity of that part of the organ in question; A, its overall sectional area; and I its length. The value of E may be determined in various ways and Jensen et af. (1969) have described a useful pycnometer method for assessing root porosities. The geometry of the intercellular space system also contributes to the overall impedance of the diffusion path: the effective linear path along an organ such as a root may be made considerably longer than the organ itself because of the tortuosity of the channels. By convention tortuosity is considered as a modifying influence on Do the diffusivity of the respective gas in air. The porosity factor E is considered likewise and the modified diffusivity term is known as the effective diffusion coefficient (De). To allow for tortuosity the fractional porosity may be raised to a power m, and From this we get the relationship :

Alternatively Do may be multiplied by a tortuosity factor value < 1. D, is then given by

T

having some

For diffusion through a system occluded by glass beads, Penman (1940) has established the relationship De/Do = 2/36, while Jensen et al. (1967) have suggested that tortuosity may reduce gas diffusivity in roots by almost 60 % : for roots of low porosity they estimate a value of T = 0.433. Clearly however as porosity increases the value of T must approach unity. In the following sections T will be adopted as the tortuosity term with De derived as inequation (20).

250

W. ARMSTRONG

6. Radial Diflusion: The Simple Case For radial diffusion between cylinders the simple source-sink steady-state solution of equation (4) (analogous with planar solution 14) has been of considerable value in the study of soil and plant aeration (Lemon, 1962; Kristensen and Lemon, 1962; Letey and Stolzy, 1964; McIntyre, 1970; Armstrong and Wright, 1975). It forms the foundation for the assessment of data obtained using the cylindrical platinum technique reviewed in Section 111. Consider a medium in the form of a long hollow cylinder such that at the inner and outer radii (r = a, and r = b), a diffusible species is maintained at concentrations C, and C, respectively. The differential equation describing the steady-state condition for radial diffusion may be written :

=o

-1- -d( r . D E ) r dr

Assuming that D is constant integration gives : C

=

A

+ Blog r

(22)

where A and B are constants to be determined for the boundary conditions: C

=

Co on r

=

a, C

= C,

on r = b and a b B = - (Co - C,)/lOg-

< r < b. Hence

a

and b (Co log b - C, log a)/loga On substituting for A and B in (22) we get A

=

co 1%

(b/d + C,log (r/a) 1% ( b b ) and differentiating with respect to r: C=

_ dC dr log(b/a)

(++?)

For the radial system the analogue of equation 6 is:

Q =-DD.dC

-

tA

dr

and substituting from equation (24) we get

Q -- - D . _ tA

1 log (b/a)

Finally on rearrangement this becomes:

co c, (-y T) +

AERATION IN HIGHER PLANTS

Q - = D.2rrh. t

25 i.

(Co - C,) r.log (b/a)

By analogy with equations (14) and (15) it may be noted that the resistance to radial diffusion is given by the term r.log (b/a)/DA,, where A, is the surface area of a cylinder of radius r. If our observations concern the diffusion incident upon the inner cylinder r = a, the resistance term becomes a.log (b/a)/DA,, while for diffusion incident upon the surface r = b the term becomes b.log (b/a)/DAb. The numerical value of the two terms is of course ~. we equal and has the standard units of diffusive resistance s ~ m - However, may also note that the terms a.log (b/a) and b.log (b/a) are analogous with 1 in equations (14) and (15); they may be thought of as the effective path lengths when diffusion is considered with respect to the particular surfaces concerned. Hence diffusion to or from the inner cylinder appears to be controlled by a path length shorter than the observed path length b-a. Diffusion to or from the outer cylinder is less than would have been expected for planar ffow and the effective diffusion path length b.log (b/a) is greater than b-a. If we were concerned with flux only, our resistance terms would be a.log (b/a)/D and b.log (b/a)/D respectively and the units would be s cm-l. For equal increments of the path b-a, the resistance component of r.log (b/a)/DA is distributed in a curvilinear manner and hence at equilibrium the concentration profile between b and a is also curvilinear (Fig. 3c). This contrasts with the linear profile in the corresponding planar system. The simple case for spherical diffusion has been little used in plant aeration studies and its derivation will be omitted here. Mathematically it is similar to the linear case from which it can be derived by simple transformation. The final solution is:

Q=D

-

t

ab A r y (CO - C J r (b- a)

where a < r < b. For diffusion to or from a sphere of radius r = a, the effective length of the diffusion path will be given by a(b - a)/b. The resistance in the radial direc~ ) be tion (measured in s ~ m - will b-a ab

--/4

T

D

7. Respiration, Synergism and Eflective Diflusive Resistance (a) Concepf. Consider a simple electrical circuit such as that in Fig. 4a in which we have a source of potential V (300 volts) and two conductors, B and C in series, each conductor having a resistance of 100 52. The current through the conductor path M-L given by V/R will be (300/200) or 1.5 amperes and the potential which will fall linearly from M to L will be 150 volts at N. If a

252

W. ARMSTRONG

M

t

B

OoQ

1

N

C

D lOSl

t

10051 L

Fig. 4 (a) (b). See Section IIB (7).

third conductor (D) of 10 SZ resistance is now positioned “laterally” as shown in Fig. 4b the current will diverge at point N and flow in the directions indicated by the arrows. As conductors C and D are in parallel we find from equation (18) that their combined resistance is 9.09 SZ. The current through conductor B must therefore be (300/109*09) or approx. 2.75 amperes and hence the potential drop across B will be approx. (2.75 x 100) or 275 volts. The potential at N must therefore be 25 volts and the current flow through C is now only c. 0.25 amperes. If we could measure the current through C, (i.e. Ic), but were unaware of the existence of conductor D, and had no prior knowledge of the resistance values for B and C we should conclude that the combined resistance B C was equal to V/Ic, i.e. 300/0-25, or 1.2 kSZ. In other words because of the shunting of electricity through D we gain the impression of a much greater resistance between M and L and this we could consider t o be the effective resistance between the two points. This simple principle is of major consequence in the plant aeration process whether we are concerned with oxygen uptake or carbon dioxide output. Such processes can be thought of respectively as lateral shunts or sources of potential which in effect contribute additively (in both instances) to the linear diffusive resistance. In the case of carbon dioxide, lateral sources of carbon dioxide would effectively reduce the rate at which the gas could escape from the sites more remote from the sink. The lateral “leakage” of oxygen whether it be for metabolic usage or, as in the case of radial oxygen loss (p. 281284), for lateral diffusion to the external environment, acts synergistically

+

AERATION IN HIGHER PLANTS

253

with the pore-space resistance producing what we might term the eflective diffusive resistance of the linear diffusion path. We may express this effective diffusive resistance in the conventional resistance units s ~ m - ~ . It must be remarked upon at this stage that respiratory uptake of oxygen is not concentration dependent in the way that current flow through D was dependent upon a difference in potential: the rate of oxygen uptake can remain unaffected by concentration down to extremely low values (p. 286). Hence although the simple electrical model adequately serves to illustrate the principle of synergism in the diffusion path, if we truly are to simulate respiratory activity it becomes necessary to replace our lateral conductor with a constant current device. Furthermore as respiration can be more or less homogeneously distributed along the diffusion path a single constant current device could be a very inadequate means of simulating activity (p. 268). Mathematical expressions which embrace the synergism between physical resistance and loss or gain of diffusible species can be derived from the differential equations (3), (4) and (5). A number of the solutions are particularly relevant to the problems of plant aeration and are considered at some length below. (b) Planar flow. The differential expression for planar flow (equation 3) can be used to derive a solution describing the steady-state condition for linear diffusion through a medium in which the physical resistance to diffusion has an approximately homogeneous distribution and in which the sites for absorption of the diffusible species are also distributed homogeneously. The longitudinal acropetal gas-phase diffusion of oxygen which takes place in submerged roots can approximate in certain circumstances to this fairly simple diffusion model. Greenwood (1967b) considered such a case where the roots of intact plants were embedded in oxygen-free agar to the root/shoot junction and the stems were exposed to air. If, (a) no oxygen transfer occurs between the roots and the surrounding agar-medium, (b) the rate of oxygen uptake by the metabolic processes is homogeneously distributed in a linear direction and is unaffected by lowering of the oxygen concentrations until extremely low values are reached, and (c) there is a homogeneous distribution of pore space resistance in the longitudinal direction, then, when equilibrium is established between oxygen transport from the leaves and consumption by root metabolism, the distribution of oxygen concentrations along the root is given by solving

M D,

-

d2C dx2

- ~and D, is the where M is the rate of oxygen uptake by the root (g 0, ~ m s-l) effective diffusion coefficient for oxygen transport along the root (see equation 20).

254

W. ARMSTRONG

On integrating with respect to x, dC Mx + A dx De where A is a constant of integration. On further integration with respect to x

Mx2 C=-+Ax+B 2De where B is a second constant of integration. If the root is of length I, then on x = I

M1 -+A dx De MI and A = - De

-dC= o =

If oxygen enters the plant at concentration C,, we have C and C,

=

(ii)

=

C, on x

=

0,

(iii)

B

Substituting in (i) for A and B gives

c-c O -

Mx2 2De

MIX - Mx (x x 21) De 2De

and on rearrangement

c=c,-

MX(21 - X) 2De

from which the concentration C at all distances of x < l may be found, and when C = C, at the root apex (x = I),

MI2 c, = c, - 2De

from which the concentration C, may be determined. If De is replaced by D,TEand C, can be measured experimentally it becomes possible to find the value of T E ; if E is known the tortuosity factor can be established. If root growth stops at some apical concentration C,’ then the maximum length of root growth (1’) supported by longitudinal oxygen transport from the leaves will be given by:

AERATION IN HIGHER PLANTS

and if oxygen becomes zero on x = I" and root growth stops when C the maximum length of root (I")is:

255 =

0

Similarly, if the oxygen concentration becomes zero at some distance (x,) from the entry point the boundary conditions are C = Co on x = 0, and

on x

=

xl, and the distance x, is given by (34)

Unfortunately these equations are of rather limited application experimentally. Although techniques are available for measuring the concentrations C , and detecting the location C = 0 (Greenwood, 1967a, b ; Armstrong, 1967a; Armstrong and Wright, 1975), in practice there are probably few instances in which all the prior assumptions hold. Respiration perhaps never truly approximates to a uniform linear distribution in roots as apical respiration is invariably higher than elsewhere, porosity can vary considerably with length and, in oxygen-free media there will always be some oxygen leakage through the root wall. However, despite these inadequacies the equations can be used to illustrate a number of fundamental principles concerning the aeration of submerged roots and stems: (a) If we solve equation (30) for C over a range of values of x we find that the distribution of oxygen varies curvilinearly from x = 0 to x = 1 (Fig. 3d). This is the characteristic pattern for a system in which the diffusible species enters at one end of a tube and is then consumed over the whole length of the diffusion path. It contrasts with the linear distribution which was characteristic of the more simple planar system (cf. Fig. 3a). (b) Equations (33) and (34) serve to illustrate how the synergism between oxygen uptake and pore space resistance can effectively make the diffusive resistance between x = 0 and x = I appear to be infinite: in equation (34), the effective diffusive resistance has become infinite when x = x,. (c) If in equation (31) we were to substitute a zero value for M the whole of the right hand term becomes zero and hence C, = C,, and this would be so even if I -, co. In other words, no matter how great is the physical diffusive resistance in a submerged organ, provided that there is no oxygen usage or leakage the oxygen concentration throughout the gas space system must remain equal to that at the point of entry. In practice of course such a situation can never arise although respiratory activity and leakage may reach extremely low values. (d) Equation (33) also provides the opportunity to observe in some detail

256

W, ARMSTRONG

1

I-

1

Piant oxygen consumption (ng C' ~ r n -tissue ~

Fig. 5. Data computed from equation (33) predicting the maximum distance to which oxygen will diffuse longitudinally through plant organs in which there is uniformity of respiratory oxygen consumption and porosity along the diffusion path.

the relationship which exists between pore space resistance, respiratory activity and l', the potential length of aerated diffusion path. The line plots in Fig. 5 derived by substituting for M, De and 1' in thisequation are valuable as a general reference to the interaction of these parameters over the normal physiological range. For example a mean respiratory rate of c. 120 ng ~ m - ~ s-1 and an effective porosity of c. 3 % would be fairly representative for the roots of many non-wetland species; a glance at Fig. 5 is sufficient to demonstrate the potential inadequacies of internal aeration in such roots. Further references to this figure will be made later. (c) Radialflow: the cylindrical case. Steady-statesolutions forthe cylindrical case in one dimension in which respiratory activity and pore space are distributed uniformly along the radial diffusion path are helpful in assessing certain aspects of root and rhizosphere aeration. Two basic situations have been considered: (1) the radial diffusion of oxygen from soil to root (Lemon,

AERATION IN HIGHER PLANTS

257

1962; Lemon and Wiegand, 1962; Kristensen and Lemon, 1962; Griffin, 1968; Greenwood, 1969 and see below), (2) the converse of this, radial oxygen loss from root to soil (Armstrong, 1970). It may be noted that solutions in case (1) are also suitable for separately considering the oxygen diffusion across the stele. Case ( I ) : Consider a root segment of unit length within which diffusion is entirely radial and let the potential rate of oxygen consumption (M) and the effective diffusion coefficient (De) for oxygen be constant throughout the segment. For steady-state radial diffusion in cylindrical coordinates the diffusion equation is:

Multiplying throughout by r we get

!(rg)

=

De Mr

dr and on integrating both sides with respect to r then: dC Mr2 r- = -- + A dr 2De Dividing throughout by r gives dC - _ _-_Mr dr 2De

+ -Ar

and integrating both sides with respect to r we get C

Mr2 4De

= __

+ A logr + B

(39)

If C = C, at the perimeter of the root where r = b then for roots in well g ~ m at- 23°C. ~ stirred aerated water, C, would approximate to 8.57 x Synergism between respiratory activity and diffusive impedance will lower the oxygen concentration radially across the root. If at some inner radius r = a the oxygen concentration falls to some critical value C = C, such that on r < a M becomes zero, then from r = a to r = 0 the ratio dC/dr = 0. Summarizing these boundary conditions we get r 2 0 < a < b 2 r a n d C, on r = b dC C=C1,-=O,onr=a dr C

From equation (39) we get

=

258

W. ARMSTRONG

Mb2 C --+Alogb+B O - 4De and

Ma2 C --+Aloga+B - 4De

Subtracting (41) from (40) gives C, - C Now as C = C,,

-

M (b2 - a2) 4De ~

+ A log (b/a)

dC - = 0 on r = a dr

from (38) we get

-Ma +

A -=o a

2De

(43)

and A=-

Ma2 2De

(44)

Substituting from (44) into (42) gives: C,-C,s-

(bZ- +

Ma2 4De a2

2log--l b

which may be solved to derive the radius r = a. If we consider the situation where C = C, on r simplifies to

c,-c

=

(45)

a = 0 then equation (45)

Mb2 - 4De --

and as oxygen consumption appears to continue at a constant rate down to extremely low concentrations such that C, NN 0 then C, may be eliminated from equation (46) and on re-arranging we get:

where b is now the critical radius of the root when the root is just wholly aerobic (e.g. see Table 111). For fixed values of De, C, and M any increase in radius b will cause the development of a core of anaerobiosis within the root. If C, = 0 and is eliminated from equation (45) then the radius r = a will be the estimated radius of this anaerobic core. If we wish to predict the distribution of oxygen across a cylinder in which

AERATION IN HIGHER PLANTS

259

all the prior assumptions have been met we require a solution in terms of C,, C,, C, r, b and a. The solution is:

but if C ,

=

0 and dC/dr

=

0 on r

=

a the equation simplifies to:

This solution may be suitably modified for a = 0 (Equation 94, Appendix 2). Numerical solutions of equations (48) and (49) are best carried out by computer. Equation (46) appears extensively in the literature but to my knowledge solutions (45,) (48) and (49) have not previously been published in connection with root aeration although Currie (1961a) has presented spherical analogues of (45) and (49) (see Appendix 2). Equations (45)-(49) must of necessity be treated with caution and they have very obvious limitations when applied to diffusion across the whole section of root: while respiratory activity might frequently approximate to a uniform distribution, diffusivity most certainly will not unless for some reason the intercellular space system has become flooded. In the intact root there will be a major change in diffusivity at the stelar boundary. Within the stele the effective diffusion coefficient may approach the diffusivity of oxygen in water; in the cortex it will more closely approach that for oxygen in air but modified by pore-space volume and tortuosity of the diffusion path. The latter is probably greater in the radial path but again, as with longitudinal diffusion, definitive values appear to be unavailable. Greenwood (1968, 1969) has used equation (46) to predict an overall value of De of approximately 1-2 x cm2 s-l for “average” roots and has suggested that the critical value of C , for roots of radius 0.037 cm is very low, c. 2.8 x ml/ml in water (approximating to a partial pressure of c. 0-01 atm). More will be said of this later (p. 313). Case (2): Outwardflux. Radial oxygen loss from roots into saturated soil encourages the formation of an aerated rhizosphere in which microbial activity may be aerobic and in which the by-pro’ducts of anaerobic decomposition and metabolism may be oxidized. The diffusion equations describing the outward diffusion of oxygen from a root into a medium in which the potential rate of oxygen consumption (M) is constant radially and in which oxygen diffusivity (De) is also constant are identical in form with those for inward diffusion but the a, b notation is reversed : if we take as our boundary conditions: (a) C = C, on r = a (root radius) and (b) C = C,, and dC/dr = 0 on r = b, where 0 < a ,< r < b the solution equivalent to equation (45) is: a

260

W. ARMSTRONG

Analogous solutions are available for equations (48) and (49); there are no corresponding analogues for equations (46) and (47). Equation (50) in an alternative form has been used to confirm the likelihood and extent of rhizosphere oxygenation in soils (Armstrong, 1970) and has provided the basis for an electrical simulation of soil sink activity around the root (p. 270). C. THE OXYGEN SOURCE

1. Leaf Resistances During the hours of darkness oxygen reaches the mitochondria of the leaf tissue by diffusive flow along a negative gradient of concentration which builds up between the external atmosphere and the tissues. In daylight the circumstances change: the leaf becomes an oxygen generator and this leads to a net efflux of oxygen across the leaf surface. Although the diffusion path between the leaf mesophyll and the turbulent atmosphere is in some parts non-planar (Brown and Escombe, 1900; Bange, 1953), when the various resistances in the path have been quantified the diffusion of carbon dioxide and water vapour across leaf surfaces can be treated in planar terms (Meidner and Mansfield, 1968; Nobel, 1974). The movements of oxygen can be treated likewise. Consequently, provided that we can identify and quantify the diffusive resistances and are aware of the rate of oxygen consumption, the simple steady-state solution for planar diffusion should enable us to compute the approximate gradient of oxygen concentration across the leaf surface during periods of darkness. Conversely, the net output of oxygen into the atmosphere during daylight can be used to predict the gradient of oxygen pressure from mesophyll surface to atmosphere. The recognizable resistances to oxygen flow and their distribution in one dimension are indicated diagrammaticallyin Fig. 6; they are quantified below. (a) The boundary layer, The thickness of the boundary layer of still air adjacent to the leaf varies with position on the leaf, with leaf shape, and with wind-speed and direction. Following normal convention we denote the average depth of the unstirred layer as S& and we may compute its approximate value from the expression: 68 x 0-4

(k)

where I is the linear dimension of the leaf in the downwind direction, and v is the ambient wind velocity in cm SKI. Equation (51) is based on hydrodynamic theory for laminar flow adjacent to a flat surface, but has been modified from experimental observations to fit the leaf model (see Nobel, 1974). The magnitudes of 6a for a wide variety of wind speeds and leaf dimensions are given in tabular form by Nobel and others: as I varies from 0.2 to 50 cm, and v from 10 cm s-l (“still air”) to 1000 cm s-l (22 m.p.h.) S*

261

AERATION IN HIGHER PLANTS

Mesophyll surface

Turbulenl air

Stomota

7

Unstirred boundary layeradjacentto leaf

lnlercellulor space

Epidermis and

cuticle

Fig. 6. Resistances involved in oxygenexchangebetween the leaf and external atmospheres.

ranges from 0.0057 to 0.89 cm. Accordingly by analogy with equation (12) we find that the resistance Ra must be given by S / D and for oxygen in air at 23°C will vary between (0.0057/0*205)or 0.0278 s cm-l and (0-89/0.205)or 4.34 s cm-l. If we wish to express the resistance for a defined portion of the total leaf area we may use the expression [IDA (cf. equation 14) and for unit area of surface the two resistances would become 0.0278 and 4.34 s ~ m - ~ respectively. (b) Leafwall resistance. Leaf-wall resistance is dependent upon stomatal numbers, the degree of opening and the length, depth and other geometrical characteristics of the stomatal pore (Bange, 1953), but as the stomata approach the point of closure it becomes increasingly a function of cuticular and epidermal resistances. The relationships I/D or 1/DA for resistance to onedimensional planar flow (equation 12) can both be used to derive leaf-wall resistance. (i) The stomatal component: Stomata1 pore space (or rather the lack of it) modifies the value of D for oxygen transport across the leaf surface just as cellular occlusion of potential gas-space modifies the value of D in longitudinal transport (p. 249). Total resistance per leaf will be reduced still further if stomata occur on both leaf surfaces. In the following paragraphs it will be assumed that the stomata are confined to one surface and for convenience that the non-stomata1 surface is impermeable to oxygen. The fractional porosity (c) of the stomatal leaf surface (area, Az') is given by the expression n . ast/Al', where n is the number of stomata and ast is the average area per stomatal pore; the effective diffusion coefficient D& is therefore given by Do . n . ast/Az', (i.e. Doc). For leaves with fully open stomata the value of E generally varies from c. 0-004 to 0-02 depending upon the species and the corresponding maximum values of De(ox)at 23°C are (0.004 x 0.205) or 8.2 x cm2 s-l and (0.02 x 0,205) or 4.1 x 10-3cm2s-1. In practice we calculate E from the relationship n'ast where n' is the number of stomata per unit area of stomatal surface, and ast is the average area per

262

W. ARMSTRONC

stomatal pore. The measurable depth of the stornatal pore, dst, is not strictly analogous with 1 in the expression l/DA and it is necessary to make an end correction to allow for the funnel-like diffusion patterns which occur immediately beyond each end of the pore. It has been shown that the pore usually has an effective depth equal to the observed depth (d") plus a distance equal to the mean radius of the pore rst. The latter is given by

and if there was but one-stoma per unit area of leaf its effective resistance R' (s cm-l) would be given:by :

However stomata occur in extremely large numbers over leaf surfaces (Meidner and Mansfield, 1968) and behave collectively as a network of parallel resistances such that the greater their number and degree of opening the less will be the total resistance of a given leaf surface. If there are n stomata over a given leaf surface then the total stornatal resistance of that surface R& (measured in s cm-l) will be:

or by analogy with equation (18) we may write

If the fractional porosity of the leaf surface were 0.02, dst, lOpm and rst, 5pm, the value of R& would be 15 x 10-4/4.1 x or 0.365 s cm-I. cm2 s-l and dst is 50 pm Taking the more extreme case where De is 8.2 x the resistance becomes 6.7 s cm-l. Again, if we wish to use the analogue IIDA then for unit area of leaf surface the above resistances would take the units s ~ m - ~ . (ii) The epidermis. The leaf epidermis itself with its waxy cuticle forms a substantial impedance to the diffusional flow across the leaf surface. We may make an approximate lower estimate of the epidermal resistance by appropriate substitution in the expression I/D. If the liquid path across the epidermis is 40 pm and the coefficient for liquid phase diffusion of oxygen across the cell is approximately that in water, then at 23°C the resistance would be (40 x 10-4/2.267 x or 176 s cm-l. Epidermal resistance per se is thus substantially greater than the stornatal term and as a resistance in parallel with the stomata will be insignificant providing that the stomata remain open. (c) Gas-space resistance. This is probably the least significant resistance in

AERATION IN HIGHER PLANTS

263

the diffusion path into the leaf. The effective length (i.e. average distance between the mesophyll surfaces and stomatal pores) of the diffusion path ranges between 100pm and 1 mm for most leaves and hence the resistance Re*will range from (100 x 10-*/0.205), 0.049 s cm-', to 0.49 s cm-'. The gas-space resistance will be ignored in the examples which follow. 2. Photosynthetic versus Atmospheric Oxygen Source Having regard to the potential diffusional impedances outlined above it is now possible to perform a few simple calculations and from these to make a number of general statements of principle concerning the role of the leaf in whole plant aeration. Consider firstly a leaf in which the diffusive resistances to planar flow are the greater of those computed above. Let the leaf have but one permeable and stomatal surface and let the non-stomata1 path through the epidermis have a resistance of 200 s. cm-l. The combined epidermal and stomatal resistance (Rv)is 6.48 s cm-l and is given by 1/Rv = 1 /6.7 1 /200. If R&is 4.34 s cm-l, the resistance external to the gas space, Re, will be 10.82 s cm-l. If the leaf is in darkness, the stomata fully open, and respiration relatively high such that the oxygen flux into the leaf is 0.2 nmol cm-2 s-l we can calculate the oxygen concentration in the gas-space of the leaf C, from a derivative of equation (12):

+

where J is the oxygen flux and C, is the oxygen concentration in the atmosphere beyond the boundary layer. If we assume the atmosphere to be water g ~ m (20.41 - ~ %) and C, will saturated then C , may be taken as 269 x - [ 10-82(32 x 0.2 x or 268.93 x g~m-~(20.40%), be 269 x a fall in concentration of only 0.01 %. If the stomata had been fully closed the total resistance would have been 204.34 s cm-' and C, would have become 267.7 x (20.3 1 %) which is still a very small fall in concentration (0.10 %). Consider now the same leaf illuminated, supposing it to have a net inward flux of carbon dioxide from the atmosphere of 1.8 nmol cm-2 s-l. If this represents approximately nine-tenths of the amount of carbon dioxide fixed by photosynthesis the other tenth being supplied as a respiratory by-product, then the photolysis process will supply the equivalent of 2.0 nmol cm-2 s-l of oxygen and nine-tenths of this (1.8 nmol cm-2 s-l) will escape to the atmosphere. If the stomata are now fully open the concentration of oxygen or [ 10.82 (32 X 1.8 X in the intercellular spaces will be 269 x 269.62 x g ~ m (20.46 - ~ %) a rise in concentration of as little as 0-05% and hence the total fluctuation in the oxygen percentage within the leaf from darkness (stomata fully closed) to full illumination is only 0.15. If leaf respiration can go unchecked at internal concentrations 2 2 % (see p. 286 and Yocum and Hackett, 1957) it will be clear from these examples

+

264

W. ARMSTRONG

that for leaves in isolation the stornatal and other lateral resistances are of no consequence in the leaf aeration process during darkness, and in daylight they have virtually no restraining influence whatsoever on the escape of oxygen produced by the photolysis process. How then might these characteristics affect the aeration of submerged underground organs dependent for their oxygen supply upon the aerial parts of the plant? Clearly this must depend upon the morphology and growth characteristics of the plant concerned and upon habitat circumstances. If the water table is coincident with the soil surface and the growth habit of the plant is such that both stem and roots are below ground then during darkness the oxygen can only enter across the leaf surfaces. Because of the greater flux now required to supply the respiratory needs of the below-ground parts the fall in oxygen concentration across the leaf surface may be greater than previously indicated. How significant this might be will naturally depend upon the oxygen demand, leaf area, the degree of stornatal opening and upon cuticular and epidermal resistance; it could also depend to some extent upon resistances to “longitudinal” flow within the leaf, However, the concentration drop across aunit area of leaf having a total dark resistance of 204-34 s ~ m - ~ and , supporting 125 cm of root (r = 0.05 cm and mean respiratory rate 60 ng ~ m s-l) - ~would be only 1.01 %. If the stomata were a tenth open (dark resistance 58.87 s ~ m - ~ ) the corresponding value would be only 0.29 %. If the leaf area was doubled the respective values would be 0 5 7 % and 0.16%. If the stomata were 50% open, resistance would be 16.9 s ~ m and - ~the concentration drop across unit area of leaf would be 0.08 %. A doubling of leaf area would give a concentration difference of as little as 0.05 %. It should be noted that the highest oxygen flux across the leaf will tend to occur at the point of least resistance from the sink, i.e. at ground level, consequently there will be a tendency for the fall in concentration across the leaf to be greatest at this point. However, this tendency will be immediately counteracted and minimized by longitudinal oxygen flow from more distal parts of the leaf. If the stem is emergent, lenticels or stem-borne stomata will provide the route of least resistance to below-ground parts and again there will be the tendency for the surface oxygen flux and concentration drop to be greatest at soil level. From what has already been said we might again predict that relatively little emergent stem may be required to sustain the respiratory activity of submerged parts. Conversely if longitudinal resistance to flow is high, compensating movements from more distal parts will be restricted and this will tend to limit the effective entry point to basal regions. Since all the examples chosen above have been deliberately weighted in favour of high leaf resistances one is forced to conclude that surface resistances at the point of oxygen entry to the plant must normally have an almost undetectable restraining influence on gas flow to more remote parts. Experimental support for this view is illustrated in Fig. 7. We can conclude also that

265

AERATION IN HIGHER PLANTS

I

0.0 0

A

2

6 0 10 12 Length of leaf lanolined ( c m ) L

14

16

, 10

Fig. 7. Sequential basipetal occlusion of stomata in a single-leafed specimen of Eriophorum nngusrifolium and its effect upon the oxygen status of the root apex. Stomata occluded by painting the leaf with lanolin down to the submergence level. Oxygen status of the root apex indicated on the ordinate by the electrolysis current caused by polarographic electro-reductionof oxygen leaking from the root (after Gaynard, 1979).

unless some extra restraint can be placed upon the escape of oxygen from the leaves in daylight (see p. 297) photosynthesis will do no more than ensure a daytime oxygen source almost identical in pressure with that in the external atmosphere and the switch from atmospheric to photosynthetic source will be almost undetectable; this is borne out experimentally (T. J. Gaynard and W. Armstrong, unpublished). With increasing respiratory demand there will be a tendency for the daytime oxygen supply to include a significant atmospheric component and in parts of the leaf system there might be a net efflux and in others a net influx of oxygen. In species with non-photosynthetic woody stems atmospheric oxygen will normally remain the sole oxygen source at all times. D . THE AERATION MODEL

I . Introduction Some indication has been given already of the usefulness of mathematical modelling in the analysis of aeration problems. However, whilst the relatively straightforward mathematical solutions for planar and radial diffusion are valuable for evaluating certain specific aspects of the aeration process, individually they are somewhat limited in their scope and the development of more elaborate models which might integrate the many facets of plant aeration is a desirable goal. In recent years two such models have been evolved, both intended primarily

266

W. ARMSTRONG

for the study of root aeration by the internal path. The first of these (Luxmoore et al., 1970, 1972) relies on the linking together into a continuity equation the individual equations describing diffusion in the various segments of the aeration path. The final continuity equation is manipulated using modern computation methods. This model devised by Soil Physicists at Riverside, California overcomes various of the deficiencies inherent in equation (27) : uniformity of porosity and respiratory activity is no longer a requirement and both can vary in successive root segments. Radial oxygen loss from the root is allowed for, as is radial intake in the presence of a soil oxygen source, while provision is made for a concentration-dependent respiratory rate. The latter was perhaps an untimely refinement for the concentration dependence characteristics used (10% oxygen for 3 max. respiration) have since been proved wrong. Recent studies indicate values considerably lower than this (< 2.5 % oxygen for maximum respiration, p. 286) and hence the published data from this model may need to be treated with some reserve. The principle limitations of the model lie in the provision made for “simulating” the radial oxygen loss from the root in a wetland environment and the radial oxygen intake in an unsaturated soil. In both instances simulation is based on the simple case for radial diffusion (equation 27). For inward diffusion (nonwetland soil) it is assumed that the root (radius r = a) for the whole of its length is surrounded by a shell of water (r = b) such that on r = b, C = C, = 18 % oxygen. No allowance is made for oxygen consumption within the liquid shell which would be analogous with soil oxygen consumption. For radial diffusion from the root (the wetland condition) it was assumed that on r = b, C = C , = zero. Potential sink activity external to the root is therefore intensified by reducing the input radius of the liquid shell (r = b). This device is not a strict analogue of soil oxygen demand for no allowance is made for the distribution of oxygen demand along the diffusion path a -, b and no soil respiratory rates were specified. An inbuilt assumption is that soil oxygen consumption will vary linearly with the internal oxygen concentration of the root while in reality the relationship is more likely to be curvilinear (p. 272). However the importance of such limitations has yet to be established and the errors introduced may yet turn out to be marginal. If not the model can undoubtedly be modified to accord more closely with reality. The second of the two models, a functional electrical analogue was developed specifically to simulate root aeration in the wetland condition. In its original form (Armstrong and Wright, 1976a) it embodied the same deficiency found in the mathematical model, i.e. to simulate the soil “sink” it was assumed that the root for the whole of its length was bounded by a shell of water (radius r = b) such that on r = b, C = C , = zero. This deficiency has now been rectified and the model has also been modified to accommodate the soil oxygen demand of the unsaturated (non-wetland) soil condition (p. 321).

AERATION IN HIGHER PLANTS

267

For the average biologist the electrical analogue is undoubtedly the easier of the two models to understand and operate. The principles underlying its design are outlined below and in subsequent sections reference is made to its application. 2. The Electrical Analogue (a) The basic unit. The similarities between electrical and diffusion laws demonstrated earlier provide the basis for the electrical modelling of diffusive aeration. These similarities are such that in a functional model electrical resistors may take the place of diffusional impedance (p. 247), resistors with “leakage” to “earth” can behave as diffusion sinks, and electrical “pressure”, (EMF), substitutes for partial pressure and concentration differences of diffusate. In the functional model appropriate values are assigned to these simulators and both flow and partial pressure of diffusate at any point in the system can be respectively monitored by ammeter and electrometer (voltmeter) suitably scaled. We can electrically simulate unit length of root-wet-soil system as shown in Fig. 8a; further identical circuit units are added in series to simulate an increasing length of root but, it may be noted that in the unit representing the apical segment of root the resistor RP, becomes superfluous (see Fig. 8b). Pore space resistance to longitudinal oxygen flow is represented by the equal resistors Rp, and Rp,, and, on the assumption that respiratory activity is homogeneously distributed with length, the oxygen consumption by the root tissues is simulated by a single lateral current leak from between Rp, and Rp,,. This may be controlled by a simple variable resistor RR as indicated but in practice it is more satisfactory and realistic for RR to operate through a constant current (compensating) device. In this way the model automatically simulates the natural insensitivity of respiratory activity towards oxygen concentration (p. 286). Consequently once respiration has been programmed it is not readily upset by the changes in concentration induced by programming adjustments made elsewhere in the model. Radial oxygen loss to the soil is also simulated by a lateral tapping between Rp, and Rp,,, controlled by another variable resistor R,, and a realistic simulation of soil sink activity in the wetland environment is quite feasible with this simple device. The impedance of the root wall is simulated by resistor RWL.It is recognized that this may not be an entirely satisfactory way of simulating the root wall, since it fails to take account of sink activity within the wall itself. It is felt that a truer simulation must await the results of further experimental and theoretical studies into the nature of root wall resistance. Oxygen concentrations at positions P, G and T are measured by electrometers Vp, VGand VT.The use of the high impedance electrometer is necessary to ensure an insignificant lateral current loss through the measuring device. Respiratory activity and radial oxygen loss are measured on ammeters AR

268

W. ARMSTRONG

and As respectively and it is especially important in this model that these lateral current tappings should be taken at a point midway along the root segment. Only then will the concentration recorded at T be a true reflection of the concentration drop along a root segment in which respiratory activity and pore space resistance are distributed uniformly with length. The truth of this statement is easily demonstrated electrically as follows. In unit length of root in which respiratory activity and pore space are distributed uniformly with length (and where the oxygen source is at one end) the oxygen profile along the root may be determined by solving equation (30). In electrical terms we could in theory simulate this root as a longitudinal resistor bearing an infinite number of lateral tappings extracting equal currents and summing to the total respiratory consumption. However it is impossible to construct such an analogue and in practice some compromise must be reached. With this in mind consider a lOD electrical resistor R, with a source of potential Vp (100 volts) applied at one end, and let the current i, taken from each of a finite number of lateral tappings (n) set equidistant along R, be equal and, be such that the total current taken (Xi,) is 5 amperes. In the arrangement shown in Fig. 8c, n = 5, R,, = R,, = R,, = R,4 = R15= 2 Q and i, = 1 ampere. The voltage drop across RI1,given by (6V‘ = Zi, x R,,) is (5 X 2) or 10 volts, and the voltage at Q is therefore (100 - 10) or 90 volts. The voltage drop across R,, is given by SV“ = (Xic x ic) - R,, and hence the voltage at S is 90 - (4 x 2) or 82 volts. Continuing with this procedure it can be shown that the voltage at T (VT) must be 70 volts. Again, by Ohm’s law we can calculate that this same potential would have been realized had the current, Xi,, been taken from a point R p along resistor R,. The value of R *, given by (VP - VT)/Zic, is (100 - 70/5)or 6 D and it may be noted that the ratio R,*/R, is 0.6. When n < 5 the ratio is larger and as n increases the ratio diminishes. If we plot R,*/R, against n we obtain a curve which is fitted by the equation y = (O*5/n) 0.5 where y = R,*/Rl from which it is evident that as n +-co then y --+ 0-5. Hence provided that our major concern is with total concentration drop along unit length of root rather than with the concentration profile it is obvious that this may be satisfactorily achieved by extracting the total oxygen

+

Fig. 8 (a). Detailed arrangement of resistances and meters to simulate the diffusion of oxygen in unit length of root-wet soil system: Rp’ = Rp” = fRp, where Rp is the longitudinal pore-space resistance of the root segment; RR, a variable resistor for setting the respiratory uptake of the root segment; RWL. the diffusive resistance of the root wall; Rs, variable resistance for the control of soil sink activity. Electrometers, V, indicate oxygen concentration;meters, A, register oxygen consumption by root and soil. (b). Alignment of root-soil units in a functional electrical analogue. (c). Five electrical resistances in series, representing together the diffusiveresistance in a unit length of root in which respiratory activity is represented by five equal current tappings (ic). See Section 11. D.2 (a).

n

u

Y

b

n Y

c V

-

i U

t2

270

W. ARMSTRONG

consumption at a point halfway along the pore-space impedance of the segment (H in Figs 8a, b and c). (b) Calibration. To calibrate the model it is necessary simply to equate some value of electric potential (e.g. 20 V) with the atmospheric oxygen source, and similarly to choose some resistor value (e.g. lo3 SZ) to equate . electrometers with a particular diffusive resistance (e.g. lo4 s ~ m - ~ )The may now be calibrated so that 20 volts will read atmospheric oxygen concentration near full scale deflection. Choosing the lesser of these two values, 20 volts might therefore equate with an oxygen concentration in the gas-phase g cm3 at 23°C (see Table I). If this source (C = C,) is separated of 269 x , from a sink (C = C, = zero) by the diffusive resistance, lo4 s ~ m - ~the diffusion rate will be (269 x 10-s/104) or 26.9 ng s-l. The ammeters will read (20/103 amperes) or 20 mA and the linear scale of the meters may then be calibrated directly to read 26.9 ng s-l at 20 mA or shunted so that 25 ng s-l will read on an appropriate scale division. The diffusive resistances of the plant, RP and RWL(calculated from the expressions l/DA (p. 249) or a. log (b/a)/DA, (p. 251) or derived experimentally (p. 276), may be assigned their respective analogue values (ohms) from the relationship R(ohms) = R(s ~m-~ )104/103.Resistor values chosen for RR and R, are those necessary to programme for the appropriate ranges of root respiration and soil sink activity. The potential assigned to the atmospheric oxygen source is applied at the top of the series of current units as shown in Fig. 8. (c) The wetland soil sink. To simulate the activity of the soil oxygen sink (whether wetland or non-wetland) it is convenient to adopt the oft-made assumption that the potential respiratory activity of the soil is constant along the radial diffusion path and is unaffected by oxygen concentration until this approaches zero (Greenwood, 1961, 1962, 1963). Similarly let us assume that the effective diffusion coefficient in the radial direction is a constant. Having made these assumptions we may approach the electrical simulation of wetland soil sink activity as follows: Consider a unit length of internally aerated root (radius r = a) lying within a wet soil having a potential rate of oxygen consumption, M (g ~ m s-l). - ~If the oxygen concentration at the root wall is CWLthen from equation (50) the radial distance (r = b) from the centre of the root a t which the oxygen concentration in the soil must fall to zero is given by a and in unit time the quantity of oxygen, Q, consumed by the soil surrounding the root will be Q =

t

MTI(b2 - a2)

(57)

AERATION IN HIGHER PLANTS

27 1

Q/t is also the rate of oxygen diffusion from root segment to soil (i.e. the radial oxygen loss, g s-l) and this value may be programmed into the analogue by adjusting the resistor R,. For any given value of M and a, the radial distance b-a varies as a curvilinear function of CWL(and hence with the concentration within the root) (see Fig. 9a); radial loss from the root also varies in a curvilinear manner (Fig. 9b). The necessary programming adjustments for sink activity in any particular root segment are made by consulting the graph of CWLand radial oxygen loss appropriate to the particular root radius and potential soil activity in question. The modelling of soil sink activity in this way is clearly a simplification although not necessarily a serious one; it is also a compromise. Regarding the simplification it is by no means certain that soil respiration and effective diffusivity will be uniformly radially distributed. As values for M and De become known from experimental observation it may become necessary to adopt the method developed for simulating root aeration to the unsaturated soil (see p. 321) and separately model soil resistance and respiration in successive shells around the root. If soil oxygen consumption varies in a curvilinear manner with CWLin theory it must vary also in a curvilinear manner along a root segment. It would be a difficult matter to simulate this circumstance exactly and the compromise adopted, referred to above, is the tapping of soil activity midway along each circuit unit (cf. root respiration). It is felt that the error introduced by this procedure is small enough to be disregarded. ( d ) Programming. Normally, one circuit unit represents unit length of root-soil system and to simulate the conditions required the model is programmed by making the necessary adjustments to the various resistors. Pore space and wall resistances are usually programmed first, followed by root respiration. As the oxygen consumed by the soil must depend on the oxygen concentration at the root wall this parameter is programmed last of all. Depending upon the procedure adopted t o simulate sink activity (above and p. 321) itmay or may not be necessary to make severalsuccessive adjustments to the resistor R, to bring the model to equilibrium. When fully programmed the model automatically integrates the interactions between the various impedances and sinks acting on the linear diffusion path, and the oxygen profile along the root is obtained by plotting the readings from meters VL; VLOrepresents the concentration at the root base, VL,one centimetre from the base, and likewise VLn the concentration at the apex, where n = the number of circuit units.

272

W. ARMSTRONG

Equivalent gas-phase oxygen concentrotion in soil solution a t mot’wall (%)

Fig. 9 (a). Thickness of oxygenated rhizosphere (b - a, equation 56) as a function of the oxygen concentration at the root wall and the root radius, a (a = 0.05 cm or 0.01 cm). Values computed assuming a uniform respiratory activity in the aerated soil of 5.27 x g 0, cm4 s-l, and a uniform oxygen diffusivity of 1 x cma s-l. (b). Radial oxygen loss from a root (radius, 0.05 cm) into wet soil, as a function of the oxygen concentration at the root wall. Soil characteristics as above.

111. THE CYLINDRICAL PLATINUM ELECTRODE TECHNIQUE 1. Introduction

The inadequate methods of analysis which for many years hampered the study of diffusive aeration in plant and soil have now been superseded by the more sophisticated tracer, GC and polarographic techniques (Lemon and Erickson, 1952, 1955; Barber et al., 1962; Armstrong, 1964, 1967a; Greenwood, 1967a,b; Greenwood and Goodman, 1967; Jensen et al., 1967; Armstrong and Wright, 1975,1976b; Smith, K. A., 1977). Theconstruction and use of cylindrical Pt electrodes for polarographically assaying the oxygen diffusion from roots in anaerobic media was first reported as a method for quantifying the differences in the rhizosphere-oxygenating activity of wetland plants (Armstrong, 1964, 1967a). Since then it has become apparent that “flux” data yielded by quite simple procedures can be successfully manipulated to quantify many aeration properties in both wetland and non-wetland plants (Armstrong and Wright, 1975). At present the cylindrical Pt electrode technique probably provides the least expensive and most versatile method for assessing the diffusive resistance to oxygen transport in roots: pore space resistance, root wall resistance, tortuosity, the synergism between root respiration, lateral leakage and pore space resistance may all be quantified,

AERATION IN HIGHER PLANTS

273

U potentialof Pielectrode (volts)

Fig. 10 (a). Cylindrical platinum electrode for assaying the oxygen flux from roots. Key: a, celluloid guide; b, perspex tube; c, platinum cylinder; d, epoxy-resin; e, root through electrode; f, solder joint; g, araldite; h, sleeved copper wire. (b). Curve (i) represents a typical current voltage curve (polarogram) resulting from the electrolytic reduction of oxygen at a platinum cathode. Reactivity of the electrode increases from A to B when it achieves maximum efficiency. In the plateau region, B-C the rate of oxygen diffusion is the controlling factor. H+ ion reduction begins at C and this process accelerates from C-D. In the absence of a supporting electrolyte the oxygen polarogram characteristics are lost (curve ii).

while aeration parameters in shoot and leaf and the dynamics of oxygen transport can also be assessed. The essential features of the technique, described below, provide a useful background to much of the data presented in the following two sections. 2. The Polarographic Method The polarographic determination of oxygen diffusion from roots is based on the characteristics of the current-voltage (c-v) curve obtained (Fig. lob) when oxygen in aqueous solution is electrolytically reduced in a cell in which one electrode, the cathode, consists of a sleeve-insulated thermo-pure platinum tube (Fig. lOa) while the other is some standard half-cell (e.g. saturated Ag/AgCl reference electrode). The reduction of oxygen at a Pt surface is thought to proceed in two stages (McIntyre, 1970). At p H 3.5 or above the overall reaction follows the equation: 0, 2H,O + 4e- + 40H-, and for each molecule of oxygen reduced there is a current transfer of 4e-. At low potentials (applied EMF) this reaction is voltage dependent but with increased potential it becomes dependent on the rate of oxygen diffusion to the electrode surface. The c-v curve then assumes the form of a plateau (curve (i), Fig. lob). If the applied voltage in the plateau region is sustained the current equilibrates to a value which is related to the rate of oxygen diffusion to the electrode according to the equation:

+

274

W. ARMSTRONG

it where it n

=

nF fxeO,t

(58)

= the

diffusion current in amperes at the time of equilibration, t, number of electrons required for the reduction of one oxygen molecule = 4, F = The Faraday, 96 500 coulombs, and fx=O,t= the oxygen flux at zero distance (x) from the platinum surface at time t (mol cm-2 s-l). Oxygen diffusion from roots is measured in oxygen-free liquid medium (+supporting electrolyte, see Fig. lob). Roots are inserted through the Pt electrode as shown in Fig. 10a, and a shell of liquid (of uniform thickness) separates the root from the inner (reactive) electrode surface. Under the appropriate polarizing voltage the platinum acts as a sink for oxygen, (the oxygen concentration at the electrode surface is effectively maintained at zero) and a diffusion gradient is set up between root and electrode. At equilibrium the rate of oxygen loss from that portion of root lying within the electrode can be calculated from equation (58) which simplifies to the expression : = the

ROL

4.974 it 60

=-

(59)

where ROL = the radial oxygen loss in ng s-l, and it = diffusion current (PA) with the root within the electrode, provided that the root is the only significant oxygen source. 3. Manipulation of “Flux” Data (a) Calculating the difusive resistance oflered by the liquid shell between root and electrode. Reference to Section II.B.6 will show that at equilibrium the boundary conditions of the root-surface/electrode diffusion-system are those of the simple case for steady-state diffusion along radial coordinates. It follows that diffusion must conform with equation (27) and as C1(concentration at the electrode surface) is equal to zero the diffusion rate given by equation (59) above must also be that given by the expression: D ;ARCWL Q’t = a log (b/a)

where Q/t is the diffusion rate in g s-l, DG is the diffusion coefficient for oxygen in water at the temperature T, AR is the surface area of the root within the electrode (cma), CWLis the dissolved oxygen concentration at the root surface (g ~ m - ~ ) , a is the root radius, and b the electrode radius (cm). In equation (60) the resistance of the liquid path between root and electrode

AERATION IN HIGHER PLANTS

275

is expressed by the term a log (b/a)/DG AR.However, the liquid shell between root and electrode may be considered as a lateral extension of the diffusion path within the root ( = t o R,, Fig. Sa). To quantify the resistance of the shell relative to transport in the gas-phase of the root allowance must be made for the fall in oxygen concentration which occurs across the “air”-liquid inter~ the ) liquid shell becomes: phase. Accordingly the effective resistance (s ~ m - of

) where Cz is the oxygen concentration in air at temperature T (g ~ m - ~and C;feWis the oxygen concentration in air-saturated water ( g ~ m - ~ )If. a = 0.05 cm, b = 0.1 125 cm and electrode length 0.5 cm, the value of Rsh is 3.570 x lo5 s ~ m at- 23°C. ~ (6) The overall dijiisional impedance apparent at the apical root wall. If the oxygen diffusion rate (Q/t) is measured over the submeristematic apical root segment (1 = 0.5 cm) it is a simple matter to determine the total effective internal diffusive resistance (Rt) between the atmosphere and the root surface. From what has been stated previously it follows that diffusion rate from the root must be given by the expression:

and if Q/t is computed from equation (59) and Rsh from equation (61) the equation can be solved for Rt. The term Rt is a measure of the total effective diffusive impedance between the oxygen source and the surface of the root apex. However, if the oxygen enters the plant at a point very close to the root base the term Rt can effectively represent the total synergistic resistance of the root at that particular temperature plus a wall resistance component. If the wall resistance component is insignificantly small Rt is then a measure of the total effective resistance to longitudinal diffusion within the root itself. (c) Root wall resistance. What little data there is available suggests that the oxygen permeability of root walls can be surprisingly high in apical regions (Armstrong, 1967; Greenwood, 1967a; Luxmoore et al., 1970). As the root “wall” forms only a small part of the lateral diffusion path between root and electrode cyclosis within the wall layers may considerably reduce their apparent diffusive resistance (p. 240). The diffusive resistance of the liquid shell will effectively enhance the effects of any streaming component in the wall and the natural diffusive resistance of the root wall will be masked. Wall resistance can be estimated by measuring first the oxygen diffusion from the root apex and then extracting and analysing the gas from the intercellular spaces of the cortex (T. J. Gaynard and W. Armstrong, unpublished). The wall resistance is given by solving the equation:

276

W. ARMSTRONG

where Ciasis the oxygen concentration within the root, and RWLis the apparent resistance of the wall.* The relationship found between wall resistance and total resistance in Eriophorum angustifohm using this method is shown in Fig. 11. u 4.0ul

In

--

0 a + .-05B2.0UI

-r

r ” -

L

0

B

0

/

9”.

Fig. 11. Eriophorum angustifohm. Apical root wall resistance to radial oxygen diffusion, as a function of total plant resistance at 23°C (y = 0.6487~-1,089)(after Gaynard, 1979).

(d) Synergism in the longitudinal path. If effective wall resistance is derived by the procedure outlined above, the term Rt - RWL will give a total effective internal resistance to longitudinal transport between the oxygen source and the apical segment of the root. It is thus a measure of the synergism between pore-space resistance and the two “lateral” sinks : respiratory activity and subapical oxygen leakage through the root wall. If radial oxygen loss cannot be detected at the root apex then provided that there is negligible wall resistance one may conclude that the effective internal resistance has become infinite (p. 255). In these circumstances some indication of wall resistance may be obtained by raising the concentration of the oxygen source until it is possible to apply procedure 3. If the lateral leakage from subapical parts can be curtailed then Rt - RWL quantifies the synergism between the pore-space resistance (R,) and the respiratory activity of the longitudinal path. (e) Pore-space resistance. Since Rt is a corporate resistance in which lateral leakage and respiration mask the expression of R, it becomes necessary to suppress these influences in order to quantify R,.

* Provided that Cia is sufficiently high ( > 5 %), RWLwill normally approximate to the non-metabolic resistance of the wall.

AERATION IN HIGHER PLANTS

277

In wetland plants lateral leakage is naturally suppressed by the impermeability of the root wall in subapical regions; with non-wetland plants leakage can be severely curtailed by embedding the subapical regions in thick agar (p. 303). Respiratory activity can be curtailed by cooling (Armstrong, 1971a, 1971b) but for plants which still respire significantly at low temperatures an alternative method is required. Greenwood (1969) has maintained that the respiratory sites in the root will be oxygen saturated at very low gas-phase concentrations (c. 1 % or less) while Armstrong and Gaynard (1976) have shown that respiratory rate in whole roots remains constant provided that the concentration of oxygen maintained in the cortical gas-phase is in excess of 2-3 %. Assuming that these observations are generally applicable then if lateral leakage can be satisfactorily suppressed the respiratory component may be masked and the magnitude of R, obtained as follows. Where radial oxygen loss from root to electrode (1') indicates a cortical gas-phase oxygen concentration > 3 % we may write V' - Resp. (64) (Rp + RWL+ Rsh) where V is the concentration of the oxygen source in the atmosphere and Resp. is the respiratory component removed along the longitudinal diffusion path. If the concentration of the oxygen source is now raised to a new value V" we may express the new rate of apical oxygen loss I" as:

I'

==

TI"

If respiratory oxygen demand is fully satisfied by concentrations of oxygen > 3 % the magnitude of the term Resp. will be common to both equations and hence on combining with respect to Resp. and re-arranging we obtain:

(f)Effective diffusion coeficient. It follows from Section II.B.5 that the pore-space resistance (as obtained above) can be manipulated to derive an effective diffusion coefficient for longitudinal transport in the root. In appropriate circumstances this effective diffusion coefficient may then be further manipulated to determine the tortuosity of the longitudinal path. If the root is of uniform cross-section (A,) and has linear uniformity of porosity the effective diffusion coefficient will be given by the expression :

278

W. ARMSTRONG

where I is the length of the root (cm). Tortuosity may then be computed from equation 20, such that D, = DOi-ewhere E is the fractional porosity determined separately by the method of Jensen et al. (1969), T is the tortuosity factor and Do the diffusivity of oxygen in air at the temperature concerned. IV. AERATION I N THE WETLAND CONDITION A. THE WETLAND PLANT

I . Responses to Anoxia (a) Total anoxia. Although wetland plants flourish in anoxic soils there is no convincing evidence that their roots are less sensitive to anoxia (oxygen stress) during normal growth than are species which frequent unsaturated soils; indeed, the converse might be true. Vartapetian (1970) observed destructive changes in the cell organelles of excised rice roots after only four hours in anoxic culture and after seven hours the ultrastructure had become grossly impaired. This contrasted with the effects observed in the non-wetland species, pumpkin, bean and tomato where mitochondrial ultrastructure remained intact for the first 24 h of anoxia. It has since been shown (Vartapetian et al., 1976) that mitochondrial damage under anoxia can be delayed for two days in roots of freshly germinated rice if the whole seedling is kept in anoxia and similarly in excised roots if they are kept in 0.5% glucose solution. During this two day period the mitochondria develop parallel cristae and in the intact seedlings they increase in size. Changes such as these have been noted also by Morriset (1975) who described mitochondria with cristae in characteristically parallel arrangement in tomato roots after 72 h in anoxic culture. The failure of rice root mitochondria to be sustained longer than two days under anoxia contrasts markedly with the response found in the coleoptile and leaf. Intact coleoptiles kept in distilled water and excised coleoptiles in 0.5 % glucose solution remained intact even after five days in a nitrogen atmosphere and again the mitochondria enlarge and develop parallel cristae. Leaf mitochondria were persistent and contained stacks of parallel cristae after five days but were less enlarged. Although they do not offer an explanation for the changed nature of the persistent mitochondria, Vartapetian et al. suggest that the capacity of rice coleoptiles to grow under anoxia and to preserve undamaged mitochondria and other organelles is not caused by the resistance of the cell organelles to oxygen deficiency. They consider it to be due rather to the ability of the seedling to transport organic compounds easily, even under the exclusion of oxygen, from the grain to the coleoptile where they can be utilized by glycolysis. The lower resistance to anoxia in the cells of rice roots is variously explained: there is the possibility of more active anaerobic metabolism which renders the sugar supply inadequate; alternatively it might be that the early disintegration of the root mitochondria in

AERATION IN HIGHER PLANTS

279

an oxygen-free environment is caused by a failure of the root cells to develop glycolytic processes to an adequate degree; again it is possible that the root cells are less resistant to the products of anaerobic metabolism. Concerning the second of these suggestions it may be noted that at the tillering stage rice plants grown throughout in aerobic culture exhibit some glycolytic activity in the roots under anoxia (John and Greenway, 1976) but this is substantially lower than if the plants have been pretreated for several days with a supply of nitrogen in the rooting medium. The effects of total anoxia on rice germination and early seedling development have been studied by Kordan (1974, 1975, 1976a, b, c, d, 1977). Kordan has demonstrated convincingly that initial coleoptile growth and the laying down of the first adventitious root primordia can take place under conditions of complete oxygen exclusion. He has shown also that for the further development of adventitious roots a supply of molecular oxygen is essential. So too is molecular oxygen necessary for chlorophyll development (Kordan, 1976b; Kirk and Tilney-Basset, 1967) and for normal vertical shoot growth. Kordan’s observations are in line with the general premise (Vartapetian et al., 1976, and others) that the normal activities of higher plants require an external source of molecular oxygen. It is perhaps pertinent to stress at this stage that there is as yet no evidence to suggest that an oxygen requirement for root growth is not universally true. Anaerobic pathways of metabolism, the activity of which can be increased by oxygen stress, do not alone seem able to sustain growth. (b) Anaerobic metabolism. Although there seems little reason to doubt that anaerobic metabolism sustains the submerged overwintering leafless rhizomes of some marsh species, the role of anaerobic metabolism in the wetland condition is not well understood. Information is scanty and contradictory and a number of basic questions remain unanswered (Rowe and Beardsell, 1973): particularly is there uncertainty concerning the possible auto-toxicity from fermentation by-products such as ethanol. Apart from cyanide poisoning in species containing cyanogenic glycosides (Rowe and Catlin, 1971) the chemical basis of death from anoxia has not been established. It is still not clear whether the roots of non-wetland plants are any more o r less wellendowed with the potential for anaerobic metabolism than those of wetland plants; neither is it known whether they are more sensitive to anaerobic end products. It could be argued that because of aerenchyma formation wetland species might have less need for anaerobic metabolism. In anaerobic conditions a small net production of energy can still be gained in plant tissues by fermentation which yields ethanol and carbon dioxide as end products. However, ethanol is potentially phytotoxic and it could be envisaged that the prolonged waterlogging of plants might lead to its accumulation in damaging quantities. Although no clear evidence for this has emerged, such considerations have stimulated the search for less toxic by-products in

280

W. ARMSTRONG

wetland species, a search which has not gone entirely unrewarded. However, just as both wetland and non-wetland species possess the ability to respire anaerobically so are the less toxic products of anaerobic metabolism, e.g. lactate, succinate, malate, glycerol, shikimate, produced by both. Mazelis and Vennesland (1957) were of the opinion that malic acid should be considered a principal end-product of anaerobic respiration in many, if not all, plant tissues. However, malate was not found in anoxic Iris rhizomes (Boulter et al., 1963) and Effer and Ranson (1967) found that malate accumulation was associated with aerobic respiration in buckwheat seedlings and that its concentration declined under anaerobiosis. Ethanol and carbon dioxide were the major end-products in the buckwheat but significant quantities of lactate, succinate and free amino acids accumulated also. Crawford (1969) has suggested that malate forms preferentially in the roots of flood-tolerant plants, alcohol in the roots of intolerant species. The dark fixation of carbon dioxide which accompanies malate formation and the potentially less toxic nature of malic acid make the theory an attractive one and it has excited much interest. However, the supposition that ethanol is harmful to the plant is not borne out in practice and ethanol accumulation is known to occur in a number of tolerant species. Boulter et al., (1963) found considerable quantities of alcohol in the rhizomes of Iris pseudacorus but no adverse effects were noted. Similarly ethanol accumulation without ill-effects has been noted also in the flood tolerant Nyssa aquatica and Nyssa sylvatica (Hook and Brown, 1973; Hook et al., 1971). Alcohol production in rice is enhanced by pretreatment with lower oxygen concentrations (John and Greenway, 1976). There is no net yield of energy in Crawford’s proposed scheme and the theory was based on the failure to detect malic enzyme in flood-tolerant plants (NB malic enzyme catalyses the conversion of malic acid to pyruvic acid). Highly active malic enzyme has now been found in some flood tolerant species (Davies et al., 1974) including those studied by McMannon and Crawford (1971) and this must cast doubt on the validity of the proposals. However, it is still possible that malate might accumulate in some other way: it could be that malic enzymes may be inhibited in some flood-tolerant plants (Chirkova et al., 1973; Crawford, 1976). Nevertheless, until more is known one can only warn against an uncritical acceptance of the suggested alternatives to alcoholic fermentation. There is no evidence yet for any appreciable involvement of anaerobic respiration in growth activities, but there are grounds for believing that ethanol production is a self-regulating process. Ethanol may induce a quasidormant state in tissues (Rowe, 1966) and the rapid catabolism of ethanol and other end-products which follows re-aeration (Effer and Ranson, 1967; Rowe, 1966; D. V. Beardsell, personal communication) indicates that they can serve as a metabolic pool and may be non-toxic. In suitable circumstances ethanol vapour may be lost from tissues via the gas-space system or by

AERATION IN HIGHER PLANTS

28 1

diffusion into the soil (Hook et al., 1972), while the transpiration stream also can act as a carrier (Fulton and Erickson, 1964). (c) Gas-space development. An enlargement of the gas space within the plant body improves internal ventilation. It lowers the resistance to gas flow, it also reduces the potential respiratory demand per unit volume of tissue and in its natural habitat the wetland plant is characterized by a gas-space system of exceptional proportions which extends even into the aerial parts. Tissues having abnormal amounts of gas space are often loosely referred to as aerenchyma and for detailed accounts of aerenchymatous structure and formation in wetland plants the reader may refer to Arber (1920), Sifton (1945,1957) and Sculthorpe (1967). Aerenchyma formation is an obvious adaptation to the wetland condition ; it is at best only poorly developed by non-wetland species. Nevertheless, although we can readily demonstrate the superior ventilating efficiency of the wetland plant body (pp. 289-297) the chemical basis of aerenchyma formation remains obscure. The extensive gas-space development in the roots is an obvious response to conditions associated with soil anoxia and may be delayed, reduced or prevented if the soil is made aerobic (Van der Heide et al., 1963; Armstrong, 1971a, b; Das and Jat, 1977). However, for the most part it seems that the wetland plant will not normally experience anoxia even within the cells of the root meristem (p. 294) and it becomes difficult to believe that the triggering stimulus for gas-space enlargement is anoxia within the root cells. It might be that certain processes within the wetland plant require higher than normal oxygen levels to function as in the non-wetland plant: the reactions concerned with thepolysaccharide formation required to stabilize cell wall structure could be those affected perhaps (Van der Heide et al., 1963) and in rice there is evidence that planes of discontinuity in the middle lamella arise at an early stage in cell maturation (Boeke, 1940). It is interesting to note that in rice, aerenchyma fails to develop in those sectors of cortex which lie adjacent to the lateral root initials (Armstrong, 1971a, 1971b). 2. Radial Oxygen Loss and Phytoxin Immobilization (a) General. Aerobic conditions persist only in the surface of the wetland soil and one may illustrate this theoretically by substituting appropriate soil data for M and D in equation (33) (see Table II)." Where oxygen is unavailable facultative and obligate anaerobes proliferate. These organisms use oxidized mineral components or organic matter dissimilation products as respiratory electron acceptors and consequently the chemistry of the submerged soil differs considerably from that of its unsaturated counterpart (Ponnamperuma, 1972; Gambrel1 and Patrick, 1978). Nitrate, Mn(IV), Fe(III), S042-, and the dissimilation products C 0 2 and N, give Mn(II), Fe(II), H2S, methane, NH, and H,; a host of organic compounds which emanate from the further reduction of pyruvic acid and which eventually are degraded

* See p. 238.

282

W. ARMSTRONG

to methane, accumulate also. Notable examples of these include the Iower alcohols, the highly volatile and phytotoxic lower fatty acids (formic, acetic, propionic and butyric) and the plant hormone ethylene. Many of these substances are phytotoxic and the survival of plants in the wetland habitat is closely linked with an ability to transform soil-borne toxins to less harmful products (see Armstrong, 1975, 1978). Transformation is essentially oxidative in nature and species can be graded on the oxidizing abilities of their root systems. Those exhibiting the greatest oxidizing powers are the most tolerant of phytotoxins and prove not surprisingly to be the better ventilated. The sites and means of toxin transformation are various although ultimately oxygen from the ventilating system is the major electron acceptor. The Fe2+ (soluble) -+ Fe3+(“insoluble”) conversion which can be effected by molecular oxygen alone or enhanced by enzymatic activities (Yamada and Ota, 1958), is the most readily observed example of toxin transformation. Insoluble iron residues may line the intercellular gas spaces within the root itself (Armstrong and Boatman, 1967), and this has recently been most elegantly demonstrated by Green and Etherington (1977) who have shown that the iron is deposited within the cell walls also. Deposits of ferric iron around the root bear witness to the protective role of oxygen leaking radially from root to rhizosphere (Armstrong, 1967b). If internal aeration is adequate, metabolic activities within the root may destroy some potentially phytotoxic materials (e.g. the lower fatty acids, Sanderson and Armstrong, 1978), so too may the activities of the microbe populations within an oxygenated rhizosphere (Pitts et al., 1972; Yoshida and Suzuki, 1975). (b) The oxygenated rhizosphere. Although ventilating “power” affords protection in several ways the absolute quantity of oxygen available for phytotoxin immobilization must ultimately determine the total quantity which can be removed from circulation. However, it seems likely that the dimensions of the oxygenated rhizosphere which again depend upon the absolute quantities of oxygen available in the root may also play an important part in the protective process. The better ventilated the root the broader may this zone of oxygenation be and this should advantageously prolong the period in which slowly oxidizable compounds may be immobilized during their passage to the root. Some indication of how the dimensions of the oxygenated rhizosphere might vary with ventilating power may be gained by solving equation (50) for appropriate values of De, M and a. Unfortunately, the various assumptions upon which this expression is based may be a poor approximation to conditions in the rhizosphere and the results must therefore be treated with some caution. Equation (50) depends upon the establishment of a state of quasi-equilibrium in the rhizosphere. It does not embrace the initial phase of

AERATION IN HIGHER PLANTS

283

oxygenation which follows root penetration into reduced soil and during which any resident pool of reduced products would be oxidized (Teal and Kanwisher, 1961). It fails also to take proper account of the continuing diffusion of reduced substances into the rhizosphere. However, despite these drawbacks the predicted dimensions of rhizosphere oxygenation are comparable with the oxidized rhizosphere zones found in the wet soil (Armstrong, 1967b). The predicted relationship between root radius and the radius of oxygenated rhizosphere is interesting. Relatively broad zones of oxygenation are predicted for the narrower roots (Fig. 9a) and this again accords with practical observation: lateral roots (rs0-01 cm) often show zones of oxidation similar to those of the major root from which they originate. From considerations of phytotoxin exclusion alone it might be concluded therefore that in wetland conditions plants having narrower roots could be at a competitive advantage. Such an assumption would be incorrect, however, for since the narrow root can lose relatively more of its oxygen by leakage, the effective resistance to diffusion in the longitudinal path becomes correspondingly greater (p. 302). The oxygen balance must therefore become rapidly poorer with increasing length. Consequently, it is not surprising that the narrow roots of wetland species are laterals with a marked tendency to be short and borne on the basal regions of major roots where internal oxygen levels are relatively high. Lateral roots often display negative geotropism (positive aerotropism?) in wet soils. (c) Root wailpermeability and rhizosphere dynamics. Root wall permeability to oxygen in wetland species declines rapidly with distance from the apex (Armstrong, 1964, 1971b; Luxmoore et al., 1970) and oxygen leakage may cease at distances 2 2 cm from the apex. The effect that this has upon oxidation in the rhizosphere is evident from the changing pattern of iron deposits found there (Armstrong, 1967b). Where rhizosphere oxygenation is appreciable, major iron precipitates are usually remote from the root surface. With declining permeability the deposits may eventually be confined to the rhizoplane, subsequently to be re-solubilized if impermeability becomes complete. It is natural to question the adaptive significance of this declining permeability. It has been suggested that there will be a conserving effect on the oxygen available for longitudinal “flow” to the root apex (Armstrong, 1967) but since the wetland root is aerenchymatous and diffusive resistance low (p. 293), the synergistic effects of oxygen leakage in subapical regions could be relatively slight; only in longer roots might there be a sufficient accumulation of diffusive resistance for the conserving factor to gain in importance; a significant effect may be noted in Fig. 16. The impermeable root wall could be more important as a barrier to phytotoxin entry and this may be related to the dynamics of rhizosphere oxygenation. The oxygen supplying power of the wetland root apex can be simulated

284

W. ARMSTRONG

to some extent by the oxygen permeable tubular Si-rubber apex of the model roots depicted in Fig. 12. The leuco-methylene blue in dye-impregnated wet soil will rapidly oxidize around the apex of these roots, eventually forming an oxidized halo of significant proportions (r = 0.3-0.4 cm). However, in waterlogged loam we have found that the peak dimensions of the halo can be relatively short-lived : in time the oxidation boundary contracts and eventually approaches the root wall again. We suspect that this decline may be caused by a build-up of aerobic organisms in the rhizosphere immediately adjacent to the root. If the living wetland root did not quickly lose its permeability to oxygen and phytotoxins it is conceivable that such a contraction in the rhizosphere boundary might lead to a critical influx of phytotoxin. It is suggested therefore that the declining permeability of the root wall may be necessary to curtail the period over which oxygen is released to any fixed point in the soil. 3. Critical Oxygen Pressure The relatively high oxygen status attained in the roots of wetland species by internal longitudinal transport can be readily demonstrated. Apical concentrations c. 10% or greater are not uncommon in roots up to 10 cm long (Armstrong, 1967a; Gaynard, 1979, in preparation). However, concentrations of this magnitude are not in themselves sufficient justification for assuming adequate internal aeration; it is necessary also to establish the relationship between respiration and oxygen concentration in the intact plant. Similarly one needs to know the optimal levels of oxygen required to achieve root growth (p. 288) and adequately oxygenate the rhizosphere (p. 282). Unfortunately, little is known of the respiratory responses of intact plants for it has proved much more convenient to measure the oxygen uptake of excised blocks of tissue. In these circumstances oxygen uptake increases hyperbolically as the oxygen pressure is raised, until a point is reached, the critical oxygen pressure (COP, Berry and Norris, 1949), at which the respiration becomes constant. A search of the literature reveals few instances in which the COP obtained by in vitro methods lies below 0.10 atm (10% oxygen) and Luxmoore et al. (1970) have recorded values in excess of 0.2075 atm for rice root respiration. However, one must seriously doubt the significance of COP data obtained by in vitro analysis. These methods usually cause the intercellular gas-space of the sample to be flooded and this infilling of gas-space will substantially raise diffusional impedance. In these circumstances an abnormally high oxygen pressure at the boundary of the sample will be necessary to sustain its respiratory activity; consequently an abnormally high COP must be recorded. Where lower values of COP have been found there is usually evidence to show that surface moisture on the sample has been minimal (Yocum and Hackett, 1957; Forrester et al., 1966). In the intact plant the unflooded gas-

AERATION IN HIGHER PLANTS

285

space system greatly enhances oxygen diffusivity. Consequently the COP of the in vivo condition should, in theory at least, be substantially lower than that normally detected in vitro. Some of our recent experiments (Armstrong and Gaynard, 1976) confirm that the COP for root respiration in the intact plant may be nearly an order of magnitude lower than that found by in vitro analyses.

F9 I

1

Y

Fig. 12. An artificial root for simulating the radial leakage of oxygen from the living root. Glass micro-capillaries joined in series form the subapical parts of the root; the oxygenpermeable apex (hatched) is translucent silicone-rubber tube (O.D., 0.1 cm), sealed by a small bung at the free end. The length of the “root” and hence the internal resistance to diffusion may be varied by altering the length of glass capillary. After Armstrong, (1972).

The in vivo analysis which is entirely non-destructive is based on the premise that radial oxygen diffusion (I) from the intact root will be a linear function of the leaf oxygen pressure (V) provided that the internal oxygen concentration is everywhere greater than the COP. The relevant equation from Section 111 is I=

v

- Resp.

(RP + Rsh + RWL)

(64)

If a gradient of oxygen pressure exists between leaf and root apex it follows

286

W. ARMSTRONG

that if V is reduced it must bring the oxygen concentration in the root apex nearerto theCOP. As Resp., Rp, Rsh and RWLare constants in the short term then for any given root, I will fall linearly with V. If the concentration at the root apex falls below the COP the term Resp. will be no longer constant: it will decline, and hence the slope I : V will change. If Resp. declines linearly with V the I : V relationship will again be linear; if the decline is curvilinear the new relationship will be curvilinear. Whichever is the case the COP will be apparent from the inflexion point in the plot of I and V, and may be determined by substituting I for Q/t in equation (63). In practice we have found without exception that the relationship between I and V has the bilinear character indicated in Fig. 13B (ii). The linearity above the COP is as forecast above but below the COP the curvilinear pattern had been anticipated. COPS (atm) for root respiration (together with standard deviations) calculated, from the oxygen flux at the inflexion point were: Rice cv. Norin 37, 0.026 & 0.002 (5 plants), Rice cv. Norin 36, 0.024 0.001 (6 plants) and Eriophorum ungustifolium (cotton sedge), 0.02 ~-t0.004 (10 plants). An explanation for the unexpected bilinear character of the experimental plot and its inception with the abscissa was sought using the electrical root analogue described earlier (Section II.D.2). The experimental flux pattern was finally reproduced by making the following assumptions : firstly, that the COP was experienced within the apical 2 c m of root only; secondly, that respiratory rate and internal oxygen concentration for segments of intact root do not follow the hyperbolic in vitro form normally attributed to them, but adhere rather to the type of relationship outlined in Fig. 13A. Here it is assumed that the cortical cells accounting for 50 % or more of the respiratory demand exhibit a very low COP (e.g. 0.001 atm or less) because of their close contact with the gas phase. Experiments on the dark respiration of whole leaves (Forrester et al., 1966) and moist tissue slices (Yocum and Hackett, 1957) support this assumption. On the other hand the tissues accounting for the remainder of the respiratory activity in the apex, such as the central vascular core and meristem are devoid of gas-space; the assumption made was that the effective diffusion coefficient of these tissues is low enough for anaerobic centres to arise when the declining cortical oxygen concentration approaches 0.02-0.025atm. It was also assumed that below the COP the decline in respiratory activity from the increasing volume of anaerobic tissue is a linear one with oxygen concentration (Fig. 13A). This does not accord with the simplest applications of equation (45) but suggests rather that the major respiratory demand in the stelar core is peripheral in location. If these assumptions are correct we must conclude that the COP of the intact root is a property of extremely low porosity tissues. The indications are that COP for the intact porous cortical region per se is so low as to be extremely difficult to detect.

-

Internal oxygen concentrotion i % l 8 12 16 20 c

'r,

L

Apical centimetre

Second centimetre

r

U

e

.

a

a

a

a - 4

* 8 12 16 20 Leaf chamber oxygen concentration 1%)

Fig. 13. Analogue simulation of oxygen concentration vs apical flux relationship. (A) Relationships between root respiration and internal oxygen concentration used to programme the analogue in order to reproduce the experimental flux pattern. (B) Analoguepredicted relationships between leaf chamber oxygen concentration and apical oxygen flux for a cooled (0)and uncooled ( 0 )root. Programming details as follows: Distance from apex, cm 0-1 1-2 2-3 3-4 4-5 Effective porosity, % 6 9 12 16 24 Potential respiration, ng min-' cm-2 root surface 360 150 135 109 99 Potential oxygen leakage through root wall, % maximum 100 95 73 56 0 After Armstrong and Gaynard, 1976.

288

W. ARMSTRONG

The attainment of full respiratory activity at such low oxygen pressures will undoubtedly be beneficial in anoxic soils and this will be enhanced by high oxygen diffusivity and low respiratory demand in wetland species.

4. Oxygen Pressure and Root Growth Although Kordan’s findings (p. 279) and those of Amoore (1961a, b) and others (Banath and Monteith, 1966; Huck, 1970; Vartapetian et al., 1977) lend support to the premise that oxygen is an invariable requirement for root growth in both wetland and non-wetland species, nevertheless we know very little concerning the relationship between internal oxygen pressure and root growth. Those who have sought to establish the relationship between oxygen pressure and growth have done so usually by measuring the response to oxygen pressure in the solution culture bathing the roots. Understandably many conflicting records have thus accumulated for it has not been appreciated until quite recently (Greenwood, 1967a, b ; Luxmoore e f al., 1970; Healy and Armstrong, 1972) that even in non-wetland plants there may be a very significant (but undetected) internal oxygen supply. Furthermore, there has often been a failure to appreciate that the rate of stirring of culture solutions needs to be high if the concentration of oxygen at the root wall is to be equal to that in the bulk solution. Diffusion gradients between the bulk solution and the root can arise even with quite vigorous stirring (Greenwood, 1969). Consequently there is often something of an analogy between the studies in vitro of respiratory activity criticized earlier (see previous section) and those in which root growth is measured in relation to solution-culture oxygen-pressure. At the cellular level somewhat different results have been obtained and Amoore (1961a, b) has shown that in excised pea roots mitosis will proceed unchecked at an oxygen concentration of 0.5 % and above while mitotic activity is only arrested completely below 04005 %. Having found a way to control and monitor the internal oxygen pressure in roots (see previous section), we have recently carried out several trial experiments to examine the relationship between internal oxygen pressure and root growth in rice (T. J. Gaynard and W. Armstrong, unpublished). Although the work is at a very preliminary stage, parallels between the results obtained and the findings of Amoore are evident. Growth was apparently indifferent to oxygen pressures greater than the COP (c. 2.5 %), while at all concentrations below the COP root growth ceased; the results accord with the development of anaerobic centres in the root at values immediately below the COP (see p. 286). The rapid cessation of growth at very low oxygen pressures indicates an effect upon the elongation phase; the less rapid decline at higher pressures suggests that perhaps the meristem only is affected. Normal growth activities always recommenced within 4-7 h after the original oxygen supply had been

AERATION IN HIGHER PLANTS

289

restored, even after treatments in which oxygen was barely detectable at the root apex for 40 h. The time taken for recovery appeared to be related to the period and intensity of oxygen stress. Interesting parallels have been found also with the work of Vartapetian et al. (1976, 1977). When whole plants were subject to anoxia for 20 h by replacing the leaf atmosphere with nitrogen, it was found on re-aeration that both roots and shoots were dead. However, the roots of plants subjected to this treatment after introducing a 4 % glucose solution as the anaerobic rooting medium recommenced their growth within 1 h of re-aeration from the leaves; the leaf system itself appeared moribund. The obvious inference is that root viability had been maintained by anaerobic metabolism in the presence of a readily available substrate and, further, that this activity was insufficient for growth. 5. Aerenchyma and Aeration If plants are to successfully exploit the wetland condition diffusive theory (Section 11) demands that the total effective resistance to internal longitudinal oxygen transport should be minimized : pore-space resistance must be reduced together with the synergistic effects of respiration and lateral oxygen leakage (Fig. 5). Pore-space resistance in the wetland plant is reduced by the formation of aerenchyma, so too is respiratory activity. Lateral leakage from the root is restricted by the basipetal decline in root wall permeability; leakage from submerged portions of leaves and stems is often restricted by the absence of stomata. (a) The aerenchymatous root. Most roots, whether wetland or non-wetland, develop small but continuous intercellular gas-spaces in cortical parenchyma. In rice this pore space is first evident within a few cells of the root cap (see Plate I, 1, 4). As the cells mature the individual spaces as seen in T.S. enlarge (Plate I, 2) and when fully developed may occupy up to 12% or more of the total root cross-sectional area. In wetland grasses and sedges the normal intercellular space system rarely persists beyond 2-3 cm from the root apex under wetland conditions. At this point the separation and collapse of the cortical cells begins and the gas-space becomes considerably enlarged : aerenchyma is formed (Plate I1 and Fig. 14). Sometimes this collapse will be so extensive as to leave intact only those cells immediately adjacent to the endodermis and root cortex. However, the pattern of gas-space development does show genetic variation: the enlarged spaces may replace only the inner cortex in some species and it may be noted that the cell walls which persist in the aerenchymatous grass root are radially orientated (Plate 11, 2) while in sedges they are tangential (Plate 11, 1). In herbaceous species other than the grasses and sedges shizogeny without lysigeny is common: the number of cells is not reduced but rather a honeycomb structure is formed.

290

W. ARMSTRONG

tt

0/O-O-'

0

I

0

I

2

I

I

4

I

I

I

6 cm

t

I

B

7-"--

I

0

I

2

I

I

4

1

1

1

1

6 cm

Distance from apex

Fig. 14. Rice: A. Variation in position of initiation, and subsequent levels of lacuna production, in roots from waterlogged and non-waterlogged soil. B. Total porosity changes along some roots. After Armstrong (1971b).

Again, in rice, we find that the gas-space of the non-aerenchymatous apex is non-tortuous (Plate I, 4). It is not clear whether this might be a general feature of the wetland plants. Eriophorum angustifolium also displays the same regularity of packing of the cortical cells found in rice and the impression gained is that regularity of packing evident in T.S.may always coincide with the superimposition of intercellular spaces over appreciable distances. A lack of tortuosity in the apex together with the continuous development of high subapical porosity is a most effective recipe for minimizing pore space resistance and at the same time ensuring that the vital functions of the root are maintained. We may transform the pore space distribution pattern of the aerenchymatous root into the equivalent one of pore space resistance by the approprigiven X earlier (p. 249). If we transform the ate use of the expression I / D o ~ ~ A Plate 11. Gas-space characteristics of wetland plants. (1) and (2), Eriophorum angustifolium and Spurtinu x townsendii: respectively showing the lysigenous aerenchyma characteristic of Cyperaceous and Graminaceous adventitious roots (sections at 4 cm and 6 cm from root apex; magnifications X 62.5 and x 40). (3)Limonium vulgare:T.S. petiole to show the honeycomb arrangement characteristic of schizogenous aerenchyma (magnification x 95). (4) Spartina x townsendii: T.S. aerenchymatous leaf sheath showing lacunae (magnification x 65). ( 5 ) E. angustifolium: T.S. leaf-aerenchyma diaphragm showing the unusual gas-spaces within adjoining cell walls (magnification x 1300).

Plate I1

292

W. ARMSTRONG

data from Fig. 14 the tortuosity factor (7) may be omitted (cf. Plate I, 4, 5). The resulting plot of pore space resistance and root length over 15 cm where r = 0.05 cm is shown in Fig. 15(i). If aerenchyma did not develop at 1.5 cmit can be seen that by 15 cm the accumulated resistance would have been nearly treble that in the aerenchymatous root; if porosity had not risen beyond the apical figure (c = 0.065) and if the non-aerenchymatous path was tortuous (e.g. T = 0.66) the resistance at 15 cm would be approaching a figure seven times that in the aerenchymatous root (Fig. 15 (iii)). However, aerenchyma does not simply reduce the pore space resistance in the root: no matter how it forms it reduces also the respiratory demand of subapical parts; it is associated also (perhaps causally) with the declining permeability of the root wall. Luxmoore et al. (1970) have shown that in the rice root respiration declines from a peak of 360 ng ~ m s-l - ~in the apical 0-0-5 cm to 180 ng ~ m s-I - ~at 0.5-1 cm and becomes relatively constant (c. 60 ng ~ m s-l) - ~beyond 5 cm. The same pattern occurs in Eriophorum angustifolium and with the extremes of aerenchyma formation found in rice and Eriophorum it follows that respiratory demand in subapical regions could be almost entirely stelar. The wetland root as exemplified by rice and Eriophorum is thus most effectively adapted to the rigours of the submerged soil: the low pore space resistance in subapical regions in itself reduces synergism significantly; the low subapical respiratory activity and lateral leakage enhance the effect still further. Using a 10 cm rice root as a template and the electrical analogue as a model we can illustrate the magnitude of these synergistic effects and assess in detail how the various properties of the aerenchymatous root influence its aeration status : oxygen concentration profiles with different degrees of respiration and leakage, and in the presence and absence of the aerenchymatous structure, are given in Fig. 16. From these data the non-aerenchymatous root structure emerges as probably the biggest single obstacle to adequate root aeration in the wetland soil, although the respiratory and leakage characteristics of the non-aerenchymatous condition are of significance also. If both the respiration and leakage characteristics of the aerenchymatous rice root are imposed upon the nonaerenchymatous structure there is a moderate improvement in aeration (an increase in oxygen concentration of 3.3 % at a root length of 7 cm, cf. curves 5 and 8). However, with the leakage or respiratory characteristics of rice applied independently the improvement is approximately halved and hence is relatively small (cf. curves 6 and 8, 7 and 8). For the non-aerenchymatous root itself the indications are that growth would be limited to around 6.5 cm where the COP for growth is 2%. When the aerenchymatous structure is mated with non-aerenchymatous characteristics the oxygen status improves enormously (curves 2, 3 and 4). It

AERATION IN HIGHER PLANTS

Root length

293

(cm1

Fig. 15. Accumulated pore-space resistance as a function of root length: (i) wetland rice; (ii) a uniformly porous root (non-aerenchymatous, see Fig. 16), having an effective porosity equal to that in the apical centimetre of the rice root; ( 5 ) uniformly porous root (TE = 4.3 %).

is interesting to note that the rise in apical concentration which accompanies the change from non-aerenchymatous respiration and leakage to the equivalent rice characteristics (curves 1 and 4) is similar in magnitude to differences already noted (curves 5 and 8). This type of effect will increase in magnitude the longer the root becomes and will apply also to conditions 2 and 3. The characteristics used to simulate the rice place a limit on root growth of c. 30t cm. This is in keeping with the normal growth habit of the rice plant which has a relatively shallow root system. Rice relies heavily on new nodal roots developing successively with plant growth to maintain the high oxygen status and strong oxidative powers of its functional roots (Okajima,

294

W. ARMSTRONG

1964). At a length of 10 cm the rice root is losing approximately 30% of its oxygen supply to the rhizosphere (see Fig. 16), but this is concentrated in the apical region (41 % in the apical centimetre) where the internal oxygen concentration is 11.1 %. In contrast, the non-aerenchymatous root of length 7 cm loses approximately 41 % of its oxygen supply to the rhizosphere but the apical rhizosphere receives the smallest part of this (7-6%) and the internal oxygen concentration of the apex is only 1.5 %. (b) The leaf andstem. The aerenchyma found in the leaf and stem of wetland plants generally differs structurally from that in the roots. The spaces occur as discrete chambers (lacunae) in longitudinal array. The lateral walls of the lacunae are often very thin and may be imperforate but the gas-space is always continuous through the thin end walls or diphragms: unwettable perforate cellular plates of multifarious design and very low porosity. Although septa of such low porosity increase the overall resistance of the longitudinal diffiusion path and indeed counteract to some extent the low path resistance of the lacuna body nevertheless their presence is essential. In the event of injury they form a most effective safeguard against internal flooding; they may be considered as the cofferdams of the gas-space system. Consequently, it seems likely that the presence of a diaphragm will be desirable no matter how small the lacunae; diaphragm frequency may be determined more by the mechanical needs of the organ. Little but speculative comment concerning the diffusional resistance imposed by diaphragms is to be found in the literature. From experimental observations Teal and Kanwisher (1966) and Armstrong (1972) have concluded that they might not be a serious impedance to root aeration. The diffusional impedance imposed by diaphragms will be a function of their frequency, thickness and effective porosity. From visual observations, Coult (1964) has estimated “diaphragm” porosity in the Menyanthes rhizome as 0.6%; thickness is 40 pm and frequency c. 1 per 1.5 cm of aerenchyma channel. The diffusive resistance of the “diaphragm” thus approximates to 3.25 s cm-l and the resistance of the lacuna body 7.317 s cm-l. The resistance of a core of lacuna tissue 1.0 cm long and equivalent in cross-section to a In. the aerenchyroot (radius r = 0.05 cm) would be 0-00894 x lo5 s ~ m - ~ matous root this would be equivalent to an overall porosity of 69%. In Eriophorum angustifohm we find that diaphragm resistance in the leaf Fig. 16. An analogue analysis showing how the various characteristics of wetland (aerenchymatous rice root) and non-wetland (non-aerenchymatous)roots contribute to the oxygen status of the root in the wetland condition. The data were compiled on the assumption that the wetland soil, where aerated, consumes oxygen at the uniform rate of 5.27 x 10“ g ~ r n s-l, - ~ and that oxygen diffusivity in the soil was a uniform 1 x cm2 s-l. It was assumed also that wall permeability to oxygen of the rice root declined from a maximum of 100% at the apex, to zero at 5 cm from the apex; in the non-aerenchymatous root the minimum value (60%) was attained at 6 cm.

I

45 60 0

45 60 0

45 60

0

45 60 0

45 60 0

44 65 0

40 75 27.5

23 90 66

11 3 7 2 120 270 141 166

Effective porosity (*lo) ResDiration (na - - ,c1-6~ 5-l) Radial oxygen loss (ng crn-2 rn1n-l)

+Rice

IROL aplcot 16:ng

cm-* m1n-l)

C Rice slructure 8 leokoge, non-oerenchymolous resp C Rice structure

8 r e s p , non - oerenchymolous leokoge

C Rice structure. non- aerenchymolous l e o k o g e

.

I

72 120 164

72 72 72 7 2 120 120 120 120 1315 1125 9 2 5 78

72 72 120 270 6 7 5 53

Fig. 16

@

Non-aerenchymotous IROL apical 53ng

@

Respiration (ng ~ r n -s-l) ~ Radial oxygen loss (ng cm-* rnin-')

2 4 6 8 Distance from base (cm)

10

6

Non-oerenchymotous structure 8 resp rice leokoge Non-oerenchymolous structure 8 leokoge. rlce resp

Effective porosity (%)

1

0

@

B resp @

Non - oerenchymatous structure. rlce l e a k a g e 8 resp

01

Q

mon-ll

(3

296

W. ARMSTRONG

aerenchyma is somewhat greater (Gaynard, 1979). Leaf porosity (I 60 %) compares with an effective diaphragm porosity no greater than 0-4%, and the mean value recorded was c. 0.25%. The latter figure was obtained by assessing total pore space resistance in the longitudinal path (equation 66) before and after the excision of leaf segments. The number and area of the diaphragms within the segments were then determined and the pore space resistance attributable to the diaphragms estimated. Diaphragm thickness was c. 0.0058 cm, their frequency 2/cm/aerenchymatous channel and hence the resistance of 1 cm of leaf base of 0.05 cm2cross-section area and 63 % porosity - ~ diaphragm resistance accounts for 82% of this amounts to 874 s ~ r n and ~ pore space resistance offered by 1 cm of root (r = total. 874 s ~ m is- the 0.05 cm) having an overall porosity of 71 %. A further possible source of internal diffusive resistance is to be found within the root and shoot at the root/shoot junction. Coult (1964) was of the opinion that the cortical gas-space system of shoot and root became discontinuous at this point in Menyanthes. By assessing total pore space resistance prior to and after excision of the root/shoot junction in Eriophorum angustifolium, Gaynard (1979) has obtained values for the pore space resist. ance of the junction which range from 0.057-0.13 x lo5 s ~ m - ~Neither figure suggests a discontinuity in the gas-space across the junction, rather do they indicate an effective porosity in the range 0.9-2 %. (c) The functional significance of aerenchyma. In a stimulating essay under this title Williams and Barber (1961) assessed the merits of what they chose to label respectively the Transport and Reservoir Theories which seek to explain the need for aerenchyma. An acceptable theory, it was suggested, should satisfy each of four basic postulates: 1. The structure should be necessary for the successful growth of the plant in competition with others. 2. The structural provision should be adequate for the requirements of the function it is supposed to serve. 3. These requirements could not have been met with markedly greater economy by some other available means. 4. Provision should not be markedly more than is necessary to fulfil the functional requirements. The Reservoir Theory assumes that the aerenchymatous structure is required as an oxygen reservoir to tide the plant over periods when stomata1 closure or other events interrupt the gas-phase continuity of plant and atmosphere. For a variety of reasons Williams and Barber could not satisfy themselves on the validity of this argument, for different reasons neither can we: in our experiments with rice and other aerenchymatous species such as E. angustifolium (T. J. Gaynard and W. Armstrong, unpublished) the extensive lacuna systems support the respiratory needs of the plants for periods

AERATION IN HIGHER PLANTS

297

rarely exceeding 60 min. The oxygen is consumed to extinction and hence the second of the four postulates is not satisfied. The Transport Theory in summary states that “the normal intercellular space system of vascular plants is inadequate to transport oxygen at the necessary rate” in a wetland environment and “that the deficiency is redressed by the formation of aerenchyma”. Williams and Barber dismiss this hypothesis on several grounds. Their principle objections concern the gas-space system of the leaf and shoot. The volume of the lacunae is they suggest more than necessary to fulfil the transport role (a contravention of postulate 4); a small increase in the pore-size of the diaphragms, they argue, would be vastly more effective in reducing the diffusional resistance than a large increase in chamber size. This cannot be disputed, but if diaphragm pore size is determined by the need to form a barrier to flooding (p. 294) this objection is perhaps less valid. Williams and Barber concluded that aerenchyma affords a mechanical-cum-metabolic compromise in the plant and the starting point of this hypothesis is the assumption “that the oxygen flow within the plant is not easily increased but that the oxygen requirement of the submerged portions can be reduced”. “What is required . . . is a structure which for any given diameter provides the greatest possible strength with the least possible amount of tissue’’ and “this double requirement is met by the honeycomb, and by this only”. The hypothesis is an attractive one but perhaps embraces something of an unnecessary over-reaction to the principles of the Transport Theory. Of primary concern for the majority of aerenchymatous species is probably the need to maintain high levels of oxygen in the root systems to effect phytotoxin transformations in root and rhizosphere; perhaps also to support an aerobic microflora. In a previous section (Fig. 16, p. 295) we saw how important for this purpose was an aerenchymatous structure in the root: a reduction in metabolic activity to aerenchymatous levels in a non-aerenchymatous structure increased the oxygen status by a relatively small amount. It has been my contention that the scale of aerenchyma is perhaps primarily concerned with the achievement of high oxygen levels in the root system (Armstrong, 1972) and to this end high porosities are advantageous. If this is so, the aerenchymatous provision will not necessarily contravene Williams and Barber’s fourth postulate and we may regard the Transport and Mechanical-cum-metabolic Theories as essentially complementary. 6. Photosynthesis and Aeration As early as 1940, Laing reported increases in the oxygen pressure in the stems, roots and leaves of Nuphar advenum Ait., Petrandra virginica (L.) Kunth, Typha latifolia L., Sparganium eurycarpum Engelm and Scirpus validus (Vahl), during periods of illuminat ion. In Menyanthes trifoliata L. Coult and Vallance (1 958) recorded light/dark fluctuations in oxygen pressure

298

W. ARMSTRONG

of c. 0.048 atm in the stem cortex and 0.033 atm in the root. In all of these cases it appears that the photosynthetic activity responsible for oxygen enrichment of the plant atmosphere took place in organs which were either submerged or astomatal or both. The oxygen build-up during illumination can thus be attributed to diffusional impedances preventing the rapid release of oxygen to the external atmosphere and this accords with the theory outlined in Section II.C.2. In Eriophorum angustifolium we find that the extent to which photosynthesis enhances root aeration depends upon two factors: the degree of immersion of the leafy parts and the availability of free carbon dioxide at the submerged leaf surface. The bicarbonate ion is an ineffective carbon source where cuticular resistance is high. The sheathing leaf bases in Eriophorum are largely astomatal and non-photosynthetic and in unsubmerged plants or those submerged to the top of the outermost leaf sheath photosynthesis has an insignificant effect on root aeration even at light intensities of 100pE m-2 s-l. If submergence is extended to include the photosynthetic parts of the leaf system the oxygen pressure in the root system rises as a function of light intensity, carbon dioxide concentration and degree of immersion. At a solution concentration of 0.8 mM (CO,) I-' and a light flux of 100 p E m-2 s-l the oxygen concentrations in the root apices of fully submerged plants can rise to 150% of the unsubmerged condition; at 3 submergence the value falls to c. 140%; at 3 submergence a figure of 110% has been recorded; at 300 p E m-, s-l the latter value rose to 120 % of the unsubmerged condition. In the dark, oxygen levels in the root are depressed by all degrees of immersion which cover the stomata1 surfaces of the leaf: the length of the internal diffusion path is extended and the total effective resistance is thus increased. Consequently, the advantages accruing from submergence in daylight could be countered during the subsequent dark period. One may foresee circumstances where partial submergence in darkness could result in a lowering of the oxygen tension below the COP in more remote parts. Alternatively, the process of rhizosphere amelioration which could be enhanced by photosynthetic activity might be critically imbalanced by darkness. There are indications of the latter in the reported sulphide damage to rice during periods of sunless weather or deliberate shading (Vamos and Koves, 1972). It is possible that a thorough evaluation of the photosynthetic enhancement of root aeration in rice would indicate the need for reduced levels of submergence during periods of sunless weather. B. THE NON-WETLAND PLANT

The vast majority of higher plants are confined to well-aerated soils. In general they respond unfavourably to sudden soil waterlogging, are extremely shallow rooting and non-competitive in wet soils and grow relatively poorly in solution cultures which lack forced aeration. Until quite recently little

AERATION IN HIGHER PLANTS

299

consideration had been given to the likelihood of internal longitudinal oxygen transport in these species; it was not generally appreciated that the small cortical intercellular spaces of the roots might form a gas-filled continuum with the atmosphere. Only in the late 1950s and early 1960s was it conclusively demonstrated that oxygen could travel by gaseous diffusion through the cortical intercellular space system of the non-wetland root. Since then it has become increasingly apparent that internal oxygen transport is just as normal a feature of non-wetland species as it is of wetland plants. Where the two groups differ is in the degree of aeration afforded: root porosities are much lower and overall respiratory demand higher in the nonwetland root; there have also been suggestions that tortuosity might add significantly to the pore-space resistance (Jensen et al., 1967). Neither the respiratory demand nor the root-wall permeability in the non-wetland root show the same marked basipetal decline met with in the wetland plant (Luxmoore et al., 1970). The critical oxygen pressures for respiration and root growth in non-wetland roots are known with even less certainty than for wetland species. Huck (1970) has noted that whereas tap-root elongation in soybeans and cotton ceases abruptly if the oxygen is purged from the soil gas-space, oxygen levels of 2-5 % resulted only in a temporary reduction in growth. Recent experiments of the kind performed on rice and cotton grass (p. 288) (Webb, 1978-unpublished), indicate that primary root elongation in pea can continue below the respiratory COP and even for some time after the oxygen pressure in the root apex has fallen so low as to become immeasurable. Roots have extended for up to 80 hours and by 7 mm under these circumstances, with the growth rate showing a progressive, if somewhat stepwise, decline before growth finally halted. However, several hours after growth ceased, re-aeration of the apex via the internal path has brought an almost immediate return to the initial growth rate. This could indicate that root growth had continued until some critical distance separated the apex from the still-oxygenated parts of the root; a critical distance perhaps limiting the supply of oxidizable substrates and alternative electron acceptors, or limiting the removal of metabolic by-products. That oxygen was still present but undetected because of high wall resistance is also a possibility which cannot be precluded at this stage. At the cellular level the data of Vartapetian et al. (1977) and Morriset (1975), suggest that total anoxia may be less immediately damaging to the non-wetland root. On the other hand Huck reports that periods of soil anoxia exceeding 30 min resulted in a killing of tap-roots in cotton and soybean. Five hours of anoxia was sufficient to cause the tip-death of all tap-roots. 1. Adaptability Non-wetland species show some improvement in their ventilating character-

300

W. ARMSTRONG

istics in waterlogged soils. Nevertheless, the plasticity of response found in the wetland plant is lacking. Yu et al. (1969) grew severalnon-wetland cropspecies under a whole range of soil treatments which included full flooding, halfflooded and drained. Root porosities were always lower in the drained treatments and ranged from 3.5% (barley) to 7.5-11.5% in corn. From the diffusional point of view it is not surprising that such roots penetrated but a short distance below the water table of the half-flooded treatment. With the exception of barley most plants responded to full flooding by producing fresh roots of higher porosity. The porosity in corn rose to 15-18 % and the roots penetrated up to 17 cm. In sunflower the porosity rose from 6 % to c. 11 % and some roots penetrated the wet soil to a depth of 15 cm. Porosity in barley varied little but the roots penetrated to 12 cm; root porosity in “Pato” wheat increased from 6 to nearly 15% but penetration was limited to about 5 cm. Yu et al. interpret their data in terms of effective root ventilation; the apparent exceptions of barley and wheat they suggest were due respectively to exceptionally low and unusually high root respiration in these species. 2. Analogue Data Electrical analogue studies of aeration in the non-wetland root type yield results which accord well with the observations of Yu et al. and the supposition that root penetration into wet soil reflects the sufficiency of internal aeration. However, the analogue approach makes it abundantly clear that root radius and the “sink” activity of the wet soil may substantially influence the root’s oxygen status and potential for growth; it seems more than likely that differences in root radius will have contributed to the interspecific differences in growth recorded by Yu et al. Analogue data presented in Fig. 17 (A-H) show how the internal oxygen supply to the apex of non-wetland roots ought to vary with length, radius, effective porosity, root respiration and soil oxygen demand. For convenience it has been assumed that the roots are devoid of laterals and are of constant radius (r = 0.05 cm, By C , F and G; r = 0.01 cm, D and H). It has been assumed also that oxygen entry is at the root base; that root respiration and diffusivity ( D p ) remain constant with distance from the root apex; and, (A and E excepted), that root wall permeability declines from a maximum (100%) at the apex to a minimum (60%)at6cmfrom theapex.For(A)and(E) the root wall permeability was set at zero to give zero soil sink activity. Root respiration is programmed at two levels: 120 ng cm-S s-l (A-D) is a moderately high rate, 30 ng cm-3 s-l (E-H) is relatively low. Soil oxygen demand is represented at three levels: zero (A and E); 4 x cm3 ~ m s-l- ~ (B and F); and 4 x cm3 ~ m s-1- ~ (C, D, G and H); the oxygen demand of 4 x cm3 ~ m s-l - ~is a high level of activity. Since the effective root porosity for non-wetland plants appears to lie within the range 1-5-15%,

20

10

0 20

10

0 20

10

0 20

0'

2

L

6

8

10

0

2

L

6

Root length

8

10

12

1L

16

18 20 22

lcrnl

Fig. 17. Internal apical oxygen concentration in roots as a function of root length, root respiration, effective porosity, root radius and soil oxygen sink activity. Analogue data obtained by assuming (i) uniform respiratory activity in the roots, 120 ng ~ m s-'- (A, ~ B, C, D) and 30 ng ~ m s-'- (E, ~ F, G, H); (ii) uniform effective porosity throughout the roots: from left to right in each figure 1.5 %, 3%, 7%, 15 %; (iii) uniform oxygen consumption in aerated rhizosphere soil: B and F, 5.27 x lo-" g cm-s s-l; C, D, G and H, 5.27 x 10" g ~ m s-l, - ~but no oxygen leakage from root to soil in A and E ; (iv) root wall permeability to decline from 100% at the apex to a minimum of 60% at 6cm and beyond; (v) root growth ceases at an internal oxygen concentration of 2%. In B, C, F and G root radius (r) is 0.05 cm; in D and H, r = 0.01 cm; the data in A and E are independent of root radius.

302

W. ARMSTRONG

these and two intermediate values (3 % and 7 %) have been programmed for each set of conditions. Perhaps the most interesting point to emerge from these data is the very considerable influence exerted by the soil sink when root radius is low. Conversely, the respiratory activity of the root is of only minor importance under these circumstances; the maximum root length predicted in (H) is only 5.5 cm (r = 0.01 cm; E = 0.15; soil respiration 4 x cm3 ~ m s-l), - ~and yet a four-fold increase in root respiration, (D), reduces this figure by only 0.3 cm. However if oxygen leakage to the soil is reduced to zero (as in E), the root can in theory attain a length of c. 22 cm before the hypothetical COP (2 %) is reached. It is also interesting to note that an increasing effectiveness of the soil sink becomes apparent in a concavity of the appropriate curve (cf. E and G ) . At the higher root radius (r = 0.05 cm) soil sink activity has considerably less influence on the internal oxygen regime (cf. A, B, C and E, F, G ) ;root respiration exerts a substantial effect which is approximately equalled by the effective root porosity. Where there is no lateral leakage of oxygen to the soil, (e.g. A and E), the internal oxygen regime becomes a function of porosity and root respiration only and the internal oxygen status becomes independent of radius. Under these circumstances the maximum attainable root length supported by initial aeration is predicted here as 22 cm (e = 0.15 in E). This is several centimetres longer than the deepest recorded root penetration in the studies of Yu et al. and for non-wetland herbaceous species is possibly near the attainable limits of root penetration into wet soil; the lower limit is probably less than 2 cm (e.g. D, where E = 0.015 and r = 0.01 cm). If rhizosphere oxygenation and phytotoxin immobilization are taken into account it will be obvious that many of the predictions concerning attainable root length (B, C, D, F, G and H) may in practice never be realized. In the majority of cases the apical oxygen status declines rapidly with increasing root length and so too will the protection afforded by ROL rapidly diminish.

3. Oxygen Transport in Pea: An Experimental Study In a recently completed study of oxygen transport in pea roots Healy (1975) has provided what appears to be the first experimental record of the changes in diffusional resistance which accompany root elongation in a nonwetland plant; the plants ranged in age from 1-55 days. Cylindrical Pt electrodes were used to monitor the oxygen flux from the apices of the primary roots and the diffusional resistances in the longitudinal path were calculated as described in Section 111. Pore space resistance and the synergistic effects of respiratory activity and leakage were quantified as were the effects of secondary root production, and these results are summarized in Fig. 18 (curves a-e). Changes in the total effective resistance which could be attributed to a

AERATION IN HIGHER PLANTS

303

decline in respiratory activity were noted and plumule resistance was demonstrated by submergence experiments. No evidence was found of significant stomata1 resistance to longitudinal oxygen transport. The peas were grown for the most part in sterile 1 % agar medium in 250 ml glass cylinders and the plants were removed for assay by extracting the agar core intact. To minimize the oxygen leakage from subapical regions of the roots, only the agar enclosing the primary apex was trimmed away before assay; plant and agar jacket were immersed in anaerobic liquid medium to the cotyledon junction before moving the Pt electrode to ensleeve the primary apex (Section 111). The effects of subapical oxygen leakage were studied by trimming away the whole of the agar jacket; the influence of lateral (secondary) roots on the oxygen status of the primary root was determined by assaying the oxygen flux from the primary apex before and after excision of the laterals. The relationship found between pore-space resistance and root length (Fig. 18a) illustrates one aspect of root aeration not stressed previously. During the early stages of root elongation (3-5 cm-8.5 cm) there is no observable gain in pore-space resistance. Upon further investigation it was found that this phenomenon could be attributed to the changing shape of the developing root. The final root shape is that of an elongated inverted cone and this may be of considerable benefit to the non-aerenchymatous uniformly porous root: as basal respiration declines with age the effective resistance of the base could decline very substantially, more so than in a root of uniform diameter. The oxygen status in the root apex would be enhanced accordingly. The pore-space resistance in the pea cannot be entirely accounted for in terms of root shape and mean porosity (3.8%) alone and it is tempting to suggest that the diffusion path may be a tortuous one. However, although longitudinal sections of the roots do not show the same continuity of channels evident in rice (cf. Plates I11 and I), the impression gained is that gas-space tortuosity might be relatively slight even in this species. Unfortunately, the data in Fig. 18a are inclusive of apical wall resistance; until we can establish the magnitude of this term it is not possible with any certainty to quantify the tortuosity. Healy found that apical oxygen flux from the primary pea root always declined with increasing root length provided that the plants were no more than 10 days old (1 I10 cm approx.). In jacketed roots with excised laterals the oxygen flux showed a smooth curvilinear decrease with length. The corresponding increase in effective diffusional resistance (8.1 x lo5 s ~ m at- ~ ~ cm; Fig. 18b) is attributable to the synergism 3 cm to 27.2 x lo5 s ~ m at- 10 between primary root respiration and pore-space resistance. The retention of secondary roots was associated with a bi-modal flux pattern. The initial decrease in flux was as before but during lateral emergence there followed a

+

OD

Te

a I

I

2

4

6

8 10 12 Length of primary root (cm 1

14

16

-

Fig. 18. Oxygen transport in the primary root of Pea; collective figure showing the changes in total effective diffusive resistance found during primary root elongation under the circumstances indicated. Compiled from Healy (1975) and Armstrong and Healy (unpublished).

AERATION IN HIGHER PLANTS

305

steep decline in the apical oxygen status: effective diffusional resistance rose ~ cm, immediately prior to emergence, to a from c. 14.1 x lo5 s ~ m at- 5.5 ~ cm (Fig. 18d). When leakage is permitted value of 294 x lo5 s ~ m at- 10 the role of the lateral roots becomes even more pronounced: the apical oxygen status of the primary became immeasurably small at c. 8 c m and hence the effective diffusional resistance approaches infinity (Fig. 18e). This is the first record of the possible influence of secondary roots on the internal ventilation of a major root and the effect can be seen to be substantial indeed. In periodically degassed liquid medium pea roots cease growth abruptly at 8.5-9 cm. This accords closely with the length at which the roots have accumulated infinite diffusional resistance. Consequently we have suggested that the initial 8-9 cm of root growth in pea will be sustained by internal ventilation provided that there is some impedance to radial oxygen loss (Healy and Armstrong, 1972) (provided also that the level of the culture solution does not come above the root-shoot junction). Static oxygen-free culture solution seems to fulfil this requirement. If growth is to proceed beyond 9 cm a more effective “jacketing” appears to be essential. Agar jelly (1 %) can fulfil this role but for prolonged growth it must eventually become necessary for the rooting medium to receive forced aeration. In saturated soils it seems unlikely that pea could initially attain a root length of 8-9 cm because of soil oxygen demand and it is interesting to note that in continuously degassed medium growth ceases at 4-5cm (Healy, 1975). However, the final root length attained might depend also upon the ageing effects in the root system: in roots of 11 days and older Healy noted an increase in the apical oxygen concentration (Fig. 19). In periodically gassed culture solution this was accompanied by a resumption of growth for a short period. C. TREES

It is perhaps the economics of tree production and the deleterious effects of waterlogging which have done most to stimulate an interest in tree aeration and several reviews concerned chiefly with this topic have appeared recently (Rowe and Beardsell, 1973; Hook et al., 1972; Gill, 1970; Coutts and Armstrong, 1976). The following is intended only as a supplement to these reviews. It seems certain that much of what has been said already concerning the ventilation of herbaceous species applies equally well to woody plants. There are, however, characteristics peculiar to woody species which can create special problems for the ventilation of submerged parts and which reduce the competitiveness of tree species in wet soils. These features include: the development of secondary tissues with ensheathing and perhaps non-porous cambia; the loss of primary cortical tissues and their replacement by secondary cortex which with time occupies less and less of the total cross-section of

Plate 111

AERATION IN HIGHER PLANTS

307

the plant body; and last but not least the more massive dimensions of the woody plant body itself. The part played by anaerobic metabolism in tree roots is very uncertain (Rowe and Beardsell, 1973) and there are as yet no firm grounds for believing that internal longitudinal oxygen transport is not just as necessary a requirement in submerged tree roots as it so obviously is in non-woody species. The long distances over which oxygen might have to travel in trees when the roots or even the basal regions of the trunk have been inundated has led to speculation that internal oxygen transport may be insufficient to maintain viability in remote organs (Crawford, 1976). This may well be so (see below) but in such circumstances anaerobic metabolism is apparently equally insufficient to sustain the root system. Gill (1970) observes that few, if any, temperate species can survive an indefinite period of partial inundation and it is interesting to note (D. D. Hook, personal communication) that even the roots of the Swamp Cypress may die back to near the trunk if the mature tree is submerged by flood-waters to a depth of 100-150 cm. Root activity recommenced in Taxodium by the production of new lateral roots which had their origin close to the trunk. Secondary and adventitious root development in trees is a common response to soil flooding and there is every reason to suppose that these roots are aerated internally. Hook et al. (1970, 1971) found that swamp tupelo and water tupelo developed new lateral roots when inundated and these new roots oxidized their rhizosphere under anaerobic conditions, whereas the initial roots failed to do so. The newly formed roots differed from the initial ones in anatomy, rates of anaerobic respiration, and their ability to tolerate high concentrations of carbon dioxide in the flooded soil. Adventitious tree roots formed in response to flooding can be strongly aerenchymatous in the primary cortex (personal observation) ; they emerged from trunk lenticels which are themselves grossly hypertrophied with loose aerenchymatous tissue (Gill, 1970). Similar roots produced by cuttings receive oxygen by internal longitudinal transport (Armstrong, 1968) and will oxidize reduced media (Leyton and Rousseau, 1957). Extensive tracts of fused hypertrophied root and trunk lenticels within the soil and immediately above it are a common feature in Lodgepole Pine, one of the more wet-tolerant members of Pinaceae. The lenticel material is exceptionally hydrophobic and must ensure Plate 111. Pea: primary root apices: transverse and longitudinal appearance of cortical gas-space-plants grown in a nutrient 1 % agar. (1)-(3), transverse sections at 30pm, 90pm and 4 5 mm from the root/root cap junction (magnification x 1175, x 92.5, and 62.5); gas-spaces already visible in the differentiating cortex in (1). (4) and (S), radial longitudinal sections at extreme apex (magnification x 823), and at 0.2 cm from the apex (magnification x 67.5). Non-tortuous spaces extending for several cell lengths may be seen in both sections. The impression gained is that gas-space tortuosity is relatively slight in these roots.

308

W. ARMSTRONG

.

-:: -

e

!$50;1,* 0 D

:**

8 .

U

B

8

- *:I, *.*.* 0

I

*

"

e

0

ij:, ;:

e a

* .*

*

e

,

, ,

,

, ,

,

,

,

,

,

,

,

,

,

. .

,

Age of plant (days)

Fig. 19. Oxygen transport in the primary root of Pea: oxygen flux from the root apex as a function of plant age at 23°C. Roots jacketed in agar and laterals intact. After Healy (1975).

the gas-phase continuity between plant and soil atmospheres whenever the level of free water allows. Gas-phase continuity between root lenticels and gas-spaces in the root apices has been demonstrated by Coutts and Phillipson (1978a). There seems little doubt that the major sites of oxygen entry to the submerged roots of woody plants are the lenticels. Both Hook et al. (1971) and Armstrong (1968) have shown that obstruction of lenticels halts rhizosphere oxidizing activity and radial oxygen leakage from the roots. The unsubmerged lenticels lying closest to the water table will normally be those most closely involved with the aeration of the root system (Section II.C.2); it seems most unlikely that leaves will directly influence the aeration process unless they abut the water table or are submerged beneath it. The diffusion path from lenticel to root apex is still somewhat uncertain (Hook et al., 1972; Coutts and Armstrong, 1976): it probably very much depends upon species and age. Submergence experiments (Armstrong, 1968) suggest that the adventitious roots of cuttings may rely chiefly on transport in the secondary stem cortex but there are indications that in relatively wettolerant species gas-filled elements of the xylem might form a major route in the mature plant. Specialized xylem aerenchyma of secondary origin occurs infrequently (Arber, 1920) but extensive zones of unspecialized but gas-filled elements can form within the maturing secondary xylem of both Angiosperms and Gymnosperms.

AERATION IN HIGHER PLANTS

309

Oxygen entering the xylem must cross the cambium and recent research shows that adequate gas-space continuity across the cambium may be a feature which characterizes only the more flood-tolerant of woody species (Hook and Brown, 1972). Effective porosities will normally be exceedingly low and longitudinal diffusive resistance correspondingly high if an impervious cambium and water-filled xylem confine longitudinal oxygen movements to the narrow secondary cortex of the woody stem and root. It has recently been suggested that the deeper penetration of waterlogged soil by Lodgepole Pine (LP) is due to internal oxygen transport in the stele; Coutts and Phillipson (1978b) found that the actively growing roots of LP would penetrate the water table to a depth of 20 cm at 10°C whereas Sitka Spruce ( S S ) made only shallow growth; they also found (1978~)that LP roots effect a greater degree of rhizosphere oxidation. Gas-filled elements are a more characteristic feature of the LP and gas-filled cavities in the secondary tissues of the pericycle were evident in those pine roots which penetrated the water table. These cavities which were absent from the spruce connected ultimately with the lenticels above the water table. The roots of actively growing cuttings of LP and S S also react to soil flooding in a manner which suggests that aeration in the two species differs and is a major factor determining the responses (Coutts and Phillipson, 1978a, b) : the responses varied with the period of waterlogging, temperature and depth below the water table (Fig. 20). Waterlogging for a period of seven days at 15°C killed all of the root apices in S S but 60 % of the pine root tips survived. When the flood period was extended to 28 days only 38% of the pine root tips survived but dieback extended a relatively short distance from the apex. In spruce the only root tissues to remain alive were within 3 cm of the water table. A lower temperature (6°C) reduced dieback in both species but the greater tolerance of LP was still evident. Root regeneration from solution culture anoxia and phytotoxin treatments is also consistent with differences in the internal aeration of the species (Sanderson and Armstrong, 1978) and oxygen diffusion studies at low temperature confirm that LP is the better aerated (Sanderson, 1977). It is interesting to note however that the root apices of Taxodium distichum which produced primary root aerenchyma - ~ acid treatment in which the apices of LP are can survive a 100 pg ~ m acetic killed; the roots of the strongly aerenchymatous herbaceous species E. ungustifolium will even continue to grow under these circumstances but the response in pea parallels that of the SS and regeneration is by adventitious root production. Sanderson and Armstrong (1978) have summarized these observations as part of a more generalized scheme outlining the possible relationship between ventilating power and response to soil waterlogging (see Fig. 21). Wall permeability in the tree root decreases markedly as secondary tissues develop (Sanderson, 1977) and in large organs such as mature tree roots

Fig. 20. Sitka Spruce (SS)and Lodgepole Pine (LP) after flooding of the soil: the dieback and survival of roots at different depths below the water table. The data are grouped into four depth classes along the horizontal axis; class I represents the mean of all roots extending 0-100 mm below the water table, class 11, 101-200 mm, class 111, 201-300 mm; and class IV, 301-400mm. Four treatments (a-d) are represented. (a) Growing roots, 15"C, 7 days; (b) growing roots, 15"C, 28 days; (c) growing roots, 6"C, 28 days; (d) dormant roots, 6"C, 28 days. The left-hand column of each pair is Sitka spruce (solid columns), the right-hand, Lodgepole Pine (dashed columns). The stippled portion represents the mean extent of dieback; the unstippled portion, the length of root surviving. The vertical bars are the standard errors of the mean length of dieback, where they are of sufficient size to be represented. After Coutts and Phillipson (1978a).

AERATION IN HIGHER PLANTS

31 1

lateral leakage to the soil should have little effect upon the aeration process (p. 301). Porosity, tortuosity and respiratory activity should be of major importance. The greater penetration of anaerobic soil by LP at 10°C than at 20°C (Coutts and Phillipson, 1978b) is consistent with a reduced respiratory demand and enhanced oxygen transport a t the lower temperature. By substitution in equation (34) it is possible to make some tentative predictions concerning the aerated path length in woody organs. If the gas-filled parts of the xylem contribute to the internal diffusive path (Coutts and Armstrong, 1976) effective porosities might rise to 60 % or greater in mature trees and if so, the overall respiratory demand will be relatively low, perhaps 10 ng ~ m - ~ s-l or even less. In these circumstances we find predicted an aeration path of 250 cm or greater. This is hardly adequate for the aeration of laterally spread root systems if the roots are completely inundated, but it might in appropriate circumstances make possible a sufficient anchorage of trees in wet soils. To maintain tree stability it is necessary for the roots to penetrate deeply. If gas-filled xylem abounds, substantial sinker root development might be possible from beneath the bowl or, (water table permitting), from the large laterals of the primary root system. The increasing proportion of gas-filled elements in the xylem of maturing LP could be responsible in part for its deeper rooting on wet sites when at the pole stage and beyond. Conversely, the serious losses from windthrow where Sitka Spruce has been used for afforesting wet sites (Fraser and Gardiner, 1967) may be attributed to, among other things, the relatively small proportion of gas-filled elements in the xylem. It is interesting to note that when sinker roots develop in wet soils they can be strongly carrot shaped (see p. 303). Dormancy may help submerged roots to survive a period of flooding (Coutts and Philipson, 1978b). However, it would seem that to survive frequent inundation during non-dormant periods the laterally spread root system will require an internal oxygen path which by-passes the stem. In the mangrove Avicenniu nitidu snorkel-like lateral roots are produced (Scholander et ul., 1955). A single tree may produce several thousand of these; 20-30 cm high, 1 cm thick, soft and spongy, studded with numerous lenticels, they project from the mud and aerate the submerged radially spreading main roots. The “knees” of the Swamp Cypress may function similarly although Kramer et al. (1952) have cast some doubt on this. In a section dealing with tree aeration it would be inappropriate if something was not said concerning internal aeration and mycorrhizal roots since these are of such particular importance in forest tree nutrition. The mycorrhizal fungi are strongly aerobic organisms and in ectotrophic associations can account for >50 % of total respiratory demand. The bulk of the ectotrophic fungus lies outside the root and in aerated soils is undoubtedly oxygenated from the soil; indeed the mycorrhizal rootlets are a feature of the more aerobic soil horizons. Those hyphae which enter the root occupy the inter-

312

W. ARMSTRONG

Fig. 21. Suggested relationship between internal root ventilation and the responses of wetland and non-wetland plants to a limited period of soil waterlogging. Root growth before and during, and regrowth after the end of, the waterlogged period is indicated as follows: before (and surviving) dotted; before but killed by the waterlogging bIack; during shaded; regrowth white. The figure also records how Sitka Spruce and Lodgepole Pine respond when exposed for a limited period to anoxia or to acetic acid (100 ppm) in solution culture. NB. Mature woody roots may at times substitute for the stem in the above scheme. After Sanderson and Armstrong (1978).

cellular spaces of the outer cortex and form what is called the Hartig “net”. This must reduce oxygen diffusivity in the rootlet and although there is no direct supporting evidence we must suppose that it will adversely affect aeration of both root and fungus under conditions of submergence. Likewise, endotrophic mycorrhizal fungi must hinder the internal ventilation process in anoxic soils. Internal longitudinal oxygen transport can sustain mycorrhizal fungi in anoxic agar culture (Read and Armstrong, 1972) but it is perhaps unfortunate that the only experiments so far performed were with conifer seedlings where the internal diffusion path was short. In the same series of experiments it was found possible for the first time to induce ectotrophic “mantle” formation artificially on internally aerated Si-roots. These were introduced into an agar medium containing fungal macerate and the other essential growth factors. It is my opinion that this result could be open to

AERATION IN HIGHER PLANTS

313

misinterpretation : it most probably indicates that mantle formation is a phenomenon requiring a point (radial) source of one of the essential growth factors upon which the other essential growth factors can impinge. In the field situation the roles will be reversed, the carbohydrates and vitamins will form a point source diffusing from the rootlet while the oxygen normally in surplus will diffuse in radially from the soil.

V. ROOT AERATION IN THE UNSATURATED SOIL Whilst we now recognize that internal ventilation is a not insignificant property of mesophytic species, equally well do we recognize it as generally insufficient to sustain the activities of the extensive root systems found in unsaturated soils : in these circumstances its role is essentially supplementary to the radial movements of gases which can take place. However, despite the gas-space continuity of soil and aerial environment and the maintenance of appreciable oxygen concentrations within the soil atmosphere, root aeration in the unsaturated soil is not always adequate: critical configurations of soil structure, water distribution and oxygen demand can combine to cause oxygen stress in both root and soil (see below). The supplementary role of internal transport can be analysed only by resort to relatively complex mathematical or analogue techniques (see (3) below). However, with a knowledge of relevant soil and root diffusion characteristics it is possible to utilize relatively simple diffusion equations to assess the sufficiency of root aeration by the soil path.

1. The Soil Path The effectiveness of the soil path is determined by numerous factors: its structural characteristics, the distribution of water, the distribution and respiratory activities of the microorganisms and the distribution, internal diffusivity, diameter and oxygen requirements of the roots themselves. So far as aeration is concerned two primary structural soil types may be recognized : where sand predominates and the pore space follows a normal distribution the soils may be regarded as homogeneous; more generally, however, the soils contain some clay, their primary particles aggregate into distinct units and the pore-space distribution is strongly heterogeneous. In these soils distinct zones of relatively fine crumb pores are separated by a more continuous system of larger intercrumb pores (Currie, 1961a). Since water offers a considerable resistance to diffusive gas exchange, soil structure affects root aeration chiefly by its influence on water distribution. In the homogeneous soil gas-space continuity throughout the pore space is established even at low suctions (Currie, 1961b): 80% or more of the pore space system of sand will drain at 10-20 mb suction and a single plateau plot of suclion versus saturation is obtained. Consequently, diffusion to

314

W. ARMSTRONG

depth in these soils will usually occur freely throughout the pore system with roots and microorganisms always closely adjacent to the soil atmosphere. The final stage of the diffusion path will be a relatively uncomplicated function of the water-film thickness around the root and the effective overall soil porosity within the water film. The microorganism respiratory component might be small enough to be ignored and on this assumption Kristensen and Lemon (1962) modified and combined the two diffusion equations (27) and (47) to describe root aeration by the soil path and to predict the limiting thickness of water film at which the root just remains wholly aerobic. The equation is: log-a b

=

~

D, --2DeC, 2Di Ma2

where a is the root radius (cm); b is the critical radial distance from the centre of the root to the airwater interphase between the bounding water film and the soil atmosphere (cm) at which the root remains just wholly aerobic; D, is the effective diffusion coefficient for oxygen within the liquid film ( D O ~ ~ ~ (cmz OTE s-1) ) (NB for thin films the tortuosity factor will be effectively zero) ; C,is the equilibrium oxygen concentration in the water film where it adjoins the soil atmosphere (g ~ m - ~ ) ; Di is the overall effective radial oxygen diffusion coefficient of the root (cm2 s-l); and M is the rate of root oxygen consumption (g ~ m s-l). - ~ The equation is a useful one if a true value can be assigned to Di: it would cm2 s-l) now seem that the value chosen by Kristensen and Lemon (8 x might be substantially in error except at low temperatures and Greenwood has suggested a figure of 1.2 x cm2 s-l. A more conservative estimate, 7x cm2 s-l, is calculable on the basis of the COP data given earlier, assuming an effective root wall thickness of 35 pm. Choosing this figure and considering a root of radius 0.035 cm, then if g~ m - M ~ , = 2 x lo-’ g ~ m s-l- and ~ D, is 1 x cm2 s-l Co is 8 x (i.e. Doc where E is 0.44:Kristensen and Lemon, 1962) we obtain a value of 0.0276 cm for the “critical” path length (water film thickness: b - a). This contrasts with the figure of 0.001 cm estimated by Kristensen and Lemon. Further estimates for different values of a, M and C, are shown in Table I11 together with values for the critical root radii under these conditions calculated from equation (68). Clearly, root radius is of major importance in aeration by the soil path and in contrast with the wetland condition the narrower the root the less likely is it to be made anoxic. However if we extrapolate from the data of Kemper and Rollins (1966) we find that water film thickness rarely exceeds 6 x lo-* cm at suctions greater than 20 mb. Consequently,

AERATION IN HIGHER PLANTS

315

unless waterlogging occurs or microorganism activity within the water films is found to be too great to be neglected, we can, for homogeneous soils, predict adequate aeration in all roots narrower than the critical radius. The intercrumb pores of the heterogeneous soil occupy up to 60% of the pore space system and, like the homogeneous soil, drain at extremely low water sections, < -10 mb (Currie, 1961b). Accordingly, the atmosphere within the intercrumb pore space is characterized by high concentrations of oxygen : Currie has suggested that intercrumb concentrations < 15% O2 will be a rarity. Conversely the fine crumb capillaries drain much less easily and, at suctions less negative than field capacity, the crumbs can remain water-filled. At field capacity (pF 2-0) the crumbs will begin to empty, albeit slowly, and the bimodal nature of the system is confirmed by a double-plateau plot of suction and % soil saturation (Currie, 1961b). Since the saturated soil crumb is a respiring unit oxygen will decline in concentration inwardly from its surface as a function of respiratory demand and diffusivity. The saturation, narrowness and tortuosity of the capillaries are synonymous with low diffusivity and hence despite the high gas-phase oxygen levels of the intercrumb pore space there is a tendency for centres of anaerobiosis to develop within the wet crumb (Currie, 1961a; Greenwood and Goodman, 1967 and Appendix 2). for dry crumb material Currie (1965) obtained diffusivity values, (DCR/D~), which ranged from 0.025 to 0.156 where crumb fractional porosities lay between 0.25 and 0.41 ; such values are indicative of the considerable tortuosity of the crumb capillaries. When translated into diffusivity for wet and DCR= 0.156 x crumbs at 23°C we get DCR= 0-025 x 2.267 x 2.267 x i.e. 0.56 x lo-@cm2 s-l to 3-54 x lo-@cm2 s-l where D o 2 / ~ , is ,, 2.267 x cm2 s-l. When fitted into the Currie:Greenwood equations describing crumb aeration (see Appendix 2) we find that for spherical crumbs the critical crumb radius (i.e. the maximum radius at which the crumb is just wholly aerobic) ranges from 0.023 cm to 0.59 cm where crumb respiration - ~ and 5.27 x 10-lo g ~ r n s-l - ~ (i.e. 4 x lies between 5.27 x lo-” g ~ r n s-l ml 0, ml-l s-l and 4 x 10-7 ml O2 ml-l s-1). Since aggregates >4 cm diameter are by no means uncommon (Smith, 1977) anaerobiosis might often be considerable within some unsaturated soils and phytotoxins may accumulate; however it is of interest to note that as the volume of anaerobiosis within the aggregates increases so too will the oxygen content of the intercrumb pore space because of the overall reduction in oxygen consumption. Just as it was possible to calculate critical water film thickness in the homogeneous soil type we can readily derive an expression to describe the diffusion to a root lying within respiring wet aggregates if we assume these to be cylindrical rather than spherical : the particular solution required is the “critical thickness” of crumb material which will just bring the centre of the root to zero oxygen concentration. The problem is rather similar to that

TABLE I11 Conditions Leading to Anoxia at the Centre of Roots in Non-aggregated Soil (i) Critical soil water-film thicknesses, +a), at which roots would remain just wholly aerobic, predicted for various combinations of root oxygen consumption, MR, and root radius, a (see equation 68). Root oxygendifbivity, D ,7 x lo4 cm*s-l; soil oxygen-diffusivity, De, 1 x loJ cma s-l; G, the solution oxygen concentration at the gas-liquid interphase in the soil on radius, b, either 8.56 x lod g cm9(2O*41%)or 6.18 X lod g ad. (ii) Critical root radius, q, at which the root would be just wholly aerobic if the solution oxygen concentration at its surface was equal to C, (see equation 47).

b-a (cm) MR 1 x 10-7

MR 2 x 10-7 (g ~ m s-l) - ~

co

CO

a

(a) 0.01 0.02 0.03 0.04 0.05

0.06 0.07 0.08 0.09 0-10 ac

8.56 x

lo-@

(s cm-9 4.85 x 1-38 x 4.23 x 2.35 x 1-55 x 1.08 x 7.61 x 5-14 x 3.13 x 1.43 x

10' lo-' 10-* 1W2 lo-* lW3 lWS

0.1095

(g

I

co

6.18 x lo-"

8.56 x

(g cm-Y

(g a

4.52 6-74 x 2.55 x 1.48 x 9.62 x 6.34 x 3.95 x 2-04 x 4.48 x

-

lo-'

10-* 10-3 lO-' 10-3 10-8 10-4

0.093

cm-3

lo-@

-3

2.53 x 1*325 1-57 x 6.85 x 4.23 x 2-98 x 2.24 x 1-73 x 1-35 x 1.05 x

1V 10-' 1W2

10-' 1W2

0.1548

MR 5 x

s-1)

(g

co

cm-3 s-')

co

co

lo-@

6-18 x lo-@

8.56 x lo-''

6-18 x

cm-3 2.19 x 108 3.9 x lo-' 8-04 x 1W2 4.07 x 2-63 x 1W2 1.87 x lo-' 1.38 x lo-* 1.04 x 10-2 7-62 x 10-3 5.36 x lO-'

cm-? 6.9 x lo1* 9.71 1-22 2.76 x lo-' 1-33 x 10-' 8.46 x 611 x lo-* 4.79 x 10-2 3.79 x 10-2 3.11 x l W 3

cm-3) 5-22 x los 9-04 4.07 x lo-' 1.35 x 10-' 7.53 x 10-a 5.11 x 10-% 3.80 x 1W2 2-96 x 10-' 2.37 x lo-' 1.92 x lo-'

0.1316

0.2189

0-186

(g

(g

(g

317

AERATION IN HIGHER PLANTS

outlined in Section II.7(c) except that at the boundary of the inner cylinder (in this case the root surface) respiration does not cease but changes in intensity. The respiratory characteristics of the root can be accommodated by specifying the critical oxygen pressure required at the root wall and defining the oxygen gradient at that point. Consider a root (radius = a) which lies a t the centre of a cylindrical respiring soil crumb (r > a) in which respiration and pore space are homogeneously distributed radially. Let soil respiration be M and soil diffusivity be De, then from Section 11. 7(c) we can write: C

=

Mr2 __ + A logr 4D

+B (equation 39, p. 257)

If C

=

C, on r

=

a, then Ma2 C --+AAoga+B - 4De

Suppose that

dC -dr

=

P at r

=

a, then from equation (38) we get:

= -Ma +- A

p=(!E)

2De

r=&

a

and therefore A=

(P-- 3a

Substituting for A in equation (70) gives,

c w 4De = E (+~ - & ) a . l o g a + B

:.

B = c, - M-a2- (P 4De

-El

a. log a

(73)

(74)

If C , is the oxygen concentration at the root wall which will just maintain the root wholly aerobic, then substituting for A and B in equation (69) and solving the resulting form of (69) for r when C = C, we obtain

M C, - Cw = __ (r2 -az) 4De

+ alog a

(75)

where C, is the oxygen concentration at the outer surface of the soil crumb and r is the critical radius of the crumb cylinder. It may be noted that if M becomes zero the expression M(r2 - a2))/4Dedisappears and we are left with equation (27) while if M is the same in both soil and root, the right hand

318

W. ARMSTRONG

expression disappears and we are left with an equation which satisfies the boundary conditions C = CWand dC/dr > 0 on a > 0, and C = C, on r = r. When a = 0, dC/dr = 0 and the expression simplifies to equation (46). In practice the oxygen gradient at the root surface, P, is given by the expression : flux/De where the flux is calculated from the root’s respiratory rate and surface area, and De is the crumb diffusivity. Equation (75) is applicable both to “crumb” structured and homogeneous soil conditions but whereas in the homogeneous soil type our main concern is to predict critical film thickness (r - a), in the heterogeneous soil the critical aggregate diameter, 2 [a (r - a)] is the more important feature. The term r - a is, in itself, of little relevance and indeed for comparative purposes can be highly misleading as reference to Table IV will show. The data in this table obtained by fitting various combinations of a, P, M and D into equation (75) (together with a C, value of 8.56 x g~ m - ~ indicate ) once again that in terms of r - a, narrowness of root is perhaps an advantageous feature in wet aggregated soils. However, when account is taken of the volume of aggregate occupied by the root itself the situation canbe reversed. Forexample,when D = 0.56 x cm2 s-l, M = 5.27 x g ~ m s-l- ~and a = 0.05 cm, the critical aggregate diameter can be more than double that tolerated by the narrower and M lower (e.g. root (a = 0.01 cm). Only when D is higher (3.54 x 5.27 x 10-lo) do narrower roots achieve a significant advantage. If soil aggregates much exceed a diameter of 1 mm there would seem to be a real danger of inadequate aeration in all roots having a radius 50.05 cm. However, if roots are sufficiently thick to occlude intercrumb pores this might effectively unite the surrounding aggregates into bigger units and cause problems of root aeration where crumb diameter is not obviously critical. Against this must be set the crumb draining powers of the roots and differences in aggregate shape: in spherical aggregates critical crumb diameter will be greater whilst the crumb draining activities of roots may be sufficiently rapid to prevent anaerobiosis being sustained for sufficiently long periods to cause injury. The effects of introducing a respiratory term into the water film of the homogeneous soil types are also demonstrated in Table IV: critical water film thickness is very much reduced for the narrower roots but again there seems little likelihood of inadequate aeration if root radius is 50.05 cm.

+

2. Oxygen Flux in the Soil: Measurement and Interpretation Since adequate root aeration in the unsaturated soil depends chiefly upon the characteristics of the final water-saturated stage of the diffusion path, an evaluation of conditions at the interface between the root surface and the soil system presents the greatest possibility of ascertaining the influence of soil aeration on plant growth. It was this rationale which motivated Lemon and Erickson (1952) to develop a polarographic method (the Pt-micro-

TABLE IV Conditions Leading to Anoxia at the Centre of Roots in Aggregated and Non-aggregated Soil Critical soil water-& thicknesses, r - a (cylindrical wet aggregates and non-aggregated soils), and critical diameter, 2 [a (r - a) 1, of saturated cylindrical aggregates, for various combinationsof soil oxygen consumption, Ms,root oxygen consumption, MI(, and soil path diffusivity, De cS). Root radius, a, either 0.05 cm or 0.01 cm; G, the oxygen concentration at the gas-liquid interphase in the soil at radius r, taken as 8-56 x 10" g cm-l (20.41 %). Data calculated using equation (75).

+

M.¶ (g cm-8 s-')

De(s)

MR

(cma s-')

(g cm-8 s-l) 2 x 10-7 1 x 10-7 5 x 10-8 2 x 10-7 1 x 10-7 5 x 10-8 2 x 10-7 1 x 10-7 5 x 10-8 2 x 10-7 1 x 10-7 5 x 10-8 2 x 10-7 1 x 10-7 5 x 10-8 2 x 10-7 1 x 10-7 5 x 10-8

0.56 x 10-o 5-27 x 1W8 3.54 x 10-8 0-56 x lo-" 5.27 x lWIO 3.54 x 10-6 Zero

1 x 10-6

5.27 x

1 x 1W6

______

+

+

________~

r-a 2 [a (r -a) I r-a 2 [a (r - a) I where a = 005 cm where a = 005 cm where a = 0.01 cm where a = 001 cm

(cm) 0-0007(5) 00017(2) 0*0035(0) 0.0047 0-0107 0.0190 0.0007(7) 0-0017(5) 0-0035(5) 0-0050 0.0121 0.0289 0.0155 0.0423 0.1330 0.0143 0.0291 0.0439

(m) 1.01 1-03 1-07 1.09 1.21 1-38 1.01 1*03 1-07 1.10 1-24 1-58

(cm) 0.00153 00087 0.0117 0.034 0.036 0.039 0.0060 00156 0.048 1 0.1408 0.3046 0.3904

(mm) 0.303 0.374 0.434 0.88

0.92 0.98 0.321 0.512 1.162 3.016 6.29 8.00

48-5 Not aggregated Not aggregated

2-53 x lo6 5-22 x 108 0,0636 0.0686 0.0712

Not aggregated Not aggregated

320

W. ARMSTRONG

electrode technique), for assessing the oxygen flux within the wet phase of the soil. The Pt micro-electrode comprises a short apical oxygen “sensor” of bare thermo-pure Pt-wire ( I 5 1 cm) which is embedded basally into a strong but narrow insulated rod where it fuses with a copper wire from the polarizing circuit (Armstrong and Wright, 1976b). Oxygen is “consumed” electrolytically at polarographic electrodes (p. 273) and when embedded in the soil the polarized Pt-wire micro-electrode, being dimensionally similar to a root apex, is in a sense analogous with the respiring root apex. However, whilst the root respires throughout its volume, electrode activity is a surface phenomenon : the oxygen concentration at the activated electrode surface is effectively zero but that at the root surface is always >O provided that the effective diffusive resistance of the soil path is less than infinite. Consequently, the oxygen flux to an electrode will always be greater than to a root of equal radius lying within the same soil micro-zone. The flux at the micro-electrode surface will be the maximum possible flux to a cylindrical body of such dimensions at that particular location. Although the Pt micro-electrode technique is a valuable agronomic and ecological tool and is widely used, it has been made very clear by McIntyre (1970) and others, that it must be used with a caution which has so far been noticeably lacking. Those wishing to use it should not neglect to study McIntyre’s excellent review. A number of factors can interfere with the correct functioning of this technique; others can alter the current-voltage relations of the oxygen reduction process in a manner which is easily overlooked by those using fundamentally incorrect operating procedures. The use of constant applied voltage regardless of soil conditions is a procedure which can be particularly criticized: in poorly aerated or acidic soils plateau potentials become less negative; the fixed potentials in the range -0.6 to -0.8 V which have been employed by so many workers can then cause a reduction of Hf ions in addition to oxygen. Because of this the literature abounds with suspect data. The technique is at its most reliable in saturated conditions, and in welldrained aggregated soils appears to be applicable without complication only at moisture contents greater than field capacity (McIntyre, 1970). However, it could be argued that this is not a serious limitation since at tensions beyond field capacity aeration is much less likely to limit root activity. Efforts to establish the critical oxygen flux for root growth from soil oxygen flux measurement have not unnaturally revealed enormous variation. To some extent this is due to differences in the oxygen requirements of the roots themselves: the flux requirement at the root wall is a function of root radius and respiratory demand and, as Table V shows, one may forecast a 40-fold increase in flux requirement (15-600 ng cm-2 min-l) as root radius g ~ m s-l- ~to and oxygen demand are raised from 0.01 cm and 5 x 0.1 cm and 2 x lo-’ g ~ m s-l. - ~However, there are other complicating

AERATION IN HIGHER PLANTS

321

factors to be considered. These include the influence of electrode diameter, the use of inappropriate polarizing potentials, and aeration by the internal path (p. 305). As regards the polarizing potential McIntyre (1970) has suggested that the soil oxygen flux figure of 200 ng cm-2 min-l claimed by some (Stolzy and Letey, 1964a, b; Letey and Stolzy, 1967) to be critical for most non-wetland species could be partly a result of H+ ion reduction in nearly anaerobic media. The effect of electrode diameter is such as to widen still further the range of soil oxygen flux which might be regarded as critical for root growth (see Table V). Consider a root of radius 0.1 cm and respiratory demand 2 x lo-' cm2 s-l) of critical g ~ m s-l; - ~if the root lies within a water film (De, 1 x thickness 1.43 x cm (Table 111) the flux requirement of the root, 600 ng cm-2 min-l will be satisfied. The Pt micro-electrodes most commonly used have diameters of 1.2 mm, 0.64 mm and 0.46 mm; if these electrodes were surrounded by the critical film thickness 1.43 x 10-3cm the critical flux recorded (equation 27) would be 3634, 3671 and 3702 ng cm-2 min-l respectively. At the other extreme (root radius 0.01 cm, respiratory demand 5 x lo-* g ~ m s-l, - ~and critical flux requirement 15 ng cm-2 min-l) the same electrodes would register critical values of 2.6, 4.9 and 6.7 ng cm-2 min-l. As electrode and root approach a common radius so does electrode flux more closely register the true oxygen availability to the root and hence there are good grounds for trying to match the electrode radius with some characteristic root radius of the plant species concerned. In aggregated soils the problem of diameter becomes even more acute and hence the matching of electrode sizes with characteristic root radii is almost a necessity.

3. Internal Oxygen Transport in an Aerated Soil As soils drain, more and more oxygen will tend to enter roots via the soil path, and proportionately less will be provided by internal longitudinal transport from the aerial parts. The extent to which this will occur will depend chiefly upon the gas-phase oxygen concentration in the soil, upon soil diffusivity and metabolic activity in any saturated zone around the root, and upon root length, radius, pore-space resistance, and respiratory demand. Provided that the root is well within its critical radius, a, (see Table 111), and, in aggregated soils, the aggregates are not large or saturated, the soil path should predominate. However, as the following examples show, if aggregates remain saturated the internal path may continue to provide a significant proportion of the root's respiratory needs. These data were obtained using an electrical analogue suitably modified to simulate unsaturated as well as saturated soils. The introduction of a soil oxygen source made it necessary to devise new means for simulating soil sink activity and this was achieved by siting series of constant-current devices along the diffusion paths (diffusive resistances) between root and soil gas space; a digital-analogue-

TABLE V Oxygen Flux at Pt-microelectrode When Surrounded by the Critical Soil-water Film Thickness Appropriate to Roots Having the Characteristics Shown Predictions based upon an effective oxygen diffusivity of 1 x loJ cm2s-l within the soil-water film and a solution oxygen concentration at the gasliquid interphase in the soil of 8.56 x 10- g cm4 (20.41%).

MR. (g

s-*)

2 x 10-7

1 x 10-7

Root radius (cm)

(ng cm-a min-l)

0.01 003 0.05 0.10 0-01 0.03 0.05 0.10

60 180 300

0.01 5 x 10-8

Oxygen flux required at root wall

0.03 0.05 0.10

600 30

90 150 300 15 45 75 150

Observed flux at electrode lying within critical a m thickness (ng min-') Electrode radius 0.06 cm

Electrode radius 0.032 cm

Electrode radius 0.023 cm

160

21.9 190-5

372 3634 5.7 66.6 160 530 2.64 27.9 73 205

3671 10.1 90.4 190 565 4-86 43.7 98 236

29.2 214 433 3702 13.77

12.8

406

108.5 214 593 6.69 55.9 116.6 261

TABLE Vf Root Aeration in Saturated and Unsaturated Soil Comparative data compiled by electrical analogue simulation of a root (radius, a = 0.05 cm) lying within saturated soil aggregate (radius, b = 0.10 cm) such that the thickness of aggregate around the root is everywhere 0.05 cm (ie b - a). Other characteristics as follows: root length, 9 cm; MR = 02, 70 ng cm3 s-', ie 33 ng min-I cm-l, ~ M =R297 ng min-'; D, (soil) = 3.54 x cm2 s-l; M s =5.27 ng cm4 s-I. Internal oxygen concentrations (a) and lateral oxygen exchange (b) across the root surface (influx, ; efflux, -), at the distances indicated (*) or (**) are given for the following circumstances: (i) no lateral transfer between root and soil-impermeable root wall; effective root porosity, 7%: (ii) soil acting only as an oxygen sink -fully permeable root wall; effective root porosity, 7 %: (iii) soil oxygen-source (20.41 %) introduced at aggregate boundary; effective root porosity, 7 %: (iv) as for (iii) but effective root porosity, 5 %.

+

(ii)

(i) Distance from root base (cm)

*

**

I 2 3 4

0.5

5 6 7 8 9

1.5 2.5 3.5 4,5 5.5 6.5 7.5 8.5

a 0,

b Lateral exchange (ng min-l cm-l)

a 0 2

nil

nil nil nil nil nil

nil nil

nil

Lateral exchange

a 0,

(ng min-I cm- l)

**

17.26 14.48 12.07 10.03 8.36 7.06 6.14 5.58 5.39

(iii)

b

**

**

16.68 13.43 10.60 8.2 6.24 4.76 3.73 3.08 2.85

-7.5 -7-5 -7.5 -7.5 -7.5 -7.3 -5.9 -5.4 -4.9

b Lateral exchange (ng min-' cm-l)

17.71 15-4 13.4 11.75 10.37 9.37 8.65 8.19

8.05

-1.7 +0.6 2.2 4.0 +4.9 +6.0 6.9 +7.1 $7.6

+ +

+

a 0,

(%)

*

16.89 13.97 1 1.45 9.39 7.72 6.44 5.55

5.01 4.83

(iv) b Lateral exchange (ng min-2 ern-')

**

-1.3 +1.5 +3.6 1-5.6 +7.2 $8.3 +9.4 +9.6 10.2

+

324

W. ARMSTRONG

converter and display replaced the meters (Fig. 8a) of the original analogue system (E. J. Wright and W. Armstrong, unpublished). Consider a root (length 9 cm, radius 0.05 cm, and oxygen demand, MR, 7 x lo-* g ~ m s-l- ~ (33 ng min-l cm-l)), lying within saturated soil aggregates such that the thickness of aggregate material around the root is 0-05cm : let the radial oxygen diffusivity of the aggregates, De, be 3.54 x cm2 s-l and their rate of oxygen uptake, Ms, be 5.27 x lo-$ g ~ m SKI. - ~If there was no leakage of oxygen between root and soil the root’s respiratory needs would be met entirely by internal transport and the oxygen profile along the root could be readily computed from equation (30). For an effective root porosity of 7 % the internal oxygen regime would be as shown in Table VI(i)a. If conditions were such that the soil behaved only as an oxygen sink (and there was no restriction on oxygen leakage through the root wall), internal oxygen transport would continue to fully support the root’s respiratory requirements, but the oxygen regime would change to that in Table VI(ii)a. Radial oxygen loss to the soil would amount to 14.5% of that entering the root (Table VI(ii)b). The introduction of an oxygen source (20-41%) into the soil gas space, it is predicted, would modify the oxygen regime in the root t o that shown in Table VI(iii)a. Oxygen leakage to the soil is now confined t o the basal centimetre of the root. Thereafter the soil makes an increasing contribution to root aeration (Table VI(iii)b) but nevertheless, this does not exceed 13 % of the total oxygen requirement. However, it can now be seen that the soil oxygen source is also acting as a buffer to the root’s internal supply (see also p. 305). A decrease in the effective porosity of the root increases the contribution made by the soil source: at an effective porosity of 5 % the soil will provide c. 19 % of the root’s requirements (see Table VI(iv)b). A further lowering of porosity would lead to further increases in oxygen flux from the soil but whilst it might eventually provide the bulk of the oxygen consumed at this aggregate thickness it could never fully support the root’s respiratory needs: root aeration would become inadequate as the effective root porosity approached 3 %. ACKNOWLEDGEMENTS

I wish to express my gratitude to the following for their considerable help during the preparation of this article: Dr J. D. Smith (Winchester College), who derived the equations set out in Section III.A.3 and Appendix 1, and who has given me valuable help on other diffusion problems; Dr J. Dunning-Davies (Dept of Applied Mathematics, Univ. of Hull), who derived equations (48), (49) and (75) and willingly gave of his time to answer other mathematical queries; Mr E. J. Wright, for his efforts in developing and perfecting the electrical analogue described in Sections II.D.2 and V.3; Dr T. J. Gaynard,

AERATION IN HIGHER PLANTS

325

who provided much of the data in certain sections, assisted me in the preparation of the analogue material, and kindly read the manuscript; Miss S. Lythe, who prepared and photographed the sections shown in Plates 1-111, and carried out the bulk of the figure drawing; Messrs P. Smith and P. Meaker for photographic and drawing assistance; Miss E. M. Sharpe for typing the manuscript; and finally, my wife, for her advice and much needed encouragement.

1. The Transport of Diffusible Species in Media Moving by Mass Flow: (a supplement to Section II.A.3) (a) The Complementary Condition Consider an idealistic model in which two planes X and Y within, of equal radius to, and normal to the long axis of, a water-filled tube P provide respectively, a source of diffusible species at constant concentration C,, and an effective sink at a lower but constant concentration C1.If water flows through P in the direction X, Y and at a constant velocity V, there will be the potential for diffusive as well as mass flow of diffusible species from X to Y, and the two processes will act in conjunction. If there is no lateral leakage of diffusate from P the differential equation for diffusion in one dimension can be written: APPENDIX

where x is the space coordinate in the direction X to Y. The planes X and Y are given by x = 0 and x = 1 respectively, and D is the diffusion coefficient of the diffusate in the water. The time independent solutions satisfy aC/at = 0, and have the form C = A BeAx, (77) where h = V/D, and A and B are constants of integration to be determined for the boundary conditions C = C, on x = 0, and C = C1 on x = 1. The solution which satisfies these boundary conditions is : -1 c -c, = ( C , -C1)---eAx eAl- 1

+

(equation 1, p. 240) The diffusion rate per unit area (i.e. rhe diffusiveflux)can be expressed as follows:

Since the solutions are time independent we now use the ordinary differential notation and thus on x = 0,

and on x

=

I,

326

W. ARMSTRONG

The total flux of diffusible species, the sum of diffusive flux and CV, is derived as follows : Total flux

=

- BVeAx+ V (A

=

- Cl) c, + (C, (eAl- 1)

+ BeA")

(82)

(b) Diffusion and mass flow in opposition When the mass flow is negative the solution which corresponds with equation (78) is: 1 -e - A ~ c = c, - ( C , -C1)- 1 -e-Al where h = V/D as before. The diflusive flux is given by

and thus on x

=

0

dC dx

=

V(C, -Cl)eAz eAl- 1

-D- dC dx

=

V(C0 -C1) eAl- 1

-D-

and on x

=

1

Corresponding to (83), we have Total flux

=V

( co -cleAz) eAl

-

(NB. Numerical solutions from equations (78) and (84) are plotted in Fig. 2.) APPENDIX 2 : Radial Diflusion into Respiring Spherical Bodies

Solutions describing the radial diffusion of gases within spherical bodies have been used extensively in connection with the aeration of soil aggregates (Currie, 1961a; Greenwood and Goodman, 1967). Spherical analogues of various of the solutions for radial diffusion into cylindrical bodies are given below and in each of these the spherical body is of radius r = b, and for all values of r, De is a constant. (i) The spherical analogue of equation (49) is given by Currie (1961a) as Co -C where

=

M

6D (b2 -r2) -2a

3

(r' - t ) = A r

(89)

M is the oxygen uptake within the sphere (g cm3 s-l) a is the radius at which aerobic respiratory activity ceases such that dC/dr = 0 on r = a, and a > 0. It is assumed that respiration is unaffected by oxygen concentration until extremely low values are reached and hence C z 0 on r = a. Co is the oxygen concentration (g cm3) at the surface r = b. C is the oxygen concentration at any radius r where a L r 4 b. dr is the oxygen deficit within the crumb at radius r.

AERATION IN HIGHER PLANTS

327

(ii) The maximum radius at which the body is just wholly aerobic is given by putting r = a = 0 and is termed its critical radius bc. The oxygen deficit, dc, is then, (90) (Currie, 1961a) and is the spherical analogue of equation (47). (iii) For all b > bc we get (91) (Currie, 1961a) which is the spherical analogue of equation (45). (iv) If we combine equations (90) and (91) we obtain,

(Currie, 1961a) from which the fractional anaerobic volume can be calculated: at twice the critical radius 30 % of the sphere will be anaerobic; at four times the critical radius the figure rises to 60 % and when the radius is 10 bc the anaerobic volume is 84 %. Currie (1961a) has noted that as the radius becomes very much greater than the critical value “the volume of the crumb which remains anaerobic becomes proportional to the surface area of the crumb”, i.e. when b )) be, 4/3 rr(b3 - a3) + 4 T b2(b - a) and (v) When dC/dr:

=

0 on a

=0

equation (89) simplifies to

C

=

M Co - -(b2 - r2) 6De

(93) (Currie, 1961a)

where M(b2 -r2)/6De is the oxygen deficit between b and r. The cylindrical counterpart of this equation which has been given by Lemon (1962) is: C

=

M C, - -(b2 - r2) 4De

and defines the oxygen field within the cylinder, r buted uniformly along r.

= b,

(94) when M and De are distri-

328

W. ARMSTRONG

REFERENCES Arber, A. (1920). “Water Plants”. Cambridge University Press. Amoore, J. E. (1961a). Proc. R. SOC.B. 154, 95-108. Amoore, J. E. (1961b). Proc. R. SOC.B. 154, 109-129. Armstrong, W. (1964). Nature, Lond. 204, 801-802. Armstrong, W. (1967a). Physiol. PI. 20, 540-553. Armstrong, W. (1967b). Physiol. P1. 20, 920-926. Armstrong, W. (1968). Physiol. PI. 21, 539-543. Armstrong, W. (1970). Physiol. PI. 23, 623-630. Armstrong, W. (1971a). Physiol. P1. 24, 242-247. Armstrong, W. (1971b). Physiol. PI. 25, 192-197. Armstrong, W. (1972). Physiol. PI. 27, 173-177. Armstrong, W. (1975). In J. R. Etherington “Environment and Plant Ecology”, pp. 181-218, John Wiley Ltd. Armstrong, W. (1978). In “Plant Life in Anaerobic Environments” (D. D. Hook and R. M. M. Crawford, Eds), pp. 269-297. Ann Arbor Science Publishers Inc., Michigan, U.S.A. Armstrong, W. and Boatman, D. J. (1967). J. Ecol. 55, 101-110. Armstrong, W. and Gaynard, T. J. (1976). Physiol. PI. 37, 200-206. Armstrong, W. and Wright, E. J. (1975). Physiol. PI. 35, 21-26. Armstrong, W. and Wright, E. J. (1976a). Physiol. PI. 36, 383-387. Armstrong, W. and Wright, E. J. (1976b). J. Appl. Ecol. 13, 849-856. Banath, C. L. and Monteith, N. H. (1966). Plant and Soil 25, 143-149. Bange, G. G. J. (1953). Acta. bot. need 2, 255-297. Barber, D. A., Ebert, M. and Evans, N. T. S. (1962). J. exp. Bor. 13, 397-403. Berry, L. J. and Norris, W. E. Jr (1949). Biochim. Biophys. Acta. 3, 593-606. Boeke, J. E. (1940). Ann. Jardin Bot. Buitenzorg. 50, 199-208. Boulter, D., Coult, D. A. and Henshaw, G. G. (1963). Physiol. PI. 16, 541548. Bouyoucos, G. J. (1915). Michigan Agr. CON.Expt. Sta. Tech. Bull. 22. Bowling, D. J. F. (1973). Planta. 111, 323-328. Boynton, W. P. and Brattain, W. H. (1929). I n “International Critical Tables” Vol. V. p. 62, McGraw-Hill Book Co. Inc., N.Y. Brown, R. (1947). Ann. Bot., NS. 11, 417-437. B. 193, 223-291. Brown, H. and Escombe, F. (1900). Phil. Trans. R. SOC. Bruins, H. R. (1929). I n “International Critical Tables” Vol. V. pp. 65-66. McGrawHill Book Co. Inc., N.Y. Buckingham, E. (1904). U.S. Dept. Agr. Bur. Soils Bul. 25. Canny, M. J. (1961). Ann. Bot., N.S. 26, 603-617. Canny, M. J. (1973). “Phloem Translocation”. Cambridge University Press. Chapman, S. and Cowling, T. G. (1939). “The Mathematical Theory of NonUniform Gases”. Cambridge University Press. Chirkova, T. V., Khazova, I. V. and Astafurova, T. P. (1973). Fiziol. Rust. 21, 102-1 07. Collander, R . (1959). I n “Plant Physiology: A Treatise” (F. C. Steward, Ed.), pp. 3-102, Academic Press, London and New York. Conway, V. M. (1937). New Phytol. 36, 64-96. Coult, D. A. (1964). J. exp. Bot. 15, 205-218. Coult, D. A. and Vallance, K. B. (1958). J. exp. Bot. 9, 384-402. Coutts, M. P. and Armstrong, W. (1976). I n “Tree Physiology and Yield Improve-

AERATION IN HIGHER PLANTS

329

ment” (M. G. R. Cannel and F. T. Last, Eds), pp. 361-385, Academic Press, London and New York. Coutts, M. P. and Phillipson, J. J. (1978a). New Phytol. 80, 63-69. Coutts, M. P. and Phillipson, J. J. (1978b). New Phytol. 80, 71-77. Coutts, M. P. and Phillipson, J. J. (1978~).New Phytol. 80, 341-349. Crafts, A. S. and Broyer, T. C. (1938). Am. J. Bot. 25, 529-535. Crank, J. (1975). “The Mathematics of Diffusion”. Clarendon Press, Oxford. Crawford, R. M. M. (1969). Ber. dt. bot. Ges. 82, 111-114. Crawford, R. M. M. (1972). Trans. Bot. SOC.Edinb. 41, 309-322. Crawford, R. M. M. (1976). In “Tree Physiology and Yield Improvement” (M. G. R. Cannel and F. T. Last, Eds), pp. 387401, Academic Press, London and New York. Currie, J. A. (1961a) Soil Sci. 92, 40-45. Currie, J. A. (1961b) Brit. J. appl. Phys. 12, 275-281. Currie, J. A. (1965). J. Soil Sci. 16, 279-289. Das, D. K. and Jat, R. L. (1977). Agron. J. 69, 197-200. Davies, D. D., Nascimento, K. H. and Patil, K. D. (1974). Phytochemistry 13, 241 7-2425. Effer, W. R. and Ranson, S. L. (1967). PZ.Physiol. 42, 1042-1052. Engler, R. M. and Patrick, W. H. Jr (1975). Soil Sci. 119, 217-221. Evans, N. T. S. and Ebert, M. (1960). J. exp. Bot. 11, 246-257. Forrester, M. L., Krotkov, G. and Nelson, C. D. (1966). Plant Physiol. 41,422427. Fourier, J. B. (1822). “Theorie analytique de la chaleur” English translation by A. Freeman, Dover Publ., New York, 1955. Fraser, A. I. and Gardiner, J. B. H. (1967). “Rooting and Stability in Sitka Spruce”. Forestry Commission Bull. No. 40, H.M.S.O., London. Fulton, J. M. and Erickson, A. E. (1964). Soil Sci. SOC.Amer. Proc. 28, 610-614. Gambrel], R. P. and Patrick, W. H. (1978). In “Plant Life in Anaerobic Environments” (D. D. Hook and R. M. M. Crawford, Eds), pp. 375423. Ann Arbor Science Publishers, Inc., Michigan, U.S.A. Gaynard, T. J. (1979). “Some Aspects of Internal Aeration in Wetland Plants”. Ph.D Thesis, University of Hull. Gill, C. J. (1970). Flora, Bd. 164, S . 85-97. Goodwin, P. B. (1976). In “Intercellular Communication in Plants: Studies on Plasmodesmata” (B. E. S. Gunning and A. W. Robards, Eds), pp. 121-129, Springer Verlag. Grable, A. R. (1966). Adv. Agron. 18, 57-106. Green, M. S. and Etherington, J. R. (1977). J. exp. Bot. 28, 678-690. Greenwood, D. J. (1961). Plant and Soil 14, 360-376. Greenwood, D. J. (1 962). Plant and Soil 17, 378-391. Greenwood, D. J. (1963). Chem. and Znd. pp. 799-803. Greenwood, D. J. (1967a). New Phytol. 66, 337-347. Greenwood, D. J. (1967b). New Phytol. 66, 597-606. Greenwood, D. J. (1968). Trans. 9th Znt. Congress Soil Sci. 1, 823-832. Greehwood, D. J. (1969). In “Root Growth” (W. J. Whittington, Ed.), pp. 202-223, Butterworths, Oxford. Greenwood, D. J. and Goodman, D. (1967). J . Soil Sci. 18, 182-196. Griffin, D. M. (1968). New Phytol. 67, 561-577. Gunning, B. E. S. and Robards, A. W. (1976). In “Intercellular Communication in Plants” (B. E. S. Gunning and A. W. Robards, Eds), pp. 297-301. SpringerVerlag.

330

W. ARMSTRONG

Hall J. L., Sexton, R. and Baker, D. A. (1971). Planta 96, 54-61. Healy, M. T. (1975). “Oxygen transport in Pisum sativum L.” Ph.D Thesis, University of Hull. Healy, M. T. and Armstrong, W. (1972). Planta 103, 302-309. Hook, D. D. and Brown, C. L. (1972). Bot. Guz. 133, 304-310. Hook, D. D. and Brown, C. L. (1973). Forest Sci. 19, 225-229. Hook, D. D., Brown, C. L. and Kormanik, P. P. (1970). Bot. Gaz. 131, 217-224. Hook, D. D., Brown, C. L. and Kormanik, P. P. (1971). J. exp. Bot. 22, 78-89. Hook, D. D., Brown, C. L. and Wetmore, R. H. (1972). Bot. Gat. 133,443454. Huck, M. G. (1970). Agron. J. 62, 815-818. Humphries, W. J. (1926). In “International Critical Tables”, Vol. I, p. 393. McGrawHill Book Co. Inc., N.Y. Jensen, C. R., Stolzy, L. H. and Letey, J. (1967). Soil Sci. 103, 23-29. Jensen, C. R., Luxmoore, R. J., Van Gundy, S. D. and Stolzy, L. H. (1969). Agron. J. 61, 474-475. John, C. D. and Greenway, H. (1976). Aust. J. PI. Physiol. 3, 325-336. Kaye, G. W. C. and Laby, T. H. (1966). “Tables of Physical and Chemical Constants”. Longmans and Green, London. Amer. Proc. 30, 529-534. Kemper, W. D. and Rollins, J. B. (1966). Soil Sci. SOC. Kirk, J. T. 0. and Tilney-Basset, R. A. E. (1967). “The Plastids”, W. H. Freeman, London and San Francisco. Kordan, H. A. (1974). New Phytol. 73, 695-697. Kordan, H. A. (1975). Ann. Bot. 39, 249-256. Kordan, H. A. (1976a). Ann. Bot. 40, 347-350. Kordan, H. A. (1976b). Ann. Bot. 40, 1329-1332. Kordan, H. A. (1976~).New Phytol. 76,81-86. Kordan, H. A. (1976d). J. Cell Sci. 20, 57-59. Kordan, H. A. (1977). Ann. Bot. 41, 1205-1209. Kramer, P. J., Riley, W. S. and Bannister, T. T. (1952). Ecology 33, 117-121. Kristensen, K. J. and Lemon, E. R. (1962). Agron. J. 56, 295-301. Laing, H. E. (1940). Am. J. Bot. 27, 861-867. Lemon, E. R. (1962). Agron. J. 64, 725-729. Lemon, E. R. and Erickson, A. E. (1952). Proc. Soil Sci. SOC.Amer. 16, 160-163. Lemon, E. R. and Erickson, A. E. (1955). Soil Sci. 79, 382-392. Lemon, E. R. and Wiegand, C. L. (1962). Agron. J. 54, 171-175. Letey, J. and Stolzy, L. H. (1964). Hilgardia 35, 545-554. Letey, J. and Stolzy, L. H. (1967). Soil Sci. 103, 404-409. Leyton, L. and Rousseau, L. Z. (1957). In “Physiology of Forest Trees” (K. V. Thimann, Ed.), pp. 467-475. Ronald, New York. Luxmoore, R. J., Stolzy, L. H. and Letey, J. (1970) Agron. J. 62, 317-332. Luxmoore, R. J., and Stolzy, L. H. (1972) Agron. J. 64, 720-729. Mason, T. G. and Phillis, E. (1936). Ann. Bot. 50, 455499. Mazelis, M. and Vennesland, B. (1957). PI. Physiol. 32, 591-600. McIntyre, D. S. (1970). Adv. Agron. 22, 235-283. MacManmon, M. and Crawford, R. M. M. (1971). New Phytol. 70, 299-306. Meidner, H. and Mansfield, T. A. (1968). “Physiology of Stomata”, McGraw-Hill, London. Millington, R. J. (1959). Science 130, 100-102. Molisch, H. (1888). Sitzungsber. Akad. Wiss. Wien. Math. Nat. KI. 96. 84. Montgomery, H. A. C., Thom, N. S. and Cockburn, A. (1964). J. appi Chem. 14, 280-296.

AERATION IN HIGHER PLANTS

33 1

Morrisset, C. (1975). Abstracts XZZ Inter. Bot. Congress Leningrad, p. 366. Nelson, C. D., Perkins, H. J. and Gorham, P. R. (1958). Can. J. Biochem. Biophys. 36, 1277-1279. Nobel, P. S . (1974). “Introduction to Biophysical Plant Physiology”. W. H. Freeman, San Francisco, U.S.A. Okajima, H. (1964). In “The Mineral Nutrition of the Rice Plant”. T.R.R.I. Symposium, pp. 63-73, John Hopkins Press, Baltimore, Maryland, U S A . Peel, A. J. (1974). “Transport of Nutrients in Plants”. Butterworths. Penman, H. L. (1940). J. Agric. Sci. 30, 437-462. Pitts, G., Allam, A. I. and Hollis, J. P. (1972). Science 178, 990-992. Ponnamperuma, F. N. (1972). Adv. Agron. 24, 29-95. Quereshi, F. A. and Spanner, D. C. (1973). Planta 110, 131-144. Raciborski, M. M. (1905a). Bull. Int. de L‘Acad. Sciences (Cracovie) 338-349. Raciborski, M. M. (1905b). Bull. Int. de L’Acad. Sciences (Cracovie) 668-693. Read, D. J. and Armstrong, W. (1972). New Phytol. 71, 49-53. Robards, A. W. and Clarkson, D. T. (1976). In “Intercellular Communication in Plants” (B. E. S. Gunning and A. W. Robards, Eds), pp. 181-201. SpringerVerlag. Rowe, R. N. (1966). “Anaerobic m-tabolisrn and cyanogenic glycoside hydrolysis in differential sensitivity of peach, plum and pear roots in water-saturated conditions”. Ph.D Thesis, University of Calif. Davis. Rowe, R. N. and Beardsell, D. V. (1973). C.A.B. Horticultural Abstracts 43, 533-548. Rowe, R. N. and Catlin, P. B. (1971). J. Am. SOC. hort. Sci. 96, 305-308. Sanderson, P. L. (1977). “On the responses of Sitka Spruce and Lodgepole Pine to conditions associated with waterlogging”. Ph.D. Thesis, University of Hull, U.K. Sanderson, P. L. and Armstrong, W. (1978). Plant and Soil 49, 185-190. Scholander, P. F., van Dam, L. and Scholander, S. I. (1955). Am. J. Bot. 42, 9298. Schreiner, 0. and Reed, H. S. (1909). Bot. Gaz. 47, 355-388. Schreiner, 0. and Sullivan, M. S. (1910). U.S. Dept. Agric. Bureau Soils Bull. 73, 1-57. Sculthorpe, C. D. (1967). “The Biology of Aquatic Vascular Plants”. Arnold, London. Sifton, H. B. (1945). Bot. Rev. 11, 108-143. Sifton, H. B. (1957). Bot. Rev. 23, 303-312. Smith, K. A. (1977). Soil Sci. 123, 284-291. Spanswick, R. M. (1976). “Encyclopedia of Plant Physiology”. N.S. 2B (U. Luttge and M. G. Pitman, Eds), pp. 35-53, Springer-Verlag. Stolzy L. H. and Letey, J. (1964a). Adv. Agron. 16, 249-279. Stolzy L. H. and Letey, J. (1964b). Hilgardia 35, 567-576. Teal J. M. and Kanwisher, J. W. (1961). Limnol. Oceanogr. 6, 388-399. Teal J. M. and Kanwisher, J. W. (1966). J. exp. Bot. 17, 355-361. Tyree M. T. (1970). J. exp. Bot. 26, 181-214. Ullrich W. (1961). Planta 57, 402427. Vamos R. and Koves, E. (1972). J. Appl. Ecol. 9, 519-526. Van Bavel C. M. M. (1952). Soil Sci. 72, 3345. Van der Heide, H., Van Raalte, M. H. and de Boer-Bolt B. M. (1963). Acta But. Neerl. 12, 231-247. Van Raalte, M. H. (1941). Ann. Jard. Bot. Buitzenzorg 51,43-57.

332

W. ARMSTRONG

Van Raalte, M. H. (1943-1944). Hort. Bot. Bogoriensis Java. Syokubutu-Zho 1, 15-34. Vartapetian, B. B. (1970). Agrochemica 15, 1-19. Vartapetian, B. B., Andreeva, I. N. and Kozlova, G. I. (1976). Protoplasma 88, 21 5-224. Vartapetian, B. B., Andreeva, I. N., Kozlova, G. I. and Agapova, L. P. (1977). Protoplasma 91, 243-256. Von Fick, A. (1855). Annln. Phys. 94, 59-86. Weast, R. C. (1974). “Handbook of Chemistry and Physics”. CRC. Press, Inc., Cleveland, Ohio, U.S.A. Williams, W. T. and Barber, D. A. (1961). S.E.B. Symposium 15, 132-144. Wood, J. T. and Greenwood, D. J. (1971). J. Soil Sci. 22, 281-288. Yamada, N . and Ota, Y. (1958). Proc. Crop Sci. SOC.Japan 26, 205-210. Yocum, C. S. and Hackett, D. P. (1957). PI. Physiol. 32, 186-191. Yoshida, T. and Sukuki, M. (1975). Soil Sci. Plant Nutr. 21, 129-135. Yu, P. T., Stolzy, L. H. and Letey, J. (1969). Agron. J. 61, 844-847.

NOTE ADDED IN PROOF Since submission of this article the following relevant publications have appeared :

(a) Starch metabolism Kaiser, W. M. and Bassham, J. A. (1979). PI. Physiol. 63, 105-108. Mares, D. J., Hawker, J. S. and Possingham, J. V. (1978). J. exp. Bot. 29, 829-835. Pongratz, P. and Beck, E. (1978). PI. Physiol. 62, 687-689. Stankovic, Z. S. (1978). Plant Sci. Lett. 12, 371-377. (b) Metabolite transport in intact chloroplasts Akamba, L. M. and Siegenthaler, P. A. (1979). FEBS Letters 99, 6-10. Flugge, U. I. and Heldt, H. W. (1978). Biochem. Biophys. Res. Com. 84,!37-44. Flugge, U. I. (1978). PhD Thesis, University of Munich. Hampp, R. (1978). PI.Physiol. 62, 735-740. Huber, S. C. (1979). Biochim. Biophys. Acta 545, 131-140. Werdam, K. (1975). PhD Thesis, University of Munich.

(c) Ribulose-l,5-bisphosphatecarboxylase regulation Heldt, H. W., Chon, C. J. and Lorimer, G. H. (1978). FEBS Letters 92, 234240. Robinson, S. P., McNeil, P.H. and Walker, D. A. (1979). FEBSLetters97,296-300.

( d ) Galactosyl trangerase activity of chloroplast envelopes Dalgarn, D., Miller, P., Bricker, T., Speer, N., Jaworski, J. G. and Newman, D. W. (1979). PIaet Sci. Lett. 14, 1-6. (e) Amino acid composition of chloroplast envelopes Mackender, R. 0. (1978). Plant Sci. Lett. 12, 279-285.

(f)Import of proteins into cell organelles Maccecchini, M. L., Rudin, Y., Blobel, G. and Schatz, G. (1979). Proc. Nut. Acad. Sci. USA 76, 343-347.

Population and Community Structure and Dynamics of Fungi in Decaying Wood

A. D. M. RAYNER School of Biological Sciences, University of Bath, Claverton Down, Bath, Avon, England BA2 7 A Y and

N. K. TODD Department of Biological Sciences, Hatherly Laboratories, Prince of Wales Road, Exeter, Devon, England EX4 4PS

1.

11.

Introduction . . . . . . . . . . . . A. General . . . . . . . . . . . B. Types of Decay and Fungi Inhabiting Wood. C. Scope of the Present Article . . . . . Direct A. B. C.

Methods of Analysis . . . . . General . . . . . . . . . Recognition of Interactional Patterns Zone Lines . . . . . . . .

. . . . . . . . . . . . . . . . . . . .

334 334 335 336

. . . . . . . . . . . . . . . . . . . . . . . . . . . .

337 337 339 343

.

346 346 359 375

111.

Intraspecific Antagonism: The Delimitation of Individual Mycelia A. General . . . . . . . . . . . . . . . B. Basis of Antagonism . . . . . . . . . . . C. Significance and Potential Use of Antagonism . . . .

IV.

Interspecific Interactions: Their Role in the Development and Maintenance of Community Structure . . . . . . . . . . . . 380

. . .

334

A. D. M. RAYNER AND N. K. TODD

A. B.

C. D. E.

V.

VI.

VII.

Theories of Succession and Community Development . . . 380 Interactions and Their Significance . . . . . . . . 384 Results from Laboratory-based Studies . . . . . . . 388 Interactions in Nature . . . . . . . . . . . . 399 Concluding Comments . . . . . . . . . . . . 401

Ecological Roles and Spatial Distribution Within Communities of Fungi in Decaying Wood . . . . . . . . . . . . . A. Introduction . . . . . . . . . . . . . . . B. Factors Influencing Mode and Pattern of Growth of Fungi in Decaying Wood . . . . . . . . . . . . . . C. Discussion . . . . . . . . . . . . . . . Schema for Fungal Community Development in Decaying Sapwood of Hardwoods after Felling . . . . . . . . . . . . .

403 403 404 414 415

Concluding Comment . . . . . . . . . . . . . . 417 Acknowledgements . . . . . . . . . . . . . . . 417 References . . . . . . . . . . . . . . . . . 417

I. INTRODUCTION A. GENERAL

By its very nature decaying wood presents unique opportunities for study to those interested in the manner in which fungi grow and interact with one another to form populations and communities in natural substrata. This is partly because it is a spatially defined, often bulky resource in which many of the fungi, especially those causing decay, frequently occur as single colonies, each occupying considerable volumes, as noted by Garrett (1970). Within many a piece of wood undergoing active decay it is therefore possible, using the correct methods, to define the internal three-dimensional distribution of individual mycelia and hence their spatial organization into populations and communities; we must remember the truism that it is the mycelium and not its reproductive structures which constitutes the main body of a fungus. Thence we may begin to understand something of the nature and development of fungal populations and communities at a level approaching that already achieved for higher plants and animals: it has been emphasized by Hudson (1968) that such knowledge for fungi is almost entirely lacking. We hope t o show that ultimately populations and communities of fungi in decaying wood have much in common with those of other organisms; they may be viewed as being dynamic rather than static and consisting of individuals fulfilling their nutrient requirements, whilst being in competition and interacting with others. As our knowledge of the pattern of growth of these interacting individuals develops, so our understanding of the forces operating for stability and change within the resource will deepen.

FUNGI IN DECAYING WOOD

335

As with all research, such vision requires that we should first formulate the correct questions, and subsequently adopt appropriate methods to answer them. We believe that failure on one or other of these counts may explain previous lack of appreciation of decaying wood as a source of exciting biological information. On the one hand, much work on wood decay has, understandably, focused on the need for its prevention in timber. This work is an essential component of timber technology and its economic implications make it both admirable and important, but the approach required is unlikely to reveal the type of information we have described. On the other hand, more ecological or biological work has suffered from lack of use of natural material, such as logs, stumps and branches, and from rigid adherence to tried and tested methods of sampling such as the increment borer which removes a core of wood from which isolations onto agar are then made. As we will point out later, such methods are unlikely to reveal the full three-dimensional structure of fungal communities in wood without inordinate effort, and may be misleading in other ways. In effect we need to combine direct field observations of natural woody substrata undergoing decay with direct methods of sampling and analysis. With the background thus obtained we can set up appropriate physiological and genetical experiments in the laboratory, which may help us to understand the patterns revealed. In this article we describe how use of such an approach may provide a clearer understanding of the natural situation than might otherwise be achieved. At present relatively little such work has been attempted. B. TYPES OF DECAY AND FUNGI INHABITING WOOD

This topic has been adequately reviewed elsewhere (e.g. Kaarik, 1974) but it may assist understanding to include a brief summary here. A wide range of fungi have been found to occur in wood including certain mucoraceous species, ascomycetes, basidiomycetes and fungi imperfecti. They have varied effects on the wood, partly relating to the manner in which they obtain nutrients from it. Decay fungi are normally considered to fall into three different classes; those causing white rots, brown rots and soft rots. In white rots all components of the wood are removed, including both lignin and cellulose. At an advanced stage of decay the wood characteristically appears bleached and has a fibrous or spongy consistency. Microscopically, the rot is characterized by progressive thinning of the cell walls. White rot is usually regarded as more common in hardwoods than softwoods and can be caused by a wide range of basidiomycetes (including a variety of crusts, brackets, agarics and gasteromycetes) and a few ascomycetes (e.g. species of Xyluriu, Ustulinu). In brown rots cellulose is removed but the lignin is left virtually unchanged. At an advanced stage of decay the wood is normally some shade of brown and is characteristically cubically cracked. It will crumble readily to powder between

336

A. D. M. RAYNER AND N. K. TODD

the fingers. The characteristic thinning of cell walls found with white rots is absent, their shape being maintained until their final collapse. Brown rot is caused only by a relatively small number of basidiomycetes and is usually considered to be more common in softwoods than hardwoods. The third type of rot, soft rot is caused by a variety of ascomycetes and fungi imperfecti which are often regarded as “microfungi” in that they do not generally form large fruit-bodies such as those characteristic of many brown and white rot fungi. The attack is much slower than in most brown and white rots and unlike these latter does not normally penetrate deeply, affecting mainly surface layers of the wood. Soft rot usually only becomes of major consequence in wood subject to high water content, such as the fill of water-cooling towers. As with brown rots, cellulose but not lignin is removed from the wood. Microscopically, the rot is very characteristic, especially in conifers where chains of spindle-shaped cavities are formed around hyphae growing through the secondary cell walls of the wood elements. In addition to the cell wall polymers lignin, cellulose and hemicellulose, wood may also contain less complex carbon compounds such as sugars and amino acids. These substances may be present, for example, in the contents of recently dead cells in the medullary rays, or they may be released during degradation of cell wall polymers by decay fungi, and can provide a carbon source for fungi unable to utilize cell wall materials. A variety of microfungi found in wood including mucoraceous species, fungi imperfecti and ascomycetes probably rely on such simple carbon sources (these are sometimes called “sugar fungi” after Garrett (1963)) causing little, if any, structural damage to the wood. They may, however, cause staining due to the pigmentation of their walls; blue-stain is a common fault in timber arising from colonization by these fungi. Whilst some staining fungi do not cause any real structural damage to timber, others, under suitable conditions, may be able to cause soft rot. C. SCOPE OF THE PRESENT ARTICLE

Wood may become decayed under a wide range of circumstances: within the living tree; after felling or pruning in dead trunks, logs, stumps and branches; and at almost any stage during its conversion and subsequent varied use by man. The process and rate of decay is influenced by innumerable factors including the type of wood (e.g. whether it is hardwood or softwood, sapwood or heartwood, living or dead), its moisture content, its nitrogen content, treatment with preservatives, temperature etc. It is by no means our intention in this article to cover completely so vast and varied a field, but rather to show how decaying wood may be used as a vehicle for understanding natural fungal populations and communities. To this end we will principally be concerned with decay in dead or moribund natural woody substrata such as stumps, logs, trunks and branches rather than in living trees or with

This Page Intentionally Left Blank

FUNGI IN DECAYING WOOD

337

timber converted for use by man and which may have been variously treated or preserved. Since white rots and brown rots come to predominate in such substrata, it is in the fungi causing these, rather than soft rots which we shall mainly be interested. Further, our discussion will inevitably be biased towards our own studies on hardwoods and will centre on decay of sapwood in those trees, such as oak, which have a morphologically distinct heartwood resistant to decay.

11. DIRECT METHODS O F ANALYSIS A. GENERAL

To study fungal population and community structure in decaying wood requires methods which detect the overall, three-dimensional distribution of mycelia and, as we have already indicated, standard procedures involving isolation from small samples of wood onto artificial media are inadequate for this purpose. This may at first seem an intractable problem, but it is easily overcome by using very direct methods of sampling and analysis. If a piece of decaying wood, such as a stump or log, is sawn through, examination of the cut surface often reveals a mosaic of decayed and discoloured regions (Fig. l), which may be separated from each other by narrow dark zones appearing as lines (so-called zone lines). In many instances this mosaic is a direct reflection of the spatial distribution of individual mycelia, and hence of the structure of the community or population within the wood, the zone lines being the interfaces between different fungal thalli. Individual decayed and discoloured regions can normally be shown, for example by examination of consecutive slices sawn from the wood, to be longitudinally continuous, and the fungi present in them can often be identified simply by correlation with the position of recognizable fruiting or mycelial structures present at the surface, or in some cases by characteristic features of the decay itself. An immediate impression of the fungal population or community present in a piece of wood may therefore, with a little experience, be obtained by direct field observation of the distribution of fruiting and other recognizable fungal structures and of the internal patterns of decay and discoloration present. This can be extremely useful since it allows general patterns to be detected (before their confirmation by more detailed analysis) or confirmed (after they Fig. 1 . Section across beech log occupied by a community of decay and staining fungi. Notice the strongly mosaiced appearance of the wood. Numbers indicate the following: (1) Wood occupied by Bjerkunderu udustu; (2) Intraspecific interaction zone between colonies of B. udustu; (3) Wood probably occupied by Melunommu pulvis-pyrius; (4) Intraspecific zone lines of M . pulvis-pyrius; (5) Sites of interaction between M . pulvis-pyrius and decay fungi; (6) Sites of replacement of M . pulvis-pyrius by decay fungi; (7) Wood occupied by Stereum hirsutum; (8) Intraspecific interaction zones between separate colonies of S. hirsutum; (9) Wood occupied by Coriolus versicolor.

338

A. D. M. RAYNER AND N. K. TODD

have been suggested by detailed analysis) directly in the field. However, for greater certainty, and a more precise understanding of the fungal community, more detailed laboratory analysis is required. We have found that procedures, such as the following, which employ direct incubation of woody samples provide a useful basis for appreciating fungal population and community structure. First of all samples of individual stumps, logs or branches obtained from the field are sawn into a series of slices (c. 2 cm thick is usually suitable). The position of regions of decay and discoloration as seen in two dimensions is then recorded from consecutive slices, and from this their three-dimensional distribution is ascertained. The fungi present in various regions are then identified by incubating the slices in moist conditions: a suitable procedure is to re-assemble the samples by placing the slices together in their original sequence and then to incubate them individually wrapped in moist newspaper, in large polythene bags. 10-14 days at 15-20°C is usually adequate as an initial incubation period, after which aerial mycelium of many fungi will have grown out and be identifiable: that which is not may be so after a further period of incubation. Part of the attraction of this procedure is that it very simply and rapidly enables identification of the three-dimensional distribution of fungal mycelia in decaying pieces of wood in a manner not possible using isolation from small samples. Further, since the fungi growing out have only the wood as a nutrient source, the chance of selecting out ecologically unimportant species (e.g. those present only as dormant propagules) which is always possible when using artificial media for isolation is avoided. However, it is still possible that at least some of the fungi observed using this technique may merely be contaminants, and that others, which do not grow freely out of the wood, but which are nevertheless important, may go undetected. The first possibility can often be eliminated by consideration of distributional patterns : fungi which occur in corresponding positions on consecutive slices are unlikely to be contaminants. Nevertheless, in the light of these possibilities it is always wise to supplement direct incubation by other methods of detection: these might include aseptic removal of small pieces (c. 10-50 mm3) from specific regions of wood followed by placing onto artificial media (3 % malt extract agar is very satisfactory), or incubation of surface-sterilized segments of wood in sterile, deep Petri dishes (Rayner, 1977a). A feature of direct incubation which may surprise some readers is our contention that many of the fungi can readily be identified on the basis of mycelial characters. Whilst it is self-evident that members of the fungi imperfecti may often be identified on the basis of their spore-bearing structures, it is frequently thought that vegetative mycelium of, for example, basidiomycetes (which of course include most of the major decay fungi) cannot so readily be identified. This is by no means the case; vegetative mycelium of these fungi as detected by incubation often exhibits a wide range of characters enabling its identification and which correlate well with those observed in

FUNGI IN DECAYING WOOD

339

cultures on malt agar. These may include gross morphological features such as colour, texture and consistency. For example, Phlebia merismoides Fr. produces for the most part a rather diffuse mycelium but which at the edges of decay zones from which it is growing forms a dense aggregate of feathery mycelium, often salmon pink in colour. In contrast Coriofus versicolor (L. ex Fr.) QuC1. usually produces a very tough white mat, parts of which may eventually turn brown, whilst Hyphofomafascicufure (Huds. ex Fr.) Kummer, produces a silky white, feathery mycelium, older parts of which are normally tinted sulphur yellow. Microscopically, the mycelium of different decay fungi is often very characteristic. Some may contain distinctive hyphal structures, for example species of Hypholomu have terminal inflated cells with short lateral projections, which stain deeply in cotton blue : these resemble small lizards when viewed microscopically, and indeed are termed as such by some. Other fungi may contain characteristic crystalline inclusions, e.g. Phallus impudicus (L.) Pers. Certain microscopically observable features which are of use for identifying mycelium of some of the more common fungi found in hardwoods such as beech, birch and oak are illustrated in Fig. 2. It is clear then, that the characteristics of vegetative mycelium produced a s a result of direct incubation can be used for identification. Further the fact that fruit-bodies are normally borne at the surface in positions corresponding to those of their underlying mycelia can help to increase confidence in identifications made of the latter. Also at any one site with a particular type of wood, it is likely that a relatively small number of species will predominate in the decay community. Once these have been characterized it will be possible to continue work at the site unhindered by problems of identification. B. RECOGNITION OF INTERACTIONAL PATTERNS

Different fungal mycelia will inevitably come into contact with one another as they grow through wood. Theoretically we might expect three possibilities : they may intermingle freely, one may dominate and grow through wood occupied by the other (replacement), or neither may be able to grow into wood occupied by the other (deadlock, mutual antagonism). Which of these types of interaction actually takes place in any given combination can often be ascertained during direct analysis procedures of the type described above. Intermingling is indicated when a mixed crop of different fungal mycelia are obtained from the same portion of wood, there being no obvious spatial separation into discrete zones. This type of situation is most often found a t early stages of colonization before decay becomes established and principally involves microfungi, causing staining of the wood. Deadlock or mutual antagonism is very common between different mycelia of decay fungi, including both brown and white rotting types. When it occurs discrete portions of wood are present, each of which is occupied by a single mycelium and separated from the others by abrupt demarcation zones. Often

Fig. 2

FUNGI IN DECAYING WOOD

34 1

the demarcation zones (which then fall into the general term zone lines) are narrow, darker and less decayed than the portions of wood they separate. They may occur either between different individual mycelia of the same species (see next section) or between mycelia of different species. They vary in appearance according to the species or species-combination involved; for example in the section of a decaying birch log, illustrated in Fig. 3, intraspecific zone lines formed in wood between different mycelia of the whiterotting polypore Coriolus versicolor, are narrower and paler than the interspecific zone lines formed as a result of mutual antagonism between C. versicolor and the pyrenomycete Hypoxylon muitiforme (Fr.) Fr. When sections of wood containing zone lines resulting from mutual antagonism between adjacent mycelia are incubated, it is commonly seen that the latter grow out in such a way as to “wall themselves off” from each other along the position of the zone line: each mycelium produces a tough, skinlike crust separating it from the other. This is often particularly clear with intraspecific interaction zones (e.g. Fig. 15). Another characteristic feature of these interaction zones is that they may themselves be colonized by other fungi-notably certain dematiaceous hyphomycetes (e.g. Rhinocludiellu spp.) which are not active in decay (Rayner, 1976): this will be referred to again later. Finally we may note that there is usually a reciprocal relationship between the extent of spread of mutually antagonistic mycelia in wood, tapering columns of decay occupied by the one being associated with broadening columns occupied by the other. Replacement is usually most easily detected b y making observations of fungal communities present in wood over a period of time, as will be discus-

Fig. 2. Hyphal characteristics in mycelium of basidiomycetes common in decaying hardwoods: A. Clavate terminal cells of Phlebia merismoides. (a, b, c) Relatively young cells with dense cytoplasmic contents and without superficial liquid globules attached. (d) Cell beginning to vacuolate and with a few superficial globules. (e) Cell with numerous liquid globules attached. B. Teased mycelium of Phunerochuete velutinu. The hyphae are very large, averaging c. 10pm diameter and vary in the thickness of their walls; presence of clamp connections-which may be absent, single, paired or multiple at each septum; presence of superficial deposit of crystals, which is often very extensive, totally obscuring the appearance of the underlying hypha. C. “Lizards” of Hypholoma fusciculare. Hyphae of this and other species of Hypholoma ( H . sublateritium, H. cupnoides) are normally fairly regularly clamped, 2-3 p m wide and at a later stage heavily encrusted with minute crystals. If a portion of the mycelium is mounted in a stain, such as cotton blue, the remarkable hyphal terminations (almost invisible without staining) become apparent. As shown, these take up stain more readily than other hyphal segments, and consist of sinuous, slightly inflated cells, with few to many short lateral protuberances, and appearing remarkably like small lizards. D. Teased mycelium of PhaNus impudicus.As shown the hyphae are very heterogeneous in appearance. Often fairly numerous narrow ( 2 4 p m diameter) hyphae occur fairly densely together, interspersed with more irregularly shaped and inflated cells. Numerous crystals are usually present, and are of two types. (a) Superficial, occurring on the surface of hyphae. These may be so numerous in places as to obscure, almost totally, underlying hyphal structure; (b) Intracellular, occurring as very large, radiately striate structures within markedly inflated hyphal segments.

Fig. 3. Section across birch log extensively colonized by Coriolus versicolor and Hypoxylon multiforme. Most of the wood is occupied by numerous colonies of C. versicolor separated by relatively narrow interactionzone lines (a). Regions of colonization by H. multiforme (b) are much less decayed, and delimited by broad, dark zone lines at sites of contact with C. versicolor.

FUNGI IN DECAYING WOOD

343

sed later; however, it is often possible to detect in an individual piece of decaying wood at any one point in time. Essentially one looks for evidence for the prior presence of a fungus in wood now occupied by another species. Such evidence can take a variety of forms: 1. When identifiable fruit-bodies at the surface correspond in position to portions of wood occupied by species other than those forming the fruit-bodies. 2. Where characteristic zone lines are present in wood occupied by species different to those forming the lines. In some cases where slow replacement occurs, a succession of zone lines may be produced by the retreating species, interrupting the advance of the dominant species. Some examples of this pattern of replacement can be seen in Fig. 1. 3. When one fungus is present only at the periphery of decay columns, the proximal portions being occupied b y some other species. This information can only be ascertained if the three-dimensional distribution of mycelia within the wood is understood. It is often possible to ascertain the direction of replacement by this means. It is clear then that direct analysis of the fungal population and community pattern in a sample of decaying wood at any particular time can in itself tell us much about the interactional forces which are instrumental in developing that pattern. Whilst this is so our understanding is enhanced further if we undertake our studies over a period of time: we can then be much more definite about the changes taking place in fungal community composition and pattern, and about the occurrence of various types of interaction, especially replacement. This type of study is still very rare with fungi : whilst there are numerous examples of supposedly successional investigations on a variety of substrata, these have largely been concerned merely with detecting broad changes in the floristic composition of fungal communities, no attempt being made to examine changes in their detailed structure. The approach required here is to set up large-scale, highly replicated experiments from which whole stumps, logs or similar substrata are sampled at intervals and the internal composition of fungal communities present analysed. If the level of replication is sufficiently high the effects of various treatments on the development of fungal communities can be studied. In some cases deliberate inoculation of selected fungi at the outset of an experiment can also provide interesting information, as will be described later. An example of this type of study is to be found in recent work on fungal colonization of hardwood stumps (Rayner, 1975, 1977a, b). C. ZONE LINES

Mention has already been made of the occurrence of zone lines in decaying wood, i.e. narrow dark zones appearing as lines in transverse section, and of their use in interpreting patterns of population and community structure.

344

A. D. M. RAYNER AND N. K. TODD

Apart from being amongst the most obvious, zone lines are also amongst the least understood of features in decaying wood. This may partly result from a tendency amongst many to assume a common basis for their occurrence: in fact there is probably quite a wide variety of causes, some of which may not yet have been discovered. There is also a tendency to believe that only certain decay fungi-Armillaria mellea (Vahl ex Fr.) Kummer, and various Xylariaceous ascomycetes (e.g. Xylaria polmorpha (Pers.) Grev., Ustulina deusta Fr.) produce zone lines; in fact we believe that the majority, if not all, decay fungi produce zone lines of one type or another. In many instances zone lines are potentially useful for diagnostic purposes (cf. Hubert, 1924), but as yet there is too little published work concerning their characteristics, both macroscopically and microscopically observable, to make this viable in all but a few cases. Zone lines, are then, central to interpreting population and community structure in wood, and it is essential that we understand their nature and origin. However, since the classic pioneering studies by Campbell (1933, 1934) reports of work on zone lines have been sparse and are diffusely scattered through the literature, there being few attempts to integrate and collate existing information except perhaps to some extent by Lopez-Real (1975). This is clearly a field in which there is still much confusion and, therefore, scope for much more detailed and directed research. In our experience there seem to be two basic types of zone line, differing in their modes of origin: (a) Those resultingfrom the action of a single mycelium. The most frequently accepted explanation for zone lines is that in the course of its normal growth through wood a single mycelium lays down sheets of dark mycelium which appear as lines when viewed in section. This was the type encountered by Campbell (1933, 1934), during his classic studies with Armillaria mellea and Xylaria polymorpha. With these species the sheets were found to contain numerous dark, inflated bladder-like cells and were considered by Campbell to represent the outer rinds of pseudosclerotia (i.e. discrete structures containing an inner loosely interwoven mycelium surrounded by a tough outer rind) buried within the wood. This concept has been accepted by Lopez-Real (1975) who suggested that the term pseudosclerotial plate (PSP) be used in preference to zone line. We do not accept this entirely since whilst the term PSP is useful to describe simple, single sheets of mycelium (and we shall ourselves use it for such purposes subsequently), many zone lines do not have such a simple structure. A feature of zone lines which do consist of PSPs produced by a single mycelium is that they do not normally delineate extensive longitudinally continuous columns of decay. Rather they tend to surround numerous adjacent pockets of decay each of which is normally inextensive longitudinally, as with A. mellea and Xylaria spp. or they are often adjacent to exposed surfaces of decaying wood. In some cases, especially with A. mellea in well-decayed wood, the PSPs can easily be dissected away as whole sheets of mycelium separate from the woody tissue. A variety of

FUNGI IN DECAYING WOOD

345

physical factors including desiccation, fluctuating moisture content and gaseous composition of the wood have been mooted as causing PSP formation by single mycelia. Some interesting recent work investigating these possibilities has been based on the fact that PSPs form readily behind freshly cut surfaces in decaying wood (Lopez-Real and Swift, 1975, 1977). A totally different way in which a single mycelium may give rise to zone line formation occurs with certain parasitic decay fungi colonizing living wood, e.g., in Ganoderma adspersum (Schultz) Donk. Here a dark brown band c. 5 mm wide is usually present marking the extreme limit of advance of the fungus (Cartwright and Findlay, 1958). This band is probably produced as a host reaction and, microscopically can be seen to consist of a layer of cells in which tyloses are abundant and wound gum is present in varying amounts. (6)Those resulting from the interaction of di3erent mycelia. This type of zone line, which is very common, has already been mentioned and some of its characteristic features discussed. It is a curious fact that whilst the interactive basis of many types of zone line was clearly understood quite early on (Hubert, 1924), this seems to have attracted relatively little attention. Most of the early work on interaction zone lines indicated that they were formed between different species of decay fungi-i.e., that they were interspecific (Campbell, 1933). That many interaction zone lines actually result from intraspecific antagonism between mycelia of the same species appears only recently to have been substantiated (Rayner, 1976; Rayner and Todd, 1977), although there have been occasional hints towards this possibility (Campbell, 1938 ; Childs, 1963; Adams and Roth, 1967). The curious thing here is that intraspecific interaction zone lines are probably amongst the commonest types and are readily observed with some of the most widely occurring and extensively investigated of decay fungi such as Coriolus versicolor and Stereum hirsutum (Willd. ex Fr.) S . F. Gray. Both inter- and intraspecific interaction zone lines often delimit longitudinally continuous and extensive decay columns. Whilst pseudosclerotial plates may be present, these are often double, each antagonistic mycelium being delineated by its own PSP. In some cases, for example in intraspecific zone lines in Phlebia merismoides and Hypholoma fasciculare, PSP formation as such probably does not take place. Zone lines resulting from mutual antagonism are often hard, the wood being relatively undecayed-perhaps because the antagonism results in local disruption of the capacity of enzymes of the mycelium to degrade wood polymers.

346

A. D. M. RAYNER AND N . K. TODD

111. INTRASPECIFIC ANTAGONISM : THE DELIMITATION OF INDIVIDUAL MYCELIA A. GENERAL

The term “population”, which may be defined as an aggregate of individuals of the same species considered together because of some property or properties they share in common, is an essential unit for assessing biological significance in any group of organisms. Population units can be defined on any level appropriate to the problem in hand and given such geographical and ecological significance as may be deemed necessary. In sexual organisms the unit of greatest importance is the breeding population which can be defined as that group of individuals which, because of their proximity in space and time, are potentially capable of interbreeding. In higher plants and animals, populations and their constituent individuals are generally easily recognizable, but with fungi this is more difficult. In part this is due to the complexity of mycelial growth which in nature results in a three-dimensional network of hyphae involving many orders of branching, and the occurrence, in higher fungi, of anastomoses between hyphae either within a mycelium, or between different mycelia, which may obscure the distinction between individuals, and hence the structure of populations. Perhaps as a consequence of these difficulties, little information is available concerning the population biology of fungi. In many mycological studies, the need to distinguish between different isolates of the same species inevitably arises and in this connection, for higher fungi, there are numerous, but widely scattered, reports of the existence of antagonistic influences between mycelia, often genetically distinct, of the same species. This antagonism has been observed as so-called “barrages”, “aversion phenomena”, “interaction zones” or “demarcation zones” between colonies placed opposite each other in culture. Commonly this occurs in heterothallic species (in which the homokaryon is self-sterile but crossfertile) between homokaryons, and is related either to homo- or heterogenic incompatibility (see Raper, 1966; Esser and Blaich, 1973). Antagonism has been observed in culture during heterokaryon formation (the bringing together of unlike nuclei into the same cytoplasm) in many ascomycetes where it is known as vegetative or heterokaryon incompatibility. There are many examples of this phenomenon, including the extensive work on Aspergillus by the Birmingham group used here as illustration (for a recent survey of their findings see Croft and Jinks, 1977). It was found that wild isolates of A . niduluns Eidam collected from a variety of localities throughout England and Wales, could be divided into a number of heterokaryon compatibility (h-c) groups so that all members of any one group

FUNGI IN DECAYING WOOD

347

readily formed heterokaryons with each other but not with members of any other group. To date they have identified 19 such h-c groups. There is more variation in cultural characteristics between than within h-c groups : Grindle (1963, a, b) noted that members of one h-c group tended to have similar general morphology and demonstrated intra-group similarity and intergroup difference with respect to radial growth rate. It is difficult to conclude much about their geographical distribution since certain h-c groups are very widely scattered and, often, members of more than one group are found in any one area, sometimes from the same soil sample. It has also been shown that in this fungus heterokaryon formation is under nuclear control and that this is heterogenic in nature; that is, heterokaryon incompatibility results from heterozygosity at one or more series of incompatibility loci. Croft and Jinks (1977) conclude that in the wild A . nidulans is divided into a number of sub-populations, each of which is a clonally related group of strains in which evolution may be proceeding independently, little gene exchange occurring between groups. They are, however, unclear about the role of heterokaryon incompatibility in maintaining these divisions. One possibly significant observation has been made in this direction by Caten (1972) who demonstrated that heterokaryon incompatibility acted as a barrier to virus infection in A . amstelodami (Mangin) Thom & Church, and has suggested that this form of incompatibility may be important as a cellular defence mechanism. A similar system of vegetative incompatibility has also been found in another ascomycete fungus, Endothia parasitica (Murr.) A. & A. which causes a serious canker disease of chestnut (e.g., Anagnostakis, 1977). It has been shown that the pathogenicity of certain strains can be overcome by the transfer of cytoplasmic determinants, via hyphal anastomosis from a nonpathogenic strain, and this method has been adopted as a possible control for the disease. The occurrence of vegetative incompatibility in this fungus is therefore of great concern in its potential for restricting transmission of cytoplasm between strains. So far, 28 compatibility groups have been identified and it is suggested that this is governed by a heterogenic system. As mentioned earlier, heterokaryon incompatibility in Aspergillus effectively prevents exchange of cell contents between members of different h-c groups. However, in E. parasitica there is some evidence for the system being slightly “leaky”, allowing some exchange in the host between strains of different compatibility groups. It has often been assumed in homothallic species (in which the homokaryon is self-fertile) such as A . nidulans and E. parasitica, that heterokaryon formation is a necessary prerequisite for outcrossing via parasexual or sexual reproduction. Following on from this it was suggested that vegetative incompatibility was by necessity a barrier to this occurring. However, it has been found that in both these fungi outcrossing still occurs irrespective of

348

A. D. M. RAYNER AND N. K. TODD

vegetative incompatibility (Butcher, 1968 ; Anagnostakis, 1977) being effected by fertilization of trichogynes of one strain by spermatia of another, a process requiring only very limited exchange prior to sexual fusion. Turning now to wood-decaying basidiomycetes, there are several reports of antagonism between paired isolates, both mono- and dikaryotic, of different origin in culture and here also heterogenic incompatibility has been implicated. This phenomenon has again been used by various workers as a guide to identifying genetically distinct isolates from nature and amongst some of the earlier reports are those of Mounce (1929) for Fomes pinicola (Swartz) Cke. and Campbell (1938) for Oudemansiella radicata (Relhan ex Fr.) Sing. In the literature the distinct mycelia identified by antagonism as being members of one group have been variously referred to as constituting “clones”, “strains” or having the same “compatibility genotype”. Childs (1963) used antagonism to determine whether isolates of Poria weirii Murr., a fungus causing root-rot in conifer species, from different trees belonged to the same or different “clones”. By “cross-plating” he found that a particular local sample consisted of many clones, and that each clone generally occupied at least several adjacent trees. Earlier, Childs (1937) reported antagonism between isolates of Phaeolus schweinitzii (Fr.) Pat. and Barrett and Uscuplic (1971) used this as a basis for studying the field distribution of different strains of this fungus in conifer plantations. By pairing wild isolates they showed that almost all trees sampled contained a different strain, but that there was never more than one strain in an individual tree, and that all isolates of a single strain were morphologically similar. In addition they carried out pairings between dikaryons synthesized from monokaryons and found that the intensity of interaction, as assessed by the density and width of the interaction zone, increased as the relatedness between dikaryons decreased. Adams and Roth (1967, 1969) investigating Fomes cajanderi Karst., did find more than one antagonistic isolate per tree, but unfortunately from the point of view which we shall shortly be discussing, they did not sufficiently correlate their field sampling procedures with their genetic studies in culture; these latter were carried out with four isolates derived from heartwood of two glazed-damaged Douglas firs, which were obtained without reference to their position in decay columns. By using pairings between synthesized dikaryons it was shown here also that the frequency of demarcation line formation was a predictable event when the relationship between colonies of a pair is known. They arrived at the same conclusion as Barrett and Uscuplic in that the more distantly related strains were, the more intense the antagonism, and concluded that antagonism from paired colonies collected from the field is evidence of genetic differences. In a recent publication (Adams, 1974), “clones” of A . mellea in young-growth Ponderosa pine have been identified on the basis of antagonism in paired culture tests. A total of 243 isolates of

FUNGI IN DECAYING WOOD

349

this fungus were collected from a field site and found to belong to three readily distinguishable clones. Adams found that members of each clone were very similar morphologically, an observation which contrasts with the dissimilar appearance within compatible lines found among isolates of F. cujunderi (Adams and Roth, 1969). Whilst intraspecific antagonism has therefore been used as a means for detecting differences between isolates obtained from the field, its significance as a mechanism operating to delimit individuals in nature has not until recently been fully appreciated. It is perhaps for this reason that some of the better known “population” studies with higher fungi have used criteria other than intraspecific antagonism. Notable amongst these is the work of Burnett and his colleagues who used the distribution of mating-type factors as a guide to studying the nature and significance of the mycelium within natural populations. Burnett and Partington (1 957) determined the incompatibility factors of closely associated individual fruit-bodies of five different basidiomycetes. In three of these species, Coprinus comatus (Mull. ex Fr.) S.F. Gray, Flammulina velutipes (Curt. ex Fr.) Karst. and Hypholoma fasciculare, the same incompatibility factors were recovered from all fruitbodies from each specific location. Similar findings were later obtained for fruit-bodies of Marusmius oreades (Bolt. ex Fr.) Fr. within individual fairy rings by Burnett and Evans (1966). In all these cases it was therefore concluded that the numerous fruit-bodies occurring at a specific location each originated from single extensive dikaryons. Such was not the case, however, for the remaining two species investigated by Burnett and Partington, Piptoporus betulinus (Bull ex Fr.) Karst. and Coriolus versicolor both of which cause wood-decay. With P . betulinus individual fruit-bodies from single birch tree trunks and collected over a period of years were obviously derived from different dikaryons but these latter contained mating-type factors in common. Similarly, adjacent fruit-bodies of Coriolus versicolor growing on a stump contained different permutations and combinations of the same mating-type factors. Unfortunately, they did not investigate the detailed structure of the mycelium within the wood, the essential step, we claim, for identifying the components of natural populations. These findings obtained with P . betulinus and C. versicolor have repeatedly been interpreted by Burnett (e.g., Burnett ,1976) as providing evidence for the existence in nature of a mycelium acting physiologically and ecologically as a unit, whilst being genetically heterogeneous, the so-called “unit mycelium”. This interpretation was based largely on the knowledge for higher fungi that there is generally little restriction on hyphal anastomosis in nature; indeed such fusion has even been observed at the intergeneric level (Kohler, 1929, 1930). Observations made with C. versicolor and P . betulinus gave Burnett no reason to believe that there was any restriction on anastomosis in these species. This intriguing concept of the unit mycelium would distinguish the

350

A. D. M.

RAYNER AND N. K. TODD

fungi from all other organisms, in which separate individuals in close contact inevitably compete and render the terms “individual” and “population” difficult or impossible to apply. Its far reaching implications affect our understanding of the behaviour of fungal mycelia in nature including, for example, the processes involved in the colonization and exploitation of a resource, the opportunities for recombination of genetic material and the effect of natural selection which would now operate on a group rather than the individuals themselves. In spite of its deep implications, little work has been carried out to establish the extent to which such fusions may occur in nature. The classic study in connection with the unit mycelium concept is that of Buller (1931) who argued that the size of fruit-bodies produced by Coprinus sterquilinus (Fr.) Fr. on dung balls could not be accounted for unless fusion occurred between numerous colonies derived from separate basidiospores. Buller claimed that the spores of the fungus, which are embedded in the dung ball and germinate in sifu, will a t first develop into numerous separate colonies and initially compete for food reserves. If this were to continue no one colony would obtain sufficient material to form a complete sporophore, but fusion between mycelia then occurs producing a single physiological unit capable of producing sporophores. From examination of mycelia in culture it was found that when young mycelium derived from a single spore grows older the hyphal fusions within it become increasingly numerous until at length the ,whole mycelium is converted into a closely woven three-dimensional network. Buller observed hyphal fusions within and between several mycelia of C. sterquilinus and many other coprini, confirming earlier reports (Brefeld, 1877; Falck and Falck, 1924). Interestingly enough, Buller draws the analogy between the compound (unit) mycelium and the social organization of many insects, such as the hive-bee, and further suggests that a unique level of social organization exists in many populations of higher fungi. Largely on the basis of the work of Buller and Burnett, and knowledge of hyphal anastomosis, the concept of the unit mycelium has achieved widespread acceptance amongst mycologists and has coloured much of the discussion about the nature and significance of mycelia within natural populations. Bearing this in mind, it was surprising to find (Rayner, 1976) that, when cutting through white rotted wood extensively colonized by Coriolus versicolor, a network of narrow, dark, relatively undecayed zones was present delimiting decay columns corresponding in position to separate groups of fruit-bodies at the surface (Fig. 4). With subsequent work (Rayner and Todd, 1977; Todd and Rayner, 1978) using field sampling procedures of the type illustrated in Fig. 5, we have demonstrated that when opposed in culture, dikaryotic isolates from wood or fruit-body tissue corresponding to separate decay columns were invariably antagonistic whilst those from the same column merged imperceptibly (Table I). We noted that the antagonistic response in culture, namely a clear zone with sparse hyphae developing

FUNGI IN DECAYING WOOD

351

Fig. 4. Surface and underlying wood of a portion of a birch stump colonized by Coriolus versicolor. (A) Surface with numerous polymorphic fruit-bodies. (B) Underlying wood with large bleached regions separated by narrow dark interaction zones (arrowed).

between mycelia and usually accompanied by brown pigment (Fig. 7), closely resembled the interaction zone (zone line) seen in the wood. The interaction paralleled those observations of intraspecific antagonism mentioned above, although at the time of first observation we were unaware of these other studies. We next demonstrated that each decay column contained a genetically homogeneous dikaryon using dedikaryotization (the purpose of which was to resolve dikaryons into their monokaryotic components, see Todd and Rayner (1978)) (Table 11) and other methods, and that the monokaryons derived from basidiospores of fruit-bodies corresponding to separate decay columns were interfertile (i.e., they produced dikaryons with clamp connections when paired in culture), thus demonstrating the intraspecific nature of the antagonism. These findings differ markedly from the nature of the mycelium of this fungus in the wild advocated by Burnett (e.g., Burnett, 1976). From such detailed analysis of the three-dimensional structure in the wood of mycelia of C. versicolor and also Bjerkandera adusta (unpublished) we have suggested

3 52

A. D. M. RAYNER AND N. K. TODD

Fig. 5. Diagram illustrating procedure for analysis of a population of Coriolus versicolor in a stump. The stump is first cut into transverse slices as indicated by the dotted lines. The position of the different decay columns, delimited by narrow dark zones, on separate slices is noted and correlated with that of the fruit-bodies at the surface. The decay columns and their corresponding fruit-bodies are labelled accordingly, and isolates made from them as follows: (i) from fruit-body tissue; (ii) from single basidiospores from fruit-body used for tissue isolation; (iii) from the decayed wood at different levels (after Todd and Rayner, 1978).

that in nature the antagonism is a vegetative mechanism operating to delimit individuals within freely interbreeding populations. If, as we suspect, intraspecific antagonism or similar mechanisms operating to prevent efective mycelial fusion, are general in fungi between genetically distinct mycelia, then this throws into question the unit mycelium concept and has significant consequences for the role of antagonistic phenomena in delimiting individuals and hence determining and maintaining natural population structure. In addition to the work with C. versicolor and B. adusta, both heterothallic species, we have carried out similar detailed analysis of the three-dimensional structure in wood of a population of Stereum hirsutum which is homothallic

353

FUNGI IN DECAYING WOOD

TABLE I Pairings between wood and fruit-body tissue isolates of Coriolus versicolor from different decay columns in a stump (a)Pairings between wood isolates Y2 X1 01 N1 L1 J1 H1 G1 C1 A1

A1

O

A

a

A

O

a

O

0

0

c1

A

a

a

n

n

A

n

O

.

GI HI

n L a

a A n

a A a

0 a o

0 n 0 o O . n .

.

A O

A A

A A

A

n

a

J1 L1 N1 01

x1

C l xC 2. G1 x G 2 0

H1 x H 2 0 J1 x J2 N1 x N 2 0 01 x 0 2 . Xl x X 2 .

O

.

.

0

( b ) Pairings between fruit-body isolates

~~~

Al x A 2 0

a .

Y2

A C F H N 0

.

O

N

H

a

o

o

a

F

n

C

o

a

o

a

.

a

a

n

.

(c) Pairings between fruit-body and wood isolates fruit-body wood

A

A C H 0 N A C H 0 C

.

o a . A . 0

~-

~

x x

A1

0 0

x

0

X

H1 01 N1 J1 01 J1

X

N1

A

x

L1

n

x x

x x

c1

0 0

A n

A

~~

Letters indicate the decay column and numbers 1 and 2 refer to different levels within a decay column. 0, Antagonism; A , antagonism accompanied by pigment production; 0, complete fusion of isolates. After Rayner and Todd (1977).

and exhibits a nuclear behaviour pattern known as holocoenocytic*. Here again narrow dark interaction zones delimit separate, mutually antagonistic mycelia occupying discrete decay columns in the wood. The main difference from heterothallic species is that antagonism is expressed directly, without the need for formation of a secondary mycelial phase following conjugation, between monospore isolates. Often antagonism can readily be observed between monospore isolates from the same fruit-body (Fig. 6 ) . Other fungi in which antagonistic influences have been observed in culture between isolates derived from opposite sides of narrow dark zones in wood

* This pattern of behaviour occurs when a single basidiospore germinates to produce a mycelium with plurinucleate hyphae segments. This coenocytic state is often maintained throughout the life cycle and extends into the fruit-body, where in extreme circumstances, the basidiole alone is binucleate (Boidin, 1971). This behaviour is frequently associated with the occurrence of multiple clamps at the septa on some of the hyphae (e.g. Fig. 2 B).

3 54

A. D. M. RAYNER AND N. K. TODD

TABLE I1 Number of compatible pairings between monokaryons derived by dedikaryotization, and fester strains, in Coriolus versicolor

Tester strains Dedikaryotized isolates

Total number of derived 7 monokaryons tested 03

\

04

06

011

Ofb

31 35 26

0 0 0

0 2 11

0 0 0

31 33 15

N1 N2 Nfb

28 47 20

N8 0 0 0

N4 0 0 0

N36 0 0 0

N31 28 47 20

A1 Afb

35 52

A29 0 0

A36 0 52

A26 0 0

A23 35 0

Ff b

10

F2 0

F4 8

F40 0

F38 2

J1

23

J3 0

56 0

J13 0

J19 23

01 02

For dedikaryotized isolates letters indicate the decay column, numbers 1 and 2 refer to different levels within a decay column and f b indicates that the isolate was derived from fruit-body tissue. For tester strains letters indicate decay column and numbers are isolate numbers. Tester strains for each column are monokaryotic basidiospore isolates each representing one of the four possible mating-types (after Todd and Rayner, 1978).

include Hypholoma fasciculare, Phlebia merismoides and the ascomycete Daldinia concentrica (Bolt.) Ces. & de Not. We feel sure that the list will be extended with further study. The expression of antagonism between paired cultures on agar varies with different species of decay fungi (Fig. 7). Most often a narrow clear zone of demarcation develops between the colonies (with P . merismoides sufficient time for development of aerial mycelium must be allowed before this becomes apparent) which subsequently may, if the interaction is intense, become pigmented (colours according to Rayner, (1970) : sepia with C. versicolor, orange to brick with P . merismoides, luteous with H . fasciculare, ochreous with B. adusta and various shades of yellow with S. hirsutum). Apart from pairing isolates in culture, which must be the ultimate test for intraspecific antagonism, the phenomenon can often be detected readily by more direct methods. In the field these can include observation of mycelial' or fruiting structures and their correlation with zone line patterns in the wood. For example, Fig. 8 shows two subcortical mycelial sheets of Hypholoma

FUNGI IN DECAYING WOOD

355

Fig. 6. Petri dish on to which several monospore isolates of Sfereurn hirsutum have been placed. Sites of antagonism between the resulting colonies are arrowed.

fusciculure growing on an oak log. At one point there is a clear zone of antagonism between these mycelia, and this correlates with a narrow, dark, relatively undecayed zone in the wood. For species such as Stereum hirsutum which form resupinate fruit-bodies in suitable circumstances, it can often be seen that whilst sometimes several different sporophores may fuse to form a single structure, on other occasions barrage lines form between them where fusion has not occurred (Fig. 9). Where barrages do occur between fruitbodies, it is often possible to correlate their position with that of narrow dark zones in the wood below (Fig. 10). Finally in numerous examples where we have examined decaying wood colonized by C. versicoior we have found no exception to the general rule that where fruit-bodies differing in appearance are present at the surface, interaction zones occur in the wood delimiting individual decay columns equivalent in number and position to the various morphological types. In contrast, where uniform collections of fruit-bodies occur, few, if any, interaction zones are likely to be found, the wood normally being uniformly decayed. These features are clearly seen in Figs 11-13. In the case of C. versicolor, a fungus noted for its polymorphism, we have found that on a particular substratum considerable variation in morphology often .exists between groups of fruit-bodies more or less identical in appearance (Fig. 12), and in a recent paper (Rayner and Todd, 1978) we provide strong

Fig. 7. Intraspecific antagonism exhibited by isolates of different decay fungi when opposed in culture. (a) Coriolus versicolor. (b) Phlebia merismoides. (c) Stereum hirsutum. (d) Hypholoma fasciculare. (e) Bjerkandera adusta.

FUNGI IN DECAYING WOOD

357

Fig. 8. Bark (right) and underlying wood of an oak log colonized by Hypholomu fusciculure. Two extensive subcortical mycelial sheets are present (most noticeable on the bark) which have failed to merge along the line arrowed. At this point a narrow dark zone is present in the decayed wood.

evidence that the polymorphism exhibited by this fungus in nature largely reflects genetic differences between dikaryons occupying different decay columns. From this and work with other fungi it is becoming apparent that genetic differences between individuals are often externally manifest by the polymorphism between their fruit-bodies and therefore an estimate of the number of morphological types present can act as a general guide to the number of individuals. The technique of direct incubation of wood sections (see Section 11) also provides a useful basis for assessing antagonism in situ between individual mycelia of a single fungus species. Figure 14(a) shows a transverse section through a beech log, colonized largely by Bjerkandera adusta, before incubation : several zone lines are clearly visible. Following incubation (Fig. 14b) aerial mycelium has grown out and the extent of each individual colony can readily be seen. By comparing the “before” and “after” incubation photographs the pattern produced by the mycelia can be seen to correlate well with the position of underlying zone lines. If a series of sections is analysed in this way the threedimensional distribution of individual decay columns can, as noted in Section 11, rapidly be ascertained. In the case of Fig. 15, which shows an incubated wood section containing Stereurn hirsutum, each individual mycelium can be seen to be bounded by a pseudosclerotial plate in the region of interaction, so that neighbouring colonies fit together like pieces of a jigsaw.

Fig. 9. Resupinate fruit-bodies of Stereum hirsutum on a birch log. Certain columns of fruit-bodies have fused up almost imperceptibly, but between these are sites where there is no such fusion (arrowed).

FUNGI IN DECAYING WOOD

359

Fig. 10. Section through a portion of the log shown in Fig. 9 showing correspondence in position of barrage zone between fruit-bodies (a) and narrow dark zone in underlying wood (b). B. BASIS OF ANTAGONISM

Recently, we have been concerned to discover the underlying genetic and physiological bases of intraspecific antagonism in Coriolus versicolor and several other common wood-decaying basidiomycetes. Whilst there is limited information of this type for certain ascomycetes (especially in relation to heterogenic incompatibility) there is virtually none for basidiomycetes (see Esser and Blaich, 1973). Apart from our own work, the only two reasonably detailed genetical studies on antagonism that we know of have been those of Adams and Roth (1967, 1969) and Barrett and Uscuplic (1971) on Fomes cajanderi and Phaeolus schweinitzii respectively. Other than a few incidental observations, no work has been carried out on physiological aspects. For the purpose of the present article we will concentrate mainly on our results obtained with C. versicolor".

* The strains of C. versicolor used were derived from a population obtained from six adjacent decay columns occupying a single birch stump, details of which have been presented previously (Rayner and Todd, 1977). For each of the six decay columns, four monokaryotic tester strains, each representing one of the four possible mating types (C. versicolor is tetrapolar), were obtained from basidiospore isolates of fruit-bodies corresponding to each column, and for coding the letters indicate the decay column and the numbers the isolate number.

Fig. 11. (a) Birch log bearing numerous, morphologically uniform sporophores of Coriolus versicolor. (b) T.S. showing almost uniform white rot. A few dark zones are present (arrowed) but these were either associated with the presence of Stereum hirsutum, or were

discontinuous indicating that they were merely relics (after Rayner and Todd, 1978).

Fig. 12. A. Top of birch stump bearing clumps of C . versicolor sporophores. Within each group the sporophores are of uniform appearance, but there is considerable variation in morphology between groups. Arrowed sporophores are of Bjerkanderu udusta. B. Underlying wood showing several decay regions separated by narrow dark zones arrowed. The position of the larger decay regions could readily be correlated with that of different groups of sporophores at the surface (after Rayner and Todd, 1978).

Fig. 13. A. Top of birch stump bearing numerous C. versicolor sporophores. There is very considerable variation in morphology between the fruit-bodies, many of which are relatively small. B. Underlying wood containing very numerous, small decay columns separated by interaction zones (after Rayner and Todd, 1978).

Fig. 14. Section through beech log extensively colonized by B’erkunderu udustu. (a) Before incubation. (b) After incubation. Notice correspondence between position of zone lines in wood and antagonism between aerial mycelium (arrowed).

364

A. D. M. RAYNER AND N . K. TODD

Fig. 15. Section through oak log, extensively colonized by Stereum hirsutum, after incubation. Notice the crust-like areas (pseudosclerotial plates) formed between the interacting mycelia.

I . Genetical Basis This has been studied by means of interactions using dikaryons of known constitution and by mating tests, both between monokaryons, and between dikaryons and monokaryons (di-mon matings). The experimental details may be obtained from Todd and Rayner (1978). The results of pairings between synthesized dikaryons are summarized in Table 111. Antagonism occurred in all instances (except in control pairings between the same dikaryons, these fusing readily), but there was a marked tendency for decreased intensity of interaction, associated with decreased pigment production, with increased relatedness. That the antagonism may be regarded as a dikaryotic phenomenon was made obvious by its occurrence in pairings between dikaryons in which all the monokaryotic components were fully compatible. In pairings between non-sibcomposed dikaryons, pigment production varied according to the type of pairing. Pairings in which all the components were of different origin (type A.a) showed a greater incidence of pigment production (80 %) than those involving either common, or sibrelated monokaryotic components. In pairings between sibcomposed dikaryons (type B.a), as previously mentioned, antagonism occurred in all pairings with the incidence of pigment production being 78%, a value corresponding closely to that obtained in type A.a pairings. Antagonism occurred in all pairings between sibcomposed dikaryons from the same decay column (type B.b)

365

FUNGI IN DECAYING WOOD

TABLE I11 Occurrence of antagonism in pairings between synthesized dikaryons of

Coriolus versicolor No. of

Pairing type

pairings

No.

180

No. 0

%A

144

36

80.0

79

48

31

60.8

160

104

56

65.0

13

8

5

61.5

11

78-0

64

4.5

A

(A) Dikaryons not sibcomposed

(a) all monokaryotic components from different decay columns (b) one pair monokaryotic components in common (c) one pair monokaryotic components sibrelated (d) two pairs monokaryotic components sibrelated

(B) Dikaryons sibcomposed (a) dikaryons from different 50 39 decay column (b) dikaryons from same decay 67 3 column Symbols as in Table I. After Todd and Raper (1978).

regardless of whether the component nuclei were of identical mating type, and regardless of relatedness of dikaryons, demonstrating that antagonism between dikaryons is independent of the homogenic incompatibility mechanism. However, there was striking evidence for reduced occurrence of pigment production in these pairings (4.5 "/o) when compared with those involving less closely related dikaryons. In certain cases, particularly in pairings involving dikaryons with nuclei in common, only rather faint interactions, but which were nonetheless present, were observed. Almost identical results (unpublished) to the above have been obtained from pairings between synthesized dikaryons of Bjerkandera adustu, which is bipolar. Antagonism was again shown to be principally a dikaryotic phenomenon independent of homogenic incompatibility mechanisms, since it was observed in all pairings between genetically different dikaryons regardless of origin. Also, the degree of relatedness affected the intensity of the interaction, closely related dikaryons being less strongly antagonistic than others. Therefore, in both fungi antagonism is inevitable between genetically distinct mycelia, but the intensity diminishes with increased relatedness-findings which agree with Adams and Roth (1967) and Barrett and Uscuplic (1971). To further the genetic information on this phenomenon in C. versicolor a series of di-mon matings were carried out to establish the nature of the resulting interaction in such pairings and to ascertain whether the dikaryon was able to dikaryotize the monokaryon (a process referred to as the Buller Phenomenon, see Raper, 1966). The results obtained are summarized in

366

A. D. M. RAYNER AND N. K. TODD

Table 1V and some examples of the different types of pairings shown in Fig. 16. In matings between non-sibcomposed dikaryons and non-component non-sibrelated monokaryons (type A pairings) antagonism was observed in all cases, and in the majority (71.2%) was accompanied by pigment production. In most cases this antagonism resembled that observed between dikaryons (Fig. 16 (a)) but it was not always clear whether the interaction finally observed resulted directly from the confrontation between monokaryon and dikaryon or whether it developed subsequent to dikaryotization. In many type A pairings, whilst typical antagonism was observed in distal portions along the line of contact between the colonies, at the initial point of contact mycelial build-up occurred instead (Fig. 16 (b)). In the majority of pairings, irrespective of the intensity of the interaction, the monokaryon became dikaryotized. There was evidence, however, that two of the monokaryons tested, A26 and especially C1, persisted as monokaryons when confronted with a theoretically compatible dikaryon. In several type A pairings we observed what we have called “trackformation”. Here the process of dikaryotization appeared to be such that different dikaryotic sectors occurred, separated from each other by narrow lines of antagonism (“tracks”); these can easily be seen in Fig. 16 (b) and (e). On subculturing an isolate which crosses several tracks, such as the dotted portion on Fig. 16 (e), the pattern of tracking is maintained in the newlygrown mycelia (Fig. 16 (f)). This observation would be expected if both nuclei of the opposing dikaryon are able to dikaryotize the monokaryon, a point which will be referred to again later. Matings between both non-sibcomposed and sibcomposed dikaryons and their corresponding monokaryotic components (type B pairings) showed antagonism in virtually all pairings, at least initially (i.e. within 3-4 days at 18-20°C). However, in many cases this was superseded by complete fusion of the isolates. In these instances traces of the original antagonistic interaction were still discernible on the plates from below, In other cases definite zones of antagonism developed, which were persistent, sometimes accompanied by pigment production. Even here, however, the antagonism was only clear along distal portions of the interaction interface; at the point of initial contact the mycelia fused imperceptibly (see Fig. 16(d)). This may suggest the operation of some kind of cytoplasmic factor and/or the production of some persistent inhibitory substance. In virtually all cases the monokaryons were dikaryotized, the exceptions being those involving C1 components; in these instances the interaction with the dikaryon was of a different type from normal in that instead of a clear zone developing between the isolates, a narrow zone of mycelial build-up occurred (see Fig. 16 (c)). In type B pairings, as might be expected, since any dikaryon formed would be identical in nuclear constitution to the parent dikaryons, no instances of track-formation were observed.

TABLE IV Numbers of pairings between synthesized dikaryons and nwnokaryons showing antagonism and other features in Coriolus versicolor Status of monokaryon following subculturing after 10 days Pairing type

No. of pairings

No. A

No. o

111

79

32

0

87

21

20

57 23

7

29 12

21 10

55

1

2 1

0 0

No.

clamps

aTrack noclamps formation

(A) Dikaryon not sibcomposed, monokaryon neither component nor sibrelated (B) Dikaryons paired with component monokaryons (a) dikaryons not sibcomposed (b) dikaryons sibcomposed

22

aFor explanation see text. Symbols as in Table I. The discrepancy between total numbers of pairings and observations regarding status of monokaryon is due to practical difficulties resulting from comparatively slow growth of monokaryon in certain pairings. After Todd and Rayner (1978).

Fig. 16

369

FUNGI IN DECAYING WOOD

The phenomenon of track-formation is an interesting one for a variety of reasons. The simplest interpretation is that tracks represent lines of antagonism between different dikaryons formed following movement into the monokaryon of the two different nuclear types from a parent dikaryon. This being so, then the pattern of track-formation is likely to reflect the pattern of dikaryotization from the point of initial contact between mycelia, and may provide us with a convenient, easily observed feature by which we can follow dikaryotization. As a start to studying the phenomenon in detail we chose the di-mon pairing 0 4 J19 x F40, which is shown in Fig. 16 (e), and isolated mycelium from the points labelled I, 11,111 and IV. As expected, all the resulting isolates were dikaryotic. Following subculturing each isolate was both dedikaryotized and tested in pairings with dikaryons of known constitution (making the assumption that any dikaryon will inevitably be antagonistic towards other dikaryons unless they are genetically identical). The results of these tests are given in Table V. The results of dedikaryotization (Table V (a)) show the consistent feature which we have noted previously (Rayner and Todd, 1977), namely that this sort of experiment often results in strong selection for one nuclear type. With three of the isolates, 11, 111 and IV, both nuclear types were recovered, allowing their constitution to be assigned, but with isolate I the only conclusion which could be drawn was that nuclear type F40 was a component of the dikaryon. It was confirmed that as expected isolate IV contained only the nuclear types present in the original parent dikaryon. The results of the pairing experiment (Table V (b)) show that each isolate only fused with one of the three possible dikaryons (that is, the one of presumably identical genetic make-up), and could therefore be genetically characterized. As can be seen from these results the two tests correlate perfectly serving to confirm our notion that such pairing tests are an easy and unequivocal way of identifying the constitution of a particular dikaryotic sector within a tracked area. Currently we are exploiting these techniques as a novel approach to investigating the Buller Phenomenon in wood-decaying basidiomycetes. From di-mon matings with C. versicolor and more recently Bjerkandera adusta (where the Buller Phenomenon also occurs and we have evidence of track-formation), it seems obvious that the ability of dikaryons t o dikaryotize -

~~

~~~

~

~

.

-~

~

-

Fig. 16. Examples of di-mon matings in Coriolus versicolor. (a) 0 4 313 x F40 (type A pairing) showing marked antagonism and pigment production between colonies. (b) A26 0 6 x F40 (type A pairing) showing antagonism in distal portions only and trackformation (arrowed). (c) C1 0 3 x C1 (type B.a pairing) with narrow zone of mycelial build-up between colonies. (d) CI A23 x A23 (type B.a pairing) showing merging at point of initial contact of colonies and very faint signs of antagonism in distal portion. (e) 0 4 J19 x F40 (type A pairing) with marked antagonism and pigment production and a sector delimited by tracks. Points I, 11,111 and IV represent sites of isolation for further analysis. (f) Result of subculturing from dotted portion shown in (e). Notice antagonism between newly grown mycelium in predictable fashion.

370

A. D. M. RAYNER AND N. K. TODD

TABLE V Identi$cution of dikaryons isolated after a di-mon mating in

Coriolus versicolor (a) Using dedikaryotizution No. of nuclei No. of derived

Dedikaryotized isolate

identified

monokaryons tested

04

J19

F40

10 14 10 9

0 13 0 1

0 0 9 8

10 1 1 0

I I1 111 IV

Constitution ofdikaryon ? F40 0 4 F40 J19 F40 0 4 J19

( b ) Using pairing test (see text)

Possible dikaryons Paired isolate

0 4 J19

0 4 F40

J19 F40

A

0

0 A

111

A A A

IV

0

I I1

A A

0 A

Constitution of dikaryon J19 F40 0 4 F40 J19 F40 0 4 J19

The di-mon mating was between 0 4 J19 and F40, as illustrated in Fig. 16. Isolates I-IV were obtained from the locations indicated in the figure. Symbols as in Table I.

monokaryons must play a very significant role in establishing a population and indeed this has often been suggested by other authors (e.g. Raper, 1966). In contrast, in di-mon matings between isolates of F. cujunderi, Adams and Roth (1967) reported that dikaryons were completely unable to dikaryotize monokaryons against which they were opposed ; they only used microscopic examination and not subculturing as their means for assessing this however. In one experiment we followed a series of identical di-mon pairings ( 0 4 J19 x F40) and identified the constitution of any dikaryons formed in the monokaryotic sector by means of the pairing test mentioned above. Some of the findings are shown in Fig. 17 (a), (b). In every mating the dikaryon successfully dikaryotized the monokaryon. However, track-formation did not occur in all pairings but where it did was very obvious. It is significant that there was no evidence for internuclear selection (defined by Raper (1966) as the “differential affinity of the monokaryon for the two nuclear types present in the dikaryotising dikaryon”) because both nuclear types seem to be able to pass over the interaction zone equally well. It can be seen that where several different dikaryotic sectors, separated by tracks, occurred, these alternated in their nuclear constitution. In all cases we only found a maximum of two dikaryon types in dikaryotized sectors namely J19 F40 and 0 4 F40. We did not find the original parent 0 4 J19, and would not have expected to

Fig. 17. (a) Nine identical di-mon matings (04 J19 x F40, a type A pairing) several of which show marked track-formation. Symbols ( 0 )indicate sites of isolation for pairing tests. (b) Interpretation of matings shown in (a) on basis of pairing tests.

372

A. D. M. RAYNER AND N. K. TODD

if the mechanism of antagonism is as we suggest a direct consequence of genetical difference between dikaryons preventing exchange of nuclei. Close examination of Fig. 17 (a) reveals different patterns of track-formation, such that in some cases the tracks originate from the initial point of contact, but in others they are far removed from this region. This might indicate that the nuclei may sometimes migrate significant distances before true dikaryon formation and subsequent antagonism occurs. 2. Physiological Basis If, as our previous work suggests, intraspecific antagonism operates in nature to delimit individual mycelial systems within the population, the interaction zone should act as a barrier to exchange of nuclei and/or cytoplasm. We have carried out experiments to explore whether the integrity of individual dikaryons of C . versicolor is maintained following interaction with other mycelia. Mating-type factors were used as markers for tracing the movement of nuclei and radioactive rubidium (*sRb) (an ion which acts essentially like potassium and is rapidly translocated throughout the mycelial system) for movement of cytoplasm. To test for movement of nuclei across the interaction barrier, isolates were made from either side of an interaction zone formed between two genetically distinct dikaryons after four weeks together in culture. Each isolate was then dedikaryotized and the derived monokaryons checked for mating-type : the results are presented in Table VI. Despite the differential selection of nuclear type, these results clearly support the notion that no nuclear exchange between the dikaryons occurred, since, for any one side, only one or both nuclear types originally present were recovered. The procedure for tracing the movement of cytoplasm was to allow various radioactively labelled dikaryons to grow towards each other from either side of a plastic partition (which acts as a diffusion barrier) on a Petri dish. After four weeks in contact the gross distribution of labelling was assessed by autoradiography. Figure 18 shows the autoradiographs obtained following the pairing of two nonsibcomposed dikaryons, the monokaryons of which were from different decay columns namely, A26 F38 paired with N8 56. The distribution of labelling shows that when a dikaryon was confronted with a genetically identical mycelium, cytoplasmic exchange occurred (Fig. 18 (a), (b)). However, when confronted with a genetically distinct dikaryon, very little, if any exchange occurred (Fig. 18 (c), (d)). Together, we believe these results provide strong evidence that the interaction zone does act as a barrier to exchange both of nuclei and cytoplasm. By microscopic examination of developing interaction zones in culture we have gained some idea as to how this may be achieved in C. versicolor. It is known from the work reported by Burnett (1965) with this fungus that there was little restriction on hyphal anastomosis between genetically alike or

FUNGI IN DECAYING WOOD

373

TABLE VI Identification of dikaryons of Coriolus versicolor by dedikaryotization following four weeks interaction in culture

Original pairing between dikaryons

Isolate No. derived No. of nuclei identified from rnonokaryons side tested A26 F38 N8 A23 C1 J19 513 A29

A26 F38 x N8 A23 A26 F38 N8 A23 Cl J19 x 513 A29 C1 J19 513 A29

30 11 28 68

0 3 0 0 0 011

0 - - - 0 - - - _ _ - _ 2 6 2 0 - _ _ _ 0 0 6 8

0 0

different dikaryons. One feature we have noticed which is likely to be relevant is that when antagonistic isolates are opposed in culture, hyphal anastomosis does at first occur, but subsequently the fusion segments change in refractive properties and swell to form spindle-shaped cells, often constricted at either end at the septa (Fig. 19). From closer examination of this process a number of key features become apparent. When the hyphae of two antagonistic dikaryons meet, the leading hyphae grow past each other and do not fuse. Fusion usually only occurs between lateral branches which have a characteristic morphology, and often lack clamps. The fusion segments then may show irregular growth in the form of swelling and possibly abnormal branching, and this continues until they form the large structures shown in the photomicrograph (Fig. 19). Whilst these events are occurring in the fusion segments, the hyphae at either side of this zone begin to lose their contents and die back leaving only hyphal “ghosts”. The final stage in formation of the interaction zone is generally the accumulation of pigment which is probably produced as a result of the death of hyphae in the immediate vicinity. Spindle-shaped cells have never been detected in the fusion zone between identical dikaryons. Our findings show similarities with those of Barrett and Uscuplic (1971) who carried out a microscopic examination of the interaction zone between antagonistic isolates of P . schweinitzii. They found that when colonies meet, their leading hyphae intermingle and anastomosis occurs, followed by disruption of hyphae and the accumulation of dark pigment.

3. Concluding Comments Taken as a whole, the results presented above provide good evidence that antagonism in C. versicofor is primarily a dikaryotic phenomenon which serves to “protect” dikaryons from nuclear or cytoplasmic exchange when in contact with other mycelia; that is, it maintains the integrity of the individual. Antagonism is inevitable between genetically distinct mycelia, regardless of relatedness, which suggests that the genetic basis is heterogenic in nature as has been proposed for the findings with other wood-decaying fungi (Esser and Blaich, 1973). The fact that the interaction zone acts as a barrier to

374

A. D. M. RAYNER AND N. K. TODD

Fig. 18. Autoradiographs showing the gross distribution of radioactive label (*"b) following interaction between dikaryons. Symbol * indicates the side to which radioactive label was added at start of test. (a) N8 56 x N8 36. (b) A26 F38 X A26 F38. (c) N8 56 x A26 F38. (d) A26 F38 x N8 36.

exchange of nuclei or cytoplasm is essential to our thesis, and it is one which ties in with the results of Caten (1972) who suggested that heterokaryon incompatibility in Aspergillus was also acting as a barrier preventing exchange. It does not, however, preclude exchange occurring between very closely related strains, especially those with nuclei and cytoplasm in common, such as would be the case with many laboratory strains. Antagonism is apparently expressed as a post-fusion event due to an incompatible reaction between nuclei and cytoplasm. This leads to the formation of spindle-shaped cells, which although sparse, nevertheless efficiently close off the immediate area from the rest of the mycelial system and act effectively as a trap for hyphal tips. The work in culture on the interaction zone correlates well with the field observation that interaction zones are relatively undecayed as compared with the rest of the wood, disruption of the normal functioning of hyphae in this region probably resulting in loss of their capacity for enzymic degradation of woody cell walls.

FUNGI IN DECAYING WOOD

375

Fig. 19. Spindle-shapedcell (arrowed) produced following hyphal anastomosis between antagonistic dikaryons of Coriolus versicolor.

The phenomenon of tracking shows us that if a dikaryon containing different dikaryotic components is formed it is likely that this could be readily observed on subculturing: different sectors would be expected to grow out separated by zones of antagonism. It is very significant that such a phenomenon has never been detected in isolates derived from the field, or from isolates subcultured after interaction in culture, further confirming that such isolates are individual dikaryons and that no nuclear exchange is taking place between them. We may conclude that, at least in C . versicolor and B. adusta, intraspecific antagonism does delimit individuals and obviously plays the major role in maintaining population structure in the wild. C. SIGNIFICANCE A N D POTENTIAL USE OF ANTAGONISM

If we may briefly reiterate the results of our studies given above: we have shown that in wood occupied by populations of decay fungi, different decay columns occur separated by narrow, dark, relatively undecayed interaction zones. These columns are occupied by genetically homogeneous, mutually antagonistic mycelia (dikaryons for heterothallic species) between which little, if any, exchange occurs, and which can therefore be regarded as individuals. Monokaryons derived from antagonistic dikaryons are normally interfertile, and we have therefore proposed the thesis that in nature the

376

A. D. M. RAYNER AND N. K. TODD

antagonism is a purely vegetative mechanism which operates to delimit individuals within freely interbreeding populations. We are satisfied that intraspecific antagonism is widespread in natural populations of higher fungi because we have seen phenomena which can be explained on such a basis in a wide variety of common wood-decaying and other fungus species, we have demonstrated antagonism in every wooddecaying species that we have examined in detail so far and because of additional reports of antagonistic phenomena in the literature. It may be worth noting that the phenomenon is likely to be less obvious in laboratory stocks where strains are often deyived from a single wild isolate or source, and are therefore usually closely related. However, when laboratory strains of different origin are examined, antagonism is usually more apparent. Intraspecific antagonism, or similar mechanisms maintaining individuality in fungi, have considerable significanceboth in relation to possible population studies and to our understanding of the behaviour of fungi in nature. Not least amongst the questions it raises is that of the validity of the unit mycelium as envisaged by Buller (1931) and advocated by Burnett (e.g. Burnett, 1976). The widespread acceptance (conscious or unconscious) of the concept of the unit mycelium has, apart from perhaps preventing further work on its occurrence (the need for which has however been repeatedly expressed by Burnett), caused many to consider the advantages of such a system, possibly forgetting any disadvantages. The ecological advantages of the unit mycelium in terms of communal exploitation of the substrate are of course obvious and highly attractive, implying a unique level of social organization. The possible opportunity for somatic recombination and hence production of new variants to exploit the substrate (Burnett, 1965) could also be seen as an advantage. In a fungus such as C. versicolor which is a highly effective agent of wood decay, individual isolates being capable of bringing about the virtually complete destruction of the wood on their own (Findlay, 1940), we believe that such production of new variants may not be particularly necessary in terms of exploitation of the substrate. In spite of these possible ecological advantages we believe that the unit mycelium may have certain inherent disadvantages, for example in terms of genetic variation. One obvious circumstance where such variation might be reduced is where one component of a unit mycelium begins to fruit, drawing on the rest of the mycelium for the necessary nutrients, at the expense of the fruiting capacity of other components. In such instances fruit-bodies of only one or a few components might predominate with consequently greatly reduced expression of the variation potential of the population. Buller (1931) counters this by arguing that where all the mycelia remain independent no single one would be large enough to give rise to a fruit-body of even minimum size ; therefore, all the mycelia would remain sterile and no contribution would be made to the next generation. However, we have suggested that in

FUNGI IN DECAYING WOOD

377

the cases where mycelia remain physically separated, each would be able to fruit on its own, albeit that in a population with numerous individuals, and hence separate decay columns, the fruit-bodies might be reduced in size. We have found evidence that this may be the case with C. versicolor (Rayner and Todd, 1977). With the unit mycelium, competition between individuals has been replaced by co-operation and so natural selection is no longer acting on the individual but on the group or population. Many population geneticists would probably not be inclined to favour the unit mycelium concept on abstract theoretical grounds since group selection is now a somewhat discredited concept (see Williams, 1971). Another possibility which needs exploration is that the formation of even a limited unit mycelium by blurring the distinction between individuals, may reduce, rather than enhance, genetic variability in the population. The reasons for the disparity between our own interpretation and those of Burnett and Partington (1957) obviously need to be questioned. Unfortunately, whilst we can explain their results by assuming extensive di-mon mating, we cannot be sure of this without knowledge of the nature of the local spore source from their sampling sites, or the presence or absence of polymorphism between the fruit-bodies sampled, or the structure of the mycelial population within the wood. Clearly we require further study, along the lines we have suggested, to clarify the issue. In view of the occurrence of intraspecific antagonism in all of the fungi we have examined we would not be too surprised eventually to discover that the unit mycelium does not exist, certainly not to the extent envisaged by many mycologists at the present time. It is of interest to compare our results concerning intraspecific antagonism in wood-decaying basidiomycetes with those of other workers on related or similar phenomena in other fungi. The occurrence of heterokaryosis which can best be explained in terms of hyphal fusion between genetically distinct mycelia in ascomycetes has been very well documented and much discussion has focused on its possible significance and advantages. Many of the arguments proposed parallel those used for the unit mycelium, indeed the two concepts, i.e. of heterokaryosis and the unit mycelium have much in common. It is interesting in this respect that researchers interested in h-c mechanisms as described above for Aspergillus spp. and Endothia parasitica, have come, in a fashion parallel to ourselves, to question how extensively heterokaryosis occurs in nature, and the significance of h-c mechanisms in preventing its occurrence. Particularly interesting here is the article by Caten and Jinks (1966) in which they state that the significance of heterokaryosis in nature still seems questionable, arguing that many laboratory investigations have been based on forced heterokaryons (complementary auxotrophs), and that numerous studies have shown that heterokaryons are only formed readily between strains of closely similar genetic background. How similar to our own arguments!

378

A. D. M. RAYNER AND N. K. TODD

Both h-c mechanisms as described in ascomycetes, and intraspecific antagonism between dikaryons as we have described in wood-decaying basidiomycetes seem to us primarily to operate as systems for vegetative but not reproductive isolation, and hence for the delimitation of individuals. Since all the very few systems which have so far been analysed in sufficient detail appear to be based on genetic dissimilarity they can be regarded as essentially heterogenic in nature. However, we are concerned that much confusion may arise as a result of considering such vegetative heterogenic systems together with heterogenic systems resulting in reproductive isolation (such as described by Esser, 1965; Burnett, 1975)under the general heading of “heterogenic incompatibility”. It is important to be clear about this since there has been a strong tendency to regard all intraspecific antagonistic phenomena as systems resulting in reproductive isolation (Burnett, 1975) and hence ultimately leading to speciation. This is clearly not the case unless we stretch the point considerably to include prevention of somatic recombination (which may be the only opportunity for any sexual or parasexual behaviour in fungi without a perfect state) as a mechanism operating for reproductive isolation. More evidence is, in fact, still required in support of many so-called isolating mechanisms, and it seems not unlikely that some of these at least may in fact be similar to the situation we have described, resulting merely in vegetative isolation. As more information comes to hand concerning the basis and extent of antagonistic phenomena in ascomycetes and basidiomycetes, it seems likely that this may throw light on possible evolutionary relationships between these groups. It may even be that some explanation for the mysterious appearance of multiple allelomorphic homogenic incompatibility systems exclusively in the basidiomycetes (see Raper, 1966) could become available. Digressing to pure speculation, the type of argument might be as follows. In ascomycetes which are gametangiogamous (i.e. fertilization occurs only between specialized sexual organs) individuality is maintained by a polygenic h-c mechanism and effective hyphal fusion and heterokaryosis prevented. The advent of somatogamy, in which hyphal anastomosis is essential for sexual conjugation, and is characteristic of basidiomycetes, will require that any mechanism preventing such fusion is overcome at least between homokaryons. Perhaps the multiple allelic incompatibility system in basidiomycetes could provide the necessary over-ride mechanism. Once sexual conjugation between homokaryons has been achieved (in heterothallic species), expression of antagonism may again occur between dikaryons. It is significant in this respect that a structure markedly resembling a clamp-connection (this has been noted by many authors) develops at the base of the ascus initial-the only true dikaryotic stage found in ascomycetes. In basidiomycetes, of course, the dikaryon becomes dominant in the life cycle, following its inception at hyphal anastomosis between compatible homokaryons. We must emphasize, of course, that this

FUNGI IN DECAYING WOOD

379

is not necessarily going to be the pattern of events, but it does illustrate a possible avenue along which these arguments may lead. One point which does need to be borne in mind during these considerations is that the wooddecaying fungi which show intraspecific antagonism probably have a wide variety of mating and nuclear behaviours. These range from homothallic, holocoenocytic as with S. hirsutum and P . velutina; through bipolar, astatocoenocytic as with P . merismoides, to tetrapolar normal with C. versicolor (Boidin, 1971).* There needs to be much deeper understanding of mating and nuclear behaviour, especially in nature, of a wider range of species than we possess at present. Certain work (Kiihner, 1977) has shown that these features may show more variability and complexity than many suppose. Given that our arguments are correct and populations of wood-decaying basidiomycetes are made up of dikaryotic individuals, they may be considered equivalent to populations of diploid individuals in higher organisms in respect of genetic flexibility and with the opportunity to maintain heterozygosity (genetic differences). Indeed this situation does offer certain advantages that no diploid could achieve because new dikaryotic associations can be attained by the dikaryotization of monokaryons. From our work with C. versicolor, it seems likely that dikaryon formation is controlled by the homo- and heterogenic incompatibility systems, and that the role of intraspecific antagonism is to maintain the integrity of the dikaryon once it has been established. Up until now we have been considering the biological significance of intraspecific antagonism from the point of view of the organisms. The phenomenon also has obvious potential significance for ourselves, both academically in terms of the way in which we conduct and interpret experimental studies and, more practically, in connection with cultivation of fungi for food or in industrial fermentations and in control of diseases caused by fungi. Considering more practical aspects first, we may note for the future, that in cultivation of fungi, use of a large number of different strains together (as with a basidiospore inoculum) rather than a single strain (a mycelial inoculum) could result in lower yields. The possibility that antagonism prevents transfer of viral infection between strains is also of potential significance, as is the fact that it might preclude biological control methods relying on heterokaryosis, as with Endothia parasitica. From the academic standpoint, studies of antagonism may have considerable significance in experimental taxonomy. For example polymorphisms may often be explained on the basis of combined studies of antagonism and interfertility, of the type we have conducted with Coriolus versicolor (Rayner

* Astatocoenocytic behaviour describes the situation where a single spore germinates to produce a strongly coenocytic mycelium. The secondary mycelium, developed following sexual conjugation is only binucleate and clamped under conditions of sufficient aeration, otherwise it is strongly coenocytic. Normal behaviour describes the classical situation where a uninucleate spore germinates to produce a monokaryon with uninucleate hyphal segments; the secondary mycelium is regularly binucleate.

380

A. D. M. RAYNER AND N. K. TODD

and Todd, 1978). Other fungi exhibiting polymorphism, and aggregates of morphologically closely similar species which could benefit from such study abound, e.g. Phlebia merismoides, and the Hypholoma fasciculare, sublateritium, cupnoides group amongst wood-decomposers. Some may favour delimitation of species on the basis of antagonism in culture, but our observations clearly indicate the dangers of such practice. For example antagonism has been advocated as an aid to species delimitation in Stereum spp. (LCger, 1968), although in this case it was also said to be associated with failure of hyphal anastomosis. In our own work, reported above, with S. hirsutum we have observed marked antagonism, even between single spore isolates from the same fruit-body (Fig. 6). Finally, we return to the theme of the article, population and community structure, the area in which intraspecific antagonism can potentially be most useful. It plays a major role in determining and maintaining the structure of natural populations and on the basis of mutual antagonism it is possible to identify the individuals within a population. This opens up the possibility of much more defined and rigorous studies of population biology (including population genetics) in fungi. For example, provided the necessary genetical experiments have been carried out, antagonism may frequently provide a sound basis for identifying the spatial distribution of individual mycelia in the field. Thus, isolates made from a range of fruit-bodies or woody material within a particular site could be paired together and the consequent expression of antagonism (or lack of it) used as an indicator of their relationship. Such procedures have of course already been used to a very limited extent (e.g. Childs, 1963; Adams, 1974), but usually without the necessary genetical background knowledge. This type of study offers an almost totally fresh approach to investigating the nature and significance of mycelia in the field. This is an aspect of the biology of the fungi, which, perhaps because of the hidden nature of mycelial systems, has been neglected far too long. IV. INTERSPECIFIC INTERACTIONS: THEIR ROLE IN THE DEVELOPMENT AND MAINTENANCE O F COMMUNITY STRUCTURE A. THEORIES OF SUCCESSION AND COMMUNITY DEVELOPMENT

There often appears to be a progressive change in the species composition of fungi in decaying wood, from the initial stages of colonization, to final degradation. A succession may therefore be said to occur. The term succession as employed by mycologists is, however, worth some discussion; we believe it is often used too readily, without sufficient rigour. Further we think that neither are its real value as a concept, nor its relation with the term as envisaged by other biologists sufficiently questioned. As a fairly rigid definition

FUNGI IN DECAYING WOOD

38 1

of fungal succession the following seems potentially useful : “the sequential occupation of the same site by thalli (normally mycelia) either of different fungi or of different associations of fungi”. However, several problems arise in attempting to apply such a definition rigorously during successional studies. Ideally, to study succession with any degree of certainty requires that changes in the mycoflora present at the same site be studied over a period of time, preferably, in decomposition studies, from the stage of initial colonization through to complete degradation. The requirement for studies at the same site, without disturbance, can be fulfilled by observation and identification of fruiting structures, and this has indeed formed the basis of many successional studies. Such practice is doubtful, however, since, as mentioned previously, it is the mycelium and not its fruiting structures which constitutes the main vegetative body of a fungus. Studies of succession using fruit-body production as their sole criterion are therefore subject to error, relating, for example, to the following: variation in the amount of time taken by different fungi to fruit; variation in the durability of fruit-bodies-tough fruit-bodies may persist long after the mycelium producing them has been replaced, or vice versa for ephemeral fruit-bodies; seasonality in fruit-body production; uncertainty in deducing the position of mycelium purely from that of the fruit-bodies. Some of these variations have been highlighted by experimental analysis of the succession of fungal reproductive structures observed on herbivore dung; here sporangia of mucoraceous fungi such as Pilaira and Pilobolus appear first, followed by fruit-bodies of discomycetes (e.g. Coprobia, Ascobolus), pyrenomycetes (e.g. Sordaria) and finally basidiomycetes (e.g. Coprinus, Stropharia). At first this succession was considered to be nutritionally based, reflecting differing capacities of the succeeding groups of fungi to utilize simple sugars, cellulose and lignin respectively. However, work by Harper and Webster (1964) showed that the observed succession could at least partly be explained simply by the increasing amounts of time required by succeeding fungi to fruit. Clearly then, we need to detect mycelium and for this purpose two techniques are available. We may, on the one hand, isolate from small samples of wood, for example cores removed with an increment borer. The limitations of such techniques as guides to the structure of the fungal community have already been emphasized. We must remember that to be certain a successional event is taking place requires precise knowledge of the spatial distribution of mycelia. Otherwise, the possibility exists of isolating different fungi at different times from different positions and interpreting the results as succession rather than mere physical separation! It seems to us that in rather too many supposedly successional studies involving isolation from small samples of wood, insufficient regard has been paid to this simple source of error. Further, whilst they involve removal only of small samples of wood, enabling successive sampling from the same substrate, the sampling procedure is

382

A. D. M. RAYNER A N D N. K. TODD

nevertheless disruptive and could lead to artefacts. The second alternative, which we favour, is to remove whole stumps, logs etc., and analyse them destructively as described in Section 11. Such a procedure is inevitably destructive but this problem can be surmounted by sufficient replication. A particularly difficult problem with successional studies in wood is that the time-span required for rigorous studies along the lines we have suggested is often considered too great to be practicable, especially if full studies, starting with initial exposure to colonization and ending with final degradation, are to be attempted. As a result much use is often made-especially in review articles on succession in wood-of studies which are not primarily successional in their objectives. For example, successional interpretations are often made of results of studies involving identification at any one point in time of the fungi present in wood of different ages, and at different sites. Differences in the mycoflora could then be attributable to factors other than succession, including site differences, differences in fungal inoculum available at the time of exposure, etc. Experience may help to eliminate some of these alternative interpretations, but their possible existence must always be remembered. Otherwise we may become oversimplistic or even inaccurate in our views on succession and complacent about our level of knowledge on the subject. Given that we study and interpret succession correctly, we must go further to question its general use as a concept. Often our understanding is crucially affected by how broad or narrow a view we take (Fig. 20). On the one hand we often envisage in successions broad waves of organisms, each following the other-but without reference to their origins. We may then greatly oversimplify the events taking place: for example we might fail to distinguish between the situation where one fungus literally only colonizes after another, and that where a mycelium established in one portion of wood is able, perhaps because of its greater competitive ability, to replace mycelium of another fungus present in an adjacent portion. We may on the other hand, take too narrow a view, being concerned only with individual cases where one fungus replaces another. Perhaps one of the greatest difficulties with succession as a concept is that it encourages belief in single, simple underlying causes, such as changes in nutrient status: in nature such simplicity seems unlikely to occur. These difficulties become less intractable when we realize, and this is a major theme of the present article, that, especially in relatively large, bulky substrata such as wood, fungal succession is but part of, and cannot be separated from, a much deeper issue, namely the nature and development of fungal communities, Rather than to ask “what is the succession?”, a more pertinent question is “what is the fungal community: how is it maintained and what forces may cause it to change?’ What we are recommending here is the really quite obvious fact that our studies should incorporate both space and time dimensions (community structure and development) rather than time alone (succession). The concept of community development and some

&I\

\

A

L

C

R

?

D

.

N

Time a

b

C

Fig. 20. Diagram illustrating the concept of fungal community development within a spatially defined resource. The total available colonizable space is represented by the gap between the top and bottom horizontal axes. The changing distribution, within this space, of a variety of different fungal mycelia A-N, over a period of time starting from the point at which the resource first becomes susceptible to fungal colonization (represented by the longitudinal axis at the left-hand edge of the diagram) is shown as a series of tapering and expanding bands. Horizontal lines between adjacent bands indicate deadlock interactions, whilst oblique ones indicate replacement. Community development in this hypothetical example can be summarized as follows. Mycelia A-K all colonize more or less from the outset and spread through the resource until eventually they all come into contact and the available space for colonization is filled. B and J occur for only a short period before being replaced. A and C, C and D, D and E, G and H, H and I and I and K are all involved in deadlock interactions, but F is able to replace C, D, E and G and becomes one of the dominants in the final community. Subsequently L, M, and N invade and replace other mycelia before becoming deadlocked, so that in the final community only L, F, H, M and N are present. The development of M in wood occupied initially by I could result from a specific mycoparasitism, such as occurs in nature between Pseudotrametes gibbosa and Bjerkunderu adusta. Successional interpretations of this complex pattern of events could easily lead to oversimplification and distortion. For example, at point a there is a “succession” from B to A to L. Whilst this does tell us of the relative competitive abilities of these three mycelia, it does not tell us when and where they colonized and, of course provides only a narrow view of events in the whole resource. Even narrower would be the view at point 8, where mycelium H persists throughout, no “succession” in fact occurring. Alternatively, if the species assemblages present in the entire resource at different points in time a, b and c were analysed, a broader picture would be obtained, yielding some information about when different mycelia colonized, but revealing little detail about specific instances of replacement and deadlock.

384

A. D. M. RAYNER AND N. K. TODD

of the limitations of purely successional interpretations are illustrated diagrammatically in Fig. 20. B. INTERACTIONS AND THEIR SIGNIFICANCE

To ask how a fungal community is maintained or changed is tantamount to asking how its constituent mycelia may interact with one another. We have already seen that three developments are theoretically possible : intermingling, deadlock and replacement. In decaying hardwood in the open, deadlock and replacement appear to be most important. Whilst deadlock may be expected to operate for maintenance of the community pattern, replacement will lead to change (Fig. 20). Theoretically, both deadlock and replacement could occur either as a result of direct interaction following contact between mycelia, or by indirect mechanisms involving some intermediary effect or agent; it is not always easy to distinguish between these alternatives. The emphasis in the following account will be on the role of direct interaction between mycelia since many instances of deadlock and replacement do seem to operate through this mechanism: we have already seen (Section 111) how direct mutual antagonism often leads to deadlock between different mycelia of the same species. Nevertheless, we must not forget various more indirect mechanisms which may also result in deadlock or replacement phenomena in wood. Amongst these we may list the following: I . Nutrient Eflects Whilst growing in wood, mycelia of different fungi will inevitably remove nutrients necessary for growth. An established mycelium is likely t o be efficient in removing and utilizing such nutrients before they become available to other non-established colonies, which are thereby excluded. If efficient utilization of the cell wall polymers was all that was important, we might expect fungi which are slow or ineffective in decay to be replaced rather readily : this is by no means the case-moderate decay-causers such as Hypoxylon muhiforme can exclude more active decay fungi (Fig. 3) for prolonged periods so that often relatively hard and undecayed columns occupied by this fungus can be lifted free from surrounding wood which has been virtually completely degraded by others. Indeed inoculation of wood, prior to its being put into service, with certain fungi inactive in decay has been suggested as a possible means for biological control of decay : these include Trichoderma spp. (Hulme and Shields, 1975) and a strain of Scytalidium (Ricard and Bollen, 1968). One mechanism which has been suggested for the mode of action of at least some of these fungi in preventing decay is pre-emption of nutrients (Hulme and Shields, 1970), i.e. that as they colonize the wood they remove from it certain nutrients, such as non-structural carbohydrates, necessary for permeation by decay species.

FUNGI IN DECAYING WOOD

385

A nutritional explanation sometimes advanced for replacement is that whilst it can still obtain sufficient nutrients a fungus will continue to occupy woody tissue, but as soon as it uses up available sources it will tend to die back, allowing entry of fungi which can obtain nutrients from the residual material. This is certainly a possibility, but again it does not account for all examples of replacement. Fungi such as Coriolus versicolor which have the enzymic capacity to degrade wood virtually completely (Findlay, 1940), are usually replaced well before they could be said to have exhausted all available nutrient reserves. 2. Production of Antibiotics Several fungi which colonize wood are believed to produce water-soluble diffusible antibiotics which inhibit growth of other fungi. Several nondecay fungi, including some of those advocated as biological control agents have been shown to produce such antibiotics, e.g. Scytalidium (Ricard and Bollen, 1968) and Cryptospoviopsis spp. (Stillwell et a]., 1969), which are inhibitory to decay fungi. Whilst production of diffusible antibiotics may be significant in a few cases of antagonism between fungi in wood, we feel that its importance may be overestimated. It is usually detected by the formation of antibiotic zones between colonies growing towards each other on artificial media, and which may result from growth inhibition of one or other, or often both (!) participants. Too often formation of such zones is taken as the main criterion for antagonism: deadlock and replacement reactions following contact (not necessarily preceded by inhibition of growth of either participant) are more commonly observed and are just as much examples of antagonism. Indeed, as we shall see, the latter may prove more reliable for understanding interactional events in the field, depending as they d o on the outcome of mycelial contact, which is an inevitable feature of fungi growing in juxtaposition in nature, rather than diffusion of inhibitory substances. It seems, intuitively, more likely that the outcome of mycelial contact will, because of the probable mechanisms involved (see below), show less inherent variability than antibiotic production. Certainly production and stability of antibiotics is likely to be markedly influenced by environmental factors, nutrient levels and isolate variability (Etheridge, 1971). This makes extrapolation from results obtained on artificial media to possible events occurring in the field particularly difficult. 3. Removal of Toxic Materials Whilst some fungi in decaying wood produce substances inhibiting growth of others, it seems likely that the opposite may also occur, i.e. that some microorganisms may prepare the way for others to follow by removal of toxic materials. This seems probable in colonization of heartwood, where in some cases colonization by non-decay species actually seems necessary for subsequent colonization by decay fungi (Shigo, 1967).

386

A . D. M. RAYNER AND N. K. TODD

4 . Removal of Host Resistance A somewhat parallel system to the last involves prior colonization of living wood by species which are able to kill living tissue, rendering it subject to colonization by purely saprophytic fungi. Chondrostereum purpureum (Pers. ex Fr.) Pouz. may act in this way (Rayner, 1977b). 5 . Incidental Environmental Factors A variety of environmental, or microenvironmental factors influence growth of fungi in wood, and if these differ in their effects according to the fungus concerned, they may favour some fungi more than others, causing selection. For example during the hot dry summer experienced in Britain in 1976, exposed portions of oak logs colonized by Stereum hirsutum became very dry, apparently leading to poor growth of the fungus which was subsequently extensively replaced by Phlebia merismoides colonizing from presumably moister portions of wood in ground contact (Carruthers and Rayner, 1979). In addition to desiccation, other factors which could lead to similar selection include, for example, presence of inhibitory or stimulatory microorganisms and temperature effects. Leaving aside whether interactions between different fungi are based on direct or indirect mechanisms, let us now turn to how they may be studied. Essentially two approaches are available, one involving laboratory experiments and the other field observation ; ideally both should be followed together rather than one in isolation. However, the majority of interaction studies involving wood-decaying fungi have, perhaps inevitably, been carried out in the laboratory. Whilst, as we shall see later, some attempts have been made to use more natural substrata, such as wood-blocks or sawdust, most experiments have involved simple pairing of cultures on purely artificial agar media. These studies began with the pioneering work of Harder (191 l), and have continued very intermittently to the present time, some of the more detailed contributions including those of Oppermann (1951), RyphEek (1966), Henningson (1967) and Rayner (1975, 1978). Many of the interactions observed during these studies are pronounced in character, especially where basidiomycetes are involved; for example they may sometimes involve marked lysis of one mycelium by another, or, very commonly, intense pigment production occurs. There is evidence that at least some of these interactions may be of use for diagnostic purposes, quite apart from their possible ecological significance. In this light it seems remarkable that so few detailed studies have been carried out during the course of so many years, and that the interactive capacities of even the most common wood-decaying species are so poorly understood. There has been little attempt to collate results obtained from different laboratories, indeed there is so little readily accessible published information that any such attempt would be premature. Except in a few specific instances we are still at the earliest stages of describing events observed

FUNGI IN DECAYING WOOD

387

at the gross morphological level between interacting colonies and information is lacking regarding, for example, physiological and structural changes at the hyphal level and the influence of isolate variability. Such extensive information as is required can only be obtained by at least several investigators, in a wide range of laboratories, pooling their results. Ultimately our aim should be to have as deep a knowledge of the interactional characteristics of wood-decaying fungi, as we already have for their other cultural characteristics. Perhaps part of the reason for the scarcity of these remarkably simple studies is justifiable apprehension that results obtained on agar may not correlate with events occurring in the field. These feelings are intensified by reports of some investigators suggesting variability of results obtained using different media, temperatures, p H etc.: if interactions are so subject to environmental influences it seems unlikely that results obtained in such obviously different situations as an agar plate and in wood will correspond. Some studies reported by Etheridge (1971) involving inhibition of decay fungi by non-decay-causing microfungi have indeed suggested fundamentally different interactions in the field to those observed in culture (e.g. Glaser et a/., 1959; Pomerleau and Etheridge, 1961). Further, other studies (e.g. Keyes, 1968) have indicated variability in the kind and degree of antagonistic response among different combinations of isolates of the same interspecific pairing. It is hardly surprising then that people are suspicious of these experiments, and the feeling of many is probably exemplified by Kaarik (1974) who states bleakly “the mutual action between two organisms is entirely dependent on external factors such as nutrition, temperature and substrate acidity. . . changes in these factors can easily result in conversion of mutual effects, and the results on artificial media must therefore be regarded with great caution in the discussion of interactions in natural substrates.” Understandable though such attitudes may be, we feel that they are unduly defeatist, focusing as they do on difficulties, rather than noting instances where a basis for understanding interactional events as they are known to occur in nature has successfully been provided. As a result, the approach may be abandoned before it has been given a sufficiently fair trial. Of course the acid test for the worth of interaction studies is whether they can be used successfully to interpret events in the field, and we are greatly restricted here by our lack of knowledge of the latter. Amongst the few, attempts to investigate simultaneously interactions in the field and in the laboratory have been made by Henningson (1967), Mangenot (1952) and Rayner (1975, 1978). As representative results of this type of study we wish, in the next two sections, to discuss in some detail those obtained by Rayner and his collaborators. Apart from purely personal bias, we do this because we feel that these results epitomize the range of interactional phenomena likely to be found between fungi in decaying wood and because in several

388

A. D. M. RAYNER AND N. K. TODD

instances they demonstrate how events occurring in the field can be interpreted satisfactorily in terms of direct interactions between different mycelia. C. RESULTS FROM LABORATORY-BASED STUDIES

1. On 3 % Malt Agar

In these experiments the interactional relationships of some 26 different fungi colonizing decaying hardwood were examined. The majority of these fungi were decay-causing basidiomycetes, but a few common non-decaycausing ascomycetes and fungi imperfecti (Botrytis cinerea Pers. ex Pers., Scytalidium album Klingstrom & Beyer and Coryne sarcoides) were also included. The experiments were conducted simply by placing inocula of different pairs of fungi at opposite sides of 9 cm Petri-dish plates; in cases where the fungi differed considerably in growth rate, inoculation of the faster growing one was delayed. In all, some 200 different species-combinations were tested, mostly at 15°C (a figure chosen as being ecologically representative in Britain) but some over a range of temperatures (i.e. 10"-22°C). For some species more than one different isolate was tested. The following interactional phenomena were observed : 1. Merging of mycelia between identical strains. 2. Growth of one fungus into or over another (replacement). 3. Deadlock, in which neither fungus was able to grow past the other. 4. Formation of coloured contact zones in the medium. 5 . Lysis of one mycelium by the other, usually prior to or accompanying replacement. 6. Development of dense, sometimes coloured zones of mycelium in the region of contact between colonies. 7. Stimulation of fruiting. 8. Development of clear zones between colonies. 9. Production of leathery crusts of mycelium between colonies, equivalent to PSPs described earlier (Section 11). Similar phenomena have been observed in other interaction studies involving wood-decay fungi (e.g. Henningson, 1967; RypBEek, 1966). (a) Replacement. Ofparticular interest in regard to possible forces for change in fungal communities in wood was the degree to which each fungus was capable of replacing, or susceptible to being replaced by other species. This is shown for a selection of eight hymenomycetes and one hyphomycete (Scytalidium album) at 15°C in Tables VII and VIII. It can be seen that these fungi varied markedly in their competitive abilities. Hypholoma fasciculare, Phanerochaete velutina (DC ex. Pers.) Parmasto, Phlebia merismoides and Scytalidium album were all highly competitive, replacing most other species. In contrast Chondrostereum purpureum was non-competitive, being readily overgrown. Bjerkandera adusta was of interest in that although it replaced

389

FUNGI IN DECAYING WOOD

TABLE VII Ability of fungi commonly colonizing hardwoods to replace others on 3% malt agar at 15°C

No.

Fungus

interactions tested

Bjerkandera adusta

25

Chondrostereum purpureum Coriolus versicolor

19 23

Hypholoma fasciculare

18

Phanerochaete velutina Phlebia merismoides

9 17

Pseudotrametes gibbosa Scytalidium album Stereum hirsutum

18 9

25

Species replaceda AM, BC, CP, CV, DC, FH, GA, (HA), HR, LS, MG, OR, PO, PV, SG, SH MG, PV CP, CS, DC, (HA), LS, OR, PO, PV, SH AM, BA, BC, CP, CV, DQ, DC, HA, OR, PO, PV, PG, (SA), SG, SH AM, BA, CP, GA, HF AM, BA, BF, BC, CP, CS, DC, MG, OR, PO, PV, SH BA, BF, LS AM, BA, CP, CV, GA, PVe, PM, SH. GA, LS

@AM,Armillaria mellea; BA, werkandera adusta; BF, Bjerkandera fumosa; BC, Botrytis cinerea ; CP, Chondrostereum purpureum ; CV, Coriolus versicolor; CS, Coryne sarcoides; DQ, Daedalea quercina; DC, Daedaleopsis confragosa ; FH, Fistulina hepatica ; GA, Ganoakrma adspersum ; H A , Heterobasidion annosum ; HR. Hymenochaete rubiginosa; HF, Hypholoma fasciculare; LS, Luetiporus sulphureus; MG, Mycena galericulata; OR, Oudemansiella radicata ; PVe, Phanerochaete velutina ; PM, Phlebia merismoides; PB, Piptoporus betulinus; PO, Pleurotus ostreatus; PV, Polyporus varius; PG, Pseudotrametes gibbosa; SA, Scytalidium album; SG, Stereum gausapatum ; SH,Stereum hirsutum. Parentheses indicate partial replacement only. After Rayner (1978).

many species, it was itself susceptible to replacement by several common fungi. The others, Coriolus versicolor, Stereum hirsutum and Pseudotrametes gibbosa (Pers.) Bond & Sing. were most usually involved in deadlock interactions (see below), although the latter replaced Bjerkandera spp. very readily. There was evidence in a few cases that the pattern of interaction could be altered by a change in temperature, for example Bjerkandera adusta overgrew Daedalea quercinu L. ex Fr. at 10°C but was replaced by the latter species at higher temperatures. In general this type of effect was uncommon, and the pattern of interaction for each species-combination was similar under different temperature regimes. The mechanism of replacement varied in different combinations and, as we will discuss later, this may be of importance in assessing whether events observed on agar are likely to be repeated in nature. In some cases replacement was by simple overgrowth of skin-like areas of closely interwoven mycelium (equivalent, perhaps to PSPs) produced by the species dominated. The basis for this pattern seems likely to be little more than that the skin-like areas provided smooth, relatively inert surfaces over

390

A. D. M. RAYNER AND N. K. TODD

TABLE VIII Susceptibility of fungi colonizing hardwoods to replacement by others on 3% malt agar at 15°C No.

interactions tested

Fungus Bjerkandera adusta Chondrostereumpurpureum

25 19

Coriolus versicolor Hypholomafasciculare Phanerochaete velutina Phlebia merismoides Pseudotrametes gibbosa Scytalidium album Stereum hirsutum

23 18 9 17 18 9

~-

25 _.. .

Replacing speciesa DQ, HF, PVe, PM, PB, PG, SA BA, BF, CV, (DC), HF, OR, PVe, PM, PB, SA BA, HF, SA PVe SA

(PW, SA HF -~ ~~~~

(HF) BA, BF, CV, HF, PVe, PM, (PB), SA ~-

a For key see Table VII. After Rayner (1978).

which mycelium of the opposing species could grow relatively uninterrupted, provided they were supplied with sufficient nutrients from behind. Species producing such skin-like zones, and which were sometimes replaced in this way included Armillaria mellea and Ganoderma adspersum. Whilst these fungi were sometimes replaced as a consequence of their relatively dense colony form on agar, the reverse also occurred i.e. species with a particularly diffuse colony form were often grown into rather readily, e.g. Laetiporus sulphureus (Bull. ex Fr.) Murr. Although the cases of replacement just mentioned, can be interpreted merely as a consequence of physical attributes of colony form, the majority seemed more likely to be the outcome of direct physiological challenge between confronting mycelia in close contact with one another. Most often in these cases replacement was rather slow relative to the growth rate, on malt agar, of the dominant species when grown in pure culture. The usual pattern in these interactions was for both species to grow towards each other at their normal growth rate, until contact was achieved. Thereafter radial growth of both would be inhibited and dense wefts of mycelium formed along the line of contact. Gradually the dominant species would then grow through and over the other as relatively dense mycelium, probably killing its opponent’s hyphae in the process, One of the most competitive species, Hypholoma fasciculare, replaced others in one of two ways. Either it grew across as a broad band of dense mycelium, or as separate, cord-like hyphal aggregates. The former was observed against, for example, Chondrostereum purpureum, Daedaleopsis confragosa (Bolt. ex Fr.) Schroet., Pleurotus ostreatus (Jacq. ex Fr.) Kummer and Stereum hirsutum, and the latter against Coriolus versicolor and Pseudotrametes gibbosa. The difference between these two modes of

FUNGI IN DECAYING WOOD

39 I

replacement may be explained in terms of nutrient availability. Where high concentrations of nutrients are available, perhaps from disrupted hyphae of the mycelium replaced, overgrowth will occur as a broad band, whilst if low levels of nutrients are available, hyphal aggregates will form: this would fit with the nutritionally-based model for mycelial cord* morphogenesis proposed by Day (1969). Another fungus commonly producing mycelial cords whilst replacing others was Phanerochaete velutina (Fig. 21). In addition this fungus caused very marked lysis of some of the species it replaced (Fig. 21) a fact which may be related to its extensive production of extrahyphal crystals (Fig. 2B). One of the most impressive examples of lysis was found in interactions involving Phlebiu merismoides (Fig. 22). This fungus lysed mycelium of Bjerkandera adusta, B. fumosa (Pers. ex Fr.) Karst., Chondrostereum purpureum, Heterobasidion annosum (Fr.) Bref., Oudemansiella radicata, Piptoporus betulinus and Pleurotus ostreatus prior (with the exception of P . betulinus) to replacing them. The usual pattern was that following contact with mycelia of these other fungi P. merismoides rapidly produced a dense zone of salmon-pink mycelium immediately behind the point of contact. Once this developed, lysis of the opposing mycelium began, usually only for short distances at a time. Overgrowth of the lysed zone would then occur until contact with living mycelium was re-established, whereupon the process was repeated, replacement thus occurring slowly and intermittently. Microscopic examination of the advancing zones of P . merismoides revealed a very high density of hyphae exuding liquid globules, which conceivably may have caused the lysis. Other examples of lysis were detected in the following combinations, the first named species being the lysing agent in each case: Bjerkandera adusta against Heterobasidion annosum and Oudemansiella radicata, Daedalea quercina against B. adusta, Hypholoma fasciculare against B. adusta and H. annosum, and 0. radicata against Chondrostereum purpureum. Whilst replacement was usually a gradual process, when Pseudotrametes gibbosa was paired with Bjerkandera adustu, replacement by the former was very rapid at all temperatures tested, its radial growth rate if anything increasing after contact. Furthermore P . gibbosa produced very much denser mycelium in the region of replacement (Fig. 23) where it also was eventually stimulated to produce fruiting structures. By growing the two fungi opposite one another on a medium containing mineral nutrients but no carbon source, sufficiently sparse growth was obtained to enable detailed microscopic examination. It was found that P . gibbosa produced hyphal branches which coiled around hyphae of B. adusta, causing coagulation and vacuolation of

* We should mention here that we regard the term “mycelial cords” for filamentous hyphal aggregates as being more descriptive and hence preferable to that of “mycelial strands” which is used by many mycologists. A “strand” is a single filament amongst many, and as such a single hypha constitutes a mycelial strand.

392

A. D. M. RAYNER AND N. K. TODD

Fig. 21. Interaction between Phanerochaete velutina (top) and Chondrostereum purpureum on 3 % malt agar. P. velutina is overgrowing C. purpureum, producing mycelial cords and causing marked lysis in the process.

their contents (Fig. 24). Staining with dilute phloxine, which is more quickly taken up by recently dead than living cytoplasmic contents (Griffith and Barnett, 1967), preferentially stained the dead B. adusta hyphae. This was regarded as a case of active mycoparasitism by P . gibbosa on B. adusta and this was supported by the fact that beyond the first point of contact with P . gibbosa, B. adusta hyphae often appeared normal and still to be growing. A similar situation to the above was found when P . gibbosa was grown against Bjerkandera fumosa. With the exception of Laetiporus sulphureus (see above) none of the other fungi tested against P . gibbosa were replaced by it, indicating a very specific relationship between the latter and species of Bjerkandera. The full significance of this finding will be disclosed shortly. (b) Deadlock. Very often the mycelia of opposing pairs of fungi were mutually antagonistic, so that neither was able to invade the other. Table IX lists those cases for our selected group of nine common hardwood-colonizing fungi where such deadlock occurred at 15°C.

FUNGI IN DECAYING WOOD

393

Fig. 22. Interaction between Phlehia merismoides (left) and Bjerkandera adustu on 3 % malt agar showing marked lysis by the former prior to overgrowth of B. adustu.

Sometimes deadlock was due to the production of an obvious physical barrier by one or other of the fungi which prevented penetration by opposing hyphae. This either took the form of zones of dense mycelium, or of leathery, skin-like crusts almost certainly equivalent to PSPs (see Section 11). The latter were produced particularly often by Ganoderma adspersum and Daedaleopsis confragosa, and occasionally by Oudemansiella radicata. In some cases “antibiotic” zones of clear agar developed between the interacting colonies. These were of two types. In one they developed only after contact between opposing hyphae. This led to mutual death of the hyphae over a short distance, often accompanied by vacuolation and granulation of their cytoplasm. Semi-clear zones traversed by dead hyphae thus developed between the colonies. Examples of this type of interaction, which almost certainly involved a similar mechanism to the “hyphal interference”

394

A. D. M. RAYNER AND N. K. TODD

Fig. 23. Interaction between Pseudotrametes gibbosa (left) and Bjerkandera adusta on 3 malt agar showing rapid and vigorous replacement by the former.

first described by Ikediugwu and Webster (1970a, b), were found between Stereum hirsutum and Heterobasidion annosum and between Pseudotrametes gibbosa and Coriolus versicolor. The second type of antibiotic zone was produced when growth of both fungi halted before contact. This was frequently observed with Armillaria mellea, which is known to produce antibiotics (Oduro et al., 1976). It was also observed when Daedalea quercina was paired against Pseudotrametes gibbosa or Daedaleopsis confragosa. D . guercina in fact inhibited growth of several fungi well before contact (often when colonies were separated by 2 cm or more) including Bjerkandera adusta and Heterobasidion annosum, although its own growth rate remained unaffected. (c) Formation of coloured zones. As reported by Henningson (1967) formation of coloured contact zones in the medium between colonies was a

FUNGI IN DECAYING WOOD

395

Fig. 24. Photomicrograph showing coiled hyphal branches of Pseudotrumetes gibbosu around an empty hypha of Bjerkunderu udustu (arrowed).

frequent and striking result of their interaction. Examples of pairings where such zones occurred are given in Table X. So striking were these zones in some instances that they may in time prove of diagnostic va1ue-e.g. the blue-green zone formed by Chondrostereum purpureum when interacted against Coriolus versicolor, Daedaleopsis confragosa and Pleurotus ostreatus : a similar zone was detected by Henningson (1967) in interactions between C. purpureum and C . versicolor. Whilst in some cases the coloured zones were more or less restricted to the region of contact between opposing mycelia (e.g. the yellow zone produced by Stereum Izirsutum), in others it diffused throughout the medium (e.g. the brown pigment produced by Daedalea quercina and Piptoporus bet ulinus). The brown pigments produced by D . quercina and P . betulinus just mentioned were often associated with inhibition, or even death of opposing mycelia. In pairings between D . quercina and Heterobasidion annosum, growth of the latter was inhibited long before contact, but directly it (i.e. contact) occurred, D . quercina produced copious pigment, after which neither fungus grew further. Microscopic examination revealed considerable vacuolation and granulation in hyphae of both species in the contact zone. The progress of the interaction between P . betulinus and Phlebia merismoides was also of interest in this connection. Initially P . merismoides produced a dense

396

A. D. M. RAYNER AND N. K. TODD

salmon-pink zone of mycelium in the contact zone, and this was followed by lysis of P . betulinus. However, the latter then began to produce brown pigment, and grew back across the lysed zone, ultimately partially replacing P . merismoides. TABLE IX Deadlock interactions at 15°C on 3% malt agar of fungi commonly colonizing hardwoods

Fungus

No. interactions tested Species with which deadlock occurreda

Bjerkandera adusta Chondrostereum purpureum Coriolus versicolor Hypholomafasciculare Phanerochaete velutina Phlebia merismoides Pseudotrametes gibbosa

25 19 23 18 9 17 18

Scytalidium album Stereum hirsutum

25

9

BF, CS AM, BC, CS, GA, PG, SH AM, BC, DC, GA, PVe, PM, PB, PG PM, PB CV, PM, SH CV, HF, PVe AM, BC, CP, CV, CS, DQ, FH, GA, HR, OR, PB, PO, PV, SH

None AM, BC, CP, CS, DQ, DC, FH, HA, MG, OR, PVe, PO, PG, SG -

a For key see Table VII.

After Rayner (1978).

Changes in temperature usually had little effect on production of coloured zones, except sometimes on their intensity. (c) Isolate variability in interactions on malt agar. The possibility that the interrelation between two species may vary with different isolates is of course an important one, affecting both the likely taxonomic and ecological value of interaction studies, and requiring far more extensive investigation. Such variability has indeed been reported by some, mostly concerning antibiotic activity; for example Etheridge (1957) noted marked variation in the inhibitory powers of a range of isolates of Coryne sarcoides. Keyes (1968) noted variability in the direct interactional relationships of several wood-decaying polypores, but unfortunately her report is insufficiently detailed for us to assess the real extent of such variability. In Rayner’s original investigations (Rayner, 1975) practical considerations militated against use of more than a single isolate in all but a few species combinations. Subsequently he and his colleagues have increased the number of isolates tested, especially for species such as Hypholoma fasciculare, Phlebia merismoides and Phanerochaete velutina which were considered as potentially useful biological control agents for Armillaria mellea (Woodgate-Jones, 1977). It was found that although slight variations occurred, the type of interaction was consistent between different isolates. This is supported by the reasonably good agreement

FUNGI IN DECAYING WOOD

397

TABLE X Formation of coloured contact zones by fungi as a result of interactions on 3 % malt agar

Fungus Bjerkandera adusta Botrytis cinerea Chondrostereum purpureum Coriolus versicolor Daedalea quercina

Type of coloured zone@ Species provoking responseb Pale ochreous Fuscous black Pale ochreous Amber Verdegris-dark cyan blue Olivaceous Sienna-sepia

Daedaleopsis confragosa Apricot Sepia Flesh Fistulina hepatica Cinnamon Heterobasidion annosum Ochreous Hypholoma fasciculare Luteous Piptoporus betulinus Sienna-sepia Pleurotus ostreatus Ochreous Polyporus varius Sienna-sepia Scarlet-sienna Stereum gausapatum Ochreous-luteous Sulphur (crystals) Orange Umber Stereum hirsutum Ochreous-luteouspure yellow

BF, CS, HA GA, PG, SG, SH BA, CS, PB DC, SH CV, DC, PO PM, PG AM, BA, BC, CV, FH, HA, HF, OR, PV, PG, SG, SH LS DQ

SG DQ, DC, PG PB, SH BA, CV, DC, DQ, PB, PO, PG CV, CS, HA, HF, PM, SH AM, BA, BF, GA BA, GA, HA, PB, PO, PG, SH CP, PM BA, CV AM, FH, HR LS DQ, HF AM, BA, BF, BC, CP, CV, CS, DQ, DC, FH, GA, HA, HF, LS, MG, OR,PM, PB, PO, PV, PG, SG DQ

Flesh a Colours are described in accordance with Rayner (1970).

* For key see Table VII.

between Rayner’s results and those obtained by other workers (e.g. Henningson, 1967; RypBhk, 1966) in the few cases where the same species combinations have been tested.

2. In Wood Lengths and Sawdust In view of the problems associated with extrapolating from results obtained on artificial media to the field, it is important to find ways of testing interactions on more natural substrata in the laboratory. Curiously, few such attempts have been made, and technical difficulties still abound. In his original studies Rayner attempted to study interactions in short lengths of stem cut from living shoots of birch, beech and oak. These were either inoculated at opposite ends with different pairs of fungi, or allowed to become permeated

398

A . D. M. RAYNER AND N . K. TODD

with mycelium of different fungi, and then held adjacent to each other. As other workers have found, difficulties were experienced in obtaining establishment of the inoculated fungi. Nevertheless, confirmatory evidence for a number of the results obtained on agar was obtained, notably the replacement of Bjerkandera adusta and Chondrostereum purpureum by Phlebia merismoides, and of B. adusta by Pseudotrametes gibbosa. Production of PSPs by Ganoderma adspersum as a result of interaction with other fungi was also demonstrated (Fig. 25).

Fig. 25. Short stem lengths of beech six months after inoculation at opposite ends with Ganoderma adspersum (left) and Pseudotrametes gibbosa respectively. The lengths have been cut into half longitudinally: one half is shown intact and the other cut into serial transverse sections. Notice the marked zone lines resulting from PSP formation between the two colonies.

A rather more successful approach subsequently used (Carruthers and Rayner, 1979) was to pair fungi opposite each other in Petri dishes on oak sawdust with moisture contents varying from 80-300% by dry weight. In fact the moisture level had no effect on the type of interaction, which is shown, for the six common hardwood-decaying fungi tested, in Table X I . Phlebia merismoides and Hypholoma fasciculare replaced most other species; Phallus impudicus replaced Coriolus versicolor and Stereum hirsutum ; C . versicolor replaced S. hirsutum. The interaction between P . merismoides and H . fasciculare was variable, deadlock occurring in most trials (32), but replacement by P . merismoides occurring in three, and by H .fasciculare in one. The mechanism of replacement varied in different species-combinations. P . merismoides produced little aerial mycelium, but replaced other species rapidly, separate zones of advancement being visible across the dish. Cord formation in P . impudicus and H . fasciculare was inhibited. Replacement by H . fasciculare was slower than P . merismoides ( H . fasciculare has a much slower radial growth rate), and much aerial mycelium was produced-often aggregating into cords. Again replacement was apparently intermittent,

399

FUNGI IN DECAYING WOOD

separate zones of advancement being visible across the dish. P . impudicus rarely occupied a large volume of sawdust, but produced a dense zone of mycelium around the colony margin, effectively preventing penetration by hyphae of other species. From this base mycelial cords grew across the surface of opposing mycelia and penetrated the sawdust at the opposite edge of the dish. New colonies of P . impudicus were thus established surrounding opposing mycelia, which were gradually overwhelmed. TABLE XI Results of interaction experiments between jive common wood-decay fungi on sawdust plates (36 trials per pairing) ~.

Fungus .~

Coriolus versicolor Hypholoma fasciculare Phallus impudicus Phlebia merismoides Stereum hirsutum

~

Species replaced Replacing species ~~~. __ SH PM, HF, PI CV, PI, SH CV, SH CV, SH -

PM (3 trials) HF HF (1 trial) CV, HF, PI, PM

Species with which deadlock occurred -

PM (32 trials) HF, PM HF (32 trials), PIa -

__ a Cord formation by P. impudicus inhibited. Abbreviations as for Table VII except PI-Phallus impudicus. After Carruthers and Rayner ( 1979).

D. INTERACTIONS IN NATURE

We have already indicated criteria whereby interactional events can be inferred from community patterns at any one time in natural substrata. Further information may become available if we observe changes in communities occurring over a period of time within the same substratum or its equivalent; for example if we know that fruit-bodies of a certain fungus were present on a log or stump at some stage in its history, but that that species can no longer be detected during detailed analysis at a later stage, its replacement by others can often be inferred. Direct inoculation can also provide useful information. Such inoculation can either be of single species, whose interaction with fungi colonizing naturally can be followed, or deliberate inoculation of more than one fungus in closely adjacent positions can be made. The latter type of experiment has hardly been attempted, but its potential i s clear, as has recently been shown by preliminary results obtained by Woodgate-Jones (1 977). Little information is available relating to interactional events in natural woody substrata, possibly because it is not generally appreciated how this may be obtained. Rayner (1978) reported interactional patterns observed in hardwood stumps inoculated with a variety of decay-causing basidiomycetes, and found many examples of replacement (Table XII). Further evidence for replacement occurring in nature was obtained by Carruthers and Rayner (1979)

400

A. D. M. RAYNER AND N. K. TODD

TABLE XI1 Replacement of fungi in stumps Fungus replaced

Fungus responsible

Phanerochaete velutina, Phlebia rnerismoides, Pseudotrametesgibbosa, Scytalidium album Chondrostereum purpureum Bjerkandera adusta, Coriolus versicolor, Hypholoma fasciculare,Phanerochaete velutina, Phlebia merismoides, Stereum hirsutum, microfungi Hypholoma fasciculare, Phlebia merismoides Coryne sarcoides Armillaria mellea, Hypholomafasciculare Coriolus versicolor Scytalidium album Heterobasidion annosum Phanerochaete velutina Hypholoma fasciculare Scytalidium album Phlebia merismoides Phlebia merismoides Stereum hirsutum Bjerkandera adusta

After Rayner (1978).

TABLE XI11 Evidence for replacement in cut branches of oak and ash Fungus

Hypholomafasciculare

Fungus replaced Stereum hirsutum Coriolus versicolor Xylaria polymorpha Heteroporus biennis Phlebia merismoides

not identifiable Phlebia merismoides

Yellow basidiomycete

Stereum hirsutum Coriolus versicolor Xylaria polymorpha Hypholomafasciculare Heteroporus biennis Stereum hirsutum Stereum hirsutum

Coriolus versicolor Phallus impudicus Total (30% of total samples observed)

No. of samples where replacement detected 3 2

2 1

3 8 14 2 1 1 1

2 1 41

After Carruthers and Rayner (1979).

during an analysis of fungal communities present in cut hardwood branches, and is given in Table XIII. The rather remarkable feature of these results is their close correspondence with events observed in the laboratory, including those on 3% malt agar. Particularly striking individual cases were the replacement of Bjerkandera adusta by Phanerochaete velutina, Phlebia merismoides, Pseudotrametes gibbosa and Scytalidium album, and of Hypholoma fasciculare by P. velutina. More generally, the marked replacing ability of H.fasciculare, P. merismoides, P. velutina and S. album all of which tend to be predominant at later stages in community development (Rayner, 1977a, b ; Carruthers and Rayner,

FUNGI IN DECAYING WOOD

401

1979), and the susceptibility of Chondrostereum purpureum to replacement by a wide range of species were also notable. More cautiously we may note that replacement of C . purpureum by Stereum hirsutum, and of Coriolus versicolor by Armillaria mellea reported in stumps had not been observed on agar. Further we might observe that replacement of C . versicolor by P. merismoides observed by Carruthers and Rayner in both logs and sawdust, was not observed in Rayner’s earlier studies where deadlock was indicated for this combination. As a final note of caution we must remember that whilst the majority of these instances of replacement in nature appeared to result from direct mycelial interaction, more indirect mechanisms (see above) could not always be completely eliminated. To end on a more optimistic note, the evidence for field replacement of Bjerkandera adusta by Pseudotrametes gibbosa was amongst the most satisfying, and was provided by the results of Rishbeth (1976) obtained in 23 sycamore stumps. Of these, nine had been inoculated with Coriolus versicolor and after two years contained only that fungus, six had not been inoculated and contained either C . versicolor or Xylaria hypoxylon, and eight had been inoculated with B. adusta. Seven of these latter were extensively colonized by P . gibbosa. The mycoparasitic activity noted previously for P . gibbosa against B. adusfa will be remembered. E. CONCLUDING COMMENTS

In this section we have tried to show that studies of interactions between fungi inhabiting decaying wood represent an exciting and as yet surprisingly little explored field for research. We have emphasized the lack of information which is available at all levels, but especially in the natural resource. Even simple laboratory experiments on agar media have been surprisingly infrequently reported, especially when it is remembered that the first of these, by Harder, were carried out as long ago as 1911. Whilst the possible ecological relevance and use for diagnostic purposes of interaction studies has been stressed, so also has the need for caution in interpretation, especially when extrapolating from results obtained on artificial media to the field. Some might question altogether the value of experiments on artificial media, and indeed have done so in the past. Here we must remember that such experiments are by far the easiest to perform on a large scale, and also that they allow direct observation of possible mechanisms of interaction-the relation between Pseudotrametes gibbosa and Bjerkandera adusta could not so clearly have been understood in their absence. This last example is one of the few cases where we may have glimpsed the true physiological basis for an interaction: such information is largely lacking, and this presents much scope for further research. It seems probable that in many cases mechanisms similar or identical to “hyphal interference” (Ikediugwu and Webster, 1970a, b) may be involved, whereby coagulation, vacuolation

402

A. D. M. RAYNER AND N. K. TODD

and death of cytoplasm occurs in hyphal segments in contact with other species. Such hyphal interference has indeed been shown to occur when hyphae of Phlebia gigantea (Fr. ex Fr.) Donk ( = Peniophora gigantea (Fr.) Massee) contact and kill those of Heterobasidion annosum (Ikediugwu et al., 1970). Both hyphal interference, and the type of reaction seen between P . gibbosa and B. adusla are examples of direct mycelial interactions, requiring contact between hyphae; other such direct interactions may have different physiological bases, for example the lysis caused by Phlebia merismoides and Phanerochaete velutina when in contact with mycelia of other species. These direct types of interaction have been distinguished by some (including Etheridge, 1971) from indirect types involving production of diffusible antibiotics, and we have more or less followed this. However, we do not entirely agree with the tendency of these same authors to describe all cases of direct mycelial interaction as “mycoparasitism” : to us this term must be restricted to those instances where one fungus clearly uses (at least initially) living hyphae of another as a nutrient source. Such is clearly not the case in deadlock interactions between mycelia involving direct interference phenomena. It was encouraging to find, in spite of expressed doubts, that in Rayner’s studies results obtained on 3 % malt agar correlated well with those in the field. Further, his results do not stand in isolation since, like Rayner, both RypBEek (1966) and Henningson (1967) found that those species predominant at later stages of decay were those most competitive on agar. It is interesting to consider why results on 3 % malt agar should have been so directly predictive of field events in so many, but not all, instances. Is there any way in which we can detect whether an event observed on agar is likely to be repeated in the field? To answer this requires that we consider the likely pattern of growth of fungal mycelia in wood, and the degree to which this might be reflected on agar. One of the main physical differences is of course that fungal mycelia in wood grow in a three-dimensionally constrained environment; there is therefore no opportunity for simple physical overgrowth, and we can therefore assume that such a manner of replacement as observed on agar (e.g. simple overgrowth of PSPs of Ganoderma adspersum, Armillaria mellea) should not necessarily be predictive of events in nature. What is important is the ability of one mycelium to grow straight into an opposing colony. Here the sheer density of hyphae of opposing mycelia may be important: on 3 % malt agar fairly dense growth is usually found and expression of any antagonism between hyphae would occur rapidly. In wood it might be thought that growth would be more diffuse, and antagonism less clearly expressed. This seems not to be the case: whilst in the central parts of decay zones hyphae are often sparse, at their periphery, especially adjacent to decay zones occupied by different fungi, dense wefts of mycelium are often present. A further comforting feature is the close similarity often found between mycelium on 3 % malt agar, and that produced by incubation of decaying

FUNGI IN DECAYING WOOD

403

wood (Rayner, 1975). This latter observation also indicates that there may often be a closer similarity physiologically between fungal mycelia in wood and on malt agar than might have been expected. The final possible point of difference between wood and agar that we might consider is diffusion of soluble antibiotics-diffusion distances seem likely to be greater in agar; if so this makes direct mycelial contact likely to be most important in determining the outcome of fungal interactions in wood. In summary, then, it seems likely that the most predictively valuable interactions on agar will be those which involve direct mycelial antagonism resulting from physiologically close contact between hyphae, such as occurs between Bjerkandera adusta and Pseudotrametes gibbosa, and where the pattern of growth of the fungi on agar most closely resembles that in wood. In conclusion, we believe that interspecific interactions between fungi in decaying wood are potentially highly significant, both from an academic standpoint in understanding community development, and also economically in developing means for prevention of disease and decay. A clear demonstration of both these points is the use of Phlebia gigantea for biological control of Heterobasidion annosum, which is a serious root-infecting pathogen of conifers (Rishbeth, 1963). H . annosurn normally enters into conifer plantations via basidiospore infection of freshly cut surfaces of stumps produced following thinning or clear-felling. The pathogen subsequently grows down into stump tissues and out into the roots, from which it may spread, at sites of root contact, into surrounding trees. Ecological studies by Meredith (1959, 1960) revealed that H . annomm and P . gigantea were amongst a small group of basidiomycetes which were pioneer invaders of pine stumps, and that colonization by P . gigantea militated against infection by H . annosum. Subsequently, this finding was developed into the method for biological control which is extensively practised in Britain. The interaction between these two fungi on 3 % malt agar seems usually to be deadlock, and this appears to be repeated in the field (Gibbs and Smith, 1978). As mentioned previously, the interaction between these fungi involves hyphal interference. Whilst we can say, then, that cautious interpretation of interaction experiments is justified, and indeed essential, neglect of these studies is not. V. ECOLOGICAL ROLES AND SPATIAL DISTRIBUTION WITHIN COMMUNITIES O F FUNGI I N DECAYING WOOD A. INTRODUCTION

We have established that communities composed of different fungal mycelia do often occur in decaying wood, and have seen how knowledge of interactions between these mycelia, be they via direct or indirect mechanisms, may help us to understand how population and community structure is

404

A. D. M. RAYNER AND N. K. TODD

maintained or changed. We need now to consider the fungal community in more functional terms, and to concern ourselves with its operation as a whole (i.e. the co-ordinated action of its component parts) during decomposition processes. For this purpose, in the tradition of biological dissection, we need first to understand in more detail how communities are constructed from their components, and what the individual activity and ecological requirements of each of these (components) may be. We may expect that such activities and requirements will vary and that this will lead to differences between fungi in their ecological roles and spatial distribution within communities. We will examine this possibility in relation to a variety of factors which may be considered likely to influence the mode and pattern of growth of fungi, and hence community structure and function in decaying wood. B. FACTORS INFLUENCING MODE AND PATTERN OF GROWTH OF FUNGI IN DECAYING WOOD

I . Mode of Nutrition We have already mentioned that fungi colonizing wood vary in their ability to utilize available carbon sources, from those which remove only sugars and similar relatively simple carbon compounds, to those removing cellulose and lignin, and that this affects the type and intensity of decay. The ratio between fungi relatively inactive in decay and active decomposers in any particular community is, of course, likely to determine the decay capacity of the community as a whole. To what extent this will be so, will depend in turn on how long the relatively inactive species can persist: where this is considerable the overall rate of decay may be greatly diminished. One hopes that such would be the case for fungi, such as Scytalidium and Trichoderma spp. advocated as possible biological control agents for decay. Certain ascomycetes causing only slight or moderate decay such as Hypoxylon multifovme and Melanoma pulvis-pyrius appear able to penetrate wood deeply and to persist for considerable periods. Where, on the other hand, non-decay species occur only transiently, there may be little diminution in the overall decay capacity of the community, and indeed it may often be that via other effects, such as removal of toxic substances or host resistance, they facilitate subsequent colonization by fungi more active in decay. The possible effect of their mode of nutrition on the temporal and spatial distribution of fungi raises some interesting questions. It might be imagined that relatively simple carbon compounds would be used up rapidly, and that fungi relying on these would have to colonize early. Whether this is so could, however, depend on how efficiently lignin and cellulose decomposers absorb monomers resulting from the action of their enzymes on the cell wall polymers: Garrett (1 963) in his general schema for fungal successions in plant remains envisaged the likely occurrence of secondary sugar fungi associated with

FUNGI IN DECAYING WOOD

405

lignolytic and cellulolytic fungi and utilizing excess sugars produced by these latter. Regarding spatial distribution, it is sometimes stated that active decay fungi causing brown and white rots penetrate deeply into wood, whilst less active forms, such as those causing staining or soft-rot are often restricted to surface layers. This is certainly not true at early stages of colonization in natural hardwood substrata when both decay and non-decay species may be present and penetrate deeply. Subsequently, however, colonies of the decay fungi may expand and replace other species which become restricted to those sites where mycelium of decay species cannot grow freely, and these may include superficial tissues. A fascinating group of fungi imperfecti, which by implication are not active in decay, has recently been discovered in decaying hardwoods (Rayner, 1976). These fungi grow in the narrow, relatively undecayed interaction zones which are formed between mutually antagonistic mycelia of such white-rotting species as Bjerkandera adusta, Coriolus versicolor, Hypholoma fasciculare, Phlebia merismoides and Stereum hirsutum. Not all these fungi have been identified with certainty, and doubtless some remain to be discovered, but amongst the most important are several Rhinocladiella species, the imperfect Catenularia state of Chaetosphaeria myriocarpa (Fr.) Booth, an Endophragmiella sp. and species of Cladosporium, Phialophora and Margarinomyces. These fungi are generally dematiaceous (i.e. have dark hyphae) and this often intensifies the dark appearance of interaction zones. In wood undergoing active white rot they are often the only imperfect fungi present to any depth internally. Microscopic examination of sections through wood containing interaction zones occupied by these fungi (Fig. 26 (a) ) often shows dense wefts of dark hyphae growing predominantly longitudinally, and often very narrowly restricted to the zones which may be as little as two or three xylem elements wide. Incubation of freshly cut sections on moist filter paper often serves to allow branches from the dark hyphae to grow out and sporulate, facilitating identification (Fig. 26 (b) ). The precise ecological status of these fungi which colonize interaction zones is uncertain and requires further investigation. Firstly we need to establish if they are truly specialized for growth in interaction zones, or whether they can colonize relatively undecayed wood more generally and are simply restricted to sites into which decay fungi cannot penetrate effectively(as in interaction zones). Secondly, their mode of nutrition is of interest: do they obtain nutrients from the wood itself, or do they perhaps obtain breakdown products resulting from the activity or death of hyphae of the decay fungi? If the latter possibility pertains, then they might truly be regarded as secondary sugar fungi as defined by Garrett (1963), albeit of a rather unusual kind. Apart from the interaction zone mycoflora there are few other candidates for the status of secondary sugar fungi in actively decaying wood because the

406

A. D.

M. RAYNER AND N. K . TODD

decay zones are usually so densely and uniformly filled by the decay species themselves.

Fig. 26. Photomicrographs showing L.S. through an intraspecific interaction zone in wood decayed by Hypholoma fasciculare (that shown in Fig. 8). (a) Before incubation showing presence of dark hyphae running along xylem elements (arrowed). (b) After incubation showing mycelium and conidiophores (arrowed) of a Rhinoeladiella sp.

2. Moisture Conrent and Aeration The moisture requirement for growth of fungi in wood is rather high: in general 26-32% on a wet weight basis is necessary for initiation of decay, and optimal growth occurs at about 40%. However, these requirements vary somewhat with different fungus species and in different types of wood; for example Coniophora puteana (Schum. ex Fr.) Karst. and Paxiffuspanuoides (Fr. ex Fr.) Fr. grow better at somewhat higher moisture levels than other fungi (Cartwright and Findlay, 1958). Below c. 26 % decay is not likely, then, to be initiated, but once established, many species of decay fungi can survive in wood at much lower moisture levels for a considerable time. For example Theden (1961) incubated colonized pieces of wood at well below c. 26% moisture and found that all ten species tested survived for over a year, and that the three hardiest (Shizopova pavadoxa (Schrad. ex Fr.) Donk, Fibufoporia vaillantii (DC) Bond. & Sing. and Lentinus lepideus (Fr. ex Fr.) Fr.) survived over six years. Above about 40% moisture, growth of most fungi in wood declines as conditions become increasingly anaerobic : here we see the importance of the close interrelationship between moisture content and aeration. It is generally

FUNGI IN DECAYING WOOD

407

held that the pore space in wood must contain around 20 air before decay becomes possible, and this has been confirmed experimentally by RyphEek t 1966). In a large mass of wood the situation may be more complex as there is likely to be variation in the composition of the gaseous phase. For example, in a standing tree the content of oxygen may be greatly reduced, whilst the carbon dioxide content is raised and there may be physiologically important concentrations of other gases such as ethylene present. Hintikka and Korhonen (1970) studied the effects of high levels of carbon dioxide on some 90 basidiomycetes including both wood- and litter-inhabiting species. Whilst the latter were almost all completely inhibited by a partial pressure of 10 kPa, all the wood-inhabiting species grew at 30 kPa, and some even at 70 kPa which was the highest level tested (e.g. Lenzites betulina (L. ex Fr.) Fr., Pholiota squarrosa (Muller ex Fr.) Kummer and Piptoporus betulinus). It seems clear then that moisture content and aeration influence growth of wood decay fungi and that their effects will vary with different species. Accordingly, we might expect that in nature the species-composition of communities would be likely to depend on moisture levels. However, there has been very little direct ecological work investigating this possibility. Results obtained by Kaarik (1971, reported also in Kaarik, 1974) were considered by her to demonstrate clearly the influence of moisture content on the composition of fungal communities in wood. She showed that in decorticated spruce poles standing in soil, there was a very clear moisture gradient from 33-47 % (by wet weight) below ground to 17-23 % aerially and that this was associated with differences between fungus species colonizing the different parts. Three groups were distinguished: I . Those colonizing below ground only, e.g. Heterobasidion annosum, Fibuloporia vaillanrii, Serpula himantoides (Fr.) Bond and S. pinastri (Fr.) Bond. 2. Those colonizing aerial portions, e.g. Phlebia gigantea, Skeletocutis amorphus (Fr. ex Fr.) Kolt & Ponz. and Hirschioporus abietinus (Dicks. ex Fr.) Donk. 3. Those colonizing both above and below ground, e.g. Sistotrema brinkmanni (Bres.) Rogers and Hypholoma capnoides (Fr. ex Fr.) Kummer. These differences in species-composition are of great interest, but in principle there seems to be no reason for relating them specifically to moisture content rather than to some other variable, such as availability and type of colonizing propagule (see below), likely to differ above and below ground. Undoubtedly it is often a problem in ecological investigation of fungal colonization of wood in the field, to distinguish between the effects of different variables when more than one is operating at any one time. 3. Temperature The optimum and maximum temperatures for growth of fungi in wood vary

408

A. D. M . RAYNER AND N. K. TODD

greatly for different species. Differences in temperature may therefore be expected to influence community composition. For most European species the optimum temperature for growth is usually between 2430°C (on agardata obtained from wood are still very insufficient).The geographical distribution of some species may be related to their temperature requirements: thus Serpula lacrymans (Wulf.) Schroet. has a remarkably low maximum (2526°C) and is absent from the tropics and parts of the world with high summer temperatures. Henningson (1967) provided evidence that the competitive ability of some species varied with temperature and considered that this could affect their distribution in pulpwood piles. For example Bjerkandera adusta and Coriolus hirsutus (Wulf. ex Fr.) Qud. which in plate tests appeared most competitive at high temperatures, were almost exclusive to the uppermost parts of piles where the temperature was likely to be highest. It is difficult however, to discern whether this was a true temperature effect, or due to some other variable. Generally there is little direct evidence available relating the temperature requirements for growth of fungi in wood to their distribution in natural communities. 4. Nitrogen Content The nitrogen content of wood is often held to be very low, the C : N ratio in most cases being between 350 : 1 and 500 : 1, and sometimes as high as 1250 : 1 in Sitka spruce (Picea sitchensis) (Cowling, 1970). In addition its distribution in different tissues is variable, being maximal near the cambium and minimal in the outermost heartwood (Merrill and Cowling, 1966). It is likely then, that nitrogen content may be an important factor determining the pattern of growth of fungi and decay in wood, and this subject has recently been reviewed by Cowling (1970). It seems that wood-decay fungi may be physiologically adapted to growing at low nitrogen levels in one of two ways, judging by results obtained with Coriolus versicolor. They may on the one hand allocate nitrogen to metabolically active cell constituents such as the nucleic acids when growing on media low in nitrogen (Levi and Cowling, 1969). Alternatively, they may conserve nitrogen by autolysis and re-use of nitrogen in their own mycelium, or by lysis of other fungi present in the wood (Levi et al., 1968). Further experiments are necessary with other woodinhabiting fungi to determine how general such physiological adaptations may be. Important though nitrogen availability may be to fungi growing in decaying wood, we should mention that the contention that the nitrogen content of wood is low has recently been questioned by Park (1976), who has criticized in particular the use of C : N ratios as criteria for determining nitrogen levels.

FUNGI IN DECAYING WOOD

409

5. p H

The p H of wood varies with different species and in some cases may be an influence in determining the pattern of fungal colonization. In general, most wood samples have a p H of 4-6, which corresponds with the range for growth of many wood decay species. Some woods, e.g. oak and chestnut may have a particularly low pH, most often due to the presence of acetic acid. Hintikka (1969) has shown that some 70 wood-rotting fungi grew well on an artificial medium with 0-008 M acetic acid, but that soil basidiomycetes were usually suppressed. Fungi characteristic of dicotyledonous wood were more tolerant than those typically found in conifers. The most tolerant species were Laetiporus sulphureus and Daedalea quercina, both of which cause a brown heartrot in oak, which grew with 0.03-0.05 M acid in the medium. 6. Wood Anatomy

Wood is, of course, a highly heterogeneous resource in which there is considerable variation in the distribution and occurrence of vessels, tracheids, fibres, parenchyma, rays, annual rings, spring and autumn wood, tyloses, heartwood, sapwood and so on. With such structural heterogeneity it might be expected that some degree of specialization might occur amongst fungi colonizing wood for attack on different components, such that some might, for example, colonize predominantly medullary ray cells, others vessels and tracheids. How far such specialization occurs we are not sure and extensive review of this topic is beyond the scope of this article. Amongst decay fungi it does not seem that there is any long-lasting specialization, all the components eventually being attacked. There is evidence that attack by many fungi may initially be restricted to certain components: for example that staining and soft-rot fungi may occur first in the ray parenchyma. A factor of importance in determining the rate of spread of fungi in wood is undoubtedly the mechanical impedance imposed by lignified cell walls. This is greatest tangentially, less radially (i.e. where rays offer a relatively easy avenue for spread) and least longitudinally. Consequently, decay fungi often occupy longitudinally elongated columns of wood which are more or less wedge-shaped in cross-section. The cambial zone in dead wood offers particularly easy access for mycelial spread, but one also where thelikelihood of intensecompetition between microorganisms will be high. Recent studies in hardwoods (Rayner, 1977b) have indicated that there are certain fungi specialized to grow in this zone as mycelial sheets, fans or cords. These fungi often colonize the cambial zone first and then spread inwards. Consequently, they are often found in peripheral decay zones in fungal communities in dead stumps, logs, etc. They include such species as Hypholoma fasciculare, Phlebia merismoides, Phallus impudicus and Phanerochaete velutina, all of which appear highly competitive in interaction experiments on artificial media. Many of them colonize wood as

410

A . D. M. RAYNER AND N. K . TODD

vegetative mycelium. This aspect is of some interest and will be considered further below. 7 . Toxic Substances Heartwood of many trees usually contains large amounts of extractives scarce or absent in sapwood and which are strongly fungitoxic or fungistatic (see Hudson, 1972). In gymnosperms these include terpenoids, tropolones, flavonoids and stilbenes; whilst in hardwoods, tannins are often important. For any particular tree there are always some species which can tolerate these substances and grow, albeit slowly, in their presence. Such adaptation of just a few species to a particular heartwood may be an important reason for the specificity to certain trees often exhibited by fungi causing heartrot. During community development in heartwood, initial removal of these substances by pioneer colonizers may be important in allowing subsequent attack by other organisms.

8. Host Resistance In the living tree, as opposed to felled timber, heartrot is of greatest importance. This is because living sapwood is rarely decayed except by true parasites such as Armillaria rnellea, Heterobasidion annosurn and Chondrostereurn purpureum, which can be interpreted as being due to resistance to infection of the living tissue. As the level of resistance declines, for example following infection by a parasite such as C . purpureum, or felling or poisoning, then the sapwood becomes increasingly susceptible to colonization by saprophytes. 9. Method of Entry of Fungi

Essentially fungi may colonize wood either by spores or as vegetative mycelium. Colonization by spores may occur via any receptive surfaces on which they can germinate-most generally those which are aerially exposed. Where, as with basidiospores, there are genetic differences between the invading propagules, this is likely to result in the formation of separate columns, occupied by mutually antagonistic mycelia and delimited by interaction zone lines, as occurs with Coriolus versicolor and Stereurn hirsuturn which are typical examples of fungi colonizing by spores in hardwoods. If colonization occurs via a cut or broken surface, as with a cut stump or log, then there is usually no obvious tendency for colonization primarily of outermost tissues. This contrasts with fungi colonizing as vegetative mycelium, where the outermost wood often is preferentially colonized, at least initially. These fungi generally colonize either as a result of contact with infected wood, or very often as aggregated systems of hyphae such as mycelial sheets and cords, or rhizomorphs.* Either way, entry is usually via the soil or litter,

* Rhizomorphs are distinguished from mycelial cords by their higher level of organization as discrete structures growing apically, often in a remarkably root-like manner.

FUNGI IN DECAYING WOOD

41 1

aerially exposed parts not being susceptible to colonization in this manner. Having entered, the fungi then spread over the surface of the wood, or, very frequently as extensive subcortical systems of mycelial sheets, cords or fans. Below ground much superficial growth is often present, but above ground, where external surfaces will be subject to desiccation and simiIar adverse influences, subcortical spread predominates (Fig. 27). This pattern of growth results in very rapid occupation of peripheral tissues, which will have the advantage of surrounding the resource and preventing colonization by potential competitors prior to radial growth into deeper tissues. Since colonization is from a mycelial source, it is comparatively rare for there to be genetic differences between invading propagules, as would occur with basidiospores, thus interaction zone lines are infrequent. Several explanations are available for the preferential colonization of outermost tissues by fungi entering as vegetative mycelium. Firstly, since cut or broken surfaces are uncommon in ground or litter contact, the first point

Fig. 27. AMS-treated beech stump 30 months after felling, showing considerable colonization by Phallus impudicus which has formed extensive superficial and subcortical networks of mycelial cords (after Rayner, 1977b).

412

A. D. M. RAYNER AND N. K. TODD

of entry will inevitably be through superficial tissues. Secondly, for certain root-infecting parasitic fungi such as Armillaria mellea and Heterobasidion annosum, extensive superficial mycelial growth may as Garrett (1970) and others have suggested, be a prerequisite for infection of the wood, acting as a mechanism for diluting out host resistance. Finally, as mentioned above, in dead wood the cambial zone is a line of mechanical weakness, probably of high nutrient status, in which extensive mycelial spread is possible for suitably adapted fungi. Fungi forming mycelial cords, rhizomorphs or similar structures and which colonize wood vegetatively from the soil or litter subsequently to spread by extensive subcortical growth form a distinct and, at least in British soils, very widespread ecological group. They include both parasites, such as Armillaria mellea and saprophytes such as Hypholoma fasciculare, Phallus impudicus, Phanerochaete velutina and Tricholomopsis platyphylla. Whilst something (but by no means enough) is known about the biology and ecology of A . mellea in Britain (e.g. Redfern, 1968; Rishbeth, 1972; Morrison, 1976), the saprophytes have been almost entirely neglected. Yet these fungi are potentially of profound significance in the forest ecosystem. In the first place they are very abundant and widespread, so that their role in decomposition of wood in ground contact is probably a major one. Secondly since to a large extent they show the same pattern of colonization of dead wood as A . mellea when it is acting saprophytically, they may play a significant role in competing with the pathogen and lowering its natural incidence. It is important to stress that such competition with A . mellea can only occur in the absence of host resistance: where living tissue is present the parasite will have a selective advantage. Production of mycelial cords and rhizomorphs by decay fungi colonizing wood via the soil or litter may have a variety of advantages. Firstly, since they are effectively a way of concentrating fungal inoculum they may increase the invasive force (inoculum potential, as defined by Garrett, 1970) of a fungus, helping it to overcome host resistance if invading parasitically, or microbial competition when colonizing saprophytically. Further, since they are aggregated structures, cords and rhizomorphs may be expected to be more resilient and to resist adverse external influences better than single hyphae. Such adverse influences may include desiccation, waterlogging and antibiotic activities of other organisms, all of which may be expected to be important in soil. Finally, the structures may act as translocating organs through which water and nutrients may be channelled from a food base to the outlying points of a growing colony. A possible disadvantage of such bulky structures is that they could form an easy target for grazing by the soil microfauna. Production of external melanized crusts (as with rhizomorphs of A . mellea) or extensive deposits of crystals (as with many cord-formers) may be of adaptive value in this respect, reducing palatability.

FUNGI IN DECAYING WOOD

413

Much remains to be learned about mycelial cords and rhizomorphs, and in general, with the exception of A . mellea rhizomorphs and mycelial cords of Serpula lacrymans (dry rot fungus), little real progress has been made since the early studies of De Bary (1887). Some of the few significant subsequent contributions have been those of Townsend (1954), Grainjer (1962) and Butler (1 966). IO. Interactions with Other Organisms The probable importance of interactions between fungi in determining patterns of population and community development in wood has been stressed repeatedly in this article and we do not propose to carry the arguments any further here. However, for the sake of completeness in the context of the present section we may simply note that such interactions are important in determining the mode and pattern of growth of fungi in decaying wood and are hence likely to affect both their individual ecological roles and spatial distribution. Fungi are of course not the only organisms present in decaying wood; bacteria and invertebrates including protozoa, termites, mites, collembola, beetles and nematodes also occur to a greater or lesser extent. In some cases these other organisms may play a very important, or even principal role in wood decomposition, for example where termites are active or in waterlogged material where bacterial attack predominates. Even where they do not predominate, the way these organisms interact with fungi may be of considerable importance in shaping fungal communities. For example, invertebrates may be of significance by importing fungal inoculum into wood, or as a result of selective effects caused by grazing. Bacteria may also act selectively on fungi either by direct antibiosis or b y altering host tree tissues in such a way as to affect subsequent fungal colonization. Shigo (1963, 1965, 1967) has repeatedly stressed the important role probably played by bacteria in development of decay and stain in living trees. In some cases, as perhaps with Phellinus igniarius (L. ex Fr.) QuC1. invading poplar and other hardwoods, colonization by bacteria and non-hymenomycetes generally occurs before entrance of the decay fungus (Good and Nelson, 1962; Shigo, 1963) and may even be prerequisite for its establishment (Kaarik, 1974). A truly comprehensive appreciation of decay communities in wood must, as Swift (1977) has stated, take account of all the organisms (and their interactions) likely to be present, including fungi, bacteria and invertebrates. Such wide vision requires much further, interdisciplinary work, for little information is available at present on interactions at this level. In this article we concentrated on the fungal community as being the primary unit in decaying wood, whilst recognizing that it will be subject to the influence of other organisms. In temperate regions where termites are absent and fungi can be regarded as the principal decomposers of wood we feel that this is probably

414

A. D. M. RAYNER AND N. K. TODD

the most meaningful approach. Only time will tell whether we are correct. C. DISCUSSION

The pattern of growth of fungi in wood is likely to be influenced by a wide variety of factors, each of which may act selectively and hence affect the composition, structure and functioning of communities. As yet, very little information is available on this topic and there is much scope for further investigation. Particularly valuable would be manipulative experimental studies investigating the effects of varying such parameters as moisture content, nitrogen content, presence o r absence of bark and host resistance o n community composition a n d structure. Few such investigations have as yet been attempted, but where they have, interesting results have been obtained. As a n example we may take the work investigating effects of chemical treatment o n the mycoflora of hardwood tree stumps pioneered by Rishbeth (1971, 1976) and followed up by Rayner (1975, 1977a, 1977b). These studies showed that treatment with 3.5 M ammonium sulphamate solution (AMS) which kills stumps and prevents regrowth, markedly influenced fungal community patterns as compared with controls left untreated o r to which distilled water TABLE XIV Fungi presenta in hardwood stumps 2-2+ years after felling

% of total number of stumps colonized

7 Birch Oak

r-----

Beech

r A -

7+--77--7

Water- AMS- Water- AMS- Watertreated treated treated treated treated treated AMS-

Fungus Bjerkandera adusfa Chondrostereum purpureum Coriolus versicolor Coryne sarcoides Hypholoma fasciculare Hypoxylon multiforme Hypoxylon serpens Oudemansiella radicata Phallus impudicus Phanerochaete velutina Phlebia mevismoides Rhinocladiella sp . Stereum hirsutum Xylaria hypoxylon Total no. of stumps

41.8 3.7 11.4 40.0 14.1 0 57.1 12.9

1.8***

5.6 1.3* 15.6*** 7.0 0 7.8*** 3.9 10.0 0 12.9 1.3*** 54.3 6.5*** 18.6 O*** 3.6 0 37.1 72.7***

7.0 6.8 48.9 44.7 20.9 42.6 0 0 2.1 25.5 89.4 36.2 29.5 21.3

4.7 51.2*** 21.3** 23.4* 18.6 48.9 0 0 0 38.3 O*** 4.3*** 59.1** 53.2**

0 0 8.7 5.1

26.1 0 10.2 0 2.2 8.7 84.8 72.9 39.5 1.7

0 0 0 3.4 77.8*** 0 11.9 0 0 35.6*** O*** 10*2*** 63.4 1.7

observed~70 77 47 47 46 45 - -~ ***, **, * Differences between AMS- and water-treatment significant at P
~~~

~~

and 0.05 respectively. a A selection only of the fungi observed is listed.

FUNGI IN DECAYING WOOD

415

had been added. Another chemical, 0.06 M 2,4, 5-trichlorophenoxyaceticacid (2,4, 5-T), used to prevent regrowth had less effect. The changes induced by AMS-treatment were striking both with regard to the assemblages of species present (as shown in Table XIV) and their pattern of colonization. Colonization by Phlebiu merismoides was stimulated by AMS-treatment in all tree species tested whilst, for example, Chondrostereum purpureum predominated on birch controls, and Xyluriu hypoxylon on controls of beech and birch after 2-2t years. For all three tree species decay was more extensive in AMStreated stumps, the roots of which in beech were often colonized by cordforming fungi such as Hypholomu fasciculure, Phanerochuete velutina and Phallus impudicus (Fig. 27). The reasons for these effects of AMS-treatment on stump colonization are probably various, but almost certainly the most important factors were removal of host resistance (allowing, for example, in beech colonization by saprophytic cord-forming fungi which subsequently spread rapidly in the subcortical zone) and addition of nitrogen. These results obtained in stumps illustrate the benefits of manipulative experiments on fungal colonization of wood in the field. Since it was difficult to discriminate between effects of addition of nitrogen and removal of host resistance, they also highlight one of the major difficulties of such experiments, i.e. the difficulties of changing one variable without affecting others. This difficulty must of course be borne clearly in mind when designing future experiments.

VI. SCHEMA FOR FUNGAL COMMUNITY DEVELOPMENT IN DECAYING SAPWOOD O F HARDWOODS AFTER FELLING Generalized schemas are of little value to biologists if based on insufficient information and taken too literally : indeed under such circumstances they can be counter-productive leading to distortion and oversimplification of ideas. On the other hand if viewed critically they can provide a stimulus for further thought, discussion and experimentation. It is to act as such a stimulus, rather than as an authoritative statement, that we propose the following very tentative and somewhat speculative schema, based on our own observations in Britain, for fungal community development in sapwood of cut hardwood stumps, logs and branches, from the stage at which, following felling, it first becomes susceptible to fungal colonization. We assume that wood freshly exposed at the time of felling is free from fungal colonization. In a sense this is of course an artificial situation resulting from man’s activities in felling: in purely natural circumstances it is likely that some decay may be present before branch or trunk fall. In the pioneer phase, the virgin wood becomes colonized by a wide and up to a point, increasing, variety of fungi including both decay- and non-decaycausing species. The exact composition of the community will partly depend

416

A. D. M. RAYNER AND N. K. TODD

on the degree of exposure of the wood to fungal inoculum: aerially exposed parts will be colonized by spores, whilst those in ground or litter contact may be more subject to colonization by vegetative mycelium. Several species totally or relatively inactive in decay and which are subject to rather rapid replacement may be at their most abundant. These would include various microfungi such as Phialophora, Botrytis, Graphium and Acremonium spp., but also some relatively rapidly growing basidiomycetes which may grow best when some living tissue is still present. Chondrostereumpurpureum is one such basidiomycete, others may include Schizophyllum commune Fr., Crepidotus variabilis (Pers. ex Fr.) Kummer and Flammulina velutipes (Curt. ex Fr.) Karst. These basidiomycetes would be expected to have the general characteristics of rapid growth rate, relative ease of replacement by other fungi, rapid fruiting (S. commune and F. velutipes are recognized for their capacity to form sporophores rapidly in culture) and relatively slight effect on the wood substance. Fungi capable of more intense decay probably also colonize early, but may be of very localized distribution initially. These would include such species as Coriolus versicolor, Bjerkandera adusta, Stereum hirsutum and Hypholoma fasciculare. Colonies of these fungi will gradually enlarge and replace nondecay species until eventually they come into contact with one another. If deadlock interactions then predominate this will lead to esfablishedphase I , where relatively small numbers of decay fungi occupying decay columns delimited by zone lines occur. Very often mutually antagonistic mycelia of the same species are present at this stage, separated by narrow, dark, relatively undecayed interaction zones which may themselves be occupied by microfungi such as Rhinocladiella, Catenularia and Cladosporium spp. This phase may persist for variable amounts of time, but eventually the pattern is likely to be disrupted by “aggressive” fungi colonizing and replacing others. These replacing fungi would include such species as Phlebia merismoides, Hypholoma fasciculare, Phanerochaete velutina and Phallus impudicus -several of which may have colonized via the soil or litter. Further deadlock interactions may then occur, leading to an established phase 2 where only a few highly competitive and actively decomposing fungi are present. As the wood becomes more intensely decayed, the vigour of the mycelium of the true decay species may decline, and accompanying this may be afinalreplacementphase when microfungi such as Trichoderma and Scytalidium spp. replace the decay species. In some cases such replacement may occur actively and relatively early on. In others it occurs as the wood becomes increasingly waterlogged, subject to invasion by animals and eventually so disrupted that it effectively becomes incorporated into the soil. At this stage typical soil microfungi may be expected to become dominant. We must emphasize again that this is only a tentative, and already oversimplified schema.

FUNGI IN DECAYING WOOD

41 7

VII. CONCLUDING COMMENT In this article we have given a very personal view of the processes involved in the colonization and decomposition of wood by fungi. We have concentrated heavily on our own work with hardwoods, and attempted to convey our belief that here is a field rich in scope for future biological research. We have attempted to demonstrate the old biological principle of initial field observation followed by increasingly sophisticated experimentation in the laboratory.

ACKNOWLEDGEMENTS We wish to take the opportunity to thank all those who have helped us by their interest, encouragement and/or valuable discussion. These include (alphabetically) Dr S . L. Anagnostakis, Dr C . M. Brasier, Prof. J. H. Burnett, Dr L. A. Casselton, D r C. Caten, Dr H. J. Hudson, Prof. D. Lewis, Mr R. W. Rayner, D r J. Rishbeth and Prof. J. Webster. Further we would like to thank those who have contributed by their technical assistance including Mr Q. Cumbes, Mr R. A. Davey, Mr €3. V. D. Goddard, M r J. Havell, Miss D. Howe and Mr M. C. Seekings. Mr M. Alexander and M r J. Saunders assisted considerably with the photography. Finally we must thank Miss Caroline Lothian for typing so conscientiously from the manuscript.

REFERENCES Adams D. H. (1974). NW. Sci. 48,21-28. Adams, D. H. and Roth, L. F. (1967). Can. J. Bot. 45, 1583-1589. Adams, D. H. and Roth, L. F. (1969). Forest Sci. 15, 327-331. Anagnostakis, S. L. (1977). Exptl Mycol. 1, 306316. Barrett, D. K. and Uscuplic, M. (1971). New Phytol. 70, 581-598. Boidin, J. (1971). In “Evolution in the Higher Basidiomycetes” (R. H. Petersen, Ed.), pp. 129-148. University of Tennessee Press, Knoxville. Brefeld, 0. (1877). Unters. ii Pilze, Leipzig 3, 16-17. Buller, A. H. R. (1931). “Researches in Fungi” Vol. IV. Longmans Green, London. Burnett, J. H. (1965). In “Incompatibility in Fungi” (K. Esser and J. R. Raper, Eds), pp. 98-1 13. Springer, Berlin. Burnett, J. H. (1975). “Mycogenetics”. John Wiley and Sons, London. Burnett, J. H. (1976). “Fundamentals of Mycology” 2nd edn Edward Arnold, London. Burnett, J. H. and Evans, E. J. (1966). Nature, Lond. 210,1368-1369. Burnett, J. H. and Partington, M. (1957). Proc. R. phys. SOC.Edinb. 26, 61-68. Butcher, A. C. (1968). Heredity, Lond. 23, 443-452.

418

A. D. M. RAYNER AND N. K. TODD

Butler, G. M. (1966). In “The Fungi: an Advanced Treatise” (G. C. Ainsworth and A. S. Sussman, Eds) Vol. 11, pp. 83-112. Academic Press, New York and London. Campbell, A. H. (1933). Ann. appl. Biol. 20, 123-145. Campbell, A. H. (1934). Ann. appl. Biol. 21, 1-22. Campbell, A. H. (1938). Trans. Br. mycol. SOC.22, 151-159. Cartwright, K. St. G. and Findlay, W. P. K. (1958). “Decay of Timber and its Prevention” 2nd edn Her Majesty’s Stationery Office, London. Carruthers, S. M. and Rayner, A. D. M. (1979). Trans. Br. mycof. SOC.(in press). Caten, C. E. (1972). J. gen. Microbiof. 72, 221-229. Caten, C. E. and Jinks, J. L. (1966). Trans. Br. mycol. SOC.49, 81-93. Childs, T. W. (1937). Phytopathology 27, 29. Childs, T. W. (1 963). Phytopathology 53, 1 124-1 127. Cowling, E. B. (1970). Acta Univ. Upsal. Dissert. Sci. 164. Croft, J. H. and Jinks, J. L. (1977). In “Genetics and Physiology of Aspergiffus” (J. E. Smith and J. A. Pateman, Eds), pp. 339-360. Academic Press, London and New York. Day, S. C. (1969). Ph.D. thesis, University of Cambridge. De Bary, A. (1887). “Comparative Morphology and Biology of the Fungi, Mycetozoa and Bacteria”. Oxford University Press (Clarendon), London and New York. Esser, K. (1965). In “Incompatibility in Fungi” (K. Esser and J. R. Raper, Eds), pp. 6-13. Springer, Berlin. Esser, K. and Blaich, R. (1973). Adv. Genet. 17, 107-152. Etheridge, D. E. (1957). Can. J. Bot. 35, 595-603. Etheridge, D. E. (1971). In “Biological Control of Forest Diseases” (V. J. Nordin, Ed.), pp. 37-53. Canadian Forestry Service, Ottawa. Falck, R. and Falck, 0. (1924). Mykol. Unters. Ber. 1. Findlay, W. P. K. (1940). Ann. Bot. 4, 701-712. Garrett, S. D. (1963). “Soil Fungi and Soil Fertility.” Pergamon Press, Oxford. Garrett, S. D. (1970). “Pathogenic Root-infecting Fungi.” Cambridge University Press. Gibbs, J. N. and Smith, M. E. (1978). Ann. appf. Biol. 89, 125-128. Glaser, T. E., Tarocinski, E. and Bauza, J. (1959). Proc. Inst. Tech. Drewna 6, 45-58. Good, M. H. and Nelson, J. I. (1962). Can. J. Bot. 40, 615-624. Grainjer, J. (1962). Trans. Br. mycol. SOC.45, 147-155. Griffith, N. T. and Barnett, H. L. (1967). Mycofogia 59, 149-154. Grindle, M. (1963a). Heredity, Lond. 18, 191-204. Grindle, M. (1963b). Heredity, Lond. 18, 397-405. Harder, R. (1911). Naturw. Z. Forst-u. Landw. 9, 129-160. Harper, J. E. and Webster, J. (1964). Trans. Br. mycol. SOC.47, 511-530. Henningson, B. (1967). Stud. Forest. Suec. 53, 1-31. Hintikka, V. (1969). Karstenia 10, 177-183. Hintikka, V. and Korhonen, K. (1970). Metsatiet. Tutkimuslait. Julk. 69, 5-28. Hubert, E. E. (1924). J. agric. Res. 29, 523-567. Hudson, H. J. (1968). New Phytol. 67, 837-874. Hudson, H. J. (1972). “Fungal Saprophytisrn.” Studies in Biology. no. 32, Edward Arnold, London. Hulme, M. A. and Shields, J. K. (1970). Nature, Lond. 227, 300-301. Hulrne, M. A. and Shields, J. K. (1975). In “Biological Transformation of Wood by

FUNGI IN DECAYING WOOD

419

Micro-organisms” (W. Liese, Ed.), pp. 52-63. Springer-Verlag, Berlin, Heidelberg, New York. Ikediugwu, F. E. 0. and Webster, J. (1970a). Trans. Br. mycol. SOC.54, 181-204. Ikediugwu, F. E. 0. and Webster, J. (1970b). Trans. Br. mycol, SOC.54, 205-210. Ikediugwu, F. E. O., Dennis, C. and Webster, J. (1970). Trans. Br. mycol. SOC.54, 308-309. Kaarik, A. A. (1971). Mitt. dt. Ces. Holzforsch. 57, 23-24. Kaarik, A. A. (1974). In “Biology of Plant Litter Decomposition” (C. H. Dickinson and G. J. F. Pugh, Eds), pp. 129-174. Academic Press, London and New York. Keyes, C. R. ( I 968). Phytopathology. 58, 1055. Kohler, E. (1929). Planta. 8, 140-153. Kohler, E. (1930). Planta. 10, 495-522. Kuhner, R. (1977). Trans. Br. mycol. SOC.68, 1-16. Leger, J. C. (1968). Bull. SOC.mycol. Fr. 84, 333-342. Levi, M. P. and Cowling, E. B. (1969). Phytopathology 58, 626634. Levi, M. P., Merrill, W. and Cowling, E. B. (1968). Phytopathology 58, 626-634. Lopez Real, J . M. (1975). Trans. Er. mycol. SOC.64,465471. Lopez-Real, J. M. and Swift, M. J. (1975). Trans. Br. mycol. SOC.64, 473481. Lopez-Real, J. M. and Swift, M. J. (1977). Trans. Br. mycol. Soc. 68, 321-325. Mangenot, M. F. (1952). Rev. gen. Bot. 51, 381-399, 439472, 477-519, 544-545. Meredith, D. S. (1959). Ann. Bot. 23, 445476. Meredith, D. S. (1960). Ann. Bot. 24, 63-78. Merrill, W. and Cowling, E. B. (1966). Can. J . Bot. 44, 1555-1580. Morrison, D. J. (1976). Trans. Br. mycol. SOC.66, 393-399. Mounce, I . (1929). Bull. Dept. Agric. Can. 1 1 1. Oduro, K. A., Munnecke, D. E., Sims, J. J. and Keen, N. T. (1976). Trans. Br. mycol. Sor. 66, 195-199. Oppermann, A. ( 1 95 I ). Arch. Mikrobiol. 16, 364-409. Park, D. (1976). In “The Role of Terrestrial and Aquatic Organisms in Decomposition Processes” (J. M. Anderson and A. Macfadyen, Eds), pp. 41-59. Blackwell Scientific Publications, Oxford, London, Edinburgh, Melbourne. Pomerleau, R. and Etheridge, D. E. (1961). Mycologia 53, 155-170. Raper, J. R. (1966). “Genetics of Sexuality in Higher Fungi” Ronald Press, New York. Rayner, A. D. M. (1975). Ph.D. thesis, University of Cambridge. Rayner, A. D. M. (1976). Trans. Br. mycol. SOC.67, 546-549. Rayner, A. D. M. (1977a). Trans. Br. mycol. SOC.69,291-302. Rayner, A. D. M. (1977b). Trans. Br. mycol. SOC.69, 303-312. Rayner, A. D. M. (1978). Ann. appl. Biol. 89, 131-134. Rayner, A. D. M. and Todd, N. K. (1977). J. gen. Microbiol. 103, 85-90. Rayner, A. D. M. and Todd, N. K. (1978). Trans. Br. mycol. SOC.71 ( I ), 99- 106. Rayner, R. W. (1970). “A Mycological Colour Chart” Commonwealth Mycological Institute and British Mycological Society, Kew, Surrey. Redfern, D. B. (1968). Ann. Bot. 32, 293-300. Ricard, J. L. and Bollen, W. B. (1968). Can. J. Bot. 46, 643-647. Rishbeth, J. (1963). Ann. appl. Biol. 52, 63-77. Rishbeth, J. (1971). In “Biological Control of Forest Diseases” (V. J. Nordin, Ed.) pp. 29-35. Canadian Forestry Service, Ottawa. Rishbeth, J. (1972). Eur. J. For. Path. 2, 193-205. Rishbeth, J. (1976). Ann. appl. Biol. 82, 57-70. RypiEek, V. (1 966). “Biologie holzzerstorender Pilze” G. Fischer, Jena.

420

A. D. M. RAYNER AND N. K. TODD

Shigo, A. L. (1963). PI. Dis. Reptr 47, 820-823. Shigo, A. L. (1965). Phytopathology 55, 957-962. Shigo, A. L. (1967). Intern. Rev. For. Res. 2, 237-299. Stillwell, M. A., Wood, F. A. and Struntz, G. M. (1969). Can. J. Microbiol. 15, 501-507. Swift, M. J. (1977). Sci. Prog., Oxf. 64, 175-199. Theden, G. (1961). Angew. Bot. 35, 131-145. Todd, N. K. and Rayner, A. D. M. (1978). Genet. Res. 32,55-65. Townsend, B. B. (1954). Trans. Br. mycol. SOC.37, 222-233. Williams, G. C. (ed.) (1971). “Group Selection” Aldine, Atherton, Chicago. Woodgate-Jones, P. (1977). M A . thesis. University of Exeter.

Aspects of the Physiology of Orchids

JOSEPH ARDITTI Department of Developmental and Cell Biology. University of California. Irvine. California 92717. U.S.A.

I . Introduction I1. Seeds . . A . History

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . B. External Morphology . . . . C . Structure and Ultrastructure . . D. Longevity . . . . . . . . E. Asymbiotic Germination . . . F. Symbiotic Germination . . . . G. Summation . . . . . . .

111. Phytoalexins

A. B. C. D.

. . . . . . . .

. . . . . . . .

423 423 438 438 441 441 489 506

. . . . . . . . . . . . . . . . . . . . . . . .

508 508 510 512 517

History . . . . . . . . . Chemistry Production and Distribution Action Spectrum and Activity . . . Biological Role . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

422

. . . . . . .

. . . . . . . . . . . . . .

IV. Carbon Fixation . . . . . . . . . . . . . . . 519 A . History . . . . . . . . . . . . . . . . 519 B. Stomata1 Rhythms . . . . . . . . . . . . . 521 C . Crassulacean Acid Metabolism . . . . . . . . . . 521 D. C, Photosynthesis . . . . . . . . . . . . . 529 E. C, Photosynthesis . . . . . . . . . . . . . 529 F. Carbon Fixation by Different Plant Organs . . . . . . 530 G . Photorespiration . . . . . . . . . . . . . . 532 H . Summation . . . . . . . . . . . . . . . 532

422

J. ARDITTI

. . . . . . . . . . . . . . . . . . 533 History . . . . . . . . . . . . . . . . 534 Introduction . . . . . . . . . . . . . . . 534 Pollination . . . . . . . . . . . . . . . 552 Post-pollination Phenomena . . . . . . . . . . . 566 Induction of Phenomena . . . . . . . . . . . 617

V. Flowers

A. B. C. D.

E.

VI. Tissue Culture VII.

. . . . . . . . . . . . . . . . 634

Epilogue . . . . . . . . . . . . . . . . . . 637 Acknowledgements . . . . . . . . . . . . . . . 637 References . . . . . . . . . . . . . . . . . 638

I. INTRODUCTION

Orchidaceae is by most estimates the largest of all flowering plant families (although some claim this “honour” for the Compositae) with 600-800 genera and 25 000-35 000 species (Garay, 1960; Schultes and Pease, 1963). Orchids vary in size and weight from a few millimetres and grams [the Australian Bulbophjdlum minutissimum is 1-1.5 mm across (Nicholls, 1969) and probably weighs no more than a gram or two] to several metres and tons [ Grammatophyllum speciosum (Sumatra to the Philippines) stems can reach three or more metres (Grant, 1895; Holttum, 1957) and mature plants can be huge, weighing several tons]. They can be found in dense tropical jungles; open tropical grasslands; hot and dry deserts; cold and damp areas; on trees which hang over the water, or on rocks subject to constant sea spray; and all other terrestrial habitats where flowering plants can grow. Orchids can be epiphytic, lithophytic, or terrestrial and even subterranean [like Rhizanthella gardneri, a monotypic genus from Western Australia (Nicholls, 1969) 3, but none are aquatic. The pollination mechanisms of orchids range from cleistogamy to pseudocopulation. Their flowers may be bizzare, insignificant or beautiful and emit revolting stenches, have no discernible smell, or be the source of very pleasing fragrances. After pollination, orchid flowers undergo changes which are no less elaborate and fascinating as the “contrivances” (to quote Darwin, 1862, 1904) which ensure pollination. Fertilization does not immediately follow pollination in orchids and may occur days, weeks, or even months after the flowers have been pollinated. The fruits develop slowly and may take a year or more to ripen; they can contain from around 1000 up to 4 000 000 seeds (for reviews see Arditti, 1967a; Darwin, 1904). Orchid seeds are tiny, most of them being 0.25-1.2 mm long, 0.09-0.75 mm wide and weighing 0.3-14 pg (Harvais, 1974; for reviews see Arditti, 1967;

ASPECTS OF ORCHID PHYSIOLOGY

423

Burgeff, 1936). They have complex germination requirements which in nature are provided by mycorrhizal fungi. Once infected by the appropriate fungus, orchid seeds can germinate but may also be parasitized and destroyed. It is not surprising, therefore, that phytoalexin production evolved in orchids (for reviews see Arditti, 1966b, 1975; Arditti et al., 1975; Fisch et af., 1972, 1973a). The great diversity of orchids and their different “lifestyles” have led to or been made possible by structural and physiological adaptations. Many of these are not very well known, or are poorly understood, because orchids have generally been neglected as subjects for scientific investigation (Dressler and Williams, 1970). Even when information is available it may often be scattered and/or published in journals of interest to those few scientists who work with orchids. Consequently, few general reviews have been written on orchid physiology and the result is a vicious circle: since little is known or has been put into perspective, not much is being done. The aim of this review is to provide at least a partial remedy. Like most reviews this one is subject to the limitations, preferences, idiosyncracies and style of the reviewer. The topics I have covered are one reflection of these. Another is the citations: to provide the maximum number of relevant citations in the minimum amount of space, I will: (i) cite previous reviews whenever they exist, even if this may lead to repetitions; (ii) hold citations of literature mentioned in previous reviews to a minimum; (iii) list in greater detail papers published since the latest reviews; and (iv) exercise the usual prerogatives of a reviewer in selecting which papers to cite (and consequently omit some recent articles or cite older ones). A third is the nomenclature in that I have chosen in most cases to use the taxonomic designations and spellings given in the original literature. I have not attempted to select “correct” names or “proper” spelling since orchid taxonomy is unusual, fluid and often subject to controversies. 11. SEEDS

Orchid seeds are unique in several respects. They are exceedingly small, but produced in large numbers. The embryos have no endosperm (Savina, 1974), no cotyledons, and no root initials. To germinate in nature they require fungal infection. It is not surprising, therefore, that orchid seeds, their nature and germination requirements were slow to be discovered and understood. A . HISTORY

For many years European botanists apparently failed to note and/or appreciate the dust-like orchid seeds and realize that they were capable of germination (Arditti, 1967a). Fanciful ideas on orchid reproduction resulted from the presence of caproic acid in Sutyriurn flowers (Anonymous, 1864;

424

J. ARDITTI

Chautard, 1864) which “. . . emit a strong and unpleasant odor of goats.” (Summerhayes, 1968). Anastasius Kircher (1601-1680) suggested that these orchids originated from “. . . escaping spermatic fluid (of animals, i.e., goats) that fell on the soil (and) underwent fermentation together with moisture of the earth . . .” (Ames, 1942; for reviews see Arditti, 1972; Shechter and Arditti, 1973). Another idea advanced by Kircher was that orchids can originate from “. . . rotting corpses of animals which still contain some seminal virtue . .” (de Wit, 1977). Tragus (Bock), another European botanist (c. 1600), suggested that orchids originate from “. . . . the ‘seed’ of thrushes and blackbirds, who copulate in spring meadows and pastures” (de Wit, 1977). A later (1855-1890), more plausible, even if erroneous, belief was that orchid seeds were not viable and incapable of germination (His, 1807). Orchid reproduction was assumed to be through bud or gemma-like structures which underwent a series of metamorphoses leading to the formation of mature plants (for a review see Arditti, 1967a). Rumphius (c. 1627-1702), the blind seer of Ambon (de Wit, 1959), may have been the first botanist to recognize orchid seeds: “The ripe [fruit] . . . opens up readily . . . and then the yellow flour is largely shed and is blown away on the wind; but whether this is endowed with seed-virtue, and settles to grow on other trees, is still unknown,” (de Wit, 1977). The answer became known 150 years later when a British botanist first described germinating seeds of Bletilfa, Orchis morio and Limodorum verecundum (Salisbury, 1804). This was followed by the description of germinating seeds and seedlings of many other orchids (for reviews see Arditti, 1967a; Beer, 1863; Burgeff, 1909, 191 1, 1932, 1936; Pfitzer, 1882; Ramsbottom, 1922; Veitch, 1887-1894). However, even following these descriptions the germination requirements of orchid seeds remained unknown until 1899. In that year a French botanist, Noel Bernard, published an account of his observation that seeds of Neottiu nidus-uvis require fungal infection for germination (Bernard, 1899, 1909; for reviews see Arditti, 1967a; Burgeff, 1909, 1911, 1932, 1936; Warcup, 1975). The genius of Bernard lies not in observing the fungus (that had been done before without appreciation of its importance), but in recognizing its role as being a mutualistic symbiont rather than a parasite. He succeeded in demonstrating the role and importance of mycorrhiza before dying at an early age in 191 1 . His discovery was put to practical use in a method for commercial germination of orchid seeds. Lest the world forget, his mentor, J. Costantin, and friend, J. Magrou (1922) described it in a paper with the pompous title “Applications industrielles d’une grande dtcouverte francaise”. For years after that J. Costantin inveighed against and maligned everyone who in his (narrow and chauvinistic) mind threatened the “. . . grande dkcouverte . . .” even if that man happened to be Prof. Lewis Knudson (for reviews see Arditti, 1972, b, d ; Shechter and Arditti, 1972).

.

ASPECTS OF ORCHID PHYSIOLOGY

425

Hans Burgeff, working in Wurzburg, made major contributions to the understanding of orchid mycorrhiza, starting in 1909 (Burgeff, 1909, 1911 , 1932, 1936, 1959). However, he made two errors of commission and one of omission. The former are his assertions that: (i) all mycorrhizal fungi of orchids belong to a separate group he called Orcheomyces; (ii) there is high specificity between orchids and fungi. His omission was the failure to discover asymbiotic seed germination. Consequently he was not very gracious when Lewis Knudson at Cornell University discovered asymbiotic seed germination and, together with Prof. John T. Curtis (working independently at Wisconsin University), showed there was no specificity (personal communications from several students, friends and collaborators of Knudson during that era). Knudson was interested in the effects of carbohydrates on green plants (for reviews see Arditti, 1967a, 1972a, b, d ; Shechter and Arditti, 1972) and this probably led him to study orchid seeds. He analysed salep (a preparation from Ophrys tubers used by Bernard to germinate orchid seeds) and using Bernard’s report that the fungus could invert sugar concluded that the fungus stimulated germination by breaking down starch and other large molecules. Next, he added sugar to Pfeffer’s solution and a modification of it and germinated Cuttleya, Luelia and Epidendrum on both of them. Since he was spending 1919-1921 in Spain, his first publication on the subject was in Spanish in a relatively obscure publication (Knudson, 1921); his next paper was in English and a major journal (Knudson, 1922). Costantin perceived Knudson’s findings as being a threat to Bernard’s discovery (“Tout cela semblerait indiquer que toute la thkorie de la symbiose imaginee per Noel Bernard est un pur roman”) and attacked, claiming that Knudson’s seedlings were not normal since they contained starch. He also asserted that without mycorrhiza they will not flower and wrote grandly about the teachings of nature. Knudson replied by pointing to the weakness in Bernard’s work and presenting his data which disproved Costantin’s claims. With this the debate ended. J. Ramsbottom, the British mycologist, was also sceptical “. . . if these methods should prove capable of general application . . .”, unimpressed “. . . no really new facts have been added to our knowledge . . .”, diplomatic “. . . such work . . . is extremely valuable in that it approaches the subject from a physiological standpoint. . .”, and erudite as well as humorous “. . . an orchid seedling without its fungus is like Hamlet without the Prince of Denmark.” (Ramsbottom, 1922). In subsequent years Knudson improved his initial medium. Others developed additional solutions for use with species which do not germinate well on Knudson’s media. In addition the discovery spurred a certain amount of basic research.

Fig. 1 . Orchid seeds, x 5 3 5 except Nos 1-8. (Plate I1 from Beer, 1863; seeds are listed in alphabetical order in joint caption for Figs 1, 2, 3.)

Fig. 2 . Orchid seeds, x 52. (Plate I11 from Beer, 1863; seeds are listed in alphabetical order in joint caption for Figs 1, 2, 3.)

Fig. 3. Orchid seeds, x 52.5. (Plate IV from Beer, 1863; seeds are listed in alphabetical order in joint caption for Figs 1, 2, 3.)

429

ASPECTS OF ORCHID PHYSIOLOGY

Orchid Seeds in Figs. I , 2, 3 (Beer, 1863 unless indicated otherMfise) Acan fhophippium bicolor, Lindl. . . . Acroclaene punciaia, BI. . Acropera citrina, Hort. . Acropera luieola, Hort. . . . Acropera loddigesii, Hort. . Acropera maculaia, Hort. . Aerides odoraium, Lour. . Aerides sp. Mont. Kashia . . . Agrostaphyllum sp. . . . . . Anaectochylus seiaceus, Blume . . Angraecum bilobum, Lindl. . Apaturia senilis, Lindl. . . . . Barkeria melanocaulon, A. Rich. Gal. . BIeiia sheperdii, Hook , . . Brassavola cordafa, Lindl. . . . Brassia cowanii, Hort. (caudaia R. Br.) Calanihe veratrifolia, Rr. Br. (Clifford and Sniith, 1969). Cattleya ameihysiina, Morr. . . . Caitleya bicolor, Lindl. . . Caiileya erispa v. purpurea, Lindl.' . . Caftleyaforbesii, Lindl. . . . . Cafileya harrissonii, Bat. . . . Caifleya lobaia (Hoehne, 1949) . , . Caiileya loddigesii, Lindl. . . . . Caiileya tigrina, A. Rich. . . . Cattleya irianaei (Hoehne, 1949) . Ceraihandra chloroleuca, Eckl. . Cirrhaea viridi purpurea, Lindl. , . Coelia alba, Lindl.? Hort. . . . . Corycium crispum, Sw. . . . . Corycium orobanchoides, Sw. . Corallorrhiza innafa, R. Br. . . . . Cymbidium odoniorrhizon, Willd. . , Cymbidium sinense, Lindl. . . Cyriosia lindleyana, BI. . . . . Cypripedium barbaium, Lindl. . Dendrobium cretaceum, Lindl. . . . Dendrobium plicaiile, LindI. . . Dicrypia bauerii, Lindl. . . Dicrypia glaucescens, Hort. . . . . Dicrypia glaucescens var. Hort. . Disa cernua, Sw. . . . . . Disa cornuia, Sw. . Disa pulchella, Sw. . Disa tenella, Sw. . . . Disperis villosa, Sw. . Epidendrum ciliare, L. . . Epidendrum cinnabarinum, S. . Epidendrum cochlearum, L. Epidendrum crassifohm, L ind . Epidendrum lancifolium, Hort. . Epidendrum papillosum, Batem. . Epidendrum ramosum, Jacq. Epidendrum siamfortianum var. carnea, B. . Epidendrum siamforiianum, Batem . Epipaciis lafifolia, Swartz. . . Episiephium parviflorum, B1. . Eulophia strepiopetala, Hort. . Gamoplexis orobanchoides, BI. Glossodia minor (Clifford and Smiih, 1969) Gongora bufonia, Lindl. . . .

.

.

.

.

.

. . . . . . . . .

Fig. 3 3 2 2 2 2 2 3 3

Sebed No.

29 52 53 30 24 31 1s 60 32 18 ~-

. .

. . . . . . . .

3 2 I

15 55 19 6 54 25

L

2 -

2 2 3 3 3 1 3 2

18 26 17 40 31 6

39 58

.

1

7

. . . . .

3 3 2 2 2

13 52

.

1

. . . . . . . . . . . . .

2 3 3 1 3 3 3 2 2 3 3 3 2 2

. .

3 3

. .

1 1

. . * .

2 2 3 1

.

2

1

38 50 30 27 30 16 46 46 12 56 59

19 14 23 42 56 54 21 21 37 8 46 37 2 45

44 33 28 20 7 1 29

430

J. ARDITTI

Gongora maculata v. pallida, Lindl. . Goodyera discolor, h o d . . Goodyeraprocera, Hook. . . Goodyera repens, R. Br. . , Goodyera semipellucida, Lindl. . Govenia lilacina, Lindl. , Gymnadenia conopsea, Rich. . Gymnadenia longifolia, Lindl. . Habenaria dilatata, Hook. . Habenaria hispidula, Lindl. . Habenaria tridentata, Hook. , Haematorchis altissima, BI. . Himantoglossum hircinum, Spr. . Huntleya violacea, Lindl. . . Isochilus linearis, R. Br. . . Luelia anceps, Lindl. . . Laelia galeottiana, A. Rich. Laelia perinii v. major, Lindl. . Leptotes bicolor, Lindl. . . . Listera ovata, R. Br. . . . Lubia teretifolia, Gaud. . . . Lycaste harrissonii, Hort. . . Malaxia lilifolia, Swartz , Maxillaria crocea, Lindl. . Miltonia morelliana, Hort. . Mormodes buccinator, Lindl. . Mormodes viridiflora,Hort. . Mormodes pardina, unicolor Hook Nigritella angustifolia, Rich. . Neottia aestivalis, Dec. . . Neottia nidus avis, Rich. . Neottia orchioides, Sw. . . . Neottiapubescens, Willd. . . . Neoftia (Stenorrhynchus)specioss, Jacq. Neottia vitalis, Lindl. . . Octomeria lancifolia, Hort. . Odontoglossum bictoniense,Lindl. Odontoglossum pescatorei (Hoehne, 1949) Odontoglossum pulvinatum, Lindl. Odontoglossum sphacelatum, Lindl. . Ophrysfunerea, Vivian . . Ornithocephalus sp. Java , Orobanche! . . Orchis acuminata, Desi. . . . Orchis brevicornu, Vivian . Orchis coriophora, Lin. . Orchis fragrans, Pollin. . Orchis intacta, Lin. . . . Orchis latifolia, Lin. . . Orchis longicornu, Poir . Orchis maculata, Lin. . . Orchis maculata, var. Lin. . Orchis secundiflora, Pertol. . Orchis speculum, Tenor. . . Otochilus fusca, Lindl. . Otochilusporecta, Lindl. Paphiopedilum curtisii (Hoehne, 1949) ' Paphiopedilum specierum (Hoehne, 1949); Paphiopedilum parishii . . . Pelexia adnata, Spr. . , Pyrola rotundifolia! Phajus albus, Lindl., Thunia alba i .f i l . 1 Phajus bicolor, Lindl. . .

. . .

.

. . .

.

.

.

.

. .

.

.

.

. .

.

?edNo. 13 39 23 3 A

12 14 45 10 48 36 34 26 25 38 20 6 36 10

17 32 58

41 I2 22 18 ~~

20 23 42 51 41 51

5

45 43 11 35 8

41 .-

35 26 29 38 33

39 ..

35 38 43

40 28 32 13 36 37 11 59 3 4 5 49 34 16 43

43 I

ASPECTS OF ORCHID PHYSIOLOGY

Phajus grandifoh, Lour. . . Phajus maculatus, Lindl. . . Phajus wallichii, Lindl. . Pholidota rubra, BI. . Pleurothalis sessiliporum, Lindl. Promenaea rollissonii, Lindl. Promenaea stapelioides, Lindl. . Pterygodium Caffrum, Sw. . Pterygodium catholicum, Sw. . Pterygodium inversum, Sw. . Pterygodium voluere, Sw. . Sarcanthus rostratus, Lindl. . . Satyrium bicallosum, Thumb. . Satyrium carneum, R. Br. . Satyrium nepalense, Don. . Scaphyglotis vestita, Brong. . Scaphyglotis violacea, Brong. . Selenipedium schlimmii, Rchb. fil. Sobralia decora, Batem. . Sobralia liliastrum, Lindl. . Sobralia macrantha, Lindl. . . Stanhopea aurea, Lodd. . Stanhopea insignis, Forst. . Stanhopea oculata, Lindl. . Stanhopea tigrina, Batem. Stanhopea tigrina, var. superba, Ba’tern. Stanhopea violacea, Nort. . Stanhopea warczewirzii, Hort. . Sturmia loeselii, Reichbeh. Tetragamestus modestus, R. f i l . . Thelymitra ixioides, Sw. . Trichocentrum fuscum, Lindl. . Trichopilia albida, H. Wendl. . Triphora pendula, Nutt. . Vanda coerulea, Griff. Vanilla sp. . Vanilla planifolia, Andr. . Xylobium squalens, Lindl. . Zygopetalum intermedium, Lodd. . . Zygopetalum mackaii, Hook. .

1

1

. .

.

.

.

. .

.

.

.

.

Fig. 2

.

2

. . . . .

3 3 3 2 2

.

2

. . . . . . .

3 3 3 2 3 3 3

. .

1 1

. . . .

3 3 2 3

. .

1 2

.

2

.

1

. .

I 3

. . .

1 1 1

. . . . .

2 3 2 2 3 1 2 3 2 2

.

. . .

.

Seed No. 3 42 34 1

27 57 II 47 51 48

50 55 57 28

22 31

31 35 9 14 24 24 7 8

22 17 33 21

9 27 9 44 44 49 53 2

40 19 5 4

Fig. 4. Scanning electron microscope photograph of Pleurothallis racemiflora seed.

x 1050. (Courtesy Dr W. Barthlott.) Fig. 5 . Scanning electron microscope photograph of Vanilla planifolia seed. (Courtesy Dr W. Barthlott.)

X

210.

Fig. 6. Scanning electron microscope photograph of Cynorkis fusfigiufu seed coat. > 2100. (Courtesy Dr W. Barthlott.) Fig. 7. Scanning electron microscope photograph of Eulophiu guinensis seed coat. x 525. (Courtesy of Dr W. Barthlott.)

Fig. 8. Scanning electron microscope photograph of Gruphorkis lurida seed coat. X 1059. (Courtesy of Dr W. Barthlott.) Fig. 9. Scanning electron microscope photograph of Calypso bulbosa seed coat. x 2036. (Courtesy of Dr W. Barthlott.)

Fig. 10. Scanning electron microscope photograph of Limnodorum abortivum (L.) Swartz seed coat. x 1050. (Courtesy of Dr W. Barthlott.) Fig. 11. Scanning electron microscope photograph of a Coryanthes species seed colt from Costa Rica. (Courtesy of Dr W. Barthlott.)

Fig. 12. Scanning electron microscope photograph of Cyrtorchis arcuata seed. X 210. (Courtesy of Dr W. Barthlott.) Fig. 13. Scanning electron microscope photograph of Catasetum russellianum seed. x 210. (Courtesy of Dr W. Barthlott.)

D

Fii 14. The five basic shapes into which orchid seeds have been classifieL [Clifford and Smith, 1969).

TABLE 1 Weights of Orchid Seeds (Koch and Schulz, 1975)

Species Galeola sp. (lindleyana?) Gymnadenia conopea Sobralia macrantha Epidendrum radicans Limodorum abortivum Dendrobium antennatum Stanhopea oculata Grammatophyllum speciosum Goodyera repens Cephalantherapallens Haemaria discolor x H. rubrovenia Angraecum eburneum x A. sesquipedalea Acanthophippium sylhetense Didymoplexis pallens Anguloa ruckerio Schomburgkia undulata a

This hybrid is known as Angraecum Veitchii. 3 932 948 seedslcapsule.

Weight4,( 140 8.0

6.3 6.0 5.73 5.65 3.0 2.48 2.0 2.0 0.85

0.70 0.66 045 0.39 0.30

438

J. ARDITTI

B. EXTERNAL MORPHOLOGY

Rumphius described orchid seeds as being dust- or flour-like. This is still the best description if one looks at the seeds with the naked eye, except that it is necessary to add that their colour may be white, cream, pale green, reddish, orange or dark brown (for reviews see Arditti, 1967a; Barthlott, 1976; Beer, 1863; Burgeff, 1936). The diversity of seed shapes is considerable (Figs 1-13), but it can be reduced to five basic forms (Fig. 14; Clifford and Smith, 1969). Orchid seeds are small and may measure down to 0.75 x 0.2 mm (Calopogon pulchellus). They may vary in length from 0.3 mm (Satyrium odorum) to 5 mm (Sobralia species, Figs 1, 2). Their weight can be 0.3 pg (Table 1). Embryos are much smaller and may consist of as few as ten cells. They are situated in the middle of the testa, being attached to it by several fine strands. Testa cells are dead, vary in size and have longitudinal and transverse walls of different thickness (Barthlott, 1976; Clifford and Smith, 1969; Rauh et al., 1975; Wildhaber, 1972, 1974). This gives them a net-like appearance. The cells may be isodiametric or elongated, polygonal, tetragonal, hexagonal or have a somewhat convoluted shape (Figs 1-3). Some testa cells appear sculptured due to variable and discontinuous thickenings on the walls (Figs 6, 9, 13). Since these characteristics can occur in a variety of combinations, orchid seed morphology can be a useful systematic tool (Barthlott, 1976; Clifford and Smith, 1969; Wildhaber, 1972). An additional characteristic of orchid seeds is that many are difficult to wet. Because of this as well as their size, shape, weight and construction orchid seeds are very buoyant in both air and water. Consequently they can travel long distances in air and water currents and this, of course, aids in their dispersal (de Lehaire, 1910; Sanford, 1974). C. STRUCTURE AND ULTRASTRUCTURE

Orchid embryos consist of relatively undifferentiated, mostly isodiametric cells with dense granulated cytoplasm and conspicuous nuclei. Cells of the posterior end are large and sometimes vacuolated. A suspensor, consisting of very large, vacuolated, dead cells is attached to the posterior (Alvarez, 1969; Veyret, 1969; for reviews see Arditti, 1967a; Withner, 1959). The anterior, chalaza1 or meristematic portion of the embryo is composed of smaller cells which in Cattleya auranriaca are 8-10 pm in diameter (Harrison, 1973). There have been few ultrastructural studies of orchid seeds due to the great difficulties encountered in their fixation and sectioning. One of the more recent (and few) studies was carried out in my laboratory and the following description and electron micrographs are taken from it (Harrison, 1973):

.

". . The ovoid embryo of the Cattleya aurantiaca seed has an average length of 260 pm, exclusive of the suspensor which is composed of dead cells. Average

Fig. 15. Seed of Cartleya aurantiaca seed, whole. Figs 1618. Electron micrographs of cells in ungerminated Cattleya auranfiaca seed. 16. Cell in ungerminated seed, x 2700. 17. Cell in ungerminated seed, x 3375. 18. Basal cell of an ungerminated seed, x 12 375. (Explanation of symbols: CW, cell wall; L, lipid body; M, mitochondrion; N, nucleus; P, protoplastid ; PB, protein body).

Figs 19-21. Effects of galactose on cells of orchid seedlings. 19. Nuclear chromatin appears dispersed and nuclear envelope evaginates (arrows) into cytoplasm, X 9338. 20. Nuclear envelope evaginatesinto cytoplasm, x 45 080.21. Nuclear envelope folded back on cytoplasmic side, x 41 160. Explanation of symbols: N, nucleus; NE, nuclear envelope (Ernst et af., 1971a).

ASPECTS OF ORCHID PHYSIOLOGY

44 1

width is 80 pm. An orange-colored, fusiform seed coat surrounds the embryo. It is 400pm long and 80pm wide. Total weight of a seed is approximately 1.5 pg (Figs. 15-18).” “Cells of the dry, ungerminated seed vary in size depending on their positions within the embryo. The chalaza1or meristematic end of the embryo is composed of smaller cells, generally 8-10pm in diameter. Larger cells comprise the remainder of the embryo and will be referred to here as the basal cells (Figs. 1 9-2 1 ).” “Ultrastructurally, all of the cells, regardless of their position in the embryo, are heavily packed with food reserves (Figs 15-21). Lipids constitute the largest portion of the reserve material and are present as lipid bodies in all cells (Figs 16-18; Wiesmeyer and Hofsten, 1976). Protein bodies are also found (Fig. 17) but are restricted to cells in the upper two-thirds of the embryo, including those in the meristematic region and upper half of the basal region. Throughout the embryo the cytoplasm is dense and stains heavily for proteins. Aside from an occasional small grain within a proplastid (Fig. 18), no starch or other carbohydrate reserves were found in the dry seed. Also, plasmodesmata are not discernible in the ungerminated seed.” “Presumptive organelles are found in the cytoplasm of the dry seed but generally cannot be classified as to type because of their immature stage of development and poor hydration state. Most of the visible organelles are positioned next to the cell wall or else in the cytoplasm immediately surrounding the nucleus. Distinguishable proplastids and mitochondria do occur in the cytoplasm adjacent to the nuclei of the larger basal cells (Fig. 18). Endoplasmic reticulum is not discernible.” D. LONGEVITY

Depending on species or storage conditions the longevity of orchid seeds may vary from two months to 18 years (Janke, 1915; for a review see Arditti, 1967a). Most seeds will lose their viability if stored at room temperature (21-22°C) without desiccation. However, longevity can be extended considerably if the seeds are permitted to dry and then stored over a desiccant (calcium chloride, for example) under low temperature (0-10°C). Orchid seeds can also survive quick freezing (Fehlandt, 1960), lyophilization in autoclaved coconut liquid as suspending fluid (Svihla and Osterman, 1943) and storage at -79°C for 351,365 and 465 days (Ito, 1965). The information on high temperature tolerance of orchid seed is more limited, but it is known that at least one species (Dendrobiurn nobile) germinates well under temperatures up to 40°C. Soaking seeds for up to four hours at 39°C is also without deleterious effects (for a review see Arditti, 1967a). E. ASYMBIOTIC GERMINATION

Knudson demonstrated that at least some orchids can germinate asymbiotically on suitable media (Knudson, 1921, 1922, 1946). Others followed and tried to improve his media or design new ones and to germinate additional species (for reviews see Arditti, 1967a; Withner, 1959). Knudson’s media are suitable for most epiphytic and/or tropical orchids. However, some terrestrial

442

J. ARDITTI

species, especially those from temperate regions, are more difficult to germinate. Consequently, additions, modifications or new media and procedures must be employed for terrestrial species (for review see Stoutamire, 1974), from Europe (Borriss, 1971; Fast, 1976; Hadley, 1970; Hadley and Harvais, 1968; Harbeck, 1963, 1964; Harvais and Hadley, 1967; Mead and Bulard, 1975; Veyret, 1969; Voth, 1976), Canada (Harvais, 1972, 1973, 1974), Australia (McIntyre et al., 1971, 1972a, b; Veitch and McIntyre, 1972; Wrigley, 1973, 1976), South Africa (Collett, 1971; Harbeck, 1968), and the USA (Stoutamire, 1974). In addition to attempting the germination of various species, investigators have also studied the effects of or requirements for a variety of factors in the germination of orchid seeds (for reviews see Arditti, 1967a; Fast, 1964, 1967; Stoutamire, 1974; Withner, 1959). I . Mineral Nutrition Most media used for orchid seed germination are more concentrated than tree trunk effluates which nurture epiphytic seedlings in nature (Table 2; Curtis, 1946; for reviews see Arditti, 1967a; Withner, 1959). Epiphytic species germinate well on these media, but some terrestrial orchids do not. Their germination is much better on more dilute media (Voth, 1976; personal communication from Robert Ernst regarding medium RE in Table 2). In the case of Dactylorhiza purpurella germination can take place on distilled water and a modified Pfeffer medium (Harvais, 1972). Arundina bambusifolia (a terrestrial orchid) seeds germinate better on the more dilute Raghavan and Torrey medium than on Vacin and Went (Table 2; Mitra, 1971). Seeds of Vanilla planifolia germinate best on 0.1 x the normal concentration of Knudson B medium (Lugo-Lugo, 1955);Cypripedium calceolus germinates on a medium (Fast, 1976) which is 2-5 times as dilute as the one used for C . reginae (Harvais, 1973), which in turn is less concentrated than Knudson C (Table 2). Thus, it seems that the concentration of media is an important factor. Unfortunately, the available information is limited and there are no clear patterns. However, a guarded generalization can be made: Lady slipper orchids (Cypripedium, Paphiopedilum) and temperate climate terrestrials seem to germinate better on more dilute media (Fast, 1976; Harbeck, 1963; Harvais, 1973; R. Ernst, personal communication). Variations in the proportion of ions in culture media have been reported to have little effect (Wynd, 1933). The same investigator also suggested that anions may play a more vital role in the germination of orchid seeds than cations. More recent work with Bletilla striata seeds (Ichihashi and Yamashita, 1977) has determined the optimal range of ions and led to the formulation of a culture medium (Tables 1, 2, 3). Before the advent of chelating agents, precipitation of iron presented a problem and a variety of substances were used to ensure its availability (for

ASPECTS OF ORCHID PHYSIOLOGY

443

reviews see Arditti, 1967a; Withner, 1959). At present, EDTA or commercial preparations of chelated iron are used widely and the problem has been solved (Fast, 1976; L. Koch, 1973; U. Koch, 1972; Mead and Bulard, 1975; Mitra, 1971; Miyazaki and Nagamatsu, 1965; Mukherjee et al., 1974; Thompson, 1977 are some examples; for reviews see Arditti, 1967a,Thompson, 1977). Comparative studies have revealed that reduction of phosphate levels can increase germination. The reasons for this are unclear. It is possible, of course, that orchid seeds may be sensitive to phosphate. However, it is also possible that high phosphate levels may lead to iron deficiency since an insoluble complex is formed during autoclaving. On media lacking phosphorus, Cymbidium protocorms become ". . . yellow grey-green and are covered with black spots, but do not die even after 100 days . . ." (de Bruijne and Debergh, 1974). Excellent germination of several species has been obtained on media low in calcium (Table 2; Fast, 1976; Ernst, personal communication; Raghavan and Torrey, 1964; Thompson, 1977; Vacin and Went, 1949; Wynd, 1933). This is especially true for diandrous (i.e., Lady slipper) and terrestrial species. Omission of calcium from culture media is without notable effect on development of the normal green colour (de Bruijne and Debergh, 1974). Precipitation of calcium as phosphate salt is also without deleterious effects (Storey et al., 1947). Orchid tubers and seeds contain limited amounts of calcium (Tienken, 1947; L. C . Wheeler, personal communication ; Wheeler and Ramos, 1965; for a review see Arditti, 1967a) and may, therefore, have low requirements. Soaking the seeds of Galeola septentrionalis, a terrestrial orchid lacking chlorophyll, in solutions of several potassium salts increased their germination. Concentrations of KCl at 5 x 10-l M, or higher, in culture media were supraoptimal and 5 x proved to be suboptimal. Similar treatments of Cymbidium virescens failed to improve germination (for a review regarding the effects of potassium, lithium and sodium on orchid seeds see Arditti, 1967a). Numerous small protocorms develop on potassium-free media and leaves are stunted (de Bruijne and Debergh, 1974). Microelements are not always added to orchid seed germination media because enough may be present in the agar, sugar or other salts. However, in some cases improved seedling growth has been reported following the addition of microelements such as boron, cobalt, copper, iodine, manganese and molybdenum. Therefore, incorporation of microelements in culture media is often recommended, probably on the assumption that even if not beneficial they are most probably not harmful either (Arditti, 1967b; Harrison and Arditti, 1970; Harvais, 1972, 1973, 1974; Ichihashi and Yamashita, 1977; Kusomoto and Furukawa, 1977; Miyazaki and Nagamatsu, 1965; Mukerjee et al., 1974; Kaewbanrung, 1967; Koch, 1972; Thompson, 1974a, b, 1977; Ueda and Torikata, 1972).

TABLE 2 Major Element Composition of Several Media Which Have Been Usedfor Orchid Seedling Culture Media (somefrom Withner, 1959) and Orchid-nurturing Tree Trunk EApuute (Curtis, 1946) [Amounts expressed in millimoles]

Ion Nitrate Ammonium Nitrate : Ammonium ratio Phosphate Sulphate Chloride Potassium Magnesium Calcium Citrate Iron

Murashige and Skoog

Fast, 1971

Sladden modif Burgeff

39.4 20.61 1-9

3.96 7.57 0-52

1*24 1-50 5-98 20.03 1-50 2.99

1*84 4.80

6.50

5.79 1.01 0-13 10 mg chelate

REa

B

C

Ichihashi and Yamashita

8.40 10.60 0.8

11-57 7-27 1.59

848 7.66 1.10

8-40 7.60 1.10

14.38 3.39 42

8-40 3-80 2.2

2.94

2.20 1-22

2.16

1a80

3-39

4.79

4-80

0.70

3*20 2.90

6.16 0.67 0.63

1-83 1.01 4.24

1-80 1-00 420

7-38 0.70 3.49

4.60 1-00 420

0.09

0.33

0.09

2-94 1-20 4.80 1-89 0.67

Knudson

Manganese Sodium Urea Ammonium: Urea ratio Total concentration

Burgeff Eg-1

0.07

0.034

93.25

6.91

39.94

30-46

30.5

29.72

33-43

28.17

-

Ion Nitrate Ammonium Nitrate: Ammonium ratio Phosphate Sulphate Chloride Potassium Magnesium Calcium Citrate Iron Manganese Sodium Urea Ammonium: Urea ratio

Vacin and Went 5.19 7-56 0.69 3.14 4.83 7.03 1.01 1-95 019 0.04

_

_

-

Burgeff N3f

Thomale GD

8.40 3-80 2-2

2.00 2.00 1

10-06 5.50 1-82

8-73

1-40

2.76 1*43

2.20 2.16

1.98 0.97 0.85

6-16 0.74

1*46 0.8 1 1*34 4.77 0.8 1 3.38

2-90 3.40 6-20 1a 0 0 4-20 0.43 0.07

Pfeffer

0.07

30.94 31.8 12.99 24.3 Total concentration Used as given for seedling culture and half strength for seed germination.

a

_

Raghavan and Torrey

~

.

_

-.

Tree trunk effluate 0-0025 0.0880 0.02 0.0105 0.0052 0.1430 0.0770 0.1770 0.0250

Thompson

Wynd 9.8

7.8

2.99 1.49

2.5 1.23

3.8 9.7

3.99 1-49 0-49

2.5 1.23 4.9

19.4 3.9 1-9

22-16

46.5

2.99

00073 0.1310 8-99 0.33 21.3

0.686

22.43

446

-

I. ARDITTI

TABLE 3 Optimal Range of Ions and Levels at a Total Medium Concentration (Ichihashi and Yamashita, 1977) of 20 mg I-'

Cations NH,+ K+ Ca2+ Mg2+

Range (% total cation concentration) 16-20

35-41 34-37

Range

Anions NO; H,PO,-

(% total anion concentration) 66-88 1-23 4-14

10

2. Nitrogen Nitrate, ammonia and urea can all be used as nitrogen sources by orchid seeds. However, several species have been reported to grow better on ammonia (for a review see Arditti, 1967a). Several species appear unable to utilize nitrate during the early stages of germination (de Bruijne and Debergh, 1974). Cattleya seedlings can utilize nitrate only after a 60-day growing period (Raghavan and Torrey, 1964). The ability to grow in nitrate develops in parallel with the appearance of nitrate reductase. This suggests that seedlings differentiate biochemically as well as morphologically (Raghavan, 1976). Urea has given contradictory results even with species in the same genus. Cymbidium seedlings were reported to have been inhibited (Cappelletti, 1933) and stimulated (Burgeff, 1936) by urea. Growth of Dendrobium phalaenopsis and Phalaenopsis was inhibited by urea (Burgeff, 1936) whereas development of Laeliocattleya (Magrou et al., 1949), Vanilla planifolia (Lugo-Lugo, 1955), Cattleya (Curtis, 1947), and Vanda was enhanced. A recently developed culture medium contains only urea and ammonia nitrogen (Thompson, 1977). Some of the reported differences in response to or requirement for nitrate, urea and ammonia may be due to the genera and/or species used in the experiments. Others may be a reflection of plantlet age, and the absence of nitrate reductase in very young seedlings. Hence, the suggestion that NH,NO, is the best nitrogen source (Mitra, 1970) seems reasonable. Cymbidium protocorms cultured on hydroxyurea, a DNA replication inhibitor, were malformed (Rucker, 1975). Almost all amino acids as well as urea and related substances have been incorporated in orchid seed culture media either as supplements or nitrogen sources. Arginine, ornithine and urea were capable of replacing NH,NO, in Cattleya cultures. Phenylalanine, citrulline, tyrosine, aspartic acid, glutamic acid, glutamine, asparagine and phenylurea could not. Proline and yaminobutyric acid were moderately good sources of nitrogen (Raghavan, 1964, 1976; Raghavan and Torrey, 1964). Glycine, a-alanine, valine, Uaminobutyric acid, leucine, phenylglycine, hydroxyproline, canavanine and threonine were inhibitory (Raghavan, 1964; Raghavan and Torrey, 1964).

ASPECTS OF ORCHID PHYSIOLOGY

447

Arginine and aspartic acid increased shoot formation in Cymbidium pwmilum and C. goeringii (Ueda and Torikata, 1968, 1969, 1972). Results with other amino acids have been extremely variable (for reviews see Arditti, 1967a; Sanford, 1974; Withner, 1959, 1974). Casein and lactalbumin hydrolysate, peptone and tryptone are often used in culture media (Harvais, 1972, 1973, 1974; Mead and Bulard, 1975; Pages, 1971; Rao and Avadhani, 1963; Torikata et al., 1965; Zeigler et al., 1967; Voth, 1976; Ueda and Torikata, 1968, to mention several reports of many; for reviews see Arditti, 1967a; Withner, 1959, 1974) with varying results. Since these substances are complex mixtures which may be changed chemically during autoclaving, it is difficult if not impossible to speculate regarding the reasons for their effects. In experiments with Orchis (Mead and Bulard, 1975) casein hydrolysate could be replaced by a reconstitution of its amino acids or glutamine only. These findings may be due to a specific requirement by Orchis since glutamine did not have a similar effect on Cattleya (Raghavan and Torrey, 1964). Nucleic acids and related compounds have had varied effects on orchid seed germination. Their effects are as difficult to evaluate as the complex nitrogen sources because they too may be changed by autoclaving (Arditti, 1967a).

3. Carbohydrates The first attempt to study the suitability of a sugar as a carbon source for germinating orchid seeds was made before the discovery of the asymbiotic method (Bernard, 1909). However, meaningful comparisons between sugars became possible only when defined media became available. It is not surprising, therefore, that the first comparative studies were published soon after Knudson’sinitialpublication(LaGarde, 1929;Knudson, 1916,1921;Quednow, 1930). Numerous studies followed (Arditti, 1967a; Ernst, 1967; Ernst et al., 1971a; Raghavan, 1976; Withner, 1959, 1974) and identified sugars and carbohydrates which can or cannot support germination. Not all species germinate on sugars listed as suitable, but in these cases the problems are more complex than the availability of an appropriate carbon source. Neither list (suitable or unsuitable in Table 4) is surprising. Higher organisms generally use D-sugars and orchids are no exception. Galactose is toxic to most plant tissues and this is also the case with orchids. It inhibits the growth of orchid seedlings and other plants a t levels as low as 0-69 mM (0.0125%; Burstrom, 1948; Arditti et al., 1972a; Ernst, 1967, 1974; Ernst et al., 1971a; Ordin and Bonner, 1957; Thimann, 1956). Its metabolic effects may be to interfere with cellulose synthesis (Ordin and Bonner, 1957), or hexokinase activity (Hele, 1953) or both. Nuclear chromatin (Fig. 19) in galactose-treated cells appeared dispersed throughout the nucleus rather than concentrated around the nuclear envelope.

TABLE 4 Suitability of Sugars, Polysaccharides, Other Carbohydrates, and Carboxylic Acids as Carbon Sources for Germinating Orchid Seeds (Arditti, 1967a; Ernst, 1967; Ernst et al., 1 9 7 1 ~ ;Withner, 1959, 1974) ~~

Suitable Monosaccharides c-5 D-ribose D-xylose

Remarks

In some cases

C-6 D-fructose aD-glucose /.?D-glucose D-mannose Disaccharides c-12 cellobiose 1actose maltose melibiose sucrose trehalose turanose Trisaccharides C-18 melezitose raffinose Tetrasaccharides C-24 stachyose Sugar alcohols

c-5

D-arabinitol ribitol xy1it ol C-6 mannitol sorbit ol Organic acids malate pyruvate

Marginal except for Vanilla

Unsuitable Monosaccharides c-5 D-arabinose L-arabinose D-xylose L-xylose C-6 D-galactose L-glucose L-mannose L-sorbose c-7 Sedoheptulosan Disaccharides c-12 D-lactose

~

~~

Remarks

Unsuitable for most species

Deoxysugars C-6 D-fucose L-fucose 2-deoxy-~-glucose L-rhamnose Sugar alcohols C-4 meso-eryt hritol c-5 L-arabinitol C-6 galact itol myo-inositol

Poly saccharides Starch Cellulose Organic acids Citric Acid Malic acid Oxalic acid Pyruvic acid Succinic acid Tartaric acid

Protocorms survive, but fail to differentiate

Figs 22-24. Ultrastructure of cells from galactose-treated orchid seedlings. 22. Unit membrane vesicles and myelin bodies, X 33 320.23. Amyloplasts appear normal, x 11 360. 24. Mitochondria1 cristae appear slightly swollen (black triangle), x 9660. Explanation of symbols: Am, amyloplast; MB, myelin body; V, vesicle (Ernst e t ai., 1971a).

450

J. ARDITTI

In many cases the nuclear envelope evaginated into the cytoplasm (Figs 19, 20, 21). This included both membranes of the nuclear envelope rather than the more commonly observed case of the evagination of the outer membrane only. These evaginations appear to be devoid of chromatin (Fig. 20; Ernst et al., 1971a). Numerous unit membrane vesicles were found in the cytoplasm of galactose-treated cells (Fig. 22). In addition, myelin bodies were present in almost all cells (Fig. 22). The unit membrane character of the vesicles and myelin bodies was clearly visible (Fig. 22). Both of these structures may have been derived from the tonoplast, which was invariably broken. No intact dictyosomes were visible in this material (Ernst et al., 1971a). Amyloplasts appeared normal (Figs 23, 24). Membranes were intact and showed little or no swelling. Large starch grains and small lipid droplets were present in the amyloplasts (Figs 23, 24). Mitochondria (Fig. 24) were intact, but their cristae appeared slightly swollen and the matrix somewhat clumped (Ernst et al., 1971a). The size of polysaccharide molecules poses permeability problems in orchids as in other plants and enzymes which can break them down are generally extracellular in nature. Permeability problems may also prevent the utilization of organic acids. It is reasonable to expect that orchid seeds (which under natural conditions require mycorrhizal infections for germination) germinate successfully and seedlings grow well on trehalose and mannitol, both carbohydrates of fungal origin (Smith, 1973). Trehalose is translocated into orchid seedlings by fungal hyphae (Smith, 1966, 1967) and may well be a product of glucose derived from the hydrolysis of cellulose by orchid mycorrhiza (Hadley, 1968). The effects of myo-inositol (Table 4) may have been due to supraoptimal concentrations of a compound which is variously classified as a vitamin or cytokinin and is beneficial in low amounts (Arditti and Harrison, 1977; Leopold, 1964). Glucose is very common in plants and fungi either free or as a component of polysaccharides; it is also an important starting point in many metabolic pathways. It is therefore not surprising that orchid seeds and seedlings can utilize glucose (Table 5 ; Freson, 1969). There are a number of reports that some orchid species germinate and grow better on fructose (Ernst, 1967; for reviews see Arditti, 1967a; Burgeff, 1936; Withner, 1959). Phalaenopsis seedlings, for example, take up and/or utilize fructose in preference to glucose (Ernst et al., 1971a). However, Cattleya aurantiaca seedlings do not grow as well on fructose as they do on sucrose or glucose (Fig. 25; Harrison, 1973; Harrison and Arditti, 1978). Sucrose is the sugar most commonly used in orchid seed and seedling culture. It can support growth equally well whether autoclaved or filter-sterilized (Fig. 26) but its effects may vary depending on concentration (Fig. 27;

451

ASPECTS OF ORCHID PHYSIOLOGY

TABLE 5 Eflect of Glucose on Cymbidium Protocorms (Freson, 1969) ~~

Concentration % mM 0 0.25 0.63 196 4

10

0 13.88 3496 88.81

222.02 555-06

Effects Protocorms do not multiply and become necrotic rapidly Poor growth, but tissues are green Improved multiplication, rhizoid formation, poor differentiation Best growth, development and chlorophyll content Increased production of plantlets, reduced protocorm multiplication and chlorophyll levels Protocorms do not multiply and become necrotic rapidly

Table 6; Hombs, 1973; Hombs et al., 1971; Homes and VansCveren-Van Espen, 1972, 1973a, b ; VansCveren-Van Espen, 1973). Organogenesis is promoted at suboptimal concentrations whereas protocorm proliferation is enhanced by supraoptimal levels (Hombs and Vansdveren-Van Espen, 1973a). Chloroplast structure is affected by the presence of sucrose (Hombs and VansCveren-Van Espen, 1972, 1973). In a sucrose-free medium the chloroplasts have peripheral vesicles, thick clusters, thylakoids forming misshapen grana, and numerous osmophilic globules, but no starch grains. On the addition of glucose, chloroplast structure becomes normal in 24 hours and maximum starch accumulationtakes place within four days. When Cymbidium protocorms are cultured on 0.5 % sucrose, chlorophyll content and photosynthetic oxygen evolution are at a maximum (VansCveren-Van Espen and Courtrez-Geerinck, 1974). Analyses of filter (i.e., cold) sterilized culture media containing different sugars (Table 7) indicate that Phalaenopsis seedlings can release extracellular enzymes which hydrolysea-D-glucopyranosyl-(1+2)-/l-D-fructofuranoside and a or /l-D-galactopyranosyb-glucopyranosebonds. This view is supported by the fact that hydrolysis of 8-D-fructofuranosides and a-D-galactosides decreases with increasing molecular weight of the sugars (Table 7); maltose (1+4-a-~glucosidicbond as in amylose), cellobiose(1-+4-/l-~glucosidic bond as in cellulose) and trehalose (l-+l-a-Dglucosidicbond) are apparently taken up whole, without external hydrolysis. This may explain, at least in part, the inability of most orchid seedlings to grow and develop on polysaccharides such as starch and cellulose. The germination of Miltonia and Odontoglossum seeds on 1% corn or potato starch (Hayes, 1969) is an exception. It may be due to the presence of impurities (i.e., hydrolysates such as glucose and maltose) in the starch. (For more detailed discussions see Ernst et al., 1971a.) Orchid seedlings obviously reach a stage at which they no longer require an exogenous supply of sugars. The length of the period during which seedlings require a source of sugar was established by growing Cattleya

TABLE 6 Efects of Sucrose Concentration on Growth and Development of Cymbidium Protocorms ( Vansdveren-Van Espen, 1973) Sucrose concentration

%

Molarity (m)

16

0.89

2-5-10

2-1.6

0'25-1.25

0

0.14-0.56

0 11-0.09

13'8849.38

Nature of growth or organ development Compact masses of very pale protocorms leaves roots rhizoids Friable masses of pale protocorms leaves roots rhizoids Masses and individual green protocorms leaves roots rhizoids Green protocorms in groups leaves roots rhizoids Small dark green protocorm masses leaves roots rhizoids

Illumination 23* hours 16 hours 5800lux 2900lux 5800lux 2900lux

0 0 0

+ 0

0

+++ +++ +++

+++ 0 +++ +++ +++ +++ ++ 0 +

Explanation of symbols: 0, no growth and/or development; + + +, maximal growth and/or development.

+ 0 0

++ 0 0

$,-

600

-

500

-

400

-

300

-

f 0

200

100

t0

I

20

I

40

I

60

I

80

I

100

Age (days)

Fig. 25. Effects of sucrose (open circles), glucose (closed circles), fructose (triangles) and inositol (squares) on growth of Cattleya aurantiaca seedlings (Harrison, 1973).

6oo

r

Age (days)

Fig. 26. Growth of Cattleya aurantiaca seedlings on autoclaved (solid line) and cold sterilized (broken line) sucrose (Harrison, 1973).

454

J. ARDITTI

Fig. 27. Effects of sucrose concentration on chlorophyll levels in Cymbidiumprotocorms. Explanation of symbols: a, 234 h at 5800 lux; b, 23&h at 2900 lux; c, 16 h at 5800 lux; d, 16 hat 2900 lux; PF, fresh weight; S, sucrose %; 2% = 0.11 M; 4% = 0.22 M; 6 % = 0.33 M (Vdveren-Van Espen, 1973).

auruntiucu seedlings on Knudson C medium (KC) with and without sucrose (KC-Suc) (Fig. 28) and transferring them from one to the other (Harrison, 1973, 1977; Harrison and Arditti, 1978). Seedlings on KC-SUCreached the protocorm stage, but did not produce roots and leaves (Fig. 28, Harrison, 1973). Those on KC developed normally. Following 21 days on KC only 13 % of swollen seeds or protocorms formed plantlets on KC-Suc. After 28-30 days on KC, 50 % of the protocorms transferred to KC-Suc produced leaves and developed into larger plantlets. Seedlings with one or more leaves also continued to develop following a transfer to KC-Suc. When the transfer from KC to KC-Suc was made after 47 days, 92% of the protocorms developed into complete seedlings. Protocorms maintained 15,30 or 60 days on KC-Suc required 21-30 days on KC before 50 % formed plantlets on KC-SUC. These findings suggest that the need for an exogenous carbohydrate is out-

TABLE 7 0-Hexose Content of Cold-sterilized Oligosaccharide Solutions Exposed for 4 Months to Phalaenopsis Seedlings (Emst et al., 1971~) Monosaccharide content, percentage of neutral fraction sugar Sucrose sucrose (cp Maltose Maltose (C) Cellobiose Cellobiose (C) Trehalose Trehalose (C) Melibiose Melibiose (C) Lactose Lactose (C) Raffinose

Rafhose (C) Melezitose Melezitose (C) Stachyose Stachyose (C)

Fractions detected by thin layer chromatography Sucrose (s)b, fructose, glucose Sucrose (s), fructose, glucose Maltotriose ? (w)b, maltose (s), glucose Maltotriose ? (w), maltose (s) Cellobiose(s), glucose Cellobiose(s) Trehalose (s), glucose Trehalose (s) Melibiose (s), galactose, glucose Melibiose (s) Lactose (s), galactose, glucose Lactose (s) Rafhose (s), melibiose (s), galactose, fructose, glucose Rallinose (s), melibiose (t)b, fructose Melezitose (s), turanose (w), sucrose (t), glucose Melezitose (s) Stachyose(s), manninotriose (s), raffinose (t), galactose, fructose Stachyose (s)

(C) = cold-sterilized nutrient solution not exposed to b TLC intensities: s = strong; w = weak; t = trace.

@

Glucose

Fructose

33.23 4.35 2.72 0.11 1-09 0.22 1-24 trace 0.63 0-07 1-26 trace 0.92

28-93 435

0.11 6.53 0-33

Galactose

Fructose

Galactose

-

-

-

-

12-19 2-18 0.10

-

3-99 0.06 9.31 0-03 1 -92

1 -0

6.3

1.o

6.9

1*o (0.08)C

13.3 (1.O)

2.1 (0.158)

0.10

-

-

0.22 4.79

068

-

0.16

0.1 7

Value in parentheses is based on fructose content being equal to 1.0,

Glucose

0.87

-

seedlings.

Molar ratios of monosaccharides

I -0 (65.3)

0.015 (1.0) 7.0 (1.0)

1*o (0.14)

456

600

-

500

-

400-

300-

200

rooo

J t

B

/-

/.

-0

.-.-~ . - o - - ~ o - - o . - - - - o - - - - o

p.’ I

20

I

40

-

a

- 0 I

60

I

80

100

0

Fig. 28. Growth index of Cattleyu aurantiuca seedlings raised on Knudson C medium with (solid line) and without sucrose (broken line; Harrison, 1973; Harrison and Arditti, 1978).

grown either with the appearance of the first leaf or the potential to generate it (Harrison, 1973, 1977; Harrison and Arditti, 1978; Figs 29, 30). 4. Lipids All cells in Cattleya aurantiaca embryos are heavily packed with food reserves in the form of lipid bodies (Figs 16-18, Harrison, 1973, 1977; Harrison and Arditti, 1978). Protein bodies are also found, but are restricted to cells in the upper two-thirds of the embryo. Aside from small grains within proplastids there are no starch or other carbohydrate reserves in these seeds. This confirms previous reports that the major food reserves in orchid seeds are lipids (Anon., 1922; Poddubnaya-Arnoldi andzinger, 1961). Analyses of Cymbidium seeds indicate that they contain 32 % lipids (Knudson, 1929; for a review see Arditti, 1967a). Such high concentrations of reserve materials are generally located in cotyledons and/or the endosperm of most plants. However, since these structures are lacking in most orchids their embryos apparently function as storage organs. This view is supported by the ultrastructural evidence (Figs 16-18). In fatty seeds the breakdown and utilization of lipid reserves is associated with glyoxysomes in the cells. These bodies which are involved in the conversion of lipids to sugars, the predominant metabolic activity during germination of fatty seeds, could not be discerned in germinating Cuttleyu

0

L2-:-& 10

20

I

I

I

,

30

40

50

60

Days on sucrose

Fig. 29. Percentages of Cattleya aurantiaca protocorms forming plantlets as a function

of the length of time grown on Knudson C medium with and without sucrose. Drawing denotes transfer sequence (Harrison, 1973 ; Harrison and Arditti, 1978).

Days on sucrose

Fig. 30. Percentages of Catrleya aurantiaca seedlings forming plantlets after an initial period on Knudson C without sucrose, transfer to sucrose containing medium and return to sucrose-free medium (Harrison, 1973; Harrison and Arditti, 1978).

458

J. ARDITTI

uurunfiuca seeds (Harrison, 1973, 1977). The Cuttleyu seeds converted only 3 % or less of the label from acetate-2-14Cinto sugars (in contrast with 90 % conversion of acetyl units in castor beans) and used their lipid reserves very slowly. Their lipid bodies were closely associated with or enveloped by mitochondria (Harrison, 1973, 1977). Clearly, then, orchids have fatty seeds, in which the appropriate metabolic pathways are severely limited. Presently, available evidence suggests that acetyl CoA released from the lipid bodies may be routed through the Kreb’s Cycle in mitochondria where it is oxidized to carbon dioxide with the production of energy. To some degree this may be a reflection of the fact that most orchids do not have cotyledons or endosperms, the organs capable of converting acetate into sugars. The disappearance of these structures has apparently led to the loss of some biochemical capabilities. 5. Hormones

Experiments with auxins, cytokinins and gibberellins and orchid seedlings have produced inconsistent and therefore inconclusive results (for reviews see Arditti, 1967a, c; Withner, 1959, 1974). The reasons for this appear to be: (i) physiological, in that the requirements and responses of genera and species vary; (ii) chemical, because different forms of each hormone were used; (iii) dosage, since a wide range of concentrations was employed; (iv) culture conditions, which vary from report to report; (v) age, because it may

Fig. 31. Seedlings of Cymbidium madidum grown on Knudson C with (above) and without (below) naphthaleneaceticacid (Straws and Reisinger, 1976).

TABLE 8 Effectsof Auxin on Some Orchidsa Orchid

Auxin

Remarks

Reference Strauss and Reisinger, 1976

Chondrorhynca discolor x Lycaste aromatica seeds

NAAb, 0.1 ppm (0.54 p ~ )

Bletilla sp.

N U b , 0.1 ppm (0.54 p ~ )

Cattleya aurantiaca

NAAb, 0.1 ppm (0.54 p ~ )

Coeloglossum viride seeds and seedlings Cymbidium sp. seeds

Wb,1 PPm

Seed germination accelerated slightly and seedling growth and development enhanced. Protocorms died after 6 months on auxin deficient media Seed germination accelerated slightly and seedling growth and development enhanced Seed germination accelerated slightly and seedling growth and development enhanced No effect

K salt of NAAb

Inhibitory

Torikata et al.,

Cymbidium sp. seedlings

NAAb, 0-1-1 ppm

Stimulates growth

Torikata et al.,

Cymbidium kanran rhizome Cymbidium kanran leaf-bud Cymbidium madidum seeds

NAAb NAAb NAAb, 0.1 ppm (0.54 p ~ )

Cymbidium virescens rhizomes and leaf buds Dactylorhiza (Orchis)purpurella seeds and seedlings

NAAb

Growth slightly promoted No differentiation Seed germination accelerated slightly and seedling growth enhanced Promotion

Sawa, 1969 Sawa, 1969 Strauss and Reisinger, 1976 Sawa, 1969

IAAb, 1 ppm

No effect

Hadley, 1970

Strauss and Reisinger, 1976 Strauss and Reisinger, 1976 Hadley, 1970 1965 1965

continued

TABLE S-continued Orchid

Dactylorhzka (Orchis)purpurella seeds and seedling Dendrobium protocorms Dendrobium seedlings

Auxin

Remarks

Reference

LAAb

Results inconclusive

Harvais, 1972

IBAb, 5 ppm; NAA, 0-5-08 ppm IAA,2 mg 1-l; NAA, 2 mg 1-l

Growth enhancement Induced callus formation, reduced percentage of normal seedlings; increased number and length of leaves Inhibition

Pages, 1971 Mukherjee et al., 1973

Dendrobium nobile seeds and IAAb, 0.1, 1, 10ppm seedlings Goodyera repens seeds and seedlings W b1,ppm Miltonia spectabilis 1 g 1-1 to 1 x 10-8 g 1-1 Miltonia spectabilis var. moreliana

l g l - I t 0 1 x 10-sg1-1

Odontoglossum grande

1 g 1-1 to 1 x 10-8 g 1-1

Odontoglossum schlieperianum

1 g 1-1 to 1 x 10-8 g 1-1

Orchis (Dactylorchis)purpurella seeds and seedlings

IAAb, 025,05, or 1 ppm

No effect Germination not “drastically stimulated”; differentiationpromoted Germination not “drastically stimulated”; differentiationpromoted Germination not “drastically stimulated‘‘; differentiation promoted Germination not “drastically stimulated”; differentiation promoted Germination impeded, elongation of protocorms

Miyazaki and Nagamatsu, 1965 Hadley, 1970 Hayes, 1969 Hayes, 1969 Hayes, 1969 Hayes, 1969 Hadley and Harvais, 1968

Phalaenopsis ovules Platanthera bifolia Spathoglottis plicata seeds

NAA, 0.5-1 ppm IAAb, 1 ppm 2,4-Db

Growth enhancement No effect Germination

Vanda cv Miss Joaquim seeds and seedlings Van& cv Miss Joaquim seeds and seedlings Vanda cv Miss Joaquim seeds and seedlings VfUl&

IAAb, 25, 50,100 ppm

Germination inhibited

Pages, 1971 Hadley, 1969 Chennaveeraiah and Patil, 1973 Goh, 1971

2, 4-Db, 0.1, 0.25, 0-5, 1, 2, 5 ppm

Germination inhibited

Goh, 1971

2,4-Db

Callus formation after 3 months

Goh, 1971

IAAb, 1 ppm

“Not very effective in the formation of seedlings”

Rao and Avadhani, 1963

a

For a table of pre-1967 findings see Arditti 1967a. indoleacetic acid; IBA, indolebutyric acid; NAA, naphthaleneacetic acid.

b 2,4-D, 2, 4-dichlorophenoxyaceticacid; IAA,

462

J. ARDITTI

have affected seed response; and (vi) interactions, due to the fact that several combinations of hormones with or without other substances, culture conditions and seedlings were employed. Auxin was the first plant hormone added to orchid cultures (Burgeff, 1934; for reviews see Arditti, 1967a; Withner, 1959, 1974). Results varied. In the majority of cases auxin (mostly IAA, IBA and NAA) enhanced germination and/or seedling growth to some extent (Fig. 31). Inhibitory effects were reported in very few cases. The same was true for “no effect” reports. In only one instance, that of excised Dendrobium ovaries, did death occur in the absence of auxin (Israel, 1963). More recent reports are also inconclusive (Table 8). Orchid pollinia are a rich source of auxin which plays an important role in the induction of postpollination phenomena (seep. 566 el seq.). However, only traces of auxin have been found in Cypripedium seeds and none at all in Calanthe and Dendrobium (Poddubnaya-Arnoldi, 1960;Poddubnaya-Amoldi and Zinger, 1961). This and the inconsistent results obtained with the addition of auxin to culture media suggest that for the most part germinating orchid seeds and developing seedlings are self-sufficient with respect to these hormones. Those that require or benefit from an exogenous supply in nature (however large or small) probably obtain it from their mycorrhizal fungi (Hayes, 1969; for a review see Arditti and Ernst, 1974). However, since at least one such fungus is enhanced by the addition of auxin (Downie, 1943), it is also possible that orchids supply hormone to their symbionts. Cytokinins have been isolated from several mycorrhizal fungi, even if not those of orchids (Crafts and Miller, 1974; Miura and Hall, 1973). Hence, it is reasonable to assume that the mycorrhizal fungi may also produce cytokinins. If so, it can also be assumed that should any orchid seeds or seedlings require an exogenous supply of these hormones, in nature this need may be satisfied by the fungi. Evidence for this in vitro would be the response of different species or stages of growth to one of several cytokinins in the culture medium. This indeed is the case: some species are enhanced by cytokinins, others are inhibited and a third group is not affected (Table 9; for reviews see Arditti, 1967a, c; Withner, 1974). Benzylamino purine (BAP) may retard development and differentiation of cells and tissues of Cymbidium protocorms (Gailhofer and Thaler, 1975). The hormone may induce the appearance of enlarged mitochondria in the light. BAP also brings about an increase in the number of young chloroplasts and delays their degeneration (Gailhofer and Thaler, 1975). In some cases the auxin : cytokinin ratio may be important (Hadley and Harvais, 1968), but the available information is not sufficient for generalizations. Germinating seeds are more sensitive to higher cytokinin concentration than protocorms. This may indicate either that the seeds have lower re-

TABLE 9 Effects of Cytokinins and Related Substances on Some Orchidsa Orchid

Cytokinin

Remarks

Kinetin, 10 ppm

Formation of protocorms not affected by low concentrations Proliferating protocorms with high concentrations Number of roots and their fresh weight and dry weight decreased with increasing concentrations Number of shoots increased by high concentrations Fresh and dry weight of protocorms and shoots decreased at low concentrations Germination retarded, growth rate of protocorms increased No root formation

Kinetin, 10 ppm

No root formation

Eenzyl adenine Benzyl adenine

0.1 ppm-development retarded 1.0 ppm-formation of roots and root

Cattkya aurantiaca, unripe seeds

Benzyl adenine

Coeloglossum viride seeds and seedlings Cymbidiumgoeringi (Cym. virescens) shoots Cymbidium insigne shoots

Kinetin, 1-10 ppm

Cymbidium cv In Memoriam Cyril Straws, protocorm

N-Benzyladenosine Kinetin

Reference Pierik and Steegmans, 1972

Hadley, 1970 Ueda and Torikata, 1972 Ueda and Torikata, 1972 Rucker, 1974

hairs inhibited 10 ppm-differentiation of buds inhibited. Chlorophyll synthesis inhibited 50 ppm-teratogenic and toxic effects, chlorophyll synthesis inhibited Mitotic and endomitotic activity stimulated by all concentrations continued

TABLE 9-continued

Orchid

Cymbidium kanran rhizomes Cymbidium virescens rhizomes Cypripedium calceolus seeds Cypripedium reginae seeds and seedlings Dactylorhiza (Orchis)purpurella seeds and d i n g s Dactylorhizapurpurella seeds and seedlings Goodyera repens seeds and seedlings Orchis purpurella seeds and seedlings Platanthera bifolia seeds and dings Spathoglottisplicata seeds and seedlings Vanda cv Miss Joaquim seeds a

Cytokinin

Adenine sulphate Adenine sulphate Kinetin or benzyl adenine, 1 ppm 6 (y, y-dimethylallyl-amino) purine Kinetin riboside, zeatin Kinetin, 1-10 ppm 6 (y, y-dimethylallyl-amino) purine Kinetin riboside Kinetin, 1-10 ppm

Remarks Differentiation of cell walls occurred earlier Effects can be reversed by transplanting protocorms on cytokinin-free medium No effect No effect Germination enhanced Growth impeded

No morphogenetic effects Germination impeded, growth rate of protocorms increased “Shoot characters” and chlorophyll formation enhanced “Root characters” suppressed No response

Kinetin, 1 ppm

Pronounced effect on growth and development Little effect Retard germination. Growth rate of protocorms increased Germination

Kinetin, 2 ppm

Some inhibition

Kinetin Adenine Kinetin, 1-10 ppm

For tables of pre-1967 reports see Arditti, 1967a, c,

Reference

Sawa, 1969 Sawa, 1969 Borriss, 1969 Harvais, 1973 Hadley, 1970 Harvais, 1972 Hadley, 1970 Hadley and Harvais, 1968 Hadley, 1970 Chennaveeraiah and Patil, 1973 Rao and Avadhani, 1963

TABLE 10 Effects of Gibberellins on Some Orchidsa Orchid Cypridium calceolus seeds and seedlings Dendrobium seedlings

Gibberellin G&, 5 ppm GA

Dendrobium nobile seeds and seedlings

GAY1, 10, 100 ppm

Orchis purpurella seeds and seedlings

GA,, 2-5, 5, 10 ppm

Vundu cv Miss Joaquim

GA, 500 ppm

a For a table listing pre-1967 findings see Arditti, 1967a.

Remarks Differentiation enhanced Induced callus formation; reduced percentage of normal seedlings, increased length, and number of leaves Enhanced rapid germination and plantlet formation. Some inhibition at 100 ppm Enhanced protocorm survival, caused abnormal elongation of emergent shoots; did not influence the growth and overall rise of protocorms “. . . to a certain extent inhibitory to . . . seedling formation.”

Reference Borriss, 1969 Mukherjee et al., 1973

Miyazaki and Nagamatsu, 1965 Hadley and Harvais, 1968 Rao and Avadhani, 1963

466

J. ARDITTI

quirements (if any, although it is hard to imagine that developing embryos d o not need these hormones), can produce all they need, deactivate the hormone at a slower rate, or combine these three possibilities. Protocorms, on the other hand, may be unable to produce enough, have the capability for faster deactivation or combine the three. In any case, the influence of cytokinins on orchid seedlings is in line with their known effects on other plants. Basidiomycetes, a group of fungi which includes orchid mycorrhizal symbionts (Smith, 1974; Strullu, 1974), contains species which produce gibberellin-like substances (Pegg, 1973). Hence, it is possible to reason (as with cytokinins) that those orchid seeds and seedlings which require gibberellins can obtain them in nature from their mycorrhizae. If so, responses to exogenous GA in culture media can be expected to vary with growth stage and species. And this indeed seems to be the case in several instances (Table 10; Arditti, 1967a, c; Withner, 1959, 1974). However, the effects of exogenous gibberellins tend to be negative. Altogether, it seems that with respect to gibberellins, orchid seedlings can synthesize whatever amounts they may need. Further, it would appear that the seeds or seedlings may have a limited ability to deactivate the hormone. Consequently, exogenous gibberellins, except when added in very small amounts, may raise concentrations in culture media to supraoptimal levels for most species. Thyroid (Withner, 1951) and “female” (Schopfer, 1943) hormones have also been applied to germinating orchid seeds and developing seedlings, but no growth-promoting effects have been noted.

6.Vitamins Since the subject has been reviewed several times, twice recently (Arditti, 1967a; Arditti and Ernst, 1974; Arditti and Harrison, 1977; Withner, 1959, 1974) only a brief summary and several additions (Tables 11, 12) will be presented here. Ascorbic acid (Vitamin C), biotin, folic acid, inositol (this polyol is variously classified as a vitamin, cytokinin, or growth factor), niacin, pantothenic acid, pyridoxine, riboflavin, thiamine, and vitamins A, BIZ,D, E and T (which can be best described as a termite extract), as well as related substances such as p-aminobenzoic acid, glutathione and other compounds have all been tested for their effects on orchid seed germination and seedling growth. As with hormones, results have been inconsistent (Table 11). In part this may be due to physiological differences between species or developmental stages. However, the inconsistencies may also be due to the presence of vitamins as impurities in sugars or other media components. In other words, basal media (i.e., controls) presumed to be vitamin-free were not and formulations presumed to include certain concentrations did in fact contain more. In such instances the results obtained would be unreliable. Proof that

TABLE 11 Effects of Vitamins on Several Orchidsa Orchid Cypripedium reginae seeds and seedlings Dendrobium seeds

European terrestrial species Serapias orientalis

a For

Vitamins Pantothenic acid, 0.5 ppm Pyridoxine, 0.5 ppm Thiamine, 5 ppm Pyridoxine HCI, 300 ppm Riboflavine, 300 ppm Thiamine HCI, 300 ppm Niacin, 2 ppm Pyridoxine, 2 ppm Thiamine, 2 ppm 1 tablet of Multivit B 1-l: Adermin, 0-5 ppm Lactoflavin, 1 ppm Niacin, 10 ppm Calcium pantothenate, 1 ppm Thiamine, 1.5 pp m

more extensive tables see Arditti, 1967a.

Remarks

Reference

Brought about increases in leaf size to “natural proportions”

Harvais, 1973

Improved germination

Mukherjee et al., 1974

Improved germination

Borriss, 1969

Enhanced germination

Voth, 1976

468

J. ARDITTI

TABLE 12 Summary of Vitamin Effectson Germinating Orchid Seeds and Developing Seedlings (from Arditti and Harrison, 1977 and Table 11)

Vitamin Ascorbic acid (Vitamin C) Biotin Folic acid Inositol

Effect Increased germination and growth in Cattleya and Oncidium.Promotes embryonic growthin Cymbidium. No effects on Cattleya and Epidendrum. Enhances growth andlor colour in Cattleya, Odontoglossum, Paphiopedilum,and Cymbidium. Mostly without effects. Unsuitable as sole carbon source for Dendrobium and Phalaenopsis. No effects on Cattleya, Epidendrum, and Goodyera. Possibly stimulates germination of Cattleya.

Niacin Pantothenic acid Pyridoxine Riboflavin Thiamine Mixtures of vitamins (Table 11)

The only vitamin reported to consistently enhance germination and development of several orchids. Generally without effects. Reported to enhance germination or growth, have no effects or be inhibitory. Does not seem to stimulate germination, enhances differentiation of plants at leaf point stage and promotes embryonic growth. The vitamin itself or only its pyrimidine moiety can enhance germination and growth. Improved germination and growth.

this may be so was obtained from early experiments with niacin (Noggle and Wynd, 1943). The available evidence indicates that symbiotic fungi provide orchids with vitamins or their precursors (Harvais and Pekkala, 1975; Hijner and Arditti, 1973; Mariat, 1948, 1952; Stephen and Fung, 1971; Vermeulen, 1947; for reviews see Arditti, 1967a; Arditti and Ernst, 1974; Arditti and Harrison, 1977; Withner, 1959, 1974) and possibly other factors (Ueda and Torikata, 1974). When incorporated in orchid culture media, folic acid has been mostly without effect (Table 12; Downie, 1949; Withner, 1951), but in one case it did stimulate germination (Mariat, 1948, 1952, 1954). Therefore, it appears that most orchids are self-sufficient in respect of this vitamin. However, some mycorrhizal fungi require folic acid or its component moiety, paminobenzoic acid (Hijner and Arditti, 1973; Stephen and Fung, 1971; Vermuelen, 1947). Arundina chinensis (Stephen and Fung, 1971), Dactylorhim (Vermuelen, 1947) and Epidendrum . (Hijner and Arditti, 1973) seedlings produce enough folic or p-aminobenzoic acid to satisfy these requirements. Niacin is reported to enhance orchid seed germination and seed development more consistently than any other vitamin (for reviews see Arditti, 1967a, b; Arditti and Ernst, 1974; Arditti and Harrison, 1977; Withner,

ASPECTS OF ORCHID PHYSIOLOGY

469

1959, 1974). This suggests that a requirement for niacin may exist. Under natural conditions the need is probably satisfied by the mycorrhizal fungi. Support for this suggestion is the production and release of niacin into culture media by two Rhizoctonia isolates, one from Dactylorhiza purpurella (Harvais and Pekkala, 1975) and another from Cymbidium (Hijner and Arditti, 1973). Perhaps the most interesting case is that of thiamine. Several investigators have reported germination and growth enhancement by thiamine (Table 11, 12; Arditti, 1967a, Arditti and Harrison, 1977; Withner, 1959, 1974). In one instance it was shown that in addition to the vitamin itself, its pyrimidine fraction alone was capable of such enhancement (Magrou and Mariat, 1945; Mariat, 1944,1948,1952). Under symbiotic conditions fungi such as Corticium catonii (Cappelletti, 1947) and a Rhizoctonia species isolated from Cymbidium (Hijner and Arditti, 1973) supply orchid seeds and seedlings with thiamine and its components. In return the fungus, which also requires thiamine or its thiazole moiety, obtains these substances from the orchid. This is coevolution on the half molecule level and a good example of nutritionally mutualistic symbiosis. 7. Complex Additives A large and bewildering number of complex additives have been used in orchid seed and seedling culture media. Coconut water, yeast extract, peptone, microbiological preparations (Kukulczanka and Sobieozczanski, 1974) and casein hydrolysate are among the common ones. Honey, sauerkraut juice, salep, fish emulsion and beef extracts are prosaic even if somewhat unusual. Malaysian beer and extracts of silkworm pupae are the most exotic additives used (Table 13; Arditti, 1967a; Ernst, 1967b; Ernst et al., 1970; Withner, 1959, 1974). The determination of which complex additive may be best is more important horticulturally than scientifically. Still, these findings may be of interest for those engaged in tissue culture and working with a plant which does not respond to normal treatments. Fractionations of fungi (Downie, 1949), tomato (Arditti, 1966) and banana (Arditti, 1968) indicate that in all three the active fraction is insoluble in water or ethanol (Withner, 1974). T,his still leaves a large number of substances to choose from, but at least suggests that distribution of the active factor(s) is not limited to fungi. The experiments with bark substrates (Frei, 1973a, b, 1976; Frei and Dodson, 1972; Pollard, 1973) do not necessarily approximate natural conditions (Sanford, 1974) for at least four reasons. First, bark samples were added to a culture medium at a pH of 5.5 and autoclaved (Frei, 1973b; Frei and Dodson, 1972). This may have modified the bark chemically, hydrolysed and/or extracted substances which would not necessarily be soluble in treetrunk effluates under natural conditions. In the forest only substances soluble

TABLE 13 Effect of Complex Additives on Some Orchidra Orchid Canadian native species, seeds and seedlings Calypso bulbosa seeds and seedlings Cymbk3um seeds and seedlings Cymbidium seeds and seedlings

Cypripedium reginae seeds and seedlings Dactylorhiza purpurella seeds and seedlings Dendrobium ovules Dendrobium protocorms

Additivds) Casamino acids, yeast and potato extracts Potato dextrose agar Banana juice; banana plus apple juice; banana juice plus peptone Tomato juice Fish extract plus peptone Extract of silkworm pupae Casein hydrolysate, yeast extract Potato extract Casamino acids Casamino acids plus yeast extract Banana homogenate plus indolebutyric acid or NAA Banana homogenate with protease peptone Banana homogenate plus coconut water and NAA

Remarks

Reference

“Intolerant”

Harvais, 1974

Best germination

Harvais, 1974

Promoted growth of protocorms

Kusumoto and Furukawa, 1977

Inferior to growth on unsupplemented Knudson C Accelerated growth of protocorms Stimulated growth “Intolerant”

Torikata et ul., 1965

“Beneficial” Superior germination and growth

Harvais, 1973 Harvais, 1972

Further improvement of growth and survival “Best response”

Pages, 1971

“Best plantlets”

Pages, 1971

Torikata et al., 1965 Torikata et al., 1965 Harvais, 1973

Dendrobium hybrid seeds and seedlings Epiphytic orchids seeds and seedlings European native species seeds and seedlings Paphiopedilum seeds Paphiopedilum seedlings Phalaenopsis protocorms

Phalaenopsis ovules Serapias parvifora and S. orientalis Vanda cv Miss Joaquim

a

Banana, coconut, tomato, fish emulsion Bark from Quercus trees

Most suitable

Mowe, 1973

Toxic or inhibitory

Coconut water

Enhancement

Frei and Dodson, 1972 Borriss, 1971

Peptone (“Fleischpeptone”) or fish meal Banana Banana Pineapple, fig and tomato fruits Coconut milk

Enhancement

Fast, 1971

Enhancement Most favourable Pronounced increase in growth

Ernst, 1967b

Strong proliferation, retarded differentiation Retarded growth, toxic “Best supplements”

Pages, 1971

“Best addition”

Voth, 1976

Tomato juice or coconut milk

Bigger seedlings formed

Rao and Avadhani, 1963

Pollinium extract, casein hydrolysate Yeast extract

“Not very effective in the formation of seedling” “. . .less effective and to a certain extent inhibitory. . .”

Grapes and raspberries Coconut water plus NAA Coconut water plus peptone Yeast with or without peptone

For pre-1967 literature see Arditti, 1967a.

TABLE 14 Effects of Light Intensity, Quality, Photoperiodr and Sources of Illuminution on Germinating Orchid See& and Developing Seedlingsa Orchid

Photoperiod (h)

Light intensity

Remarks

Light Quality

Reference

~~

Arundina bambus$olia Brassocattleya

12 16

Cattleya

16

Cymbidium

Cymbidium

3000 lux

400-5OOO lux

Philips “Natural” Cool white, warm white Standard Gro-Lux Wide-spectrum Gro-Lux Diffuse daylight

Phytor

23.)

Cymbidium Cymbidium goeringii

8200 ergs cm-* s-I

“White” (Toshiba, FL 20 SW)

73 OOO ergs cm-* s-I

Vitalux (NEC FL 20 BR)

Germination Wide-spectrum Gro-Lux is better than standard Gro-Lux or warm white. Cool white poorest

Mitra, 1971 Halpin and Farrar,1965

Shoots and terrestrial roots develop; formation of aerial roots ceases Development of protocorms “Rhizogenesis” inhibited in the dark, but etiolated shoots are formed Root formation; no root formation at 4400 and 4OOO ergs cm-* s-* under these lights or with red or blue illumination

Werkmeister, 1970a, b, 1971 Homks et al., 1971a, b Homks et al., 1973 Ueda and Torikata, 1972

Cymbidium insigne

4OOO ergs cm-2 s+

Cypripedium calceolus Cypripedium reginae Paphiopedilum seedlings Phalaenopsis Serapias orientalis Terrestrial species

Various species a

3000 lux

For a review of pre-1967 literature see Arditti, 1967a.

As above plus blue and red

Root formation. Best root formation with Vitalux. Red light not as good as blue Germinates in the dark Best germination in the dark Improved growth on darkened media Better growth on darkened media Germination in the dark Light inhibits germination Good for germination

Ueda and Torikata, 1972 Fast, 1976 Harvais, 1973 Ernst, 1974, 1975, 1976 Ernst, 1975, 1976 Voth, 1976 Stoutamire, 1974 Mukherjee et al., 1974

474

J. ARDITTI

in effluates [i.e., a dilute solution of minerals and possibly sugars and amino acids (Table 2; Curtis, 1946; Sanford, 1974) in cool usually rain, water] would affect orchid seeds and seedlings. Hence in these experiments the seedlings may have been cultured on media with limited or no resemblance to natural conditions and possibly even toxic due to substances generated during autoclaving by extraction from the bark, through hydrolysis of larger compounds and/or alteration of existing compounds. Second, orchid seedlings were usually found in association with lichens and mosses (Frei, 1973a; Pollard, 1973). This raises the possibility that their contact with the bark may be indirect. Third, the effluate which reaches the seedlings may be modified by its passage through the lichens and mosses. As the water percolates through (or washes) the bark it picks up solutes (substances which would dissolve in rain water at ambient temperature). Before reaching the orchids the effluate passes through the mosses and lichens which may add some solutes and/or remove others through uptake and/or by acting as ion exchangers. Consequently, it is possible that not all substances leached from bark reach the orchids. Fourth, it is possible that in the quadripartite association the primary relationship which could be influenced by bark substances is the one between (i) moss, (ii) lichen and (iii) tree. The (iv) orchids may depend on or require moss or lichen growth to establish themselves. If so, the effects of bark substances on the mosses and lichens would be the determining factor. Indeed, “those trees that had the most orchids had the most lichens and mosses. In each instance when the seedling was removed from the tree, I found mosses and lichens between the root system of the seedling and the bark.” (Frei, 1973a). Conversely, “. . . it was found that the trees which were the strongest in inhibition, and, as a result had no orchids, also were lacking in mosses and lichens”, and “the inhibitors in the bark . . . were a factor only in the growth or lack of growth of moss-lichens.’’ (Pollard, 1973). 8. Light Requirements for and response to light and/or photoperiods by orchid seeds vary (Table 14; for a review see Arditti, 1975a). At best, generalization can be only tentative due to insufficient information. Still, it is possible to suggest at present that epiphytic species can germinate both in the light and dark. However, they seem to require light for the induction or improvement of shoot and/or root formation. Some terrestrial species behave similarly (Ueda and Torikata, 1972; Werkmeister, 1970a, by 1971). Others germinate best in the dark (Fast, 1976; Harvais, 1973; Voth, 1976; for a review see Stoutamire, 1974). Several species grow and develop better when cultured on darkened media. This phenomenon was first observed with Cymbidium (Werkmeister, 1970a, by

ASPECTS OF ORCHID PHYSIOLOGY

FW

-Leaves Roots,

Pyr*iGlarsWooI

e

% 8

TOO

-

600

-

500

-

47 5

-

W .F

FW -. FWu Leoves Roots Leaves Roots , Laaver Roots PyrexGlossWWIt Apar+NucharC A p o r t Water :

,

Nuchor C

ealmct of NucharC

400-

0

300

-

200

-

100

-

0 FW-

Loa~esRoots,

Pyrex Glass Wool

F W - u

F W u -

Leaves Roots , Leaves Roots Cotton Wool

Pyres Gloss Wool t NuohorC

Fig. 32. Effects of glasswool and charcoal on the development of Paphiopedilum and Phalaenopsis. (a) Paphiopedilum seedlings grown on Thornale GD basal medium, 200 days under Gro-Lux illumination. (b) Phalaenopsis amboinensis seedlings grown on Knudson C, basal medium, 200 days under Gro-Lux illuminations; FW, fresh weight; a, number of leaves; b, length of leaves; c, width of leaves; d, number of roots; e, length of roots; f, diameter of roots (Ernst, 1975).

1971) and confirmed with Paphiopedilum and Phalaenopsis (Fig. 32; Ernst, 1974, 1975, 1976). A number of factors can be invoked to explain this effect. One is that the charcoal contributes microelements. However, an analysis of the charcoal (Ernst, 1975) shows that this cannot be the case (Table 15). The water extract of charcoal had no beneficial effect on seedling growth and the microelements found in charcoal are present as impurities of media components. Another explanation is that the dark medium establishes polarity which

476

J. ARDITTI

TABLE 15 A Semi-quantitative Spectroscopic Analysis of Nuchar C and Its Water Extract (Based on Weight of Charcoal, Ernst, 1975)

Element Na

Nuchar C Not extracted, % Water extract, % 0.344 0.139

cu Zn

0.35 0.17 0.034 0.048 0.020 0.36 0.11 0.006 0.041 0.001 0.002

MO

Trace

-

B

O.OOO4

K Ca Mg Mn

Si A1

Sn Fe

Trace 0.024 0.003 0.24

0.063 0.003

Trace 0.0007

Trace Trace

enhances the formation of terrestrial roots, reduces or eliminates negative-, or a-geotrophic effects and improves differentiation (Werkmeister, 1971). This appears plausible but is not supported by findings that growth on glass wool (which is white) is better than on darkenedagar (Ernst, 1974, 1975,1976). Absorption of growth inhibitors of seedling origin by charcoal has been proposed as a reason for the improved growth (Ernst, 1974; Werkmeister, 1970b). Again findings with seedling cultures on glasswool do not seem to offer corroboration. Seedlings supported on glasswool in liquid media grow better than those on charcoal-containingmedia. Therefore absorption cannot be a major factor unless dilution of inhibitors in liquid media is faster than in agar. The pH of charcoal-containing media is the same as that of controls. Clearly, then, a more appropriate pH is not a reason for the improved growth on charcoal-containing media. The equally good growth on charcoal or glasswool and charcoal suggests that a plausible explanation for the effects of both is improved aeration. The presence of charcoal granules (each a miniature sponge as it were) increases the amount of air in an agar medium; seedlings on a platform of glasswool are similarly well aerated. In nature seeds germinate and seedlings grow on rough bark, incompletely decomposed litter, rocks and other debris. Aeration under these conditions is probably excellent and it is not surprising therefore that growth in vitro is better when seedlings have plenty of air. Altogether it seems that the effects of charcoal are not merely due to the exclusion of light. The inhibitory effects of light which are more prevalent amongst terrestrial species have been explained as being “part o f . . . [a] . . .protective mechanism . . . making it impossible for seedlings to develop at the soil surface when they would be subjected to drying during the growing period” (Stouta-

ASPECTS OF ORCHID PHYSIOLOGY

477

mire, 1974). If this is so one wonders why seeds of epiphytic species, which may be subject to the same dangers, do not appear to possess such protection or have a lesser need for it. In any case light-inhibited species become more tolerant to illumination as their seedlings develop and form leaves. There are not enough comparative studies of light quality and the effects of sources of illumination to allow for generalizations (Fast, 1967; Halpin and Farrar, 1965; for reviews see Arditti, 1967a; Withner, 1959, 1974). 9. Temperature The optimal temperature for seed germination of most species is 20-25"C, but the range may extend from 6°C to 40°C (Arditti, 1967a; Mukherjee et al., 1974; Thompson, 1977; Withner, 1959). Some species may require chilling (Stoutamire, 1974) or at least tolerate cold storage (Voth, 1976), but even these germinate best at 25°C (Harvais, 1973). 10. p H All available information indicates that orchid seeds germinate best at a pH 4.8-5-2 with the range being at least 3.6-7-6 (Arditti, 1967a). Seedlings are tolerant of low pH and grow well even when the pH is 3-3-3.7 (Ernst, 1967a, b, 1974; Miyazaki and Nagamatsu, 1965). However, media of a pH below 4.5 should not be autoclaved since this may lead to the hydrolysis of the agar and release of toxic substances. Concern for the maintenance of a proper p H has led to the formulation of buffers (Burges, 1936; Harrison and Arditti, 1970) or media (Knudson, 1951; Sideris, 1950; Vacin and Went, 1949) all of which appear to have been unnecessary.

11. Atmosphere There are a number of reports that seed germination in airtight containers is equal or superior to that under conditions of ample gas exchange (for a review see Arditti, 1967a). More recent work has shown that Cymbidium protocorms develop roots and shoots in shallow layers of non-agitated liquid media. If the solution layer is deep, differentiation is inhibited but protocorm proliferation is increased (Horn& et al., 1973). Under nitrogen, differentiation and growth are inhibited. This report tends to support findings that orchid seedlings grow better under the improved aeration provided by glasswool platforms or charcoal incorporation in media (Ernst, 1974, 1975, 1976). Seeds of the achlorophyllous orchid, Galeola septentrionalis, germinate only in airtight vessels. Germination is enhanced by an air pressure of 1.8 atmospheres (Nakamura, 1962; Nakamura et al., 1975). The presence of oxygen and carbon dioxide at concentrations of 5 % (approximately 25% of normal) and 8 % (24 000% of normal) respectively is essential. Ethylene, at levels ranging from 2 to 8 pl 1-1 enhanced germination. In subsequent experiments best germination occurred under 10% 0, (50% of normal), 6 % CO,

478

J. ARDITTI

(18 200 % of normal) and 84 % N (108 % of normal) at a pressure of 1.4 kg cm-,. Tolerable ranges were 5-10% 02,2-10% CO, and 1.1-210 kg cm-, pressure (Nakamura, 1976). Subsequent development was not influenced by CO, in the range of 0-8 % and pressure from 1.0 to 1.8 kg cm-2. However, 0, levels were important with the range being 5-20 % and the optimum between 10% and 15%. Galeola septentrionalis is a chlorophyll-free holomycotrophic orchid which grows underground. Only its inflorescence appears above the soil. Hence it is reasonable to assume that if germination occurs underground, the requirements would be for high CO, and low 0, levels. Such conditions have been reported, with CO, easily reaching 16% and 12% being "quite usual''. Oxygen levels in some soils have been reported to be as low as 1 % (for a short review see Nakamura et al., 1975). The requirement for increased pressure is not easy to explain due to ". . . the difficulty of measuring atmospheric pressure in the soil . . ." (Nakamura et al., 1975). Clearly, germinating orchid seeds differ in their requirements for atmospheric conditions. This is not surprising in view of the great diversity of orchids and their adaptations to many different ecological niches. 12. Photosynthesis Cattleya aurantiaca seedlings cultured on Knudson C (KC) medium (Knudson, 1946) have detectable chlorophyll 15 days after the start of culture. Levels of chlorophyll a (Chl a) and chlorophyll b (Chl b) are nearly equal at that time and remained so for I-lt months (Harrison, 1972, 1973, 1977; Harrison and Arditti, 1978). After this period concentrations of Chl a increased, reaching a maximum by 180 days. Levels of Chl b remained unchanged (Fig. 33). On Knudson C without sucrose (KC-Suc) chlorophyll Total CHL

4.0-

c

3.01

e2 i H

0 0

E

-

/.

2.0-

,,

-

..... ........ ........

CHL b

..... . L

1.0-

OO

.fl"

0.2.

50

100

...0--

150

0

200

0

250

0

50

100

150

Age (days 1

Fig. 33. Levels of chlorophyll a, chlorophyll b and total chlorophyll in Cattleyu aurantiaca seedlings grown on Knudson C medium with and without sucrose (Harrison, 1973; Harrison and Arditti, 1978).

479

ASPECTS OF ORCHID PHYSIOLOGY

4'

"

" 50

'

' " I 100

' ' ' '150 I '

' '200 I "

'

'

I250

Age (days)

Fig. 34. Chlorophyll a / b ratios in seedlings raised on Knudson C medium with (solid line) and without (broken line) sucrose (Harrison, 1973; Harrison and Arditti, 1978).

r*-*

-

/

Age (days)

Fig. 35. Ribulose-1$-diphosphate carboxylase activity in Cuttleya uuruntiacu seedlings raised on Knudson C medium with (solid line) and without (broken line) sucrose. The asterisk denotes values found in leaves of mature plants (Harrison, 1973; Harrison and Arditti, 1978).

became measurable after 25 days. Levels of Chl a and Chl b were the same and remained constant (Fig. 33). As a result, the Chl a : Chl b ratio in seedlings on KC increased whereas that of plantlets on KC-Suc was unchanged (Fig. 34). Another component of the photosynthetic apparatus, the enzyme ribuloselY5-diphosphatecarboxylase (RuDPCase), increased rapidly between the third and ninth week in seedlings grown on KC and then reached a plateau at a level equal to that found in mature plants. On KC-Suc, concentrations of

480

J. ARDITTI

Age (days 1

Fig. 36. Specific activity of ribulose-l,5-diphosphatecarboxylase in Cuttleyu uuruntiucu seedlings raised on Knudson C with (solid line) and without (broken line) sucrose (Harrison, 1973). KC w/o Sucrose

......o.......

.-I.#---.------m--

-*.

....

.I

Age (days)

Fig. 37. Photosynthetic and dark fixation of CO, by Cuttleyu uuruntiucu seedlings raised o n Knudson C medium with or without sucrose (Harrison, 1973; Harrison and Arditti, 1978).

RuDPCase rose slowly by the ninth week and increased slightly thereafter (Fig. 35). Specific activity of the enzyme reached a maximum after 30 days on KC and decreased thereafter. On KC-Suc the peak was reached at 60 days, followed by a decline (Fig. 36). RuDPCase is the major enzyme involved in the initial incorporation of COzby the Calvin cycle during photosynthesis. Therefore it is not surprising that the photosynthetic capacity (Fig. 37) and RuDPCase (Fig. 36) increase simultaneously. Specific activity of RuDPCase rose sharply between the 15th

ASPECTS OF ORCHID PHYSIOLOGY

48 1

and 40th day. Enzyme levels increased as seedlings were maintained for longer periods on KC. Hence it seems that production of this enzyme may be one of several events which occur in seedlings when an adequate supply of carbohydrates is available, i.e. production of RuDPCase is one biochemical step towards the establishment of autotrophy, but it requires a source of energy other than CO,. 13. Miscellaneous Factors (a) Morphactins. Protocorm-like bodies develop from shoot-tip (“meristem”) cultures of Cymbidium. They develop like protocorms formed during seed germination (Arditti, 1967a, 1977; Morel, 1974). Therefore, information obtained from such cultures can be included in this section. Morphactin IT 3456, when applied at concentrations in the range of 0.01-10 ppm to protocorm-like bodies stimulates proliferation, but inhibits formation of roots, shoots and rhizoids. It also causes: (i) deformation of leaves and shoots; (ii) formation of two to three shoots on one protocorm; (iii) appearance of “. . . secondary . . .” protocorms or leaves (Kukulczanka and TwardaPredota, 1973). (6) Gamma rays. Since orchid seeds and embryos contain fewer cells than those of most other plants they are convenient organisms for studies of radiosensitivity. In one study Dendrobium nobile seeds were irradiated with 10, 20, 30,40, 60 and 80 KR gamma rays from 13’Cs and 6oCo.Germination rates of irradiated seeds were lower than those of controls during the initial three weeks and indirectly proportional to dosage. However within six weeks “. . . the rate in each treatment reached almost constant value . . .” (Miyazaki, 1968). Survival rate showed a similar tendency at 10, 20, 30 KR. At 40 KR there was an abrupt drop. Growth stopped at 60 and 80 KR. In addition, irradiation caused many deformities. (c) Surfactants. Most nonionic, ionic and amphoteric biodegradable surfactants are toxic to orchid seedlings at concentrations higher than 100 ppm (approximately 0.3-0.4 mM). Phytotoxicity cannot be correlated with the surface tension reduction properties of surfactants. However, a coincidence does appear to exist between interfacial tension reduction and phytotoxicity (Ernst et al., 1971b). This is probably because surfactants affect the cytomembranes (Healey et al., 1971). Sodium (linear) dodecylbenzene sulphonate caused severe damage after 4- and 48-h exposure to 1000 ppm (2.9 mM). Chloroplasts lost membranes and underwent drastic changes in morphology (Figs 38-40, vs. 41-44); their thylakoids were swollen and osmophilic granules became evident (Fig. 44). Other effects are the disintegration of polysomes into monosomes (Fig. 41); plasmolysis (Fig. 44); swelling of mitochondria (Fig. 44); and dispersion of chromatin (Fig. 43). These ultrastructural changes have been attributed to emulsification of membrane lipids and precipitation and dispersion of cell proteins (Healey et al., 1971).

Figs 38-40. Ultrastructure of Phalaenopsis seedlings (Healey et al., 1971). 38. Internal organization of chloroplast and nucleus with well-developed internal membranes and no starch, x 22 800. 39. Portion of cell from presumptive mesophyll with a chloroplast distended by a large starch deposit and an extensive vacuole, x 13 800. 40. Polyribosomes, x 53200.

Figs 4 1 4 . Ultrastructure of Phaluenopsis cells from seedlings treated with lo00 ppm (2.91 mmol) sodium (linear) dodecylbenzene sulphonate (Healey et ul., 1971). 41. Polysomes dissociated into component monosomes (arrows), after a 4 h exposure, x 38 OOO. 42. Chloroplast showing extensive swelling of thylakoids (T) and the indistinct nature of the outer envelope brought about by a 4 h treatment, x 38 OOO. 43. After 48 h chromatin is dispersed throughout the nucleus (N) rather than being concentrated near the nuclear envelope; Chloroplasts (C) are cup shaped and contain osmophilic dense granules at their margins, x 4840. 44. Plasma membrane (arrows) remains intact after 48 h treatment, but plasmolysis is extensive and mitochondria (M) are swollen, x 14 400.

484

J. ARDITTI

14. Development Orchid seed lacks a radicle, hypocotyl, cotyledon, epicotyl and plumule (Rangaswamy, 1968). In most species the mature embryo is a mass of cells without any distinct histogens except the epidermis (Rao, 1967). On agar media seeds swell and rupture the testa within 10-60 days (Rao, 1967; Vanseveren-Van Espen, 1971). In Arundina, Calopogon, Dendrobium and Spathoglottis (for a review see Rao, 1967) only a few apical cells divide to form a promeristem. Almost all cells divide in Bromheadia and Taeniophyllum. The promeristem gives rise to a shoot apex and structures which may be homologous to cotyledons (Goro Nishimura, unpublished work in collaboration with Drs E. A. Ball and J. Arditti). A tunica and corpus are clearly identifiable in shoot apices (for a review see Rao, 1967). The tunica is always one-layered and the corpus may have three to four layers. Just before development of leaf primordia the apex is dome-shaped. As a result the embryos turn into the top-shaped structures commonly known as protocorms. In some species (for example, Calopogon, Cymbidium, Dendrobium, Laeliocattleya, Spathoglottis and Vanda) rhizoids develop from epidermal cells. These may be long, short, simple or branching. In Cattleya, leaf primordia appear following the formation of the shoot apex. Leaves develop next and are followed by roots which are always of endogenous origin (VansCveren, 1969). Organogenesis may be inhibited by immersion of the protocorm in liquid medium-i.e., in the absence of air (Homks et al., 1971a, b). The following is cited verbatim with permission (Harrison, 1973, 1977): “During the first few days of germination the larger basal cells of the embryo in seeds of Cuttleya uurunriucu (Figs 45-50) begin to swell, primarily in a

transverse direction. The first cells to enlarge are not the outer (surface) cells of the embryo but, instead, the large basal ones located inside. These cells are the first to be activated and show the earliest ultrastructural changes. Their protein bodies, when present, become less dense and begin to enlarge and apparently fuse. The impression is gained that the enlarged (fused?) protein bodies give rise directly to the cell vacuole as their protein contents are depleted. Mitochondria are abundant in all parts of the cytoplasm of these cells by the fifth day of germination. Elongated mitochondria are more numerous around the nucleus (Figs 47-48). Proplastids are also present but are fewer in number and are located near the cell wall. Neither dictyosomes nor plasmodesmata were observed.” “Whereas the interior basal cells are the first to swell and undergo ultrastructural changes, adjacent cells often appear as quiescent as they did in the dry seed. Thus, sharp contrasts in cellular structure exist between neighboring cells.” (Fig. 47). “After 5 days of germination on KC, many mitochondria were found adjacent to the nuclei of the cells in the meristematic and basal regions. Numerous lipid bodies were also present and each one was at least partially ensheathed by mitochondria1 membranes (Fig. 48). The same situation was observed in 15-day-old seeds sown on KC-Suc. Many of the mitochondria

ASPECTS OF ORCHID PHYSIOLOGY

485

present were cup-shaped or had a bowl-shaped depression in which the lipid bodies were located. Lipid bodies farther away from the nucleus were still compressed against each other and were not enveloped by mitochondria1 membranes. Apparently, rehydration occurs first in the region immediately surrounding the nucleus and the area just inside the cell wall. No glyoxysomes were evident in 15-day-old seedlings and most of the lipid reserve was still present.”

Figs 4546. Ultrastructure of orchid seedlings raised on Knudson C without sucrose (Harrison, 1973). 45. Thick section of a 20-day-old seedling. Small starch grains are visible, x 393. 46. Basal cell from an 86-day-old seedling, x 3000. Explanation of symbols: N, nucleus; S,starch grain within a plastid; V, vacuole. “Utilization of the lipid reserves was not a rapid process but took place over a one- to two-month period depending on the culture medium used to germinate seeds and raise the seedlings. Most of the lipid was used up by 27 days in seedlings grown on K C (Fig. 49). Protocorms cultured on KC-Suc required 60-65 days to use a comparable amount of reserve lipid. In both cases, lipid bodies in the smaller cells of the meristematic region and in the interior basal ones were the first to be used.” “Well-defined granal chloroplasts were apparent in all cells (including the epidermal layer) of protocorms raised on KC for 15-20 days. The same was true for seedlings cultured 20-25 days on KC-Suc. Chloroplasts in seedlings raised on K C accumulated starch rapidly (Fig. 50). Lesser amounts were stored by the chloroplasts of KC-Suc protocorms.” (Figs 45, 46). “By 87 days, all of the lipid reserves were utilized by protocorms growing on KC-Suc. Chloroplasts contained small amounts of starch and most cells were highly vacuolated. Dictyosomes were still not evident.”

Figs 47-50. Ultrastructure of orchid seedlings raised on Knudson C medium (Harrison, 1973). 47. Basal cell from a five-day-old Cuttleyu uuruntiucu seedling, x 1105.48. The area surrounding the nucleus of a basal cell in a five-day-old seedling of Cuttleyu auruntiuca, x 8265.49. Basal cell from a 26-day-old Cuttleyu auruntiucu seedling grown on Knudson C. Lipid reserves have been depleted, x 929. 50. Electron micrograph of a basal cell from a 20-day-old seedling, x 1080. Explanation of symbols: L,lipid bodies; M, mitochondria; N, nucleus; P, proplastid; PV, presumptive vacuole; S, starch; V, vacuole.

ASPECTS OF ORCHID PHYSIOLOGY

487

Development in Vanda is similar (Ricardo and Alvarez, 1971). In this orchid trichome initials differentiate from epidermal cells. They have few conventional organelles and exhibit many membrane-bound structures which contain small crystalline inclusions.

I S . Enzymes As mentioned in the section on carbohydrates (pp. 447-456), orchid seedlings secrete a number of enzymes into the culture medium. The production of ribulose-I, 5-diphosphate carboxylase was described in the discussion of photosynthesis by seedlings (p. 479). There is also no doubt that many additional enzymes are produced by germinating seeds and developing seedlings. Unfortunately, however, only very few of these have been studied, Peroxidase activity in Vanda seedlings during various stages of development was investigated histochemically. Isozymes were studied by means of electrophoresis (Alvarez and King, 1969). Activity of the enzymes is highest during the early stages of development and lowest during differentiation. This is the exact reciprocal of IAA production by seedlings. Hence, the . . temporal and spectral activity of the enzyme in the developing seedling are in accord with expectations if this enzyme, in fact, functions to control the level of IAA . . .” (Alvarez and King, 1969). Further, the increase of peroxidase activity by exogenous IAA indicates that the auxin “. . . is capable of eliciting activity of an enzyme thought to be involved in its destruction.” (Alvarez and King, 1969). Cymbidium protocorms produce an acid phosphatase which separates in three electrophoretic zones, each containing two activity bands. Activity of an RNase produced by the same protocorms appears in two zones which differ in intensity. Production of both enzymes is influenced by streptomycin, but the effects vary with the time of application (Morawiecka, et ul., 1973).

“.

16. DNA Production Growth of Dactylorhiza (Orchis) purpurella protocorms accelerates markedly following mycorrhizal infection. Cortical parenchyma cells of these protocorms are predominantly 32C, 64C and 128C and new DNA classes are produced. Asymbiotic protocorms enlarge and differentiate at a slower rate, but there is no evidence that their nuclei undergo endoreplication (Williamson and Hadley, 1969). These observations indicate that in D . purpurella production of new DNA requires fungal infection. Autoradiography following 3H-thymidine incorporation showed that DNA synthesis is induced in fully differentiated cortical root cells of Spathoglottis plicata following infection by Tulasnella calospora (Williamson, 1970). DNA content of parenchyma cells of cultured Vundu ovules increases together with that of RNA and nuclear size. DNA as revealed by Feulgen staining in nuclei, increases to 8C (Alvarez, 1968, 1969). Hydrolysis time

488

J. ARDITTI

curves of DNA-Feulgen from senescent parenchyma cells were different from those of meristematic and normal parenchyma cells (Alvarez, 1970). Comparisons of acridine orange dye binding with Feulgen measurements indicate that no “masking” or “unmasking” of phosphate groups occur during cellular differentiation of Vanda seedlings. Therefore, the increased acridine orange binding is indicative of higher DNA content (Alvarez and Reyniers, 1970). In contrast with Dactylorhiza purpurella and other temperate zone terrestrial orchids, Vandu seeds are easy to germinate asymbiotically. Perhaps this is a reflection of the differences in the ability of seedlings to synthesize DNA with o r without mycorrhizal infection. Regular endopolyploidization also occurs in some parenchyma cells of asymbiotically germinated Cymbidium seeds (Nagl, 1972). These cells show a disproportionate increase in nuclear DNA content and volume. This disappears following inhibition of DNA synthesis with hydroxyurea. Differentiation of the protocorm is accompanied by DNA amplification and endomitotic nuclear cycles (Nagl et af., 1972; Nagl and Rucker, 1972). Further evidence that DNA synthesis is required for proper growth and differentiation has been observed from experiments with hydroxyurea (OHU). Morphological differences became apparent on M OHU. At 5 x lo-’ M protocorms were reduced in size, abnormal and necrotic (Rucker, 1975). Effects of several plant hormones on Cymbidium protocorms can also be explained in terms of their influence on DNA content of the cells. Cytokinins shifted DNA replication from diploid mitotic cells to polyploid endomitotic ones. This enhanced premature and abnormal cellular differentiation (Nagl and Rucker, 1974). Auxins bring about an increase in the AT-rich DNA fractions, the G C fractions are promoted by gibberellins (Nagl and Rucker, 1976). This differential replication of the AT and G C fractions undoubtedly affected differentiation. The AT-rich satellite DNA from Cymbidium nuclei was characterized by thermal denaturation and ultracentrifugation. It is a rare instance of a major AT fraction in plants, and limited to Cymbidium at present, not having been isolated from any other orchid (Capesius, 1976; Capesius et al., 1975). Staining with quinacrine, 4‘6-diamidino-2-phenylindoleand by the Giemsa technique indicate that this DNA is located within the centromere chromatin (Schweizer and Nagl, 1976). The available evidence suggests that the AT-rich DNA is sensitive to hormone treatment (Capesius et ul., 1975; Nagl and Rucker, 1976). Appearance of the AT-rich satellite in Cymbidium is correlated with a hormone-dependent high, but variable, amount of heterochromatin, However, it appears that DNA amplification is restricted to the non-AT-rich component of the chromatin (Schweizer and Nagl, 1976). The effects of hormone treatments and the different amplification of the two DNAs indicate that they play important roles in the growth and differentia-

ASPECTS OF ORCHID PHYSIOLOGY

489

tion of Cymbidium protocorms. What may be happening in other orchids is not clear at present. F. SYMBIOTIC GERMINATION

In nature orchid seeds germinate after being infected by a fungus. This was discovered nearly 80 years ago, but the process is still not understood very well, despite its fundamental (Smith, 1974) and practical importance (Blowers, 1966; Blowers and Arditti, 1970). I , History Orchid mycorrhizae (the term was coined by Frank in 1885) were apparently seen but not recognized by H. F. Link as far back as 1824. In 1840 he presented unclear graphical evidence of fungi in Goodyera repens root cells. The “bysoid substance” in roots of Monotropa hypopitys (Ericaceae) was recognized as being a fungus by Reissek in 1842. Four years later he suggested the presence of a fungus in the roots of several orchids including Neottiu nidus avis. Schleiden (of cell theory fame) saw the fungus in 1849, but apparently failed to grasp its significance. During the next 50 years there were several descriptions of the fungus and clump formation in roots. However its importance was not appreciated until Noel Bernard noted that seedlings of Neottia nidus avis were infected by fungi and perceived its importance. (For reviews pertaining to information presented above and below in this section see Arditti, 1967a; Bernard, 1909a; Beau, 1914; Benike, 1910; Burgeff, 1911, 1932, 1934, 1936, 1959; Hadley, 1968; Hadley and Williamson, 1972; Harley, 1969; Nicolai, 1914; Ramsbotton, 1922a, b, 1929; Smith, 1974; Warcup, 1975). Bernard died 11 years after his initial discovery, but he still managed to perform a number of critical experiments and explain the role of mycorrhizal fungi in orchid seed germination. His findings were later extended by a number of investigators including Hans Burgeff in Wurzburg (a series of books and papers from 1909 to 1959), Dorothy G. Downie, University of Aberdeen (a series of articles in the Transactions and Proceedings of the Botanical Society of Edinburgh between 1940 and 1959) and John T. Curtis, University of Wisconsin (1936-1939). More recently laboratories which devote a major effort to orchid mycorrhizae include those of Geoffrey Hadley at the University of Aberdeen, Scotland who is studying the physiology and ultrastructure of orchid germination and mycorrhizae ; Gaetan Harvais at Lakehead University, Canada, who is working on physiological and nutritional problems; Sarah E. Smith at the University of Adelaide, investigating physiology, ecology and translocation; and J. H. Warcup, also at Adelaide, who is concerned with mycorrhizal fungi and factors which affect symbiotic germination.

490

J. ARDITTI

2. The Fungi The mycorrhizal fungi of orchids vary considerably in their identity, physiology and ecology although they nearly all fall into a non-sporing group known as rhizoctonias. (a) Identity. Almost all fungi isolated from orchids have been assigned to the form genus Rhizoctonia (Arditti, 1967a; Burgeff, 1959; Hadley, 1968; Smith, 1974; Warcup, 1975). The first three isolates were named Rhizoctonia repens, R. mucoroides and R. languinosa (Bernard, 1909a; Burgeff, 1959). Subsequent isolates were designated as Mycelium radicis followed by the orchid name (M.R. Thrixspermum arachinites, for example). For a while these fungi were even assumed to be a separate group and named Orcheomyces (Burgeff, 1909, 1911, 1932, 1934, 1936). However, this classification did not persist. Another classification error which was dispelled (Gallaud, 1904) assigned orchid mycorrhizal fungi to Fusarium. A strain isolated from Orchis mascula was named Corticium masculi (Sprau, 1937). It was subsequently transferred to the nomen nudum Sistotrema (Trechispora) brinkmannii, but may not even be an orchid endophyte (see Warcup, 1975). Species bearing clamp connections have also been isolated from orchids. These fungi include Corticium catonii, Corticium octosperum (Sistotrema brinkmannii), Marasmius coniatus var didymoplexis (Didymoplexis minor) and strains from Corallorhiza innata, Epipogum aphyllum and Gastrodia sesamoides. However, uncertainties exist regarding the Corticiaceae as orchid endophytes. Several experiments with rhizoctonias which belong to this group and a number of orchids have produced tentatively negative results (Burgeff, 1959; Warcup, 1975). Hymenomycetous mycelia without clamps have been isolated from Epipogum nutans, Galeola hydra (the fungus is Fomes), G. septentrionalis (the fungus is Armillaria mellea) and G. javanica (Burgeff, 1959; Warcup, 1975). A culture isolated from Calanthe discolor was identified as belonging to the Hymenomycetes by the characters of its mycelium (Tokunaga and Nakagawa, 1974). Other basiomycetes isolated from orchids include Marasmius coniatus and Xerotus javanicus (Smith, 1974). Three different strains of Ascomycete rhizoctonias have been isolated from Pterostylis species in Australia. However, there are ". . . few data on whether [they] are orchid symbionts . . ." (Warcup, 1975). Recently the perfect states of several orchid fungi have been established (Table 16). In Hokkaido, Japan, 54 fungi were isolated from 20 orchid species. They were Ceratobasidium cornigerum, Rhizoctonia repens, R . solani (bi- and multinucleate) and other rhizoctonia. The first two were the most common (Nishikawa and Ui, 1976). (b) Physiology and metabolism. Appropriate carbon sources for orchid fungi are sugars such as sucrose, glucose, fructose, maltose, mannose,

ASPECTS OF ORCHID PHYSIOLOGY

49 1

galactose, xylose and arabinose. Glycosides may be split and their sugars utilized. In addition, starch, cellulose, wood, lignin and pectins can serve as carbon sources (Burgeff, 1936, 1959; Hadley, 1969; Hadley and PCrombelon, 1963; Harvais and Raitsakas, 1975; PCrombelon and Hadley, 1965; Smith, 1966, 1974; Warcup, 1975). Fungi which can utilize pectin as a carbon source have been shown to produce endopolymethylgalacturonase, protopectinase and endopolygalacturonase (Hadley and Ptrombelon, 1963; PCrombelon and Hadley, 1965). Other orchid fungi produce cellulase (Smith, 1974). Therefore, it would not be surprising if orchid fungi were found to produce a variety of other hydrolytic enzymes. These enzymes would enable the fungi to break down macromolecules found in the soil debris on which they live. Indeed, some orchid endophytes are soil saprophytes, others may be parasitic on a variety of hosts. Examples are Armillaria mellea, Ceratobasidium cornigerum (Rhizoctonia goodyerae repentis) and Thanatephorus cucumeris (Rhizoctonia solani). Orchid endophytes require all the usual minerals and since they can take them up from culture media there is every reason to believe that the same is true in the soil. The addition of microelements (solution of Harrison and Arditti, 1970) may depress growth slightly (Warcup, 1975). This may be a concentration effect especially since other media components usually contain some microelements as impurities. Both ammonium and nitrate are satisfactory nitrogen sources for some orchid fungi (Burgeff, 1936; Smith, 1974). Others do not grow well on ammonium as the sole nitrogen source (Hadley, 1977). Fungi of tropical holosaprophytes grow better on organic nitrogen sources such as urea, amino acids, peptones, proteins and nucleic acids. Suggestions that some orchid endophytes can fix or utilize atmospheric nitrogen have been disproved (Burgeff, 1936; Stephen and Fung, 1971). Some orchid fungi can utilize or require an exogenous source of amino acids. Two Rhizoctonia strains from Arundina chinensis grow best on glutamic acid as a nitrogen source. Other suitable sources are arginine and asparagine. Proline and methionine are unsuitable (Stephen and Fung, 1971). Asparagine, glycine and urea were good nitrogen sources for four isolates of Tulasnella calospora whereas glutamine, arginine and alanine were less suitable (Hadley, 1977). Similarly, Rhizoctonia repens M32, an isolate from Orchis militaris requires one of four amino acids for growth in vitro on a defined medium (Table 17). These reports (Hadley, 1977; Powell and Arditti, 1975; Stephen and Fung, 1971) indicate that at least some orchid fungi may have specific requirements for certain amino acids. Hence it is possible that the lack of growth on nitrate or ammonia is due to the absence of a specific requirement rather than complete inability of the fungi to utilize these ions. Aspartic and glutamic acid and glutamine have been extracted from fungi symbiotic with orchids, grown in vitro. However, ninhydrin positive substances could not be detected

TABLE 16 Perfect States of Several Orchid Fungi (Smith, 1974; Warcup, 1975; Warcup and Talbot, 1962,1963,1965,1967,1971)

Perfect state Ceratobasidaceae Ceratobasidium cornigerum

Other names or imperfect state Rhizoctonia goodyerae repentis

C. obscurum C. sphaerosporum

Orchid sources Goodyera repens Pterostylis nana, P. nutans, P. pendiculata Prasophyllum fusco-viride, P. nigricans Acianthus reniformis Pomatocaba macphersonii Robiquetia wesselli

C. sp. 0507 C. sp. El2 C. sp. 0615 C. sp. 0638

Not yet shown to be symbiotic with orchid seeds

C. sp. indet.

Dactylorchis purpurella Coeloglossum viride, Listera ovata Pterostylis mutica

c. sp.

Oliveonia pauxilla

Corticiumpauxillum Heteromyces

Thanatephorus cucumeris

Rhizoctonia solani Corticium solani

T. orchidicola T. sterigmaticus

Remarks

Corticium sterigmaticum Ceratobasidium sterigmaticum

Dactylorchis purpurella (Orchispurpurella) Orchis mascula Coeloglossum viride Thelymitra antennfera

Symbiotic with orchid seeds but not yet found in plants Not yet shown to be symbiotic with orchid seeds

ntelymitra grandiflora

T. sp. indet. Tremellaceae Sebacina vermifera

Acianthus reniformis Caladenia carnea, C. dilatata, C. latifolia, C. leptochila, C. reticulata, Glossodia major Microtis unifiora Corybas dilatatus

Tulasnellales Tulasnella allantospora T. asymmetrica T. calospora

Gleotulasnella calospora Rhizoctonia repens

T. cruciata T. viola T. sp. 0632 T. sp. 257

Uncertain Corticium catonii Thanatephorus cucumeris or Ceratobasidium

Thelymitra luteocilium, T. aristata, T. epipactoides, T. grandiipora, T. pauciipora, Dendrobium tetragonum Dactylorchis purpurella, Diuris longifolia, D. maculata, Acianthus exsertus, Caladenia reticulata, Thelymitra antennifera, Cymbidium canaliculatum, Dendrobium sp. Thelymitrafwco-lutea, T. pauciflora Acianthus caudatus Thelymitra aristata

Symbiotic with orchid seeds but not yet found in plants Corticium solani

Dactylorchis purpurella

494

3. ARDITTI

TABLE 17 Growth of Rhizoctonia repens M32 on Selected Amino Acids (Powell and Arditti, 1975) Dry weight, Amino acid na mM mg at 25 days DW m-' (DW mM-3 n-l Aspartic acid 10-2 152 2.6 4 14.9 2.6 GIycine 5.1 158 2 31.0 Serine 5.9 2.0 3 31.0 183 1 *9 Glutamic acid 5 14.9 9.5 141 Glycine 2 61.2 2.4 1 *2 144 5 61.2 0.5 Glutamic acid 2.4 149 Arginine 0.5 2.8 26 6 9.2 0.2 Leucine 6 22.6 31 1 a4 a

Number of carbon atoms per molecule of amino acid.

in the culture medium (Harvais and Raitsakas, 1975) indicating no leakage from the fungus. Several orchid endophytes require an exogenous supply of vitamins (Smith, 1974). Thiamine and p-aminobenzoic acid (PABA) are required for geographically distinct isolates of Tulasnella calospora (Hadley, 1977) and Rhizoctonia endophytes of Arundina graminifolia (Stephen and Fung, 1971). On the other hand, isolates of Ceratobasidium cornigerum and Thanatephorus cucumeris (Rhizoctonia solani) were self-sufficient for vitamins (Hadley, 1977). Therefore, at least for the present it is not possible to establish patterns for vitamin requirements by mycorrhizal fungi of orchids. Folk acid is the actual requirement of the Cymbidium symbiont, but PABA, a component of the vitamin, enhances growth equally well. This may be taken to indicate that this fungus cannot synthesize PABA but can incorporate it into the vitamin. The same may be true for the four Tulasnella culospora isolates mentioned above (Hadley, 1977). The thiamine requirement of the isolate from Cymbidium can be satisfied by the thiazole moiety of the vitamin. This means that the fungus can produce only one half of the vitamin molecule but is capable of combining the two components. Interestingly, and appropriately, both PABA and thiazole are produced by Cymbidium protocorms (Hijner and Arditti, 1973). Cymbidium seeds or seedlings require or at least benefit from thiamine or its pyridine component which is produced by the endophyte (Hijner and Arditti, 1973). This is coevolution on the half molecule level. Each partner in a symbiotic relationship produces the vitamin component needed by the other. As a consequence both can produce the vitamin thereby satisfying their own needs for the entire molecule and the requirements of the other partner. Whether the same may be true for Arundina chinensis and its endophyte is not clear. However, this could prove to be the case since endophytes of Cymbidium (Hijner and Arditti, 1973) and Dactylorhizapurpurella (Harvais

495

ASPECTS OF ORCHID PHYSIOLOGY

and Pekkala, 1975) have been shown to produce niacin and thiamine and release them into the medium (Harvais and Pekkala, 1975; Hijner and Arditti, 1973). Both vitamins enhance the growth of orchid seedlings (Arditti and Harrison, 1977). Therefore “. . . the symbiotized orchid, in agreement with Harley’s view (1 969), would obviously derive greater benefit from a metabolically active endophyte upon its digestion.” (Harvais and Pekkala, 1975). And, indeed, at least in vitro, orchid phytoalexins inhibit but do not kill the fungi. The fungi in turn appear capable of metabolizing the phytoalexins (Fisch et al., 1973a). As a consequence the fungi remain alive, for at least a certain period, and benefit the orchid by releasing vitamins and possibly other compounds. TABLE 18 Interrelationships of Orchid Mycorrhizae

Partner Tree Fungus

Orchid

Debris

Effect May be damaged by fungus and/or orchids which become parasitic Benefits from tree Benefits from and contributes to orchid Benefits from debris Benefits indirectly from tree parasitized by fungus Benefits from and contributes to fungus Derives indirect nourishment from debris Broken down and/or utilized by orchid endophyte

Condition Epiphytosis Parasite Mutualistic partner Saprophyte Epiparasite Mutualistic partner Parasite on saprophytic fungus Fungus is saprophytic

Orchid endophytes can grow as soil saprophytes or as parasites on other plants (Smith, 1974). Some may even be parasitic on the supporting trees of epiphytic orchids (Johanson, 1974, 1977) and bring about a syndrome called epiphytosis (Ruinen, 1953). Structural connections may be established between the host plant and epiphyte (Furman, 1959). And, depending on its state of health, the host may or may not be parasitized as a result (Johansson, 1977). Examples of epiphytosis damage are the killing of citrus and coffee tree branches by Zonopsis utricularioides and Leochilus labiatus (Cook, 1926) and other epiphytes (Went, 1940). Thus it appears that the orchid endophytes are both beneficial and detrimental symbionts at the same time. As a result the tripartite relationship between host tree, orchid and fungus has several ramifications (Table 18).

496

J. ARDITTI

3. Orchid-Fungus Specijicity Much has been written about the specificity of orchid mycorrhiza (Arditti, 1967a; Burgeff, 1936, 1959; Harley, 1969; Smith, 1974; Warcup, 1975). The initial belief was that specificity was high (Bernard, 1909; Burgeff, 1909, 1936). However, later work suggested that such specificity does not exist TABLE 19 Symbiotic Association of Seed of Fourteen Orchids and Strains of Seventeen Species of Endophytic Fungi (Warcup, 1975)

Fungus

Tulasnella calospora T. calospora T. calospora T. asymmetrica T. asymmetrica T. cruciata T. violea T. sp. T. allantospora T. sp. Thanatephorus cucumeris Th. sterigmaticus Th. sp. Ceratobasidium cornigerum Ceratobasidium cornigerum C. sphaerosporum c. sp. c . sp.

c. sp.

C. obscurum OIiveoniapauxilla

Isolate no. 062 0584 0689 0497 0591 0471 0353 0632 0579 257

+

+

S S 0 S S S S S S S + S S + S S S S S S + O S S S S S S

S S S

+

S

S S

S S

0 s S O S f S O O S S O S + + S + S O O + + + S O S + O + + S O O O S O

S

o s + o + + + o o o - -

0 O O

+ o

+

0 0

S O

O S + + S + S O O S p ' S - S 0 0 s 0 0 0 0 0 0 0 0 0 0 0 0 O S - b 0 0 0 0 0 0 0 0 0 + 0

T35 0708 0426

O

0167

0 0

O

0 0 0 0

O

O O O O S 0 0 0 - S

0

0 0 0 0 0 0

0

O O O + S S

+

S

+

S

S

S

0

+

O

0

S 0

S

S S S

S

- S

-

+ +

A D 1 4 O O O O O O + S S S S + 0657 0 0 0 0 0 0 0 S S S S S 0507 O O + O O O O + S S S S 0615 0 0 0 O O O O + - S - S + El3 0 0 0 0 0 0 0 + + + 08 0 0 0 o o o o o o + p + p o T330 0 - 0 0 0 0 0 0 0 - 0 0

S

+ S 0

+

+ S p S

+ + + + 0 0 + 0

Pathogenic to somela11 seed. No test. Explanation of symbols: greater germination than inoculated seed, but no shoot differentiation; 0, no germination beyond that of inoculated seed on the medium; S, formation of a shoot. a

+,

ASPECTS OF ORCHID PHYSIOLOGY

497

(Curtis, 1937; Knudson, 1929). More recent work points to different levels of specificity (Smith, 1974). Orchids and fungi may “. . . differ markedly in the range of partners with which they form effective symbiosis . . .” (Table 19), (Warcup, 1975). In Japan there was “. . . no relation among species of orchids, rhizoctonias and localities. In the inoculation of Orchis aristata Fish, with 35 different isolates of rhizoctonias, the symbiotic infection of the root was observed except R. solani and multinucleate Rhizoctonia which caused necrosis of root cells. C. cornigerum, R. solonis and binucleate Rhizoctonia tested caused the damping off of spinach, radish and cucumber seedlings.” (Nishikawa and Ui, 1976). Some groups of orchids may be associated with one or a limited number of fungi. For example the allied genera Caladenia, Glossodia, Elythranthera and Eriochilus (all from Southern Australia) are associated with Sebacina vermifera. On the other hand Diuris is associated with Tulasnella calospora (Warcup, 1971). Pterostylis species were stimulated to germinate only by Ceratobasidium cornigerum and Diuris taxa are limited to Tulasnella calospora (Rhizoctonia repens). Thelymitra seeds germinate well with several species of Tulasnella, but poorly when infected with Ceratobasidium cornigerum (Warcup, 1973). A number of Canadian species responded differently to several fungi, but “only one good symbiotic association was established. It was between Goodyera oblongifolia from British Columbia and Rs 10, a rice pathogen from Malaysia” (Table 20; Harvais, 1974). This may be interpreted to suggest lack of specificity. There was also no specificity between 15 Rhizoctonia isolates and a number of orchids (Tokunaga and Nakagawa, 1974). Symbiosis tests between orchids from different areas and 32 Rhizoctonia isolates “. . . showed no evidence of any species to species relationships between orchid and fungus.” (Table 21 ; Hadley, 1970). Altogether, the available information does not support the concept of strict specificity even if it does not exclude the possibility of preferential association. Several factors may control interaction between orchids and fungi. Seeds or protocorms may remain uninfected or the host could contain and/or eliminate the fungus (Fig. 51). Another possibility is that the fungus may parasitize and eventually kill the protocorm during one of several stages (P, sP, SP and SSP in Fig. 51). Successful symbiosis occurs when infection is compatible (s, S, SS in Fig. 51; Hadley, 1970). This series of alternatives is logical and to some extent supported by the available evidence. Unfortunately, however, the basic factors which control it are not entirely clear.

4. Structure and Ultrastructure Living epidermal hair cells (Williamson and Hadley, 1970) and basal cells of seeds (Burgeff, 1959) are the sites of infection, the fungus penetrating by single hyphae which may also grow out of the cells (Fig. 52). Following in-

TABLE 20 Interactions Between Rhizoctonia Isolates and the Seeds of Nine Orchid Species After 7-10 Months in Dixenic Culture on a Mineral Dextrose Medium, at 25°C (Harvais, 1974) Rice pat hogen

Orchid endophytes Orchid species

E,

E,

PB47

Thrl

T

RslO

Calypso bulbosa Corallorhiza maculata Cypripedium reginae Epidendrum obrienii Goodyera oblongifolia G. repens var. ophioides Habenaria dilatata H. obtusata H. psycodes

nt nt

nt nt

0 O(P) O(P) P OSP 0 P P 0

P P P P (0)P P P P P

P P O(P) P OSP P P,(OSP) P 0

P P P P S(P)SS) P PsW P P

?

Pellicularia chordulata

~.

P

0

nt

nt

(0)sP OP,OsP

(0)sP 0

nt

nt

(O)(S)P nt

nt

P

0 0

0 0 0 0 O(P) (OP 0

NOTE:Key to the symbols used: 0-no infection, seeds or protocorms healthy; P-parasitism with no signs of control by the host; s-controllec infection, pelotons formed, but no growth stimulation; S - c f . s but with growth stimulation, symbiosis; nt-not tested. Combined symbols indicatr that one condition follows the other, e.g. OSP means no infection of seeds or young protocorms, followed by controlled infection with growth stirnula. tion, followed by parasitism. Symbols in parentheses indicate a small proportion of the population in those categories.

TABLE 21 Summary of the Interactions Betweeri Protocorms of I0 Orchids and 32 Rhizoctonia Isolates (Hadley, 1970)

Fungi Ceratobasidium cornigerum C. cornigerum C. cornigerum C. cornigerum C. obscurum C. sp. indet C. sp. indet Thanatephorus orchidicola T. orchidicola T. sterigmaticus T. cucumeris T. cucumeris T. cucumeris T. cucumeris T. cucumeris T. cucumeris T. cucumeris T. cucumeris T. cucumeris

4' Rgr Thr1 0393 0479 08 F1 cs4 De 1 s2 060 0269 Rs 1 Rs6 Rsl2A Rs91 W16 W48 W82 W87

0 P (S) S

S

ss

SSP

SP

ss S P

0 S

S

sss

ss

SSP

S

SP

SP P

P 0

SP

ss sss

S*

SSP

0 0

SP

S

S*

P S*

S P

SP SSP P

sss sss P

S S

P

S*

P

S(S)

ss ss S*

S

S

S(S)

S

P

0 0

S*

S

0 0

P

ss S 0

4'

0 P S

0

0

P 0

P

P 0

S

-

-

S

P

S

S P

-

-

0 0

0 0

0 0

-

0

0 0 0

P

-

S*

-

P

0

S*

4'

-

0

ss

-

S

sss S

S

0

S

0

-

P 0

0 0 continued

TABLE 21-continued

Tulasnella calospora

Am04

0

T. calospora T. calospora T. calospora T. cruciata Rhizoctonia solani R. solani R. solani R. solani R. sp. R. sp. R. sp. R. sp.

Pb47

S

RrA 0388 0296

Rs51

S Rs16 Rs94 Rs102 SPl T RslO

S S(S) 0 S* S* S*

0

ss S* S* S*

ss ss ss ss

S

S

S

ss ss ss

SP

0

S

S SP

S

S

S S*

SSP SSP SSP

SSP

S*

0 0

ss sss

0 S 0

ss

S* S

SSP

S*

P

S

S

sss ss sss

-

S

0

0

-

-

S -

-

-

0 0

P

P

-

S S

P -

ss ss ss

sss sss sss ss

-

S S

0 0 0

(S)

-

-

-

-

S S S

0

SP

ss S

S

ss ss S S

0 0

0

(PI -

-

S*

0, No infection, protocorns healthy. s, Compatible infection but no growth stimulus seen. If symbiosis does not develop, this condition is analogous to S*. s*, Infection followed by death of hyphae, sometimes seen as hypersensitivity reaction; no growth stimulus. S, Growth stimulus; symbiosis.

P, Parasitism without any evidence of a compatible phase. sP,Parasitism following compatible phase, usually resulting in death of protocorms. SP, SSP, Parasitism after a symbiotic phase (“breakaway parasitism”). -, Inconclusive result; protocorms often moribund or dead due to dense growth of fungus but no infection seen. Symbols in parentheses indicate that a small proportion of the population was in this category.

50 1

ASPECTS OF ORCHID PHYSIOLOGY

Germinated seed (protocorm) + fungus No infection

of protocorm

S*

Death of protocorms (non-parasitic), i.e.

phase

infection otherwise contained and eliminated /

4

-1;;;;in;death

Marked growth stimulus, i.e. symbiosis c

/

---

SP Breakaway parasitism of host

1.-

Total elimination of fungus possible as old protocorm tissue becomes moribund

Normal autotroph ic phase of development to mature adult plant

As for SP (above)

i

Infection pattern may be repeated in roots? Fig. 51. Possible pathways in the development of the orchid-fungus interaction (Hadley, 1970).

fection the fungus forms hyphal clusters called pelotons (Fig. 53a; Hadley, 1975). Young intracellular hyphae are circular or elliptical in section and may stain densely (Fig. 53b). Older hyphae may be empty (Hadley et al., 1971). The hyphae are thinly enveloped by the host cytoplasm and are subsequently surrounded by an encasement layer (Fig. 53c; Hadley, 1975). In other words the hyphae are enveloped with invaginated plasmalemma which continues to produce pectin and cellulose (Nieuwdorp, 1972). The result in Dactylorhiza purpurella is a combination of host cytoplasm and organelles, host plasmalemma, encasement material and the fungal hypha (Fig. 53d). The plasmalemma of the host is usually in contact with the encasement layer (Hadley, 1975). A similar arrangement has been reported for Corallorhiza trifida (Nieuwdorp, 1972) and Dactylorhiza maculata (Strullu and Gourret, 1974). In Dactylorhiza purpurella well developed pelotons are densely packed despite the existence of spaces between adjacent hyphae (Hadley, 1975). A double cell wall surrounds the fungal cells in Ophrys insectgera (von Hofsten, 1973). During later stages the hyphae start to degenerate, collapse (Fig. 53e), become disorganized and are digested by the host. The digestion of the hyphae is accompanied by a high oxygen uptake (Blakeman e? a[., 1976).

Fig. 52. (a, b) Infection of epidermal hairs of Dactylorhiza purpurella by the fungus Thanatephorus cucumeris. (c, d) Emergence of fungal hyphae from the tip of an infected epidermal hair. (a, b) x 750; (c, d) x 1500 (Williamson and Hadley, 1970).

503

ASPECTS OF ORCHID PHYSIOLOGY

At this stage the encasement layer may thicken, become more granular and assume a reticulate appearance (Fig. 53f; Hadley, 1975). When degeneration of hyphae is complete or nearly so they aggregate into clumped masses of digested material (Fig. 53f; Borriss et al., 1971; Dorr and Kollman, 1969; Hadley, 1975). A release of acid phosphatase by both host and fungal cytoplasm may be correlated with the digestion of these hyphae (Williamson, 1971). Esterases, on the other hand, may be associated with fungal growth and differentiation (Williamson, 1973). 5 . Physiology The relation between orchids and their fungi is nutritive conjunctive symbiosis (Arditti and Ernst, 1974). It has been described as ". . balanced symbiosis . . . [but] . . . part of the time the plant acts as a parasite on the fungus." (Mejstrik, 1970). Infection of protocorms induces DNA synthesis (Williamson, 1970) and increases activities of polyphenol oxidase, ascorbic acid oxidase, peroxidase and catalase as well as higher oxygen uptake and respiration (Blakeman et al., 1976). Starch disappears from infected regions of seedlings (Burgeff, 1959) due to enhanced hydrolysis (Hadley and Williamson, 1971; Arditti, 1967a, 1972a, b, c; Arditti and Ernst, 1974; Burgeff, 1959; Hadley, 1969; Purves and Hadley, 1975; Smith, 1974). These events simply reflect the fact that growth and development are initiated by infection. Without infection there is no development in the absence of soluble sugars (for reviews see Arditti, 1967a; Withner, 1959, 1974). Protocorms can hydrolyse some but not all glycosidic bonds (Ernst et al., 1971a). Consequently, in nature the seeds (and mature plants of saprophytic species) obviously depend on mycorrhizal fungi for carbohydrates. This has been demonstrated experimentally (Smith, 1966, 1967; Arditti and Ernst, 1974; Burgeff, 1959; Purves and Hadley, 1975; Smith et al., 1969; Smith, 1974). The fungi hydrolyse polysaccharides and take up the resulting monosaccharides. In the fungus these sugars are converted into trehalose, mannitol, fructose, glucose and ribose (Downie, 1949; Smith, 1967). On being translocated into the orchid the components or trehalose are incorporated into sucrose (Smith, 1967). Movement of substances from orchids to fungi was suggested on the basis of the observation that hyphae grow from infected plantlets into soil (Burgeff, 1959) or carbohydrate-free medium (Smith, 1967; Purves and Hadley, 1975). Initial attempts to test the hypothesis showed no transport from orchid to fungus when green seedlings of Dactylorhiza purpurella were supplied with 14C0, (Hadley, 1969; Purves and Hadley, 1975; Smith el al., 1969). I n subsequent experiments label from 14C0, showed up in the fungal symbiont of Dactylorhiza purpurella. However, these experiments were inconclusive because the uninfected seedlings leaked substances into the medium. This raised the possibility that the label in the fungus may be due to uptake of

.

Fig. 53

ASPECTS OF ORCHID PHYSIOLOGY

505

these nutrients. Further, there was also a possibility that the label in the fungus could have originated from dark fixation of 14C0, (Purves and Hadley, 1975). Experiments carried out to resolve these problems (Hadley and Purves, 1974; Purves and Hadley, 1975, 1976) showed: (i) very little 14C movement to rhizomes and none to the fungus indicating that the fungus does not act as a metabolic sink; (ii) limited movement of 14C from living rhizomes to fungi, but considerable transport from killed protocorms suggesting a barrier; (iii) dark fixation by the fungus pointing to the likelihood that the label originated from this source. The obvious conclusion from these experiments is that there is no transport from orchids to their fungi. Several observations that infected seedlings have a higher nitrogen content led to the suggestion that orchid fungi can fix nitrogen (Wolff, 1927, 1933) and presumably transport it into the orchid. These suggestions were not in agreement with more careful work (Beijerinck, 1907; Burgeff, 1909, 1936; Cortesi, 1912; Hollander, 1932; Huber, 1921a, b) and the idea was abandoned (Harley, 1969; Stephen and Fung, 1971a). The higher nitrogen content of infected seedlings is probably due to fungal transport of nitrogenous substances into the orchid or digestion of the endophyte by the plant (Arditti and Ernst, 1974; Smith, 1974). Peptide and polypeptide digesting enzymes have been detected in cells which digest fungal cells (Burges, 1936; Fuchs and Ziegenspeck, 1924). Another possibility is that the infection enhances seedling metabolism and consequently nitrogen assimilation. In so far as is presently known orchid seedlings do not require specific nitrogenous substances or amino acids. Some of their fungi do (Powell and Arditti, 1975; Harvais and Raitsakas, 1975; Stephen and Fung, 1971a). Therefore it is possible that the fungi obtain these requirements at least in part from the orchids (Arditti and Ernst, 1974). Orchid seedlings do not seem to have absolute requirements for vitamins but may benefit from them (Arditti, 1967a; Arditti and Harrison, 1977; Withner, 1959, 1974). Several endophytes benefit from or require certain vitamins and/or produce others (Harvais and Pekkala, 1975; Hijner and Arditti, 1973; Stephen and Fung, 1971b). Hence, it is possible that exchanges of vitamins between orchids and their fungi do take place (Arditti and Ernst, 1974; Harley, 1969; Purves and Hadley, 1975; Smith, 1974). Germinating orchid seeds and developing seedlings appear to be largely autotrophic with respect to most (or at least non-gaseous)hormones. With one Fig. 53. Ultrastructure of Dactylorhim purpurella mycorrhiza (courtesy Dr G. Hadley): (a) cluster of hyphae (peloton), teased out from orchid cell (x 300); (b) thin section of part of a peloton with some host cell cytoplasm ( x 1500); (c) section of single hypha showing granular encasement layer, enveloped by host cytoplasni (x 20 OOO); (d) section of hypha enclosed in host cytoplasm, with host amyloplast, mitochondrion and (part of) nucleus (X 12 OOO); (e) collapsed hypha with a thick granular encasement layer (X 25 OOO); (f) partly digested hypha with reticulate encasement, being drawn into digested hyphal clump (X

25000).

506

I. ARDITTI

exception (Downie, 1943), the same seems to be true for endophytes. Traces of auxin have been found only in Cypripedium, but none was detected in Calanthe and Dendrobium (Poddubnaya-Arnoldi, 1960; PoddubnayaArnoldi and Zinger, 1961). Thus it seems that interchange of most hormones between orchids and their fungi is not an important aspect of the symbiosis if it occurs at all (Arditti and Ernst, 1974). The situation with ethylene may be different. Ceratobasidium cornigerum (isolated from Pterostylis vittata) and Tulasnella calospora (from Thelymitra aristata and Caladenia reticulata), all fungi which enhance orchid seed germination, have recently been reported to produce ethylene (Hanke and Dollwet, 1976). This suggests the interesting, but as yet untested, possibility that the germination of at least some orchid seeds is enhanced by the gas. It is possible even that ethylene is required for germination as it is by other, non-orchidaceous seeds (Hanke and Dollwet, 1976). The inability of fungal extracts to enhance germination of north temperate species tends to argue in favour of a gaseous substance (like the hormone ethylene) which may be provided by a living fungus. Additional support for this idea is provided by the general failure of efforts to isolate specific factors which can promote orchid seed germination. The extraction procedures employed in these attempts were not suitable for the trapping of ethylene. Therefore, it is reasonable to assume that, if present, the gas probably diffused out and was not incorporated (at least in sufficient amounts) in the seed germination media. As this is being written, we are initiating experiments on the effects of ethrel on the germination of several orchid species. Since both orchids and their fungi can grow axenically on mineral-containing media it is obvious that they can take up inorganic ions. However, infection may provide an advantage in this respect since 32Pis translocated into Dactylorhiza purpurella by Rhizoctonia repens (Smith, 1966; Arditti and Ernst, 1974). G. SUMMATION

In as much as orchids and their fungi were compared to Hamlet and the Prince of Denmark (Ramsbottom, 1922b), it may be appropriate to start this summation by suggesting (with apologies for the teleological and anthropomorphic implications) that orchids are victims of two tragedies : a. They have fatty seeds which lack the appropriate metabolic machinery (Harrison, 1973). b. Their seeds do not have an endosperm and are incapable of directly utilizing available substrates whereas those substances orchids can use are not usually available in nature. Therefore, orchid seeds germinate and seedlings develop only following fungal infection. As a consequence orchids have evolved a close and intricate relationship with their fungi and obtain from them carbohydrates and other substances.

ASPECTS OF ORCHID PHYSIOLOGY

507

Epiphytic orchids (especially those of tropical regions) have the simplest requirements-only sugars. And, for the most part, that is what the fungi provide. As a result, these orchids germinate very easily on media which contain no more than simple sugars and minerals. Examples of this category include, but are not limited to, Cattleya, Onciditrm,Phalaenopsis, Dendrobium, Stanhopea, Coelogyne. Orchids of tropical origin which are primarily terrestrial have more exacting requirements. In the absence of their symbionts, Paphiopedilum seeds, for example, germinate better on low calcium media which contain fructose as a carbon source (R. Ernst, personal communication). Australian terrestrial species, even if not all are of strictly tropical origin, also seem to have relatively specific requirements and are not easy to germinate asymbiotically (McIntyre et al., 1971, 1972a, b ; Veitch and McIntyre, 1972; Wrigley, 1973, 1976). Terrestrial orchids of north temperate origin (including those from Europe and North America) have the most exacting and specific requirements. Many of them germinate not at all or very poorly without infection, but the reasons for this are not understood at present. Orchid seeds cannot utilize their own fatty reserves, or do so very slowly. Neither can they hydrolyse large molecules like starch or cellulose. As a result asymbiotic germination in the absence of sugar proceeds only to the early protocorm stage. Then the seedlings come to a “resting” stage during which they survive by utilizing their reserves very slowly, “waiting” as it were for a source of simple sugars and other requirements. When these are provided (by fungal infection in nature and appropriate media in the laboratory) development continues. Some substances can be taken up directly by (at least axenically-grown) orchid seeds and seedlings in vitro. Others are transported into the seedlings by fungi in symbiotic cultures or under natural conditions following digestion of large molecules. Transport into seedlings by fungi may occur across living membranes in some cases, but digestion of the pelotons could be an even more important means by which the orchid obtains nutrients (Burgeff, 1936, 1959; Hadley, 1975; Purves and Hadley, 1975; Smith, 1974). Digestion of the hyphae is almost complete (only the wall remains) and this undoubtedly releases all of the contents into the orchid. Such transfer cannot be called translocation and favours only the orchid; it has been called “non-reciprocal” (Smith, 1974). Consequently orchids can be viewed as necrotrophic parasites on their fungus (Lewis, 1973). The available evidence supports this view, but also provides indications that some aspects of the relationship may be commensal or mutualistic (Arditti and Ernst, 1974). Orchid fungi have the potential of becoming parasitic and sometimes do. However, in cases of successful symbiosis they are kept under control, without being destroyed. Phytoalexins (Section 111) undoubtedly play an important role in this respect.

508

J. ARDITTI

The evolutionary origins of orchid mycorrhiza are lost in antiquity and are now a matter of speculation (Ames, 1948). One possibility is that having lost the ability to utilize their reserves orchids survived only by developing a symbiotic relationship with fungi. An argument against this suggestion is the small chance that the symbiotic relationship would have developed fortuitously and soon enough after loss of the metabolic machinery. Without the development of such a relationship, almost immediately (in an evolutionary time scale) the orchids could not have survived despite the considerable longevity of individual plants. A second possibility is that orchid seeds developed a symbiotic relationship with fungi before losing the ability to utilize their reserves (i.e., produce carbohydrates from stored fat) and store carbohydrates. The loss of these capabilities whether they occurred simultaneously or not, would have little or no deleterious effect on plants which already depend on or at least benefit from a symbiotic relationship. The coevolution of orchids and their endophytes (like the relationships between them and their pollinators) is clearly very successful. If it weren’t orchids would not have evolved 20 000-30 000 species in all corners of the globe and in most ecological habitats. 111. PHYTOALEXINS Compounds which ward off pathogens and are produced by plants following infection are called phytoalexins. The term and the theory behind it were proposed by K. 0. A. Muller around 1940 (Miiller, 1966). Research in the area gathered momentum slowly and reached the present levels of activity by 1960. However, phytoalexins were discovered by Noel Bernard as a result of his work on orchids approximately 35 years before Muller named them (Bernard, 1909b, 1911; for reviews see Arditti, 1968, 1975; Burges, 1939; Magrou, 1924, 1936, 1938; NobCcourt, 1923, 1938; Nuesch, 1963). A. HISTORY

In his work with orchid mycorrhiza Noel Bernard noted that following fungal infection of Orchis morio roots by Rhizoctonia repens the bulbs of these orchids appeared resistant to further fungal infections (Bernard, 191I). He later proceeded to elucidate the nature of this phenomenon: “Cesplantes ofrent au point de vue de I’immunite‘ un probl2me assez particulier . . . j e me proposais de chercher la cause, a fin de voir si elle n’avait pas une origine humorale.” To test his theory of immunity he placed bulb tissue from infected plants on agar and introduced fungi. The fungal hyphae grew in the direction of the bulb tissue, but stopped one or two centimetres short of reaching it (Fig. 54). Bernard suggested that the inhibition of fungal growth was due to a diffusible substance produced by the bulb tissue. He then tested the fungicidal effect of the substance by varying the size of bulb sections in

I.

- Culture sur

gglose du Rhizoctonia repens de

1 ' 0 r c h i s Xorio, en presence d'un fragment de bulbe de

11.

-

Culture s u r gelose du champignon de l ' 0 r c h i s !.lascula

en presence du bulbe de l a 6eme espzce.

-- L e

fragment d e

A, vue d e p r o f i l ; B, vue d e F, fragment de bulbe; l a semis i n i t i a l est

bulbe est l e r e c t a n g l e ' i l a base; l e s e m i s i n i t i a l est l e

marque par un point d'oii l ' i r r a d i e n t les filaments,

premisre legne d ' a r r e t est marquse par des r a m i f i c a t i o n s

Loroglossum hiricinum. face.

--

p o i n t d'o:

s ' i r r a d i e n t les filaments du champignon; l a

l e t r a i t o v a l a i r e marque l a l i m i t e d e l e u r dgvelop-

rEpCt6es; les f i l a m e n t s s6journant sur l e v e r r e produisent

an v o i t les f i l a m e n t s v i g i t a n t sur l e v e r r e d6passer l e fragment de Loroglossum sans

une invasion secondaire de l a g6lose; ils s o n t v i s i b i l e s p l u s bas s u r l e tube e t se terminent par une deuxizme

dependant l ' a t t e i n d r e .

ligne d'arret.

ment.

-- En A,

Fig. 54. The original figures and captions from Bernard's paper which first reported on the existence of phytoalexins.

510

J. ARDITTI

the experiments. From his data he concluded that orchid bulbs can produce a diffusible fungal inhibitor. After Bernard’s death the investigation of phytoalexins in orchids continued in France. His findingswereconfirmed (Magrou, 1924, 1936, 1938)and shown to occur in live bulbs only (NobCcourt, 1923, 1938). The phenomenon was compared by Magrou to the production of antibodies in animals: “Le phPnom2ne . . . est de tout point comparable ci la formation de anticorps . . . chez les animaux , . .”. After that interest in orchid phytoalexins remained dormant until the chemical defence reaction in Orchis militaris was described (Gaumann and Jaag, 1945). This was followed by the isolation of an orchid phytoalexin (before K. 0. Muller named them). Orchinol, the first characterized compound to which the term is properly applicable, was isolated from orchids in 1957 (Boller et a]., 1957; Gaumann, 1960, 1963-1964; Gaumann and Hohl, 1960; Gaumann and Kern, 1959a, b; Gaumann et al., 1950, 1960, 1961; Hardegger et al., 1963a, b, c; Nuesch, 1963; Urech et al., 1963). The pace of research on phytoalexins was relatively slow in the early 1960s, but during that period the first conscious isolation of a phytoalexin was carried out by Cruickshank and Perrin. A series of investigations of phytoalexins in orchids has been carried out in my laboratory (Arditti, 1968, 1975; Arditti et al., 1972, 1975; Fisch and Arditti, 1972; Fisch et al., 1972, 1973a, b). In the meantime, and independently, work on the synthesis of orchinol was undertaken by A. Stoessl in Canada (Stoessl et al., 1974). The Canadian group has since extended its investigations to structure/activity relationships and solubility phenomena (Ward et al., 1975a, b). Several structural problems are still being investigated jointly (A. Stoessl, G. L. Rock, M. H. Fisch and J. Arditti, unpublished). B. CHEMISTRY PRODUCTION AND DISTRIBUTION

Three phytoalexins have been isolated from orchids to date. All are trisubstituted dihydrophenanthrenes. Orchinol, isolated from Orchis militaris (Boller et al., 1957; Gaumann and Kern, 1959a, b), is 2,4-dimethoxy-7hydroxy-9,lO-dihydrophenanthrene(Fig. 55, I; Fisch et al., 1973a; Hardegger et al., 1963a, b; Letcher and Nhamo, 1973). Loroglossum hircinum is the source of the other two: loroglossol, 5-hydroxy-2,4-dimethoxy-9,lO-dihydrophenanthrene (Fig. 55, 11) and hircinol, 2,5-dihydroxy,4-methoxy-9,10dihydrophenanthrene (Fig. 55, 111; Fisch et al., 1973a; Hardegger, 1 9 6 3 ~ ; Letcher and Nhamo, 1973; Urech et al., 1963). Despite the similarity in names, loroglossin is neither a phytoalexin nor a dihydrophenanthrene, but a bis-p-hydroxybenzyl ester of a complex dicarboxylic acid (Gray et al., 1976). It was isolated from Gymnadenia conopea, Orchis maculata, 0 . incarnata and 0 . latifolia. Orchinol may be widespread in European terrestrial orchids although genera as well as species within a genus may differ in their ability to produce this phytoalexin (Table 22). For example, orchinol was produced by

ASPECTS OF ORCHID PHYSIOLOGY

51 1

Ip:R = p - o - Glucopyronosyl, R ' = H ; Loroglossin

Fig. 55. Orchid Phytoalexins. I. Orchinol ;11. Loroglossol ; 111. Hircinol; IV. Loroglossin.

all Serapias species tested but by none of the Ophrys. In the genus Loroglossum, L. hiricinum (L.) Rich. produced very little orchinol whereas L. longibracteatum synthesized large amounts (Table 22). In Orchis militaris incubated with Rhizoctonia repens orchinol production started 36 hours after initial contact and reached a peak after eight days (Table 23). At that time total concentration was 92Opgg-' tissue which (Nuesch, 1963). Orchinol corresponds to a level of at least 0.5 x 1 0 - 2 ~ concentration diminishes as the distance between the point of contact between fungus and orchid tissue increases (Table 23). Recently a number of phenanthrenes from a laboratory synthesis of orchinol were assayed for activity (Fig. 56, I-VII; Ward et al., 1975a). The results (Tables 24, 25) show that several phenanthrenes have high activity. Some of them are more active at lower levels which may be due to crystallization when concentrations are higher. These compounds caused stunting and distortion of the tubes formed by germinating fungal spores. The cytoplasm of affected tubes was usually withdrawn from the walls and highly granular when compared with controls (Ward et a/., 1975a). In fungus-infected Cymbidium, production of an as yet unidentified phytoalexin is accompanied by increased levels of sitosterol (SI), stigmasterol (ST) and campesterol (CA) in a 70 : 25 : 5 ratio (Arditti et al., 1975). In addition, ergosterol peroxide (EP) was found in these extracts (but see below). This raises the question of whether Cymbidium phytoalexins may be related structurally or biosynthetically to sterols. Ours seem to have been only the second isolation of sterols from orchids. The first isolation, from Cattleya and Arundina, was of SI, ST and CA, but in different ratios (Wan et al., 1971). Assays of SI, ST, CA and EP on TLC plates using Cladosporium cucumerinum indicated that the first three were not very active whereas EP exhibited considerable inhibition. In liquid cultures of Candida lipolytica all four were only slightly active, but EP was more inhibitory than the others. The EP isolated from fungus (Rhizoctonia repens M32)hfected Cymbidium pseudobulbs (Arditti et a/., 1972a) is most probably an artefact of extraction and a

512

J. ARDITTI

TABLE 22 Synthesis of Orchinol and of p-Hydroxybenzyl Alcohol in the Bulbs of Different Orchids Incubated with Rhizoctonia repens, a Strain From Orchis militaris L. (GBumann et al., 1960;Niiesch, 1963) Relative amounts of: Orchinol p-Hydroxybenzyl Alcohol

Host species Aceras anthropophora (L.) R. Br. Anacamptis pyramidalis (L.) Rich. Charmorchis alpina (L.) Rich. Coeloglossum viride (L.) Hart. Gymnadenia albida (L.) Hartm.a G. conopea (L.) R. Br. G.odoratissima (L.) Rich. Loroglossum hircinum (L.) Rich.b L. longibracteatum (Biv.) Morise Nigritella nigra (L.) Rchb.d Ophrys apifera Huds. 0. arachnites (Scop.) Murraye Orchis coriophora L. 0. latifolia L. 0. maculata L. 0. mascula L. 0. militaris L. 0. morio L. 0. sambucina L. 0. ustulata L. Platanthera bifolia (L.) Rich. Serapias lingua L. S. neglecta Not. S. vomeracea Burm.

++ ++ +++ +++ +++ ? + + +++

++ 0 0 ?

+ 0 +++ ++ +++ +++

0 0 ++f

++ ++

+ ++ ++ ++ ++ ? + + + + 0 0

++ ++ 0

++ ++ +++ ++ ?

0

++ ++ ++

Coeloglossum albidum Hartm. Himantoglossum hircinum Sprengel. C Barlia longibracteata Parlat = Aceras longibracteata Rchg. = Orchis longibracteata

a b

Biv.

Nigritella angustifolia Rich.

e Ophrysfucipora Crantz.

mixture consisting of 5a, 8a-endoperoxide and 9,l-dehydroergosterol (Fisch et al., 1973b). In addition the ergosterol itself is a fungal product. C. ACTION SPECTRUM AND ACTIVITY

Orchinol and hircinol are relatively nonspecific in their effects and can inhibit a number of fungi and bacteria (Tables 26, 27). As can be seen (Table 28) all fungi with the exception of Fusarium oxysporum are inhibited and the ED,, is 5 x lo-, M. Except for Monilinia fructicola growth is less sensitive than germination.

513

ASPECTS OF ORCHID PHYSIOLOGY

TABLE 23 The Concentration of Orchinol (pg g-l) in the Tissue-cylindersfrom the Bulbs of Orchis militaris Incubated with a Rhizoctonia repens Strain from 0. militaris (Niiesch, 1963from Gaumann and Hohl, 1960)

Time of incubation (days) 1 2 5 8 12

2 mm thick discs, numbered from the bottom to the top 1

0 28 200 920 650

3 0 0 1s 160 460

2 0 10 112 380 540

4 0 0

5 0 0 0

Traces 50 110

--__

35 100

6-

0 0 0 45 80

OCOCH3

CH30

ls!

Y

CH30

m OCOCH3

HO

Fig. 56. Structural formulae of phenanthrenes which have been tested for antifungal activity: (I) orchinol acetate; (11) dehydroorchinol; (111) dehydroorchinol acetate; (IV) loroglossol acetate; (V) dehydroloroglossol; (VI) dehydroloroglossol acetate; (VII) didemethylloroglossol (Ward et al., 1975a).

When tested against powdery mildew on cucumber cotyledons orchinol was M (Ward et d., 1975a). not effective a t 1 x Orchinol at 50 or 100 ppm is considerably more active than hircinol against Cundida Zipolyticu BY 17. With orchinol, growth is almost completely inhibited during the first six days. Following this period, growth (i.e., turbidity of the liquid culture) increases slowly, but even after 24 days absorbance of

TABLE 24 Percentage Inhibition of Spore Germination of Monilinia fructicola by Orchinol and Related Phenanfhrenes and Dihydrophenanthrenes (Ward et al., 1975a) Concentration (M x lo4) 5

Orchinol Orchjnol acetate Dehydroorchinol Dehydroorchinol acetate Loroglossol Loroglossol acetate Dehydroloroglossol Dehydroloroglossol acetate Didemethylloroglossol

2.5

1.25

0.625

0.313

83'" 98'8 OC"

51 81'8 79c

OC

21'8 21'8 OC

15crs 14

0.156

0-078

0.039

15 20'" 69C

0 0 22'8

0 0 08

OC

OC

OC

OC

0s

08

13'6

108

OC

OC

0 0 10s

18'8 0

13s 0

0

NOTE: crystals and deposit; r rupture of germ-tube tips; * stunting and distortion.

8

0 0 0 0 0

TABLE 25 Percentage Inhibition of Zoospore Germinationof Phytophthora infestans by Orchinol and Related Phenanthrenes and Dihydrophenanthrene (Ward et al., 1975) Concentration (M x lo4)

Orchinol Orchinol acetate Dehydroorchinol Dehydroorchinol acetate Loroglossol Loroglossol acetate Dehydroloroglossol Dehydroloroglossol acetate Didemethylloroglossol

5

2.5

100' 100C' 100C'

100' 100cr 100C' 100C' 10 0 c '

100C' 90C 89c

1 -25 100' lOOCr

100C' l00C'

80C

93C'* 61

OC

OC

OC

58CS

59C8

100

66 218

0625

0.313

0.156

0.078

81

40 95' 100C'

21 57 100C'

2 13 83

0 0 0

0 0 0

20 20 0

16 16

100C' 100C'

1OOC' 92C's 12 5oc

70 0

78CS

208

0 59c 59c 0

NOTE: crystals or deposit; rupture of germ-tube tips; S stunting and distortion of germ tubes.

0

0-039

516

J. ARDITTI

TABLE 26 Effects of Orchinol on Several Soil Fungi (After Gaumann et al., 1960) Fungi

Inhibitory concentration of Orchinol, molar

Phycomycetes Mucor spinosus v. Tiegh. Pythium de Baryanum Hesse Rhizopus nigricans Ehrenb.

10-3 10-2.6 10-4

Ascomycetes Alternaria tenuis auct. Aspergillus clavatus Desm. Aspergillusflavus Link Aspergillusfumigatus Fres. Aspergillus niger v. Tiegh. Botrytis cinerea Pers. Cladosporiumfulvum Cke. Didymella exitialis (Mor.) E. Mull. Fusarium culmorum (W. G. Sm.) Sacc. Fusarium lycopersici Sam. Fusarium Martii App. et. Wr. Fusarium solani (Mart.) App. et Wr. Neurospora sitophila (Mont.) Shear et Dodge Ophiobolus graminis Sacc. Penicillium citreo-viride Biourge Pencillium citrinum Thom Thielavia terricola (Gil. et Abb.) Emm. Trichoderma viride Pers.

Basidiomycetes

Rhizoctonia crocorum DC. Rhizoctonia KUhn from Pinus silvestris Rhizoctonia solani Kiihn from Solanum tuberosum

10-2.5

10-2

10-2 10-3 0 10-2 10-3.6

10-3.6 10-3.6 10-4 10-4 10-3 10-3.6

0 10-2

0 0 10-3

10-3 10-3

orchinol-containing cultures is much lower than that of the controls (Fig. 57). Cultures of Cundidu lipolyticu accelerate following prolonged exposure to orchinol or hircinol (Fig. 57). This could be indicative of adaptation by the fungus or degradation of the phytoalexins. Reinoculation of filter-sterilized media which had been used in a previous assay for 13 days indicate that the latter is taking place (Fisch et al,, 1973a). Loroglossol (Fig. 55, II) is another phytoalexin isolated from Loroglossum hircinum (Hardegger et ul., 1963a; Urech et ul., 1963). It is sparingly soluble in water and crystallizes when transferred from ethanol stock solution into aqueous culture media (Ward et al., 1975b). This reduced its concentration in several assays leading to reports that it was inactive (Fisch et al., 1973a; Hardegger et ul., 1963~).However, when a dilution series directly in ethanol was used loroglossol was shown to have antifungal activity of the same order

517

ASPECTS OF ORCHID PHYSIOLOGY

TABLE 27 Minimal Inhibitory and Lethal Concentrations of Orchinol and Hircinol (Urech et al., 1963)

(Hircinol pg ml-l) Ia Staph. aureus Escherichia coli Trichophyton interdigitale Trichophyton mentaprophytes Endomyces albicans Epidermophyton flaccosum Microsporum audouini Sporotrichum schenckii Aspergillus niger

(Orchinol pg ml-l)

Da

> 500 250 25 25 50 100

250 50 100 100

500 =.500 >500 500 =.500

Ia

Da

50 500

250 500

10 50 50 100

50 50 50

100 250

100

>500 100

500

a I, Inhibition; D, Death.

TABLE 28 Inhibition of Spore Germination and Growth of Fungi with Orchinol (Ward et al., 1975a)

ED,, Fusarium oxysporum f: vasinfectum Glomerella cingulita Monilinia fructicola Phytophthora infestms Pythium ultimum Thielaviopsis basicola Venturia inaequalis Verticillium dahliae

(M

x 104)

Germination

Growth

1.5

2.4 2.6 0.6 2,2 3.5

0.3 0.6 0.4 -

0.9 0.4 0.6

-

3.4

(minimum inhibition dose of 10-4-10-5 M) as hircinol and orchinol (Table 29; Ward et al., 1975b). D. BIOLOGICAL ROLE

The orchid dihydrophenanthrenes are probably two of the most convincingly demonstrated phytoalexins. They play important roles in protecting the orchids from fungal infections and in the establishment of mycorrhiza. Mechanical injury can induce orchinol formation in Orchis militaris (Nuesch, 1963) even if at lower concentrations than following fungal infection. Under natural conditions such induction would render the tubers resistant to infection by pathogenic fungi. For example, very fast growing fungi like Fusarium solani can invade and overcome unprotected tissues in a very short time. On the other hand, tissues which contain even small amounts of orchinol could inhibit the fungus at least sufficiently to allow infectioninduced phytoalexin production to reach protective levels.

518

J. ARDITTI

/+--

't

I

I

5

10

15 Days after inoculation

I

I

20

24

Fig. 57. Growth of Candida Zipolytica on orchinol 100 and 50 ppm, hircinol 100 and 50 ppm and ethanol 1 0 0 ppm and 50 ppm (Fisch et al., 1973a).

TABLE 29 Percentage Inhibition of Spore Germination of Monilinia fructicola and Phytophthora infestans by Loroglossol (Ward et al., 1 9 7 5 ~ ) Loroglossol (M x ~

Fungal species Monilinia fructicolae Phytophthora infestansd

5.0a

2.P

27b 226 90b 100

1.25u 0.625 2Ib 93b

0.313

0.156

0.078

Ob

Ob

Ob

206

0

0

Ob

92

.-

a Crystals present. b C

d

Distortion and stunting of germ tubes. Germination was 100% in both water and ethanol controls. Germination was 98 % in both water and ethanol controls.

Many fungi are unable to invade tissues which contain appropriate concentrations of phytoalexins (Niiesch, 1963) or can produce them fast enough. A good example of this is provided by the fact that, despite a thin cortex and limited mechanical protection, orchid storage organs rarely rot (Niiesch, 1963). This kind of protection can be overcome only by the destruction of the phytoalexin and, indeed, fungi which deactivate orchinol rapidly (e.g., Rhizoctonia solani) destroy bulbs quickly. The fungal presence in orchid roots and germinating seeds can be considered to be a localized, regulated and stabilized parasitism. This enables orchids to coexist with mycorrhizal fungi some of which are, or may become,

ASPECTS OF ORCHID PHYSIOLOGY

519

parasitic (Kusano, 1911 ; Knudson, 1929; Arditti, 1967a). A slowly degradable phytoalexin would be the most effective means of regulating such symbiosis. Compounds which could not be degraded at all or only very slowly, might inhibit the fungi to the point of destroying the association. If degradation by the fungus is too rapid the orchid would be parasitized. Indeed such extremes have been reported (Bernard, 1909; Burgeff, 1936). However, continuous production of (a) reasonably degradable phytoalexin(s) would be optimal in that the infection would be kept within tolerable limits without damaging the fungus; this is the case with orchids and their phytoalexins. Orchis mifitaris tissues start to produce phytoalexins within 36 hours after infection and continue to d o so as long as they are alive (Niiesch, 1963). Fungi (Candida lipofyticafor example), on the other hand, can destroy phytoalexins as has been shown to be the case with orchinol and hircinol (Fisch et af., 1973a). Or, if we are to adapt Ramsbottom’s analogy, it is like preventing the Prince of Denmark from killing his uncle by restricting him to the outer rooms of Elsinore Castle (of course this analogy stretches the point a bit since the orchids or fungi can hardly be compared to the murderous King Claudius). IV. CARBON FIXATION Interest in carbon fixation and related topics has its origins in early studies of orchid biology (Bendrat, 1929; Chatin, 1874-1875; Czapek, 1909; Griffon, 1898, 1899; Lindt, 1885; Magnus, 1890, 1891; Webber, 1920) and nutrition (Miwa, 1937; Tsuchiya, 1935; Went, 1946). More recent ecological, biochemical, and physiological research has elucidated the pathways of h a t i o n and their ecological importance (Avadhani and Goh, 1974; Borriss, 1967; Coutinho, 1963, 1964, 1965, 1969, 1970; Coutinho and Schrage, 1970; Dueker and Arditti, 1968; Erickson, 1957a, b; Esser, 1973; Goh et al., 1977; Hew, 1976; Knauft and Arditti, 1969; Kristen, 1965; McWilliams, 1970; Neales and Hew, 1975; Nuerenbergk, 1963; Rubenstein et al., 1976; Wong and Hew, 1973, 1975) and provided a basis for the understanding of existing horticultural applications (Davidson, 1967; Wright, 1967). A. HISTORY

Neottia nidus avis is a chlorophyll-free European terrestrial orchid. Often described as being a saprophyte, it is actually mycotrophic, i.e., a parasite on its ever-present (PekIo, 1906) mycorrhizal fungus. The plant is yellowish brown, but white forms have also been reported (Magnus, 1890, 1891). Much of the early carbon fixation research was carried out with this species. Chlorophyll was detected in N . nidus avis more than 100 years ago (Wiesner, 1865, 1871, 1874; Prilieux, 1874) but only chlorophyll a was detected (Henrici and Senn, 1925; Menke and Schmid, 1976; Montfort, 1940; Montfort and Kiisters, 1940; Reznik, 1958; Senn, 1927). Chlorophyll was also found in

520

J. ARDITTI

Limodorum abortivum, a species related to Neottia nidus avis (Chatin, 18741875; Griffon, 1898, 1899). In addition to chlorophyll, Neottia nidus avis also contains xanthophyll (Menke, 1940; Menke and Schmid, 1976; Montfort and Kusters; Reznik, 1958). The yellow-brown colour which was the subject of several early studies (among them Lindt, 1885; Wiesner, 1865 and more recently, Reznik, 1958) is brought about by a shift of the carotenoid absorption into the green. This is probably due to the binding of carotenoids to proteins as is the case in brown algae (Menke and Schmid, 1976). Interestingly, this possibility was intimated in an early study comparing Phaeophyceae and diatom coloration with that of N . nidus avis (Molisch, 1905). Bodies which contain the pigments (“Farbstoffkorperchen”) in Neotria nidus avis were observed in some of the earliest studies (Lind, 1885; Wiesner, 1865). In subsequent years these bodies were identified as chloroplasts (Montfort, 1940). With the advent of electron microscopy and development of other modern techniques these plastids were studied in greater detail. They were found to consist of coiled, branched and swollen thylakoids with stroma-like material (Menke and Wolfersdorf, 1968). Branched thylakoids have also been found in Aceras anthropophorum (Schmid et al., 1976). In addition to chlorophyll a, they were also found to contain all plastoquinones and carotenoids of normal chloroplasts but in different concentrations (Reznik et at., 1969). The thylakoids were reported to be not fully functional However, a plastid preparation from the labellum of N. nidus avis flowers was reported to “. . . perform a photosystem I-dependent photoreduction of methylviologen . . . [and] . . . photosystem 11-reactions . . . are not functioning. . . [and] . . . that no appreciable energy transfer from carotenoids to chlorophyll occurs.” (Menke and Schmid, 1976). Plastids from the labellum of Aceras anthropophorum were found to contain high carotenoid levels but little chlorophyll. They can carry out cyclicphenazinemethosulphate mediated photophosphorylation. However, they were not able to perform photosystem 11-dependent photophosphorylation or evolve oxygen (Schmid et al., 1976). The studies mentioned above inevitably led to the question of CO, fixation by orchids, at that time as a part of the broader aspect of carbon assimilation by saprophytes (for an interesting, even if older, discussion see Lebedev, 1948). Mycotrophic orchids (at that time called saprophytes by some) like Corallorhiza innata, Limodorum and Neottia were among the first to be studied (Griffon, 1898, 1899; Reznik, 1958; Webber, 1920). Others were Listera ovata, Goodyera repens, Epipactis latijiolia, E. rubiginosa, Orchis latifolia, 0. purpurea, 0. morio, 0. mascula and 0. bifolia (Griffon, 1899; Montfort and Kusters, 1940). These studies showed that Neottia nidus avis and Corallorhiza innata (mycotrophs) do not fix CO,. Limodorum abortivum (also a mycotroph) fixes very little whereas all green orchids are capable of carbon fixation.

ASPECTS OF ORCHID PHYSIOLOGY

52 1

Orchids were among the plants used in the early studies on Crassulacean acid metabolism (CAM). These studies showed that acidity increased in thick-leaved (succulent) orchids in the dark (Bendrat, 1929; Warburg, 1886-1 888). However, the significance of this observation was not realized until a quarter of a century later. When research in this area was resumed (Kristen, 1965; Nuerenbergk, 1961, 1963), modern methods confirmed that some orchids can indeed take up and fix carbon in the dark by CAM or a similar pathway (Avadhani, 1963; Hew, 1976; Knauft and Arditti, 1969; Neales and Hew, 1975; Rubenstein et al., 1976). More recently the possibility that C , fixation may occur in orchids has also been investigated (Avadhani and Goh, 1974). Altogether it seems that orchids may fix carbon by the C,, C, and CAM pathways (Table 30). Orchids are relatively slow-growing plants generally maintained in an enclosed area, i.e., greenhouses or culture tubes. Therefore, it is not surprising that attempts to accelerate growth by means of CO, enrichment of air were made or suggested some time ago (Miwa, 1937; Tsuchiya, 1935; Went, 1946). A paper (Frackowiak, 1933) often cited as reporting on CO, enrichment attempts, does not even mention carbon fixation. Efforts to accelerate growth by such means have continued through the years but results have been inconsistent (Wright, 1967). Perhaps one reason for this is that carbon fixation pathways of the orchid being fed CO, were not always considered despite suggestions that this should be done (Went, 1946; Withner, 1974). Consequently CAM orchids were sometimes given CO, during the day. B. STOMATAL RHYTHMS

An indirect method of determining which orchids fix carbon in the dark is to study stomatal rhythms. Studies of endogenous rhythms involving the “De Saussure effect” (Nuerenbergk, 1961) led to porometer measurements of stomatal opening and CO, uptake. These measurements demonstrated that a direct relationship exists between stomatal opening and CO, fixation in the dark (Kristen, 1965; Nuerenbergk, 1963). Confirmation of these findings was provided by studies of stomatal rhythms in orchids which showed that the stomata of Archnis, Aranda and Carrleya (all thick-leaved with CAM) are open at night. Stomata of thin-leaved (non CAM) orchids like Arundina, Bromlieadia and Spathoglottis are open during the day (Goh et al., 1977). Information obtained from only six orchids may be considered somewhat limited and therefore insufficient For general conclusions regarding patterns in the Orchidaceae, a family with 600-800 genera. However, a number of other factors suggest that these conclusions are plausible (see reviews by Goh e f a/., 1977; Knauft and Arditti, 1969). C. CRASSULACEAN ACID METABOLISM

Evidence that thick-leaved orchids have CAM (Table 30) was first presented

TABLE 30 Curbon Fixation and Photophosphorylation by Orchids Orchid

Organ

Probable pathway

Aceras anthropophorum

Labellum

Acropera hddgesii Angraecum sesquipedale Aplectrum hyemle Arachnis cv Maggie Oei

Pseudobulbs Leaves Leaf Leaf (1.5 mm thick)

Cyclic photophosphorylation No CAM CAM No CAM CAM

Aranah cv Deborah Aranda cv Wendy Scott

Leaf (1.5 mm thick) Leaf (1-5 mm thick)

CAM CAM

Aranthera cv James Storie Arundina graminifolia

Leaf (1.5 mm thick) Leaf (0.3 mm thick)

CAM No CAM; C,

Ascocentrum ampullaceum Brassolaeliocattleyacv Maunalani

Leaf (1 *27mm thick) Leaf

Brassavola perrinii

Leaf

Bromheadiafinlaysoniana

Leaf (thin)

CAM Nocturnal C02 assimilation and acidification Nocturnal CO, assimilation No CAM; C,

Bulbophyllum gibbosum Calanthe vestita

Leaf (1.422 mm thick) Leaves

CAM CAM

Remarks

Reference Schmid et al., 1976 Warburg, 1886-1888 Nuerenbergk, 1963 Adams, 1970 Goh et al., 1977; Lee, 1970; Neales and Hew, 1975

Species photorespires

Goh et al., 1977 Hew, 1976; Neales and Hew, 1975 Neales and Hew, 1975 Avadhani and Goh, 1974; Goh et al., 1977; Hew, 1976; Neales and Hew, 1975; Wong and Hew, 1975

McWilliams, 1970 McWilliams, 1970 Coutinho, 1964; Szarek and Ting, 1977 Avadhani and Goh, 1974; Goh et al., 1977 McWilliams, 1970 Nuerenbergk, 1963

Catasetumfimbriatum

Leaves

CAM

Cattleya sp. Cattleya sp. Cattleya autumnalis Cattleya bicolor Cattleya cv Bow Bells

Leaf (thick) Roots, leaves Leaves Leaves

Leaf (0.5 mm thick)

CAM Fixation in the light CAM CAM CAM

Cattleya forbesii

Leaf

CAM

Cattleya gigas

Roots, stems, leaves

Fixation in the light

Cattleya guttata Cattleya intermedia Cattleya labiata Cattleya loddigesii Cattleya mossiae Cattleya skinneri Cattleya trianaei Cattleya walkeriana Cattleya warneri Cattleya cv White Blossom “Stardust” X C. ov Bob Betts “Glacier” Coelogyne cristata Coelogyne massangeana

Leaves Leaves Whole plants Whole plants Whole plants Whole plants Whole plants Whole plants Whole plants

CAM CAM CAM CAM CAM CAM CAM CAM CAM

Whole plants Leaves Leaves (plicate)

CAM CAM

c,

Nuerenbergk, 1963; Borris, 1967; Goh et al., 1977 Warburg, 1886-1888 Erickson, 1957a, b Coutinho, 1969 Coutinho, 1969 Hew, 1976; Neales and Hew, 1975 Hew, 1976; Neales and Hew, 1975 Dycus and Knudson, 1957 Coutinho, 1969 Coutinho, 1969 Nuerenbergk, 1961, 1963 Nuerenbergk, 1961, 1963 Nuerenbergk, 1961, 1963 Nuerenbergk, 1961, 1963 Nuerenbergk, 1961, 1963 Nuerenbergk, 1961, 1963 Nuerenbergk, 1961, 1963 Knauft and Arditti, 1969 Nuerenbergk, 1961, 1963 Rubenstein et al., 1976 continued

TABLE 30-continued Orchid

Organ

Probable pathway

Remarks

Coelogyne muyeriana

Leaf (0.4 mm thick)

G

Species photorespires

Coelogyne rochussenii

Leaf (0.2 mm thick)

c3

Species photorespires

Cymbidium cv Chelsea

Sepals, petals, leaves

G

High rate of fixation in the dark

Cymbidium chinense Cymbidium hybrid Cymbidium cv Independence Day

Leaves (thin) Leaves (thin) Sepals, petals, leaves

CAM (?)

Cymbidium sinense

Leaves

c3

Cymbidium lowianum “Yorktown” Cypripedium acaule Cyrtopodiumparanaensis Dendrobium taurinum Encyclia atropurpurea Encyclia flabellifera Encyclia odoratissima Epiakndrum alatum Epidendrum ciliare Epidendrum elripticum

Leaves Leaves (0.406 mm thick) Leaves Leaf (1-5 mm thick) Leaves Leaves Leaves Leaf (1-397 mm thick) Leaves (succulent) Leaves

CAM No CAM CAM CAM CAM CAM CAM CAM CAM CAM

c3 G

Reference Hew, 1976; Wong and Hew, 1975; Neales and Hew, 1975 Hew, 1976; Wong and Hew, 1975; Neales and Hew, 1975 Arditti and Dueker, 1968; Dueker and Arditti, 1968 Warburg, 1886-1888 Rubenstein et al., 1976 Arditti and Dueker, 1968; Dueker and Arditti, 1968

High rate of 14Cfixation in the dark Wong and Hew, 1973 High rate of 14C fixation in the dark Nuerenbergk, 1961 McWilliams, 1970 Coutinho, 1969 Neales and Hew, 1975 Nuerenbergk, 1961, 1963 Coutinho, 1969 Coutinho, 1969 McWilliams, 1970 Bendrat, 1929 Coutinho, 1963, 1964, 1965

Epidendrum floribundum

Leaves

CAM

Epidendrum moseni Epidendrum radicans Epidendrum schomburgkii Epidendrum xanthinum

Leaves Leaves (2.159 mm thick) Leaves Roots, stems, leaves

CAM CAM CAM Fixation in the light

Eulophia keithii

Leaf

C3

Gomesa crispa Goodyerapubescens Habenaria platyphylla

Leaf (0.431 mm thick) Whole plant

No CAM CS

Laelia cinnabarina Laelia crispa Laelia Jrava Laelia millerii Laelia perrinii Laelia purpurella Laelia xanthina Lanium avicula Limodorum abortivum

Leaves Leaves Leaves Leaves Leaves Leaves Leaves Leaves All parts of the plant

Maxillaria aromatica

Leaves pseudobulb

CAM CAM CAM CAM CAM CAM CAM CAM Minimal fixation in light despite presence of chlorophyll No CAM

Neottia nidus avis

Plastids from labellum whole plant

Coutinho, 1969; Nuerenbergk, 1961 Coutinho, 1969 McWilliams, 1970 Nuerenbergk, 1961 Dycus and Knudson, 1957

Species photorespires

Hew, 1976; Wong and Hew, 1975 Coutinho, 1969 McWilliams, 1970 Raghavendra and Das, 1976

Only PS I is operative; no light or dark carbon fixation

Probably mycotrophic

Coutinho, 1969 Coutinho, 1969 Coutinho, 1969 Coutinho, 1969 Coutinho, 1969 Coutinho, 1969 Coutinho, 1969 Coutinho, 1969 Chatin, 1874-1875; Griffon, 1898, 1899

Mycotrophic species

Warburg, 1886-1888; Coutinho, 1963 Menke and Schmid, 1976; Reznik, 1958

-

continued

TABLE 30-continued Orchid

Organ

Probable pathway

Remarks Some Oncidium species have thick leaves Species photorespires

Oncidium sp.

Leaves (type unknown)

CAM

Oncidium flexuosum

Leaves (0.33 mm thick)

C,

Oncidium lanceanurn Oncidium pumilum Oncidium sphacelatum

Leaves (0.33 mm thick) Leaves Leaf (thin)

C,

No CAM; C,

Ornithidium densum Paphiopedilum barbatum

Leaf (thick) Leaf

c 3

Paphiopedilum insigne Paphiopedilum cv Mildred Hunter

Leaf (soft, leathery)

Paphiopedilum venustum Paphiopedilum villosum Phalaenopsis Phaluenopsis

Leaf (0.381 mm thick) Leaf (soft) Leaf (thick) Root, stem, leaves

Phalaenopsis amabilis Phalaenopsis schilleriana

Leaves Leaves

Pleurothalis ophiocephalus Schombocattleya hybrid

Leaves

c3

Species photorespires

CAM CAM CAM at night, C, during day

Species photorespires

Reference Warburg, 1886-1 880 Neales and Hew, 1975; Wong and Hew, 1975 Hew, 1976 Coutinho, 1969 Bendrat, 1929; Wong and Hew, 1975 Warburg, 1886-1888 Hew, 1973; Wong and Hew, 1975 Nuerenbergk, 1961 Rubenstein et al., 1976

No CAM CAM Fixation in the light

Bendrat, 1929 Borris, 1967 Dycus and Knudson,

CAM Night CO, assimilation and acidification CAM-like 8 I3C values CAM at night; C3 during day

McWilliams, 1970 McWilliams, 1970

1957

Schomburgkia

Szarek and Ting, 1977 Rubenstein et al., 1976 Neurenbergk, 1963

Schombirrgkia crispa

Leaves

CAM

Spathoglottis plicata

Leaf (0.3 mm thick)

Night CO, assimilation No CAM; C,

Spiranthes speciosa Tainia penangiana

Leaf (0.508 mm thick) Leaf

Thunia rnarshalliana Vanda

Root, stem, leaves

CAM Fixation in the light

Vanda cv Miss Joaquim Vanda tesselata

Leaves Whole plant

c 2

Vanilla aromatica Vanillafragans Vanilla planifolia

Whole plant Leaf (0-22 mm thick) Leaf (thick)

CAM CAM CAM

Vanilla sp.

Leaves

CAM

Sophrortitis cernua

Coutinho, 1969; Nuerenbergk, 1961 ; Wong and Hew, 1975 Coutinho, 1969 Species photorespires

No CAM

c,

CAM

Species photorespires

Goh et al., 1977; Hew, 1976; Neales and Hew, 1975 McWilliams, 1970 Hew, 1976; Wong and Hew, 1975 Nuerenbergk, 1961 Dycus and Knudson, 1957 Khan, 1964 Raghavendra and Das, 1976 Nuerenbergk, 1963 McWilliams, 1970 Bendrat, 1929; Warburg, 1886-1 888 Coutinho, 1969

528

J . ARDITTI

nearly 100 years ago (Warburg, 1887-1888), when it was shown that they became acidified in the dark. The acid which accumulates in orchid leaves is malate, the same one found in other CAM plants (Avadhani and Goh, 1974; McWilliams, 1970; for reviews see Arditti and Harrison, 1971; Avadhani, 1963; Benzing, 1973; Hew, 1976; Nuerenbergk, 1963). The list of orchids which undergo acidification has been extended since then (Table 30) (Avadhani and Goh, 1974; Borriss, 1967; Coutinho, 1963, 1964, 1965, 1969, 1970; Coutinho and Schrage, 1970; McWilliams, 1970; Nuerenbergk, 1963; Szarek and Ting, 1977). Orchids which accumulate malate and take up CO, in the dark are all thick-leaved and include Cattleya labiata, Encyclia atropurpurea, Schomburgkia crispa, Angraecum sesquipedale, Phalaenopsis esmeralda (Nuerenbergk, 1963), an unidentified Phalaenopsis, Cattleya bowringiana (Borriss, 1967), Vanilla fragrans, Epidendrum alatum, E. radicans, Brassolaeliocattleya “Maunalani”, Ascocentrum anipullaceum and Phalaenopsis schilleriana (McWilliams, 1970). Limited uptake and slight acidification occur in Bulbophyllus gibbosum and Spiranthes speciosa (McWilliams, 1970). CO, uptake in the dark has also been reported in the following thick-leaved orchids: Oncidium lanceanum, Phalaenopsis amabilis, Aranda cv Wendy Scott, and Cattleya cv Bow Bells (Hew, 1976). Evidence from experiments with 14C0, indicate the existence of CAM in Paphiopedilum (see below) and Shombocattleya hybrids (Rubenstein et a[., 1976) as well as Cattleya cv White Blossom “Stardust” x C . cv Bob Betts “Glacier” (Knauft and Arditti, 1969). There are no dark CO, uptake and acidification in thin-leaved orchids such as Coelogyne cristata, Cymbidium,Paphiopedilum (see below), Castasetum fimbriatum, Calanthe vestita (Neurenbergk, 1963), Cypripedium acaule, Paphiopedilum venustum (see below) and Goodyera pubescens (McWilliams, 1970). The thin-leaved orchids Arundina graminifolia, Spathoglottis plicata, Tainia penangiana, Paphiopedilum barbatum (see below), Eulophia keithii, Coelogyne mayeriana, and C . rochussenii take up carbon in a manner similar to that of C3 plants (Hew, 1976). Further confirmation for the existence of CAM in orchids was obtained from determinations of 13C : 12Cratios. During photosynthesis plants take up preferentially the lighter of the two isotopes. Therefore the 13C : 12C ratio expressed as P3C can be used as an indication of the mechanisms involved in carbon fixation (Hew, 1976; Lerman et a / . , 1974; Lerman and Quieroz, 1974; Neales and Hew, 1975). In orchids S13C is positively related to leaf thickness. The V3Cis in the CAM range, i.e., -1 5 to -16 % for thick-leaved species like Dendrobium taurinum, Cattleya cv Bow Bells, Aranthera cv James Storie, Aranda cv Wendy Scott and Arachnis cv Maggie Oei. When taken together, the available evidence indicates clearly that some orchids fix carbon via CAM. It is possible even that CAM may occur in

ASPECTS OF ORCHID PHYSIOLOGY

529

thin-leaved genera like Paphiopedilum. According to one report a Paphiopedilum hybrid “. . . is a C, plant during the day and a CAM plant at night . . .” (Rubenstein et al., 1976). On the other hand, a Paphiopedilum plant did not take up CO, in the dark and was reported to “. . . have ‘normal’ photosynthesis.” (Neurenbergk, 1963). In addition, there was no acidification in the dark in leaves of Paphiopedilum villosum (Bendrat, 1929) and P . venustum (McWilliams, 1970). These differences require further research and clarification. D. C, PHOTOSYNTHESIS

Thin-leaved orchids fix carbon in the light (Table 30; Adams, 1970; Arditti and Dueker, 1963; Bendrat, 1929; Dueker and Arditti, 1968; Erickson, 1957a, b ; McWilliams, 1970; Nuerenbergk, 1963; Warburg, 18861888). Gas exchange and 14C0, fixation patterns and P3C ratios have shown that the following are C , orchids: Spathoglottis, Arundina graminifolia, Coelogyne massengeana, C . mayeriena, C . rochussenii, Cymbidium sinensis, a Cymbidium hybrid, Eulophia keithii, Tainia penangiana, Oncidium jlexuosunt, Bromheadia jnlaysoniana, Arundina graminifolia and a Paphiopedilum hybrid (Avadhani and Goh, 1974; Hew, 1976; Neales and Hew, 1975; Rubenstein et al., 1976; Wong and Hew, 1973). The existence of the C, cycle has been interpreted as being “. . . consistent with the primitive nature of terrestrial thin-leaved orchids . . .” (Hew, 1976). This may be so. However, since some orchids (terrestrial and epiphytic) are “normal” mesophytes there may be no particular significance in the fact that they fix carbon via a pathway commonly found in such plants. E. C, PHOTOSYNTHESIS

Plants which fix carbon via the C, pathway grow best under high light intensities and warm climates. They use water efficiently, have high photosynthetic ratios and may tolerate arid conditions. C, plants are less common among dicotyledons than among monocotyledons (Edwards, 1974). Most of the evidence presented to date suggests that thin-leaved orchids are C, and not C, plants (Hew, 1976). However, some species “. . . exhibit an interesting mosaic of different types of 14C0, fixation including C-3, C-4 and CAM.” (Avadhani, 1963). More specifically there are “. . . indications that young leaves of Arundina graminifolia may photosynthesize, at least in part via the C, pathway” (Avadhani and Goh, 1974), and a Schombocattleya hybrid has been reported to be “. . . a C, plant during the day and a CAM plant at night . . .” (Rubenstein et al., 1976). Orchids are not uncommon in habitats suitable for C, plants, and most of them have not been studied. Therefore, it would not be surprising if additioiial C, species will be discovered in the future. Such species should prove of interest not only in terms of orchid biology, but also as subjects for C,

530

J. ARDITTI

research. This is especially so in view of the fact that intrageneric hybrids of orchids are easy to produce. Therefore it may be possible to produce hybrids between C3 and C4,C, and CAM, and C, and CAM species. F. CARBON FIXATION BY DIFFERENT PLANT ORGANS

As in other plants most of the carbon fixation by orchids occurs in leaves. The rate of carbon fixation in light depends on leaf age in Cattleya (Erickson, 1957a). However, these findings are difficult to interpret since Cattleya is a thick-leaved orchid which probably fixes most of its carbon via CAM. Cymbidium leaves (C,) fix more CO, per mg tissue than flowers [Figs. 58, 59 (see p. 549); Table 31; Arditti and Dueker, 1968; Dueker and Arditti, 19681. Chlorophyll-containing floral segments, stems and roots, can also photosynthesize (Table 31). 14C0, fixation by floral segments changes with age (Fig. 59). Leafless orchids like Taeniophyllum zollingeri (Wiesner, 1897), Polyrrhiza lindenii, Dendrophylaxfunalis, Sarcochilus luniferus, Doritis taenia h , Campylocentrum fasciola, C . pachyrrhizum, Harrisela porrecta and Independence day 'Yorktown' Dark control llluminatrd

:helm

-.-.-

-

i i i

i

I I

i

iI i

I

i ; P o 0 1

i

iI

i ' 0

q h

Newly dpened Fully obesed

Newly bpened Fully opened

Fig. 58. 14COzfixation by Cymbidium buds, newly and fully opened flowers and leaves. S-sepals; P-petals; 0-ovary (Dueker and Arditti, 1968).

TABLE 31 Net Carbon Fixation in the Light by Orchid Plant Organs Expressed as % Amount Fixed by Leaves (Dueker and Arditti, 1968) ~~~~

Organ

Bud Sepals Petals Ovary Newly opened Flowers Sepals Petals Ovary Fully opened Flowers Sepals Petals Ovary Lower leaves Upper leaves Stems and leaves Stem Leaf and stem Root Leaf 1 year old 2 years old 3 years old 4 years old 5 years old 6 years old 7 years old

~

~

Cymbidium cv Indep. Day cv Chelsea “Yorktown” 13.6

1 .o

5.5

0.1

1.2

04

10.7 0.5

11.9 1 .o

~

~~~~~

Epidendrum xanthinum

Cattleya gigus

~~

~

Phalaenopsis hybrids

Vanda suavis

Cattleya

0.1

1.4 0.6 0.2

4.9 1.1

0.4 100 129

100 111 80

113 100

1814 100

100

645

3689

100 100

103 103 91.6 70 50 38

532

J. ARDlTTI

Tueniophyllum jiliforme probably depend almost entirely on their roots (which are green) for carbon fixation (for a review see Churchill et ul., 1972). Orchid roots are generally fleshy ; however, the diurnal pattern of CO, fixation in roots and (thick) leaves of Arundu are different (Fig. 60, see p. 549) (Hew, 1976). C. PHOTORESPIRATION

In a systematic study, thick-leaved orchids were shown to have high CO, compensation points which are typical of C, plants (Wong and Hew, 1973, 1975). A high postillumination CO, outburst was also observed in these orchids (Fig. 60). Their photorespiratory rates are higher than those of dark respiration (Wong and Hew, 1975). H. SUMMATION

It is not at all surprising that orchids have evolved two (CAM, C,) and possibly even three (C,) carbon fixation patterns. The family is large, circumglobal, and can be found in almost all ecological habitats (including subterranean, but not subaquatic). To exist in so many habitats the Orchidaceae have evolved many adaptive mechanisms including several water conservation features (McWilliams, 1970) and carbon fixation pathways. Among higher plants in general, differences in CO, fixation pathways may exist within one family. Among Orchidaceae, as in several other families, such differences may exist within a single genus (Wong and Hew, 1973). The possibility that this may be the case with Puphiopedifum has already been mentioned. The same may be true for genera like Cymbidium (Wong and Hew, 1973), Oncidium, Dendrobium and Epidendrum which have thick- and thin-leaved species. Many, possibly most, orchids which have thick leaves and exhibit CAM live under what are essentially xerophytic conditions, often as epiphytes (Czapek, 1909; Goh et al., 1977; Knauft and Arditti, 1969; Nuerenbergk, 1963). Their roots, wrapped around or spread on their host, are usually exposed and subject to desiccation (Nuerenbergk, 1963). Therefore it is not surprising that they have evolved: (i) thick cuticles and cuticular ledges which cover stomata1 pores and form hyperstomatic chambers as in Puphiopedilum venustum (Haberlandt, 1928), Vundu tricolor (Gessner, 1956) and Cuttleyu bowringiana (Borris, 1967); (ii) nocturnal opening of stomata (Goh et ul., 1977); (iii) CAM (Nuerenbergk, 1963); and (iv) no stomata on the upper surfaces of leaves (Goh et al., 1977; for a review see Withner et ul., 1974). Another possible benefit of CAM for a terrestrial orchid growing among dense vegetation or an epiphytic one located within the canopy of a tree is the advantage of obtaining CO, when it is most abundant. In such environments CO, levels in the air could be greatly reduced during the day due to photosynthesis by the surrounding vegetation. At night most of the sur-

ASPECTS OF ORCHID PHYSIOLOGY

533

rounding vegetation does not fix carbon and in fact releases CO, due to respiration. Consequently CO, levels are higher than during the day and the ability to fix it in the dark constitutes a n adaptation which enhances survival. Measurements of CO, levels in tropical forests in Nigeria indicate that the highest CO, concentration (0.052 %) occurs at sunrise and the lowest (0.035 is at noon (for a review see Sanford, 1974). Information from reports regarding culture of orchids under controlled conditions (Krizck and Lawson, 1974), with added air (Frackowiack, 1933), or CO, “fertilization” and measurements of greenhouse or culture flask air (Miwa, 1937; Tsuchiya, 1935; Wright, 1967; for reviews see Sanford, 1974: Withner, 1974) is difficult to interpret because in most cases the carbon dioxide was added during light periods. At the time of writing there are no published reports of orchids being fertilized with CO, in the dark. The carbon fixation pathway of C, orchids also reflects their natural habitats. For example, Arundina graminifolia is found in open sunny places in the lowlands and highlands of Malaya, never in shade; Bromheadia jinlaysoniana grows in open scrub and light secondary forest whereas the preferred habitat of Spathoglottis plicata is under full sun (Goh et al., 1977; Holttum, 1953).

x)

V. FLOWERS

In 1678 Jacob Breynius wrote in his compendium “Exoticarum Aliarumque Minus Cognitum Plantarum”, on the extraordinary diversity of orchids: If nature ever showed her playfulness in the formation of plants this is visible in the most striking way among the orchids. The manifold shape of these flowers arouses our highest admiration. They take the form of little birds, of lizards, of insects. They look like a man, like a woman, sometimes like an austere sinister fighter, sometimes like a clown who excites our laughter. They represent the image of a lazy tortoise, a melancholy toad, an agile, everchattering monkey. Nature has formed orchid flowers in such a way that, unless they make us laugh, they surely excite our greatest admiration. The causes of their marvelous variety are (at least in my opinion) hidden by nature under a sacred veil (translated by Ames, 1948). John Lindley, the British botanist who established the family Orchidaceae and is therefore called the father of orchidology was n o less eloquent: Orchidaceae are remarkable for the bizarre figure of their multiform flower, which sometimes represents an insect, sometimes an helmet with the visor up, and sometimes a grinning monkey; so various are these forms, so numerous their colours, and so complicated their combinations, that there is scarcely a common reptile or insect to which some of them have not been likened. However, his scientific logic prevailed: They all, however, will be found to consist of their outer pieces belonging to the calyx, and three inner belonging to the corolla . . . (Lindley, 1830).

534

J. ARDITTI

A. HISTORY

Less than a hundred years after Breynius (in 1793) Christian Conrad Sprengel in his “Das entdekte Geheimnis der Natur im Bau und in der Befruchtung der Blumen” tried to lift the veil a bit by suggesting that the “. , . marvelous variety . .” is a method of attracting pollinators. Seventy years later Charles Darwin removed the veil almost completely (Darwin, 1862 or 1904) by describing the contrivances by which orchids are pollinated by insects. The “unveiling” was completed 100 years after Darwin by more complete descriptions of orchid pollination (Kullenberg, 1961; van der Pijl and Dodson, 1966). Interest in the mechanisms which control the blooming of orchids was apparently first generated by the gregarious flowering of Dendrobium crumenatum (Massart, 1895; Treub, 1887; Went, 1898). Interest in ecology and introduction of orchids into the commercial cut-flower industry added an economic impetus to studies of flower induction (Tables 32-36). Fritz Miiller (Fig. SOD), a contemporary of Darwin was possibly the first to become interested in the visual effects of pollination on orchid flowers (Miiller, 1868, for reviews see Arditti; 1971a, b; Moller, 1920-1921). He made interesting observations and drew conclusions which in a strict sense were erroneous, but were nevertheless accepted by Darwin. Because of that Miiller’s ideas remained unchallenged for nearly 50 years. The man who challenged them, Hans Fitting (Fig. 80C; Fitting 1909a, b, 1910) did suggest the involvement of a hormone (thereby becoming the first to use this concept in relation to plants), barely missed the discovery of auxins and became a great plant physiologist (Arditti, 1971a, b). Subsequent work on post-pollination phenomena in orchid flowers has been sporadic (Arditti 1969, 1971a, b, 1976a, b; Arditti et al., 1971 ; Knauft et af., 1970).

.

B. INTRODUCTION

There are a number of reports on the flowering dates, periodicity and rates of orchids in their natural habitats (Curtis, 1954; Dunsterville and Dunsterville, 1967; Goh, 1973; Quisumbing, 1968; Sanford, 1971; Vacin, 1952; Zotkiewicz, 1961) as well as in botanical gardens, experimental stations (Montgomery and Laurie, 1944) or private collections (including Hager, 1957; Hamilton, 1977; McDade, 1947; Wr6bel-Streminska, 1961). Such studies can be used to obtain information on the factors which induce flowering in orchids and this has been done for West African species (Tables 32, 33, 34, 35, 36) (Sanford, 1971). In other cases information on the factors which control the flowering of orchids was obtained experimentally (Casamajor and Went, 1953) or by observation and deduction (for reviews see Nuerenbergk, 1961b and Schlechter, 1977). Not surprisingly, since orchids can be found in many different

TABLE 32 West Afiican Orchids Blooming During One Fairly Consistent Period of the Year (Sanford, 1971) Genus Aerangis Aerangis Ancistrorhynchus 'Ancistrorhynchus Angraecum Chamaeangis Cyrtorchis Cyrtorchis Diaphananthe Diaphananthe Diaphununthe Diaphananthe Encheiridion Eulophia 2Eurychone Habenaria Liparis Liparis Liparis

Malaxis Plectrelminthus Polystachya Polystachya Polystachya

Species

A. Probably photo-controlled (long-day) Epiphyte Terrestrial

biloba (Lindl.) Schltr kotschyanu (Rchb.f.) Schltr metteniue (Kraenzl.) Summerh. straussii (Schltr) Schltr birrimense Rolfe vesicuta (Lindl.) Schltr arcuata subsp. variabilis Summerh. humata (Rolfe) Schltr curvuta (Rolfe) Summerh. kamerunensis (Schltr) Schltr pellucida (Lindl.) Schltr rutila (Rchb.f.) Sumrnerh. mucrorrhynchium (Schltr) Summerh. guineensis Lindl. rothschildiana (OBrien) Schltr englerana Kraenzl. caillei Finet nervosa (Thunb.) Lindl. tridens Kraenzl. katangensis Summerh. caudutus (Lindl.) Summerh. modesta Rchb.f. mukandaensis De Wild. subulata Finet

+ + + + + + + + + + + +

+ + + + + + + +

Flowering period ~

.~

July, Aug. June, July June, July (May), June, July (Aug.) June, July July, Aug., Sept. May, June, July July, Aug. June, July (Aug.) July, Aug., Sept. July, Aug., Sept. (Oct.) May, June, July

+ +

+ +

(May), June, July (Aug.) Aug., Sept., Oct. (Nov.) May, June, July July, Aug. (May), June, July (May), June, July (Aug., Sept.) (May), June, July June, July, Aug. (April), May, June, July (Aug.) (July), Aug., Sept. (June, July), Aug. (Sept.) (April), May, June, July continued

TABLE 32-continued Genus Rangaeris Solenangis Tridactyle lTri&ctyle Tridactyle

Genus Ancistrochilus Ansellia Bulbophyllum Bdbophyllum Bulbophyllum Bulbophyllum Bulbophyllum Bulbophyllum Bulbophyllum Chamaeangis Eulophia Genyorchis Graphorkis Hetaeria Poaiangis Polystachya

Species

A. Probably photocontrolled (long-day) Epiphyte Terrestrial

muscicola (Rchb.f.) Summerh. clavata (Rolfe) Schltr bicaudata (Lindl) Schltr brevicalcarata Summerh. lagosensis (Rolfe) Schltr

Species

+ + + + +

July, Aug. (Sept.) July, Aug., Sept., Oct. (July), Aug., Sept., Oct. (Nov.) Aug., Sept., Oct. July, Aug., Sept.

B. Probably photocontrolled (short-day) Epiphyte Terrestrial

rothschildianus OBrien africana Lindl. bufo (Lindl.) Rchb.f. falcatum (Lindl.) Rchb.f. fuscum Lindl. kamerunense Schltr lupulinum Lindl. mehnorrhachis (Rchb.f.) Rchb.f. rhizophorae Lindl. lanceolata Summerh. gracilis Lindl. pumila (Sw.)Schltr. lurida (Sw.) Kuntze stammleri (Schltr) Summerh. dactyloceras (Rchb.f.) Schltr afinis Lindl.

+ + + + + + + + + + + + + +

Flowering period

+

+

Flowering period

a t . , Nov., Dec. Oct., Nov., Dec., Jan. Oct., Nov., Dec., Jan. (Oct.), Nov., Dec. (Jan.) Oct., Nov., Dec., Jan. Jan., Feb., March Nov., Dec., Jan. Nov., Dec. Nov., Dec. Nov., Dec. Jan., Feb., March (April) Dec., Jan., Feb. Dec.,Jan., Feb. Dec., Jan. (Feb.) Dec., Jan., Feb. (March) Jan., Feb. (March)

Polystachya Rangaeris Tridactyle Zeuxine

golungensis Rchb.f. rhipsalisocia (Rchb.f.) Summerh. tridactylites (Rolfe) Schltr elongata Rolfe

+ + +

Nov., Dec. (Jan.) (Dec.), Jan., Feb. (March)

+

(Jan.), Feb., March, April Dec., Jan., Feb.

C. Probably day-length neutral (1) Flowering at the beginning of the growing season (on new growths)

Genus Bulbophyllum Bulbophyllum Corymborkis Eulophia Liparis" Liparisa Malaxis Malaxis Nervilia Nervilia Nervilia Nervilia Nervilia Polystachyaa Polystachya Polystachya Polystuchya

Species porphyroglossum Kraenzl. winkleri Schltr corymbosa Thou. quartiniana A.Rich. platyglossa Schltr suborbicularis Summerh. maclaudii (Finet) Summerh. prorepens (Kraenzl.) Summerh. adolphii Schltr fuerstenbergiana Schltr renifvrmis Schltr kotschyi (Rchb.) Schltr umbrosa (Rchb.f.) Schltr adansoniae Rchb.f. callunifioraKraenzl. dolichophylla Schltr odorata Lindl. var. odorata

Epiphyte

+ +

+ +

+ + +

+

Terrestrial

+ + + +

+ + + + +

Flowering period April, May March, April, May, June, July May April, May, June April, May May (June) May, June (July) April, May (June) (March), April March April, May March, April, May Feb., March (April) (April), May, June, July (Aug.) April, May (Feb.), March, April, May (March), April, May, June, July (Aug.) continued

TABLE 32-continued

Genus

C. Possibly day-length sensitive (2) Flowering at the beginning of the growing season (on mature growths) Species Epiphyte Terrestrial Flowering period

SAncistrorhynchus cephbtes (Rchb.f.) Summerh. Ancistrorhynchus recurvus Finet Bulbophyllum simonii Summerh. Calyptrochilum emarginatum (Sw.) Schltr Diaphnanthe bidem (Sw.) Schltr Diaphonanthe longiculcar (Summerh.) Summerh. Diaphmanthe plehniuna (Schltr) Schltr Polystachya saccata (Finet) Rolfe Tri&c&yIea gentilii (DeWild.) Schltr

Genus Ancistrorhynchus Angraecopsis Angraecopsis Bohiellaa Bulbophyllum BulbophyllunP Bulbophyllum Bulbophyllum

+ + + + + + + + +

April, May Feb., March, April, May Feb., March, April April, May (March), April, May (June) May, June (July) April, May, June March, April (May), June, July

C. Probably day-length sensitive via photocontrol of vegetative dormancy (3) Flowering at the end of the growing season Species Epiphyte Terrestrial Flowering period capifatus (Lindl.) Summerh. parviflora (Thouars) Schltr tridens (Lindl.) Schltr talbotii (Rendle) Summerh. buntingii Rendle fuerstenbergianum (DeWild.) De Wild. josephii (Kuntze) Summerh. magnibracteatum Summerh.

+ + + + + + + +

Sept., Oct., Nov. (Jan.) Sept., Oct., Nov. Sept., Oct. (June), Aug., Sept., Oct., Nov. (Jan.) Sept., Oct. Oct., Nov., Dec., Jan., Feb. Oct., Nov. Aug., Sept. (Oct.-Dec.)

Bulbophyllum Bulbophyllum Bulbophyllum Bulbophyllum Chamaeangis Cyrtorchis Cyrtorchis 4Cyrtorchis Habenaria Listrostachys PoIystachya Polystachya Solenangis

nigericum Summerh. pavimentatum Lindl. pipio Rchb.f. porphyrostachys Summerh. odoratissima (Rchb.f.) Schltr chailluana (Ho0k.f.) Schltr monteiroae (Rchb.f.) Schltr ringens (Rchb-f.) Summerh. macrandra Lindl. pertusa (Lindl.) Rchb.f. paniculata (Sw.) Rolfe rhodoptera Rchb.f. scandens (Schltr) Schltr

+ + + + + + + + + + + +

+ +

Oct., Nov. Sept., Oct., Nov. a t . , Nov., Dec. Oct., Nov. Oct., Nov. Aug., Sept., Oct., Nov. Aug., Sept., Oct., Nov., Dec. Sept., Oct., Nov. (May), Sept., Oct., Nov. (Dec.) (Sept.), Oct., Nov. Aug., Sept., Oct. (Nov.) Aug., Sept., Oct., Nov., Dec. Aug., Sept., Oct., Nov., Dec., Jan., Feb.

a Possibly photocontrolled as blooming season shifts to later in Cameroun-Equatorial Guinea.

Slight tendency of some clones to flower a second time in October and November.

* Reported to flower twice yearly in Uganda.

Flowering very rarely in Nov., Dec. A form of C. ringens vegetatively quite distinct from the lowlands form is found at high altitudes (5000-7500 feet) in Cameroun and Fernandos W o . This form normally flowers in May. a

540

J. ARDITTI

TABLE 33 West African Orchids Blooming During One Widely Spread Period of the Year. (Probably Day-length Neutral and Influencedby Low Temperature or Temperature Fluctuations: Sanford, 1971) ~

Genus

A. At any time during the year Epiphyte Species

Ancistrorhynchus clandestinus (Lindl.) Schltr Angraecuma chevalieri Summerh. Angraecuma pungens Summerh. Calyptrochilum christyanum (Rchb.f.) Summerh. Microcoeliaa caespitosa (Rolfe) Summerh. Polystachya caloglossa Rchb.f. Polystachya laxipora Lindl. PoZystachya supfiana Schltr

Terrestrial

+ + + + + + +

B. At any time during the growing season (Feb. or April to Oct. or Jan.)

Genus Angraecuma Chauliodona Eulophia Polystachya Polystachya

Species multinominatum Rendle buntingii Summerh. horsfallii (Barem.) Summerh. ramulosa Lindl. tessellata Lindl.

Epiphyte

+ + + +

Terrestrial

+

C. From the middle to the end of the growing season (June-July to Dec.-Jan.)

Genus

Species Epiphyte Bulbophyllum intertextum Lindl. Liparis epiphytica Schltr Polystachya cultriformis (Thouars) Spreng. a With a slight tendency to bloom twice a year, at scattered periods.

Terrestrial

+ + +

habitats, some species are long day (LD) plants; others respond to short days (SD);a number are not affected by daylength (i.e., are neutral day (ND) plants); several are induced by low temperature and a few appear to have more complex requirements (Table 35). Blooming time response is genetically controlled as suggested by observations of West African orchids (Tables 32, 33, 34; Sanford, 1971). Comparisons between hybrids (McDade, 1947) and their parents (Table 35) can also provide information on the inheritance of the response to stimuli which induce flowering. For example, the hybrid of Cattleya gaskelliana (LD at 13" or 18"C, SD at 13°C) and C. gigas (SD at 13"C), C. cv Harold flowers in early summer (McDade, 1947) which suggests that it is SD. This implies that in this case the SD characteristic may be dominant. On the other hand, C.cv Enid blooms in autumn (McDade, 1947), but is known to be unaffected by daylength (Table 35); yet it is a hybrid between C. gigas and C. mossiae both of which are SD. This

TABLE 34 West African Orchids Blooming During Two Periods of the Year. (Probably Day-length Neutral and Influenced by Low Temperature or Temperature Fluctuation: Sanford, 1971) Genus Angraecum

A. At almost any time during the year Species Epiphyte

___________

-

subulatum Lindl.

Terrestrial

__

+

B. At the beginning and at the end of the growing season (March to August and Sept. to Feb. on mature growths Genus Species Epiphyte Terrestrial -~

A ngraecopsis ischnopus (Schltr) Schltr

Angraecum Angraecum Angraecum A nsellia Bulbophyllum Bulbophyllum Bulbophyllum Bulbophyllum Bulbophyllum Bulbophyllum Cyrtorchis Diaphananthe Eulophidium Microcoelia Polystachya Tridactyle

angustipetalum Rendle distichum Lindl. podochiloides Schltr giganteu Rchb.f. var. clotica (Bak) Sumrnerh. barbigerum Lindl. calamarium Lindl. colubrinum (Rchb.f).) Rchb.f. congolanum Schltr distans Lindl. schimperanum Kraenzl. aschersonii (Kraenzl.) Schltr obanensis (Rendle) Summerh. saundersianum (Rchb.f.) Summerh. dahomeensis (Finet) Summerh. galeata (Sw.) Rchb.f. anthomaniaca (Rchb.f.) Summerh.

+ t + + +

+ + +

+

-

C. At the beginning and at the end of the growing season (March to August and Sept. to Feb.) on new growths Genus Species Epiphyte Terrestrial Bulbophyllum Bulbophyllum Bulbophyllum Eulophia Eulophidium Malaxis Polystachya Polystachya Polystachya Polystachya StOlZiQ

calyptratum Summerh. flavidum Lindl. oreonastes Rchb.f. euglossa (Rchb.f.) Rchb.f. maculatum (Lindl.) Pfitz. weberbauerana (Kraenzl.) Summerh. albescens subsp. albescens Summerh. albescens subsp. angustifolia (Summerh.) Summerh.) coriscensis Rchb.f. polychaete Kraenzl. repens (Rolfe) Summerh.

+ + + +

+

+ + +

+ + +

-

TABLE 35 Control of Flowering in Some Orchids Species Aerangis mystacidii Aerides multiflorum Arachnis cv Maggie Oei Aranda cv Deborah

Aranda cv Hilda Galistan Aranda cv Lucy Laycock Aranda cv Mei Ling Aranda cv Nancy Aranda cv Wendy Scott Blettilla striata Brassavola nodosa Brassia verrucosa Brassocattleya hybrids Bromheadia alticola B. finlaysoniana Bulbophyllum lobii Calanthe rosea Calanthe cv Veitchii

__

Factors which control or induce flowering

Referencea

~-

Bright, dry resting period Short day Salicylic acid, tri-iodobenzoic acid and coumarin. Plant growth regulators: Indifferent to daylength. Inhibited by actinomycin D and cyclohexirnide. Apical control; inhibited by auxin. Flowering gradient affected by growth regulators. Decapitation may induce flowering, indifferent to daylength. Salicylic acid, tri-iodobenzoic acid, coumarin, B995, phosphon-D and CCC enhance flowering.

Koopowitz, 1964 Bose and Mukhopadhyay, 1977 Goh, 1973, 1976a, b

Similar to Aranda cv Deborah Similar to Aranda cv Deborah Similar to Aranda cv Deborah Similar to Aranda cv Deborah Indifferent to daylight. Salicylic acid, tri-iodobenzoic acid and coumarin enhance flowering. Gibberellin treatments, 50 pprn, enhance flowering. Short days (less than 14 h) enhance flowering. Vegetative growth is enhanced by 14 h days. More flowers under short days. Low temperature stimulus. Flowering is stimulated by wet and cool days and retarded by drought. Dry periods check flower bud development. Vegetative growth is enhanced by 14 h days. Vegetative growth is enhanced by 14 h days. Vegetative growth is enhanced by 14 h days.

Goh, 1977c

Goh and Yang, 1977; Goh and Seetoh, 1973; Goh, 1973,1975, 1976a, b, 1977a, b, C, d

Goh, 1973, 1976b Sano et al., 1961 Brieger et al., 1977 Brieger et al., 1977 Sheehan et al., 1965 Holttum, 1953 Holttum, 1953; Jeyanayaghy and Rao, 1966; Sanford, 1971 Brieger et al., 1977 Brieger et al., 1977 Brieger et al., 1977

Calanthe vestita Catasetum Catt Ieya Cattleya amabilis C. cv Bow Bells C. bowringiana C . cv Dupreana C. cv Enid C . gaskelliana C . cv Geriant C. gigas C . cv Jean Barrow C. cv Joyce Hannington crosses C. labiata C . cv Los Gatos C. mossiae

C. mendelii C. cv Oenone (C. mossiae x C. Iabiata C. percivaliana

Vegetative growth is enhanced by 14 h days. High light intensities stimulate formation of female flowers; plants in shade produce male flowers. Vegetative growth under 14 h illumination. Temperatures of close to 17°C are favourable for abundant flowering. Short days. 16.) h photoperiods delay flowering. 14 h days do not prevent flowering. GA, 10 pglsheath advanced flowering by 1-2 days. Not affected by daylength and temperature. 9 h days at 13°C or 16 h days at 18°C night temperatures induce flowering. Long day plant at 1 8 T , non-photoperiodic at 13°C. GA, less than 15 pglsheath induces deformed earlier flowers. See C. warscewiczii Short days enhance flowering, but there was no peak period. See C . cv Bow Bells.

Brieger et al., 1977 Brieger, 1957; Gregg, 1975; Brieger et aI., 1977

Short days (less than 16: h) induce flowering.

Arditti, 1966c, 1967d, 1968b; Franklin, 1967; Rotor, 1952, 1959 Arditti, 1966c Arditti, 1966c, 1967d, 1968b; Rotor, 1952, 1959; Haber, 1952

Nights 8-12 h long initiate flowering. SD at 13°C and 18°C night temperature. 9 h days at 13°C induce flowering. Will not bloom under 16 h days at 13°C or 18°C night temperature; produces flowers under 9 h days at 18°C. Responds to temperature. Same as C . mossiae 16517 h light starting in late June discontinued 4 months before desired blooming date. SD (9 h) under 18°C and 18°C at night.

Tran Thanh Van, 1974 Urmston, 1949 Franklin, 1967 Arditti, 1968b Arditti, 1966c Arditti, 1968b Arditti, 1966, 1967d, 1968b; Rotor, 1952, 1959; Haber, 1952 Arditti, 1966c Arditti, 1966c Sheehan et aI., 1965 Franklin, 1967

Urmston, 1949 Franklin, 1967 Arditti, 1966c, 1967d, 1968b; Haber, 1952 continued

TABLE 35-continued Species

Factors which control or induce flowering

C. cv Pinole

Flowers form only on shoots 12-18 em long.

C. schroederae C. skinneri

Same as C. mossiae. Could be SD plant whose photoperiodic response may be modified by temperature. SD (9 h) plant under both 13°C and 18°C SD (9 h) induce flowering whereas LD (16 h) delay it.

C. trianae C. warscewiczii

SD at 13°C

Cycnoches Cymbidium

High light intensities stimulate production of female flowers. Three months of 13°C night temperature. Nights in the range of 7°C to 13°C coupled with bright, sunny days initiate flowering.

C. cv Desiree A’Legann C. cv Guelda C. lowianum C. cv Rozette

Abscisic acid did not delay flowering. Gibberellic acid causes earlier opening of flowers and larger blossoms. Flowers formed when night temperatures were 610°C. Flowers formed when plants were maintained under 14 h days, lo00 ft cdls, and 22°C and 18°C nights. Gibberellic acid increased flower size and raceme and accelerated length and flowering. Flower stalk elongation is enhanced by low temperatures. “Daylength treatments did not affect flowering. . . .” Temperatures close to 17°C are favourable for abundant flowering. Cytokinins can stimulate flowering. Their effect is enhanced by gibberellins. Flowers when grown warm and moist.

C. cv Sicily “Grandee” C. virescens Dendrobium D. cv Anne Marie

Referencea Rotor, 1952, 1959; Urmston, 1949; Johnson and Laurie, 1943, 1945 Urmston, 1949 Rotor, 1952 Haber, 1952; Rotor, 1952, 1959; Urmston, 1949; Holdson and Laurie, 1949 Rotor, 1952, 1959; Urmston, 1949 Gregg, 1975 Davidson, 1977; Arditti, 1966c, 1967d, 1968b; Rotor, 1952, 1959; Went, 1946, 1951 Brewer et al., 1969 Bivins. 1970 Arditti, 1968 Bivins, 1968 Sheehan et al., 1965; Van der Donk, 1974; Goh, 1977c

D. comatum D. crumenatum

Flowers one day before D. crumenatum. Already developed flower buds are stimulated to develop and open.

D. draconis D. jindlayanum D. formosum D. heterocarpum D. infundibulum D. cv Jaquelyn Thomas D. cv Lady Fay D. cv Merlin D. nobile

Prefers cool climate. May require cold temperatures for flower initiation. Plants flower after long dry season. See D. findlayanurn. Periods of low temperature induce flowering. Not induced by light. Not induced by light. Blooms when grown warm and moist. Flowering is induced by low temperatures.

D. phalaenopsis D. scabrilingue D. cv Thwaitesiae Epidendrum radicans

Short days and 18°C or 13°C night temperature induce flowering. See D . infundibulum. Lowering of temperature at night may induce flowering. Flowers initiated in October 1970, developed to pollen formation in early January 1971, and flowered on 15 January 1971. Behaves like Dendrobium crumenatum. Burning of grasslands may induce flowering. Alternation of wet and dry seasons may regulate flowering.

Eria Eulophia cuculata Grammatophyllum rumphianum Laelia albida Laelia purpurata Laeliocattleya cv Canhamiana Miltonia Miltonia anceps

Arditti, 1968 Holttum, 1953, 1969; also see text Kamemoto and Sagarik, 1965 Arditti, 1966c Sanford, 1971 Kamemoto and Sagarik, 1965 Sheehan, et al., 1965 Sheehan, et al., 1965 Arditti, 1966c Arditti, 1966c; Kosugi, 1952; Rotor, 1952, 1959 Rotor, 1952, 1959 Arditti, 1966c Kosugi, et al., 1973 Smith, 1926, 1927 Sanford, 1971 Holttum, 1957; Sanford, 1971

Short days may enhance flowering. LD plants. Daylength of no less than 16 h induces flowering.

Brieger et al., 1977 Arditti, 1968b Hampton, 1955

Temperatures close to 17°C are favourable for abundant flowering. Long days enhance flower formation.

Tran Thanh Van, 1974 Brieger et al., 1977 continued

TABLE 35-continued Species Miltonia ioezlii Miltonia spectabilis Mormodes Odontoglossum bictonense Odontoglossum citrosmum (0.pendulum) Odontoglossum hybrids Oncidium sphacelatum 0. splendidurn Paphiopedilum P. insigne Phojus tankervilliae Phalaenopsis

P. amabilis P. schilleriana Polystachya cultriformis Renanthera imschootiana Rhynchostylis gigantea R. retusa Thrixspermum Vanda Vanda cv Miss Joaquim

Zygopetalum

Factors which control or induce flowering

Referencea

Long days enhance flower formation. Long days enhance flower formation. See Catasetum and Cycnoches. Long days enhance flower formation. Periods of low temperature initiate flowering.

Brieger et al., 1977 Brieger et al., 1977 Dodson, 1962 Brieger et al., 1977 Arditti, 1966c.

“. . . daylength is an important factor in the production

Baker, 1968 Kosugi et al., 1973 Franklin, 1967 Franklin, 1967; Davidson, 1977 Rotor, 1952, 1959 Bose and Mukhopadhyay, 1977 Franklin, 1967; Halperin and Halevy, 1974, 1975; Halevy, 1975; Tran Thanh Van, 1970, 1974; Nishimura et al., 1972, 1976 Rotor, 1952, 1959 Rotor, 1952, 1959; De Vries, 1953 Sanford, 1971 Bose and Mukhopadhyay, 1977 Inthuwong and Watcharaphai, 1964

of spikes.” Floret initiation in late December in Japan. Flowers are initiated during short cool days. No photoperiodic response. Night temperatures of 13°C for 2-3 weeks. No photoperiodic response. Night temperatures of 13°C for 2-3 weeks. Best flowering occurred under lo+ and 13t hours. No photoperiodic response. Flowering induced by 3 weeks of 14°C nights and 20°C days. Flowering induced by low (10-15°C) temperatures. Floral initiation takes place when night temperature varies between 12°C and 17°C and day temperatures not exceed 27°C. Treatment time: 2-5 weeks. Flowering is induced by short days and 18°C. See P. amabilis. Flowering is stimulated by 2-3 weeks exposure to night temperatures below 21°C. Same as Phalaenopsis schilleriana. Short days (9 h) induced earlier flowering. Short days (8 h) at low temperature (10°C for 16 h/day) induce early flowering. See Renanthera imschootiana. A sudden drop in temperature may initiate flowering in some species. Weekly sprays of 10 ppm gibberellins induce flower formation. Longer periods in sunlight brought about more profuse flowering; endogenous gibberellins “. . . showed no obvious correlations . . . auxin levels in the shoot apex may be responsible”. See Vanda.

a Several reviews (Arditti, 1966, 1967, 1968; Rotor, 1952, 1959; Sanford, 1971) are cited for simplicity.

Holttum, 1957 O’Neill, 1958 Goh and Wan, 1973; Murashige et al., 1967 O’Neill, 1958

TABLE 36 West African Epiphytic Orchids Demonstrating Differences in Blooming Time Response Which May Be Genetically Controlled (Sanford 1971)

Genus

Species

Ancistrorhynchus Ancistrorhynchus

cupitatus (Lindl.) Summerh. clandestinus (Lindl.) Schltr

Ancistrorhynchus

struussii (Schltr) Schl tr

Angraecum

mirltinominatum Rendle

Angraecum

srrbulatum Lindl.

Ansellia

gigantea Rchb.f. var. nilotica (Bak.) Summerh. distuns Lindl.

Bulbophyllum

Culyptrochilum Chamaeangis

christyanum (Rchb.f.) Summerh. lanceolata Summerh.

Blooming period (Normally Sept .-Nov.) Jan.

Original source Fernando Po0 collections (consistently 3 years after transplant) Wamba, Iseyin, Upper Ogun, Gambari Kumba (Cameroun) Idanre, Olokemeji Erin-Ode, Ilesha, Akure

May-June March March-April June-Aug. (Normally June ; Aug.1 Oct.-Nov. June-Sept . March- May April-May; Oct. March-April-June ; Jan.

Gindiri, Olokemeji

Dec.-Jan. March-April March-Apri I June

Mongu, Nubi Mongu, Vom Sapoba (consistent after 4 years) Fernando Po0 (consistent after 2 years)

Dec.-March April-July July-Aug. NOV.-&C.

Gambari, Akure, Olokemeji Gambari, Akure, Olokemeji Mt. Cameroun (consistent after 4 years) Gambari, Sapoba

Sapoba Sapoba Gambari, Olokemeji, Lagos Idanre, Akure, Gambari, Kumba (Cameroun)

continued

TABLE 3 k o n t i n u e d

Genus

Species

Cyrtorchis

ringens (Rchb.f.) Summerh.

Diuphmmthe

plehnianu (Schltr) Schltr

Graphorkis

lurida (Sw.)Kuntze

Microcoelia

caespitosa (Rolfe) Summerh.

Microcoelia

dahomeensis (Finet) Summerh.

Polystachya

mukandaensis De Wild.

Blooming period (Normally Sept.-Nov.)

Original source

May; Nov. Sept.-Oct. Oct .-Nov. April-July March Jan-Feb. Once blooming, June. (Normally twice blooming)

Mt. Cameroun Agoi-Ibami, Sapoba Sapoba (consistent after 4 years) Sapoba (consistent after 4 years) Mongu (consistent after 2 years) Ibadan, Idanre, Enugu-Ikom Idanre (only 1 clone)

Dec.-Feb. June-July June-Jul y

Gambari Gambari Erin-Ode, Olokemeji, Upper Ogun, Iseyin Erin-Ode, Ehor, Agoi-Ibami, Sapoba, Ijebu-Ode, Oru

Aug.-Sept.

0 Independence day sepals

0 Chelsea sepals

petals 0 Chelsea petals

Independence day ovary

0

o--OChelsea

ovary

‘ A

Newly opened

Bud

lays

Fully opened

I

I

I

0

3

7

Fig. 59. Net 14C02fixation by flowers in the light as a function of age in Cymbidium cv Chelsea and C. cv Independence day “Yorktown”. The horizontal bar, representing 14C0,fixation by a mature leaf, is included for comparison purposes (Dueker and Arditti, 1968). See text, p. 530.

Liqht I

I

’5

-9

9am

I

,

I

I

I

3pm

6pm

9pm

12om

3am

6am

I 12pm

I

I

I

I

I

3pm

6pm

9pm

12om

30m

I

-6 90m

,

1

12pm

Liqht 6am

Fig. 60. Diurnal CO, gas exchange of Arunda leaf and aerial root (Hew, 1976). See text, p. 532.

550

J. ARDITTI

fact would seem to indicate a more complex pattern of inheritance than in C. cv Harold. A statement that "the blooming time of the hybrid will vary according to the strength of the influence of the parents . . ." (McDade, 1947) may imply inheritance of dominant and recessive characters, but it is clearly insufficient. More research is needed in this area. The existence of numerous orchid hybrids with carefully recorded parentages holds the promise of a fruitful undertaking. Valuable information may even be obtained from careful analysis of available data and horticultural practices. Some of the latter are unpublished having been arrived at through trial and error by commercial growers. One of the more interesting flowering habits among orchids is that of Dendrobium crumenatum, the Malayan pigeon orchid (Holttum, 1953, 1969). All plants in a certain area flower simultaneously, but the blooms last a very short time. This gregarious flowering, which seems to ensure pollination has interested investigators for a long time (Arens, 1923; BeumCe, 1927; Burkil, 1917; Coomans de Ruiter, 1930; Coster, 1925; Kuijper, 1931, 1933; Massart, 1895; Rutgersand Went, 1915; Seifriz, 1923; Smith, 1926, 1927; Treub, 1887; Went, F.A.F.C., 1898, 1917; Went, F.A.F.C. and Rutgers, 1915; Went, F.W. 1930; for reviews see Arditti 1966c, 1967; Holttum, 1953, 1969). The inflorescences of Dendrobium crumenatum are found on the terminal portion of stems covered by sheaths which are modified leaves. These inflorescences remain very short and produce several flowers each protected by a bract. Every one of the flower buds develop until all its parts are formed and the anther is ". . . almost fully grown" (Holttum, 1969). Then development ceases until reactivated by an approximately 5°C drop in temperature. In Malaya and Indonesia such a sudden drop may be brought about by a rainstorm and the subsequent evaporative cooling of the buds. Gradual cooling, if it lasts long enough (about 24 hours) can also induce flowering. Exactly nine days after the cooling (in the great majority of cases; I have also been told of eight days and counted ten for a few flowers in one case), just before dawn the flower (white with a tinge of pink) opens and starts emitting a delightful fragrance. Pollinators attracted by the combination of colour, fragrance and great masses of flowers, arrive soon after dawn. Even if the stimulus which promotes development of the flower buds has been discovered, the underlaying mechanism is still unknown. An outstanding feature of flowering in Dendrobium crumenatum is the dependence on temperature which in turn ensures an adequate water supply (Holttum, 1953). In Singapore, for example, the absolute minimum and maximum are 21°C and 34°C respectively (70"-93°F). The normal daily range may be 23"-32°C (sunny days) and 24"-27°C (rainy days). In May, the hottest month, the mean temperature is only 1.75"C (3°F) higher than in December, the coolest period (all data from Holttum, 1953). Altogether temperature fluctuations are not a major aspect of the climate. Further, drops

ASPECTS OF ORCHID PHYSIOLOGY

551

of 5°C or more, in which the temperature does not rise above 27°C are associated with rain. Thus, an ample supply of water is assured before the development of numerous blossoms because it is also the agent which brings about the drop in temperature required for flower bud development. Dependence on cooling in equatorial climates may appear to be a risky evolutionary strategy, but this is not the case. Other orchids have evolved similar mechanisms. They include Bromheadia alticola, B. jinlaysoniana (seven days from cooling to flowering) as well as several Dendrobium species some of which flower one day before D . crumenatum and others which bloom one or two days after it (Holttum, 1953; Smith, 1926). In addition, several non-orchidaceous flowers also respond to cooling including the large deciduous leguminous Malayan native tree Pterocarpus indicus; the Central American member of the Cactaceae, Epiphylhm oxypetalun and the West Indian bulbous plant Zephyranrhes rosea (Holttum, 1953). Three orchid genera, Catasetum, Cycnoches and Mormodes, can bear male, female or hermaphrodite flowers. In Catasetum and Cycnoches a plant may produce all three types of flowers on the same or different racemes during one or a number of flowering seasons. At first this phenomenon led to considerable taxonomic confusion in that plants were assigned to different genera or species. The classification problems were resolved at the turn of the century when plants in English greenhouses produced all three types of flowers. However, the physiological enigma remains despite rather interesting recent observations (Gregg, 1973, 1975, 1979). Plants of Catasetum or Cycnoches which are robust or grown under full sun produce female flowers whereas those that are smaller or maintained in shade bear male blossoms (Gregg, 1975, 1979). Production of female flowers is accompanied by increased ethylene evolution which can be inhibited by placing the plant under shade or covering the raceme tip with a foil cap (Gregg, 1973, 1975). This observation led to the assumption that femaleness is induced by ethylene. However, this does not seem to be the case since ethylene treatments and auxin applications have failed to induce the production of female flowers (Gregg, 1973). Thus it appears that ethylene evolution may be (i) a separate phenomenon brought about by strong light (a stress response?), independent of female flower induction, (ii) a byproduct of female blossom formation, or (iii) a combination of the two. The stimulation of female flower production in Catasetum and Cycnoches, by high light intensities, is an important adaptive feature because it apparently ensures sufficientphotosynthesis to support seed production. Seed capsules of Cycnoches have been reported to contain 4 000 000 seeds (Arditti, 1967a), whereas in Catasetum estimates are as high as 2000000 (Gregg, 1975). Clearly, the production of such large numbers, even of very small seeds, requires considerable amounts of photosynthates. Blossoms of Cutasetum and

552

J. ARDITTI

Cycnoches are green. Therefore, it is possible that like Cymbidium flowers (Arditti and Dueker, 1968; Dueker and Arditti, 1968) they are capable of photosynthesis and can contribute to their own energy needs. However, under low light intensities the amount of photosynthesis may be insuflkient. Consequently, a mechanism which allows for seed production only when there is enough energy to support it is of survival value. C. POLLINATION

Orchid blossoms are morphologically different from other flowers in several important details. Petals (inner whorl) and sepals (outer whorl) may be of the same colour and often bear considerable resemblance to each other (Fig. 61). The median petal is usually modified considerably and serves as a landing platform for pollinators. It is called the lip or labellum. Above the labellum (Fig. 61B, C) in the open flower of most orchids (and below it in a few) is a structure called the gynostemium (Fig. 61D, E) which represents a fusion of stamens (anthers and filaments), stigmas and styles. Monandrous orchids have only one fertile stamen near the apex of the column. Their pollen grains are united into tetrads and compressed into 2 , 4 , 6 or 8 pollinia which are attached to each other by a viscid disc called the viscidium (Fig. 61D, E). The pollinia are located on the rostellum (a modified stigma) below an anther cap (Schultes and Pease, 1963; van der Pijl and Dodson, 1966). Jn the unopened buds of monandrous orchids the labellum is above the gynostemium. However, as the flower opens it twists in a process known as resupination (Fig. 62; a, b, c, d ; Arditti, 1966d; Schultes and Pease, 1963; Ames, 1938; van der Pijl and Dodson, 1966; Zimmerman, 1932) and the labellum becomes ventral to the gynostemium (Figs 61 ; B, C; 62; a, b, c, d). But, this is not always the case. For example, Satyrium flowers do not resupinate and blossoms of Angraecum eburneum resupinate 360”. In diandrous orchids (Cypripedioideae, including the genus Paphiopedilum) the flowers are nodding and the pouch-like labellum points downward (Fig. 63). Their dorsal sepal is modified and is either erect or bent slightly forward. Anthers are located one each at either side of the staminode. The stigma is located below the anthers (Fig. 63). Orchid flowers have evolved highly specific, often bizarre pollination mechanisms (Dodson, 1967, 1975; Dodson and Frymire, 1961 ; Dodson et al., 1969 ; Kullenberg, 196 1, 1973 ; Kullenberg and Bergstrom, 1976 ; Poyanne, 1917; van der Pijl and Dodson, 1966). In all cases these mechanisms involve “manipulation” of the pollinating vector to ensure that it will carry away pollinia and subsequently deposit them on another flower (Figs 64-73). This is accomplished in a number of ways. Nectar is a major attractant in orchids (Baskin and Blis, 1969; Jeffrey and Arditti, 1968, 1969; Payne, 1964, 1968; Sunding, 1963; for reviews see

Fig. 61. Diandrous orchid flower. (A) Ground plan of an orchid flower at transitional stage. (B) Cymbidium flower, front view. (C) Cymbidium flower, side view. (D) Cymbidium gynostemium intact (top), and following removal of pollen (bottom). (E) Brussiu, column and part of ovary. (A, E) from Schultes and Pease, 1963; (B-D) from Arditti (1966d). Explanation of symbols: ds, dorsal sepal; la, labellum or lip; r, rostellum.

Fig. 62. Resupination and its result. (a, b) Vanda flower resupinating so that the lip, which is on top in the bud stage is on the bottom of the flower. (c, d) Resupinated Cattleyu (c) and Cymbidium (d) flowers (Arditti, 1966d).

ASPECTS OF ORCHID PHYSIOLOGY

555

Fig. 63. A diandrous orchid, Paphiopedilum villosum. (A) Whole flower. (B) Staminode (right), anther (centre), and stigma (left, below staminode). (C) Staminode, view from above (Williams, 1877). Explanation of symbols: a, anther; ds, dorsal sepal; PO, pouch; s, staminode; st, stigma.

Jeffrey et al., 1970; van der Pijl and Dodson, 1966). It is produced by many species and attracts a variety of pollinators including bees, moths and birds. One of the better known cases is that of Angraecum sesquipedale (Fig. 65) because Darwin predicted the pollinator on the basis of floral characteristics and spur length. The moth inserts its proboscis into the spur in search of nectar and in the process picks up the pollinia. On the next visit to another flower the pollen is deposited in the stigma. Fragrances, often confined within structural features, are not only important attractants for orchid pollinators, but also ensure specificity (Dodson and Hills, 1966; Dodson, 1975; Dodson el a/., 1969; Hills et a/.,

Fig. 64. Attachment of the pollinium to the pollinator. The column lies above the lip, forming a narrow tube through which the pollinator must enter. When the pollinating bee leaves, it must back out (a). In the process, the viscidium attaches the pollinium to its back; the anther cap is dislodged (b, c). When the bee enters another flower, the pollinium is pressed into the pocket-like stigma and retained thus fertilizing the flower (Stephens and North, 1974).

Fig. 65. Angraecum sesquipedale Thou and its pollinator Xanthopan morganii praedicta (not to scale, but proboscis length is about 30cm and equal to that of the spur). (a, b) pollinated and unpollinated flower showing the spur changes in sepals and lip. (c) Xanthopan morganii praedicta with proboscis fully extended (Straws and Koopowitz, 1973).

Fig. 66. Pollination of Cutuseturn. Trigger mechanism of Cutuseturn. (a) Frontal view; the broken lines show the position of the pollen-bearing pollinium under the anther cap. (b) Sectional side view of the interior of the flower. The trigger, when touched by a bee, releases the sticky viscidium, which twists as indicated by the upper arrow and lodges against the bee’s back. Hence, the departing bee carries the pollinium. The anther cap falls away. (c) Male bee, Euluernu cingulutu,after having triggered off the pollen of Cutuseturn platyglossurn. (d) The bees crawling inside the hood shaped labellum of a female flower of Catuseturn platyglossurn [(a, b) from Arditti, 1966b; (c and d) from van der Pijl and Dodson, 19661.

Fig. 67. Pollination of Gongora. (a) Bee gets a toboggan-like ride down the column and strikes the viscidium, which adheres to it. When the bee flies off (b) it carries the pollinia. At right (c) the bee, still carrying the pollinia, enters another flower and gets another ride down the column. On this visit the pollinia adhere to the stigma and the flower is pollinated. (d) Male bee, Euglossa viridissima pollinating Congora armeniaca. (e) Male bee, Euglossa dodsani pollinating a flower of Gongora harichiana [ (a, b, c) from Arditti, 1966b; (d, e) from van der Pijl and Dodson, 19661.

Fig. 68. Pollination of Coryunfhes speciosu. (a) Dotted line, followed from right to left, shows the path taken by the bee as it arrives at the flower and falls into the bucket, which contains a liquid produced by the gland at top centre. The only way out for the bee is through a narrow opening that forces the bee against the pollinia. (b) Male bee, Euglossa superba, scratching the lip of a Coryanthes rodriguezii flower. (c) Bee, Euglossa superbu, with pollinia on its abdomen has forced up the anther cap of a Coryanthes rodriguezii and is struggling free [ (a) from Arditti, 1966b; (b, c) from van der Pijl and Dodson, 19661.

Fig. 69. Pollination of Oncidium planilabre by a male bee, Centris geminata. (a) Bee striking the flower. (b) Bee with pollinia attached to the front of its head. (c) Bee strikes the flower, pollinium is attached to its head and anther cap is dislodged. (d) Pollinium bends downward. (e) Bee strikes another flower and pollinium is forced into the stigma [(a, b) from van der Pijl and Dodson, 1966; (c, d, e) from Stephens and North, 19741.

c

.-

10 Fecondation des Ophrys fusca, lutea et speculum. 1, Ophrys fusca, fleur vue de face; 2, coupe du labelle (grossi); 3, 0. lufea, fleur vue de face; 4, coupe du labelle (grossi); 5, fkondation operee par un petit hymknopttre (coupe du labelle); 6, l’hymenopttre s’envole avec les 2 pollinies fixees B I’abdomen; 7, 9. speculum, fleur vue de face; 8, coupe du labelle (grossi); 9, fkcondation operee par le male du Colpa auren (coupe du labelle); 10, le Colpa s’envole avec les pollinies fixbes sur la tCte. a, petit coussinet garni de poils courts sur lequel frotte l’abdomen de I’insecte; b, tache bleu metallique; c, cavite correspondant ti I’eperon des Orchis, dans Iaquelle plonge I’abdomen de I’insecte; i, pollinies; p , pilosite fauve entourant le labelle (poils epais et longs); t, petales. Fig. 70. Pseudocopulation in Ophrys. Poyanne’s original drawing and caption (Ames, 1948; Poyanne, 1916).

ASPECTS OF ORCHID PHYSIOLOGY

563

Fig. 71. Pseudocopulation in Ophrys. (a) Head of a Gorytes cumpestris male with pollinia of 0 . insectifera (left) and Listera ovatu (right, without massulae). (b) Head of a Euceru nigrilubris male with pollinia of 0 . fenthrediniferu and probably of an Orchis sp. (c) The common position of Eucera sp. male on the labellum of Ophrys bombyliflora (all from Kullenberg, 1961).

1968; van der Pijl and Dodson, 1966). In Catuseturn the pollinating bee is attracted by the fragrance to a male flower and while searching for its source triggers the pollen release mechanism (Fig. 66a, b, c). On a subsequent visitto a female blossom it deposits the pollen (Fig. 66d). Gongora flowers also attract pollinators with fragrance and have evolved a unique pollination mechanism. After scratching at the base of the lip the bee backs up (and ends up “taking” a toboggan-like “ride”) and picks up the pollinia on its back. On a subsequent visit the process results in the deposition of the pollinia into a stigma (Fig. 67). Flowers of Coryanthes are remarkable even for an orchid (Fig. 68a). Its sepals and petals fold out to form sail-like structures. The top consists of a

564

J. ARDITTI

Fig. 12. Pseudocopulation in Ophrys. (a) Andrena squalida on the labellurn of Ophrys arachnitiformisGren. et Phil. (b) Two males of Andrena squa/idajostling on the labellurn of 0. arachnitiformis.(c) Median longitudinal section of Ophrys bombyliflora Link ; note construction of distal part of labellum. (d) Ophrys bombyliflora, three flowers from different sides. Note construction of the distal part of the lab:llum (photographs and drawing coJrtesy Prof. B. Kullenbxg).

hood shaped or globular hypochile, an elongated midsection called the mesochile, and at the bottom a bucket shaped epichile. Two glands at the base of the column drip water into the bucket and partially fill it before the flower opms. Once the flower has opened Euglossa, Euphlusia or Eulaema bees, attracted by the fragrance produced by the hypochile land on it and scratch. While doing so they fall into the bucket and escape only through a tunnel formed by the apex of the column and the epichile. As the bees crawl out the pollinia are deposited on their abdomens (Fig. 68b, c). On a succeeding visit the pollinia are deposited into the stigma (van der Pijl and Dodson, 1966). The reasons why pollinating bees (which are males) collect odour sub-

Fig. 73. Pseudocopulation in Ophrys. (a) Eucera oraniensis Lep male on the labellurn of an Ophrys bombylijora Link flower. (b) Andrena sp. male on the labellurn of an Ophrys fusca Link flower. (c) Eucera longicornis L male on the labellurn of an Ophrys upifera Huds flower. (d) Colletes cunicularius infuscatus Nosk male on the labellurn of an Ophrys sphecodes provincialis Nelson flower (photographs courtesy of Professor B. Kullenberg).

566

J. ARDITTI

stances has been clarified only recently. They collect the fragrances in order to attract other male bees of the same species and form leks. These in turn attract females and mating takes place (Dodson, 1975). Pollination of Oncidium planilabre by Centris bees is based on simulation of an enemy insect by the flowers. As a consequence the blossoms are attacked by the bees and pollinated (Fig. 69). “Fear” of death, “hate” for an enemy and hunger are not the only “drives” “utilized” by orchids to ensure pollination. Sexual “passion” is also “employed”. Flowers of the genus Ophrys mimic the females of certain aculeate Hymenoptera and also produce volatile compounds which excite the males sexually and cause them to attempt copulation with the blossoms (Figs 70-73; Godfery, 1925; Kullenberg and Bergstrom, 1976; Poyanne, 1917). While they do so they pick up pollen (Figs 70-73) and subsequently deposit it on another flower. And, when all else fails orchids are capable of self pollination (Fig. 74).

Fig. 74. Self-pollination is shown in Ophrys. Before pollination (a) the column stands upright. As a result of autolysis, or self-digestion of certain tissues by the plant, the stipe bends (b), carrying the pollinia downward towards the sticky stigmatic surface. When the pollinia reach the stigma and adhere to it (c), pollination is complete (Arditti, 1966d).

D. POST-POLLINATION PHENOMENA

Ever since Aristotle first described orchids, their flowers have fascinated people, mostly because of their forms. They were assumed to be the origin of some animals and to have arisen from the semen of others (Arditti, 1972d; Fig. 75). Even science fiction writers have not been immune to their charms (Adams and Nightingale, 1976; Boyd, 1969; Clarke, 1974; Wells, 1960). The shapes of orchid flowers, their fascinating mimicry, fragrance production, food offering and colour combination all accomplish the same end : attraction of pollinators.

ASPECTS OF ORCHID PHYSIOLOGY

567

Fig. 75. Illustration by Lady Grey of Groby in James Bateman’s “The Orchidaceae of Mexico and Guatemala” (1843). It is captioned: “The hag came forth, broom and all, from a flower of Cypripedium insigne; her attendant spirits are composed of Brassia Lanceana, Angraecum caudatum, Oncidium Papilio, etc., etc. ; two specimens of Cycnoches sail majestically on the globe below, on the right of which crawls Megaclinium falcatum. In the centre, stands a desponding Monachanthus; on the left a pair of Masdevallias are dancing a minuet, while sundry Epidendra, not unlike the ‘walking leaves’ of Australia, complete the group.”

Relations between orchid species and their pollinators are most often very intimate. Each orchid is pollinated only by a specific vector (Darwin, 1904; Dodson er al., 1969; van der Pijl and Dodson, 1966). Hence the attraction mechanisms are as remarkable as they are varied and as fascinating as they are complex. But, this specificity can also constitute a serious drawback since the available pollinators may be limited in number thereby reducing the number of pollinated flowers. Indeed, several observers have noted that often a very small proportion of orchid flowers are pollinated. It becomes reasonable to assume, therefore, that survival of many orchid species would require

568

J. ARDITTI

not only adaptations that attract pollinators, but also increased longevity which would provide a longer “waiting” period. Indeed, it is not unusual for unpollinated orchid flowers to remain alive for a long time (Table 37). For example, it is said that flowers of Grammatophyllum multiforum may last nine months (Kerr, 1972); Paphiopedilum flowers can live 3-4 months and it is not unusual for those of Cattleya, Cymbidium and Phalaenopsis to remain fresh for several weeks. Few species have very short-lived flowers: Dendrobium appecdiculatum is reported to last a hard-to-believe five minutes, whereas D . crumenatum flowers live a day or less. Producing complex flowers and pollinator attraction systems and maintaining them for long periods is expensive in terms of energy resources. No wonder, then, that orchid flowers have evolved intricate mechanisms for the conservation of energy; utilization of substances from floral segments which have carried out their functions; photosynthesis by flower parts before or after pollination; cessation of scent and nectar production immediately after pollination and fast wilting. Therefore, flowers which can contribute to their own upkeep have an evolutionary survival advantage. Many orchid flowers are coloured by chlorophyll and are, therefore, green (Arditti, 1966e; Matsumoto, 1966). At least in one instance, green Cymbidium flowers have been shown to be capable of photosynthesis (Arditti and Dueker, 1968; Dueker and Arditti, 1968). This unique adaptation, which allows blossoms to contribute towards their own energy needs, most probably exists in other green flowered orchids. In these flowers pollination is the beginning of the end for the perianth. In other flowers one or more floral segments turn green and apparently become photosynthetic. If a generalization can be made about pollination-induced events (i.e., post-pollination phenomena) in orchid flowers it is that here evolution has favoured conservation. It is difficult to pinpoint the earliest description(s) of post-pollination phenomena. Conrad Sprengel ”. . . struck a new path in botanical science. . .” (Miiller, 1883) with his book “Das Entdeckte Geheimnis der Natur im Bau und in der Befruchtung der Blumen” (Sprengel, 1793) where he may have described for the first time the post-pollination phenomena in orchids. The next reports appeared soon thereafter (Bulliard, 1802; Wachter, 1799-1 801). Ludolf Christian Treviranus (1779-1 864), the German botanist who made early observations of the movement of cell contents also wrote about orchid flowers. He studied structures associated with pollination and its results (Treviranus, 1827). After that a number of studies and reports were either directly concerned with Orchidaceae (Brongniart, 1831 ; Gosse, 1863; Anderson, 1863; Riviere, 1866, 1872; Veitch, 1887; Gruignard, 1886; Anonymous, 1890,1894; Hurst, 1898; Anonymous, 1899; Malguth, 1902; Leimbach, 1911; Resvol, 1911; Morita, 1918; Pohl, 1927; for an early review see Miiller, 1883) or a number of plants including orchids (Cruger, 1851; Rossner, 1923). One observation of post-pollination phenomena led to a major biological

569

ASPECTS OF ORCHID PHYSIOLOGY

TABLE 37 Longevity of Orchid Flowers (Fitting, 1909a; Kerr, 1972; Morita, 1918; Poddubnaya-Arnoldi and Selezneva, 1957) -

Genus or species Angraecum Calanthe Calanthe discolor Cattleya Coelogyne Cymbidium virens Cypripedium

Longevity, days up to 40

Genus or species ~

up to 30 14 UP to 45-60 up to 21 14-25 up to 60 or even 90 Dendrobium up to 19 Dendrobium crumenatum 1 Dendrobium superbum 14 Epipactis erecta 8-10 E. falcata 8-12 E. papillosa 7 E. thunbergii 7-10 Crammatophyllum multiforum 270 Gymnadenia cucullata 8-1 0

~

Megaclinium Odontoglossum Oncidium Paphiopedilum Peristeria elata Phalaenopsis Phal. amabilis Phal. violacea Platanthera yatabei Rhynchostylis retusa Sobralia macrantha Spiranthes australis Stanhopea Vanda

Longevity, days several up to 20 or even 80 up to 26 or even 60 9-120 1-2 up to 35 up to 120 30 7-10 30 1 10 3 up to 60 or even 90

advance: the discovery of the cell nucleus (Brown, 1831, 1833, 1834; for a review see Arditti, 1977b). A second barely missed discovery was that of auxin. Reports from Brazil by Fritz Muller (Fig. SOD) that orchid pollen is poisonous were accepted uncritically by Darwin and this resulted in general acceptance (Hildebrand, 1868; Magnus, 1887; Moller, 1920-1921; Muller, 1868, 1869). However, a reinvestigation in Bogor using Phalaenopsis flowers convinced Hans Fitting (Fig. 8OC) that he was dealing with a special substance (Fitting, 1909a, b, 1910, 1921). Fitting called that substance Pollenhormon and thereby became the first person to even use the term “hormone” relative to plants (for reviews see Arditti, 1971a, by 1975a). Others after Fitting showed that his Pollenhormon was the auxin indoleacetic acid by studies of post-pollination (“post-floration”) phenomena (Laibach, 1930), comparisons between pollinia and growth substances (Laibach, 1932, 1933a, b; Laibach and Maschmann, 1933; Mai, 1934; Maschmann and Laibach, 1932), and application of auxins to orchids (Hubert and Maton, 1939). This has been confirmed by qualitative determinations which have shown that orchid pollen may contain as much as 1OOpg indoleacetic acid g-l (Muller, 1953). These findings have been confirmed recently (Klass, 1964). However, a report that pollinia of Cuttleya cv Enid contain naphthaleneacetic acid (Klass, 1964) requires careful confirmation. This auxin (NAA) is a synthetic substance whose isolation or even tentative identification from a

570

J. ARDITTI

natural source requires more information than used to suggest its existence (Klass, 1964). Despite these reports, Fitting (personal communication) never accepted the idea that Pollenhormon was IAA. He maintained that his was a separate and different hormone. In a way his claim may have a basis because he and everyone else who worked with pollen extracts and diffusates did not have pure IAA. Rather, the diffusate may have contained cytokinins, gibberellins (unpublished results from my laboratory) and possibly other substances. Hence, results obtained from assays of extracts or diffusates might be those that can be expected from a mixture of substances, not a pure hormone. The effects of such extracts could well have been different enough from those of pure auxin to convince Fitting that he was dealing with a different hormone. These considerations in no way detract from Fitting’s achievement. His conclusions were based on what was known during the period he was active, and this was before the discovery of auxins, cytokinins and abscisic acid, rediscovery of gibberellins and appreciation of the effects of ethylene. In fact, Fitting’s retirement preceded the important early work on these hormones all of which are known to have effects on post-pollination phenomena of orchid flowers (Arditti, 1976b). A third major research area in which early studies of post-pollination phenomena played a significant role is the effect of pollen on the initial stages of fruit formation (i.e. ovary swelling) and ovule formation (Hildebrand, 1863a, b; 1865). I . Phenomena With 20 000-30 000 species and many widely different, often bizarre, pollination mechanisms, it is only reasonable to anticipate numerous, often unique, post-pollination phenomena. Indeed, this is the case and the phenomena can be divided into a number of categories (Tables 38, 39). Unfortunately, however, reports, especially those on visual changes, are widely scattered through the systematic, ecological, pollination, horticultural, physiological and hobby literature and often difficult to trace. One must simply search through many articles, observe flowers and trust in luck and serendipity. In addition, no orchid has been investigated fully to determine which and how many phenomena it may exhibit. Therefore, compilations are in fact a list of examples (Tables 38, 39). Angraecum sesquipedale flowers are large, star-shaped, creamy-white and fragrant in the late afternoon and evening (Fig. 76D); on being pollinated the perianth segments turn yellow and scent production ceases. The dorsal sepal starts to bend down (Fig. 76E) and shortly thereafter the lateral sepals and petals bend inwards (Fig. 76F), but the lip does not move very much. Eventually the entire flower closes (Fig. 76F) and the perianth dies. In A . eburneum only the lip folds (Fig. 76A, B, C), but the perianth dies as in A . sesquipedale.

TABLE 38 Post-pollination Phenomena of Orchid Flowers Classified by Physiological or Developmental Categories Examples

Category Redifferentiat ion

Growth Nastic movement Termination of activities Morphogenesis and development Protein synthesis RNA synthesis Metabolic activity Transport

Relation to environment

Greening of sepals, petals, ovaries and gynostemia as well as changes in their nature Swelling of ovaries Hyponasty of petals and sepals Cessation of scent production Ovule development Changes in protein complement and levels Increased RNA levels in gynostemia Hydrolysis of storage and structural compounds Mobilization of substances into gynostemia and ovaries Changes of UV light reflection by flowers

Changes in pedical curvature

Closing of stigmas

Swelling of columns

Loss of curvature by columns

Changes in shape of calli

Ethylene evolution

Changes in respiration patterns Changes in protein levels

Yellow colour production by perianth

Chlorophyll synthesis or destruction

Hooding of dorsal sepals Cessation of nectar exudation Senescence and death of perianth

Anthocyanin production or destruction Altered patterns of phosphate movements Changes in flower colour

Starch accumulation in ovaries

TABLE 39 Post-pollination Phenomena in Orchid Flowers Occurrences, b

Phenomenon Cattleya

Senescence and death of perianth segments

Arditti, 1969; Hayes, 1968

Cymbidium

Arditti, 1969, 1976%b

Vanda

Burg and Dijkman, 1967; Dijkman and Burg, 1970

Greening of all or some perianth segments Swelling of column

Hayes, 1968

Swelling of ovaries

Arditti, 1969

Hyponasty of sepals and petals Changes in pedicel curvature Stigmatic closure Hood formation or folding by dorsal sepal Greening of gynostemia Termination of scent production Cessation of nectar production

Arditti, 1969

Arditti, 1969

Arditti, 1969, 1976a, b Arditti, 1969, 1976a, b

Burg and Dijkman, 1967; Dijkman and Burg, 1970

Phalaenopsis

In some species pollination accelerates senescence (Arditti, 1976a, b) Arditti, 1976a, b; Curtis, 1943; Duncan and Schubert, 1943; Ringstrom, 1968 Arditti, 1976a, b

Zygopetalum

Arditti unpublished Arditti unpublished

Arditti unpublished

Arditti, 1969, 1976a, b Arditti, 1969, 1976a, b Arditti, 1969, 1976a, b Arditti, 1969, 1976a, b

Arditti, 1976a, b

Arditti, 1976a, b

Arditti unpublished

Arditti, 1969, 19762, b

Arditti, 1976a, b

Arditti unpublished

Initiation of nectar production Production of anthocyanins or yellow pigments by floral segments Destruction of anthocyan i ns Ovule development Changes in enzyme complements or levels of corollas Hydrolysis of structural compounds Mobilization of substances from the perianth into columns and ovaries Changes in colour and shape of calli Changes in UV reflection patterns Loss of gynostemium curvature Twisting of gynostemia

Arditti, 1969, 1976a, b Arditti, 1969, 1976a, b Arditti e f al., 1973

J C

Probably occurs

Burg and Dijkman, 1967; Dijkman and Burg, 1 9 7 0 J Avadhani et al., 1971 Arditti, 1976a, b

Probably occurs Arditti, 1969, 1976a, b; Harrison and Arditti, 1976 Arditti, 1969, 1976a, b; Harrison and Arditti, 1976 Thien, 1971 Arditti, 1969, 1976a, b

Ethylene evolution continued

TABLE 39-continued Occurrenceas b Phenomenon Epidendrum

Angraecum

Phajus

Catasetum

Cycnoches ~

Senescence and death of perianth segments

Arditti, 1976a

Greening of all or some perianth segments

Arditti unpublished

Epipactis ~~~~~

~

Strauss and Koopowitz, 1973

Swelling of column Swelling of ovaries Hyponasty of sepals and petals Changes in pedicel curvature Stigmatic closure Hood formation or folding by dorsal sepal Greening of gynostemia Termination of scent production Cessation of nectar production

Arditti unpublished

J C

J C

J C

Strauss and Koopowitz, 1973 Arditti, 1976a Arditti unpublished

Arditti unpublished

Arditti. 1976a

Strauss and Koopowitz, 1973 Strauss and Koopowitz, 1973

Dodson, personal communication

Dodson, personal communication Wiefelputz, 1970

Initiation of nectar production Production of anthocyanins or yellow pigments by floral segments Destruction of anthocyanins

Colour changes Strauss and (Ames, 1947) Koopowitz, 1973

Colour changes (Arditti, 1976a)

Ovule development Changes in enzyme complements or levels of corollas Hydrolysis of structural compounds Mobilization of substances from the perianth into columns and ovaries Changes in colour and shape of calli Changes in U V reflection patterns Loss of gynostemium curvature Twisting of gynostemia

Thien, 1971

Ethylene evolution continued

TABLE 39-continued Occurrencea. Phenomenon -

Senescence and death of perianth segments

Cypripedium Rossner, 1923

Stanhopea

Schomburgkia

No effect on flower, longevity (Rossner, 1923)

Menadenium

Brazilian Miltonia species

Odontoglossum

Lip dies (Arditti, 1976a)

Hayes, 1968

Hayes, 1968

Greening of all or some perianth segments

Arditti, 1976a

Hayes, 1968

Hayes, 1968

Swelling of column

Arditti, 1976a

Hayes, 1968

Hayes, 1968

Swelling of ovaries Hyponasty of sepals and petals Changes in pedicel curvature Stigmatic closure Hood formation or folding by dorsal sepal Greening of gynostemia Termination of scent production Cessation of nectar production

J C

J C

Arditti, 1976a

Jiirgen Schrenk, 1973

Hayes, 1968

Arditti, 1976a

Hayes, 1968

J C

Initiation-of nectar production Production of anthocyanins or yellow pigments by floral segments Destruction of anthocyanins

Colour changes (Arditti, 1976a)

Hayes, 1968

Ovule development Changes in enzyme complements or levels of corollas Hydrolysis of structural compounds Mobilization of substances from the perianth into columns and ovaries Changes in colour and shape of calli Changes in TJV reflection patterns Loss of gynostemium curvature Twisting of gynostemia Ethylene evolution continued

TABLE 39-continued

Phenomenon Zygopetalum Senescence and death of perianth segments Greening of all or some perianth segments

Dendrobium, Eria

Listera

Orchis

Cypripedium, Neottia, Gymnadenia, Platanthera

Bletia

Hildebrand, 1863a, b

Hildebrand, 1863a, b

Hildebrand, 1863a, b

Hildebrand, 1863a, b

Hildebrand, 1863a, b

Hildebrand, 1863a, b

Hildebrand, 1963a, b

Hildebrand, 1863a, b

Hildebrand, 1863a, b

Hildebrand, 1863a, b

Duncan and Schubert, 1943

Swelling of column Swelling of ovaries Hyponasty of sepals and petals Changes in pedicel curvature Stigmatic closure Hood formation or folding by dorsal sepal Greening of gynostemia Termination of scent production

Cessation of nectar production Initiation of nectar production Production of anthocyanins or yellow pigments by floral segments Destruction of anthocyanins Ovule development Changes in enzyme complements or levels of corollas Hydrolysis of structural compounds Mobilization of substances from the perianth into columns and ovaries Changes in colour and shape of calli Changes in UV ieflection patterns Loss of gynostemium curvature Twisting of gynostemia Ethylene evolution continued

TABLE 39-continued Occurrences, b

Phenomenon Rhynchostylis

Mormodes

A rundina

Prasophyllurn

Bulbophyllurn

All or most orchids Arditti, 1967, 1976a, b

Senescence and death of perianth segments Greening of all or some perianth segments Swelling of column Swelling of ovaries Hyponasty of sepals and petals Changes in pedicel curvature Stigmaticclosure Hood formation or folding by dorsal sepal Greening of gynostemia Termination of scent production Cessation of nectar production

J C

J C

J C

Jones, 1972

Smyth, 1969

Arditti, 1967, 1976a, b

Initiation of nectar production Production of anthocyanins or yellow pigments by floral segments Destruction of anthocyanins

Lim et al., 1975

Ovule development Changes in enzyme complements or levels of corollas Hydrolysis of structural compounds Mobilization of substances from the perianth into columns and ovaries Changes in colour and shape of calli Changes in U V reflection patterns Loss of gynostemium curvature Twisting of gynostemia

Lim et al., 1975 J C

Lim et al., 1975

J C

Arditti unpublished

Ethylene evolution ~

a The genera and phenomena listed are only examples. Similar phenomena may occur in genera which are not listed or, some phenomena which occur in these or other genera may have been omitted. b Due to space limitation reviews are cited as much as possible. C A check-mark indicates that the phenomenon is presumed to occur.

Fig. 76. Post-pollination phenomena in Angraecum (Arditti, 1976a). Explanation of symbols: D or DS, dorsal sepal; G , gynostemium or column; L, labellurn; LP, lateral petal; LS, lateral sepal; Sp, spur.

ASPECTS OF ORCHID PHYSIOLOGY

583

A . cv Veitchi is a hybrid between these species and has inherited the folding character of both (Strauss and Koopowitz, 1973). The lip starts to fold first (Fig. 76G, H, I, J). Next, the dorsal sepal starts to bend down (Fig. 76H, I, J). After this, the petals start to fold inward (Fig. 761). At the end just before the perianth dies the flower looks like a well wrapped little package (Fig. 765; Strauss and Arditti, 1973; Strauss and Koopowitz, 1973). In Cymbidium (Figs 77, 79A) pollination brings about hood formation by the dorsal sepal; yellowing of petals and sepals; anthocyanin production by all perianth segments and the gynostemium (column); colour development, and deformation of the calli on the labella; ovule development; greening of gynostemia; stigmatic closure; enzyme production; increases in cell size; alteration in respiration patterns; changes in pedicel curvature; ethylene evolution ; altered phosphate transport among floral segments; ovary swelling; water losses; changes in dry weight and RNA synthesis (Arditti and Flick, 1974, 1976a, b; Arditti et al., 1971a, b; 1973; Arditti and Knauft, 1969; Hsiang, 1951a, b ; Hubert and Maton, 1939; for reviews see Arditti 1969, 1976a, b). Flowers of Phajus tankervilliae are fragrant and borne on flower stalks which reach two, maybe three metres. They exude a delightful fragrance and expose a purplish lip below a white gynostemium against the backdrop of brownish sepals and petals (Fig. 78A). Pollination probably occurs when the vector attempts to crawl between the lip and column. Following pollination the lip changes colour to a reddish-orange brown; scent production decreases and the angle of the flower is reduced to 40" or less. The lip also tends to tighten itself around the column (at least in some cases) and dies within a week with the rest of the perianth (Fig. 78B). A self-pollinating variety of Phajus tankervilliae also exists. Its flowers never open fully, and the pollen drops into the stigma during the late-bud stage. As a result, the flower stalks are usually laden with fruits. Pollinators play no role in this. Yet, the flower is fragrant and, once pollinated, exhibits the same phenomena as the cross-pollinated variety. Schomburgkia tibicinis produces flower stalks which may reach three metres or more. The flowers are reddish-purple with undulating sepals and petals (Fig. 78C). A white column is surrounded by a reddish-purple, translucent labellum. On a bright day, the tube formed by it (and enclosing the column) looks like a purple tunnel with alternating lighter and darker stripes. Following pollination, the flower changes very quickly (sometimes within 24 hours). Sepals and petals turn a deeper purple and fold inward forming tubes (Fig. 78D). The lip undergoes a similar colour change and folds over the column (Fig. 78D) clasping it tightly. Brassolaelia (an intergeneric hybrid) flowers retain characteristics of both parents, but look more like Brassavola than Laelia. Their labella are decorated with several lines made of purple dots (Fig. 78E). In a naturally occurring

Ethylene evolution, RNA synthesis, anthocyanin synthesis, swelling. straightening, AUXIN,

G I Stigmatic closure -

I

Stigma I

Petals, sepals

Labella

I

I

1

Anthocyanin synthesis- AUXIN, AEA, GA ETHYLENE, POLLl NATION, E M A S C U L A T I O N

General

I I

Ethylene evolution - POLLINATION, AUXIN, EMASCULATION, ETHYLENE

Mobilization of substrates from perianth into ovary and column - POLLINATION, AUXIN

I

Fig. 77. Post-pollination phenomena in Cymbidium. Phenomena (small letter) and agents or treatments which can cause them (CAPITALS) are listed in each box. Boxes are connected to the organs to which they pertain.

Fig. 78. Phajus tankervilliae (A) Unpollinated. (B) Approximately five days after pollination. Schomburgkiu tibicinis (C) Unpollinated. (D) About 26-36 hours after pollination. Brussolaeliu (E) Pollinated (F) Unpollinated. Epidendrum cv O’Brienianurn ( G )Unpollinated. (H) Pollinated (Arditti, 1976a). Explanation of symbols as for Fig. 76.

Fig. 79. Swelling of Cymbidium gynostemia. (A-D) view from 45" angle, (E-G) view from below. (H) Phuluenopsis unpollinated. (I) Three to five days after pollination. (J) Seven to nine days after pollination (Arditti, 1976a). Explanation of symbols as for Fig. 76 plus: ac, anther cap; c, calli; ct, column tip; cw, column wings; sc, stigmatic cavity.

ASPECTS OF ORCHID PHYSIOLOGY

587

species, these lines could attract pollinators. However, in a man-made hybrid, they cannot possibly have any such significance. Still, as the flower ages, or is pollinated, the spots smudge and the lines eventually disappear leaving the lip unadorned (Fig. 78E). Before senescence or pollination, the sepals and petals of Epidendrum cv O’Brienianum are a pleasant red colour, smooth and turgid (Fig. 78G). The lip has a bright yellow spot in the centre. When a flower senesces or is pollinated, the yellow spot disappears and all perianth segments shrivel (Fig. 78H). Colour changes also take place in Bifrenaria harrisoniae following pollination. Its sepals and petals turn yellow. The lip becomes dark red and the trichomes on it lose turgidity. PhaZuenopsis flowers (Fig. 79) have lobed labella and large yellow and orange calli. The column is situated above the calli, between the edges of two labellar lobes. Once a flower is pollinated, the labellum wraps itself, but not tightly, around the column (Fig. 79C). The lateral petals start to fold inward and, eventually, completely enclose the lip and column (Fig. 79D). This is hyponasty and much less common than epinasty. The dorsal sepal may bend down slightly, but there is almost no movement by the lateral sepals. The sepals and petals of Menadenium labiosum are greenish and its lip is white (Fig. 80A). On pollination, the sepals and petals enlarge, become fleshy and turn a darker green, the column swells, but the lip shrivels and dies (Fig. 80B). (a) Senescence. For a period there was a tendency among some workers to use the word ageing primarily for animals and the term senescence for plants. This is not so any more. Ageing is still applied almost exclusively to animals, but it is also used for plants along with senescence. The two words are often used as synonyms, but authors who wish to cover all bases often talk and write of “senescence and ageing” whether they mean one, or the other, or both. In other words “the term senescence has a confusing overlap in meaning with the term ageing” (Leopold, 1975). One possible way to remove the confusing overlap is to use the terms in the sense that “. . . senescence refers to the deteriorative processes that are natural causes of death and ageing refers to the wider array of processes of accruing maturity with passage of time” (Leopold, 1975 referring to Medawar, 1957). This is attractive to me, since it makes a useful distinction between events in the life of orchid blossoms. In addition, it generates several interesting questions which can be answered by using orchid flowers as the experimental organism. Another way of defining these phenomena is that ageing is “. . . the sum total of progressive changes in a whole plant or in one of its constituent organs . . .” whereas senescence is “. . . changes, caused by factors other than harmful external conditions, which are clearly degenerativeultimately and irreversibly so.” But, I see problems with these definitions in that (a) “harmful external conditions” are difficult to define (is excising an

Fig. 80. Menadenium /abiosum. (A) Unpollinated. (B) Pollinated. (C) Prof. Dr Hans Fitting, the last photograph ever taken of him (by Ms B. H. Flick). (D) Dr Fritz Miiller (Arditti, 1976a). Explanation of symbols as for Fig. 76.

ASPECTS OF ORCHID PHYSIOLOGY

589

organ or part of one during the experiment harmful, and, if so, is it harmful to the organ or to the plant from which it was excised?); and (b) it is not clear at present whether irreversibility should be part of the definition. These questions, like those generated by the Medawar definition can be answered through experiments with orchid flowers. First, there is a question regarding the aspects of ageing (as defined above), which may in fact be senescence. A deteriorative process which leads to the senescence and eventual death and elimination of a vital organ, organelle or physiological pathway would contribute to ageing. This is because the elimination contributes to the ". . . gradual accumulation of . . . changes . . ." (Leopold, 1975) which are part of the ". . . wide array of processes of accruing maturity. . ." (Leopold, 1975). In other words, one can ask whether senescence is a contributing factor to ageing. To put it differently, it is possible to ask: (a) Does ageing consist, at least in part, of steps which are senescence? (b) Could ageing proceed without senescence? (c) Can senescence take place without ageing? and (d) Does ageing result from lack of repair rather than active deterioration in contrast to senescence which is a progressive sequence ? A second question is whether senescence may not be merely accelerated ageing. A fast accruing of maturity could lead to the deteriorative processes that are senescence. Or, stated differently, the question is whether the differences between ageing and senescence are speed and rate. Unpollinated orchid flowers may remain fresh and alive for weeks or months (Kerr, 1972). Ageing and/or senescence, if any occur, are imperceptible for weeks. Even after becoming apparent the symptoms develop very slowly. In the process perianth colour may change due to anthocyanin, chlorophyll or carotenoid production or destruction (i.e., some orchids fade and others accumulate pigments as they age). Fresh and dry weight of sepals and petals and phosphorus levels in the perianth may increase or decrease depending on species (Lim et af., 1975). Gynostemia (columns) turn yellow, then grey and finally black before softening and becoming jelly-like. After that they shrivel and/or disintegrate. The ovary turns yellow and abscises. Pollination, disturbance of pollinia, emasculation, ethylene, auxin, damage and air pollution alter these events (Arditti and Knauft, 1969; Duncan and Schubert, 1943, 1947). The rate of deteriorative changes in perianth segments is greatly acclerated (i,e., senescence is induced or ageing is speeded up) in some species. Flowers of PhaIaenopsis and Cymbidium hybrids live up to eight weeks if unpollinated, but die within seven days after pollen deposition. Paphiopedilum blossoms may last up to three months but die within three weeks after pollination. When describing the differences between pollinated and unpollinated flowers Hans Fitting used the terms autonomen Blutendauer and aitionomen PostJorationsvorgange (Fitting, 1909a, b, 1910). The first (autonom flower duration), referred to ageing of flowers which was unaffected by exogenous

590

J. ARDITTI

factors including pollen. All stages of ageing occurred autonomically (i.e., “by themselves” as it were) with death coming as the logical end of a developmental process. Aitionom post-pollination events were described as induced (aitio-cause) phenomena in which some phases were modified and others eliminated by exogenous influences (i.e., the pollen of another flower). Ageing and death could be induced immediately after the juvenile phase without an intervening maturity period. In present-day terms autonom could be equated with ageing whereas aitionom may be parallel to senescence. However, it is also possible that in orchid flowers pollination merely speeds up some autonom events and also induces additional phenomena. If so the aitionom process would be a mixture of ageing and senescence in present day terms. This raises again the question regarding ageing v. senescence and also generates new ones. First, are at least some of the fast post-pollination events physiologically similar to several of the slow pre-pollination ones? Second, can pre- and/or post-pollination ageing be reversed? Third is the question whether, in unpollinated blossoms, the gynostemium, ovary, and in some cases petals and sepals age during the entire period after a flower has opened, albeit slowly. Or, do they go through a steady-state (phase during which no ageing occurs) which is then followed by a period of ageing. If this is so could the ageing period be prevented or delayed? And, once initiated is the ageing process reversible? What initiates it in the absence of pollination? The fourth question pertains to the nature of the pre- and post-pollination processes: the autonom ageing could be the result of lack of repair and maintenance, if so, events would be random. Post-pollination senescence, on the other hand, is a programmed sequence which can be characterized through standard procedures. A fifth possibility is that before pollination the process may be passive (i.e., the flowers age and die due to lack of repair or maintenance) but becomes active (i.e., tissues and substances are hydrolysed rapidly due to a marked increase in specific hydrolases) in pollinated flowers. Another, is that the process may be active in both instances but is accelerated after pollination. Sixth, in orchid flowers pollination or auxin trigger the senescence and death or transformation of the perianth, but questions remain regarding the induction of senescence: (i) what roles do plant hormones play in the process, and (ii) is an as yet uncharacterized hormone, like Fitting’s Pollenhormon (Fitting, 1909a, b, 1910) also involved? At present there is no need to implicate a new or unknown plant hormone in post-pollination phenomona of orchid flowers, but it is possible that other substances may control senescence as is the case with serine in oat leaves (Martin and Thimann, 1972). There are really no answers to these questions. One can only assume that the processes of ageing and senescence in orchid flowers are no different

ASPECTS OF ORCHID PHYSIOLOGY

591

from those in other plants. This and other factors make orchid flowers very suitable organisms for the study of ageing and senescence. The question of reversibility can be answered by determining whether unpollinated orchid flower segments (perianth, gynostemium, ovary) age constantly, even if slowly, before pollination or whether they remain at a steady state for a certain period before starting to decline. Shoud they age or senesce constantly it is obvious that pollination can reverse the process in some species and segments (see below) and accelerate it in others. If ageing or senescence are initiated following a steady-state phase, it would be feasible to determine the condition under which reversibility is possible by pollinating the flower at various times. Another advantage is that the phenomena can be initiated at will (rather than having to wait for a developmental stage), and separated from each other physiologically and studied individually. Many other studies use (or have used) excised leaves and other organs or tissues which have been removed surgically. They may be dying rather than ageing or senescing as they would in situ. Consequently, fine points may be missed entirely. The problem may even be more serious: “. . . these pieces, having been subjected to major surgery, cannot survive for more than a few days unless battered with nutrients and growth regulators which distort their physiology and delay their death. With such . . . experimental material, any effects have to be observed rapidly . . .” (Milborrow, 1974). [All this reminds me of a statement by Theophrastus in “Enquiry Into Plants”, IV, XV, 2-4 (Of Various Causes of Death): “Men try to help the tree by plastering it with mud and tying pieces of bark, reed or something of the kind about it, so that it may not take cold nor become dried up”-I think that maybe the treatments really killed the tree.] Or, the effects may be missed entirely and could be noted only in more nearly normal, un“battered” tissues. Because fruit ripening is a slower process, at times studied in whole organs under more normal conditions, the same events are more apt to be detected. Possibly, therefore, some reported differences between leaf senescence and fruit ripening are not as large or as real as they appear to be. A suitable experimental system may reconcile at least some of them. A third advantage is the relatively slow rate at which ageing or senescence occur in unpollinated orchid flowers. This would allow the detection of events which may be fleeting and therefore undetectable in other systems. For example, it is well established that during leaf senescence levels of protein, RNA and chlorophyll drop whereas the reverse may be true in some fruits. However, synthesis, or at least appearance of new proteins (perhaps proteases which lead to proteolysis) has been reported in senescing leaves and perianth segments (Martin and Thimann, 1972; Tan and Hew, 1973; Trippi and Van, 1971). The same would appear to be true for RNA. In some instances such synthesis, if it occurs, may be very low and could represent such

592

J. ARDITTI

a small proportion of the total protein content as to be undetectable. In orchid flowers these compounds could be measured. One difficult problem to answer regarding senescence, ageing and the difference between them is whether they involve lack of synthesis and/or maintenance, increased destruction (i.e., catabolism and hydrolysis) or both. Data from experiments designed to provide answers to these questions can be, and have been, interpreted in different ways. Even elegant experiments have not always removed doubts. Cytokinins can delay the senescence of rose petals for example and it has been suggested that they function by suppressing protease synthesis (Martin and Thimann, 1972). However, experiments involving hormones such as cytokinins and ABA have also been subject t o different interpretations (Milborrow, 1974). Some of the reasons for this situation lie in the systems being used. Many excised tissues (leaves for example) or specialized organs (like cotyledons) are simply dying (Arditti, 1971b; Milborrow, 1974) rather than undergoing “programmed” senescence which is a normal aspect (even if a final stage) of development (Varner, 1965). As a result “. . . the effect of ABA (and probably other hormones) recorded in such assays are usually those most readily observed . . ,” (Milborrow, 1974). And, some may be missed entirely. Not that “. . . leaf disks, explants or sections should not be used; they certainly should, but in conjunction and comparison with the same tissues in other and less damaged forms if valid information on the normal . , . is to be obtained.” (Milborrow, 1974). This Fig. 81. Effects of pollination and auxins on gynostemia. A. Longitudinal sections of the gynostemia of Coelogyne speciosu. (a) Untreated. (b)

Treated, stigmatic cavity closed; ag, agar block. B. Transverse sections of the gynostemia of an Odontoglossum hybrid. (a) Pollinated; p, pollinium. (b) Treated with 6-indolebutyric acid (cotton moistened with 50 mg 1-l) ; ct, cotton. (c) Control. C. Transverse sections of the gynostemia of an Odontoglossum hybrid. Top row: cells and epidermis at the backside of the gynostemia. Bottom row: cells between vascular bundle (hatched) and surface of the stigma. (a) Pollinated. (b) Control. (c) Treated with 8-indolebutyric acid. D. Gynostemia (backside) of Cyrtopodiumpunctutum. (a) Control. (b) Pollinated. (c) Agar block. aN: treated with a-naphthylacetic acid. agar block 2 x 2 x 1 mm, conc. 100 mg I-1 BN: treated with p-naphthylacetic acid. agar block 2 x 2 x 1 mm, conc. 100 mg I-’ FA: treated with ,%indoleacetic acid. agar block 2 x 2 x 1 mm, conc. 100 mg I-’ SP: treated with p-indolepropionic acid. agar block 2 x 2 x 1 mm, conc. 100 mg I-’ BB: treated with /3-indolebutyric acid. agar block 2 x 2 x 1 mm, conc. 100 mg I-’ E. Phulaenopsis umubilis. (a) before pollination, (b) six days after pollination. F. Aruchnunthe sulingi. (a) before pollination, (b) after pollination. G. Aerides odorutum. Unpollinated, front (a) and side (b) view; two days after pollination, front (c) and side (d) view (Fig. A-D, Hubert and Maton, 1939; Fig. E-G, Fitting, 1909a).

y

b

A

a

C

t

c

c D

b

6

@ F

a

P b

d

G

1,o,oI a

a

15.0

A

b B

c 13.0 14.0

0

3.12.0

E

I:

1.0

v 0

.-510.0 E

/--. ---.__------.cont.

f 9.0 0

6 C

C

A

8.0 0 1

2 3 4 5 6 7 Days after treatment C

Days after pollination

D

Fig. 82. Gynostemia. A. Transverse sections of gynostemia showing epidermis and cells of Cymbidium tracyanum. (a) untreated, (A) treated. Cyrtopodium punctatum. (b) untreated, (B) treated. Vanda tricolor var. suavis. (c) untreated, (C) treated. (Hubert and Maton, 1939). B.

Cymbidium finlaysonianum. (a) unpollinated, (b) five days after pollination (Fitting,

1909a). C. Effects of pollination on the swelling of Cymbidium gynostemium (Arditti and Flick, 1974). D. Effects of pollination on the swelling of the gynostemium of Cymbidium virens (plotted from data in Morita, 1918).

TABLE 40 Efects of Pollination and Excision of the Rostellum on Gynostemium Swelling and Curvature in Cymbidium Flowers (Arditti and Flick, I976) Treatments First Second description description None Rostellum excised Pollinated Pollinated Pollinated Pollinated

None None None Rostellurn excised Rostellurn excised Rostellum excised

a c, curved; sc, slightly curved; s, straight.

Time after first treatment, min

1 9.8 9.8

30

9.8 9.8

60

9.8

150

10.3

Gynosternium Swzlling, rnrn Day 2 4 7 T d

9.9 9.8 11-5

11.0 11.3 11.2

10.0 10.3 13.1 12.3 13.0 12.3

10.8 10.0 14.2 144 14.3 14.3

Curvaturea Day 1 2 4

t1.0 c

c

c

C

+O-2 +4*4 +5 0 t4.5 +4.0

c c c c

c C

C C

7

c

c c

c

s c sc sc

sc sc sc

S

TABLE 41 Post-pollination Phenomena in Cymbidium Flowers and their Parts (Arditti and Flick, 1976)

Number

Treatment

NAA in culture medium, pg13 mi

4 5 6 7 8 9 10

Whole flowers (WF) Undisturbed (control) Undisturbed (control) Pollinated Pollinated Emasculated Emasculated NAA on stigmag NAA on stigma NAA on lipg. h NAA on lip

11 12 13 14 15 16

Flower minus gynostemium ( F a ) No NAA on cuti 0 No NAA on cut 25 NAA on cut 8. 0 NAA on cut 25 Pollen on cut{ 0 Pollen on cut 25

17 18

Flower minus labellum (F-L) No NAA on cutj No NAA on cut

1 2

3

0 25 0 25 0 25 0 25 0 25

0 25

Gynostemium OVW

Swelling, mma Curvatureb 9.3 9.3 15.7 15-3 9.7 10.3 16.0 15.3 9-3 9.8

Stigmac

cur

0 0

Str

c1

Cur Str

c1

Cur Str Str Cur Cur

0 0 c1 CI 0 0

cur

diameter, mmd

Labellum callie

Perianth or labellurn wiltingf

4.23 5.23 5.2 5.4 4.7 5.1 4.9 6.1 4.4 5.4

Y

NW

4.4 5.1 6-0 5 -7 4-05 4.25 9.7 10.0

Cur Cur

0 0

4.2 4.8

YO OR OR YO YO R OR YO YO

vsw sw sw vsw Nw sw sw Nw sw

YO OR

vsw sw W sw sw sw

R R OR YO

19 20

NAA on cut Q, 1 NAA on cut

21 22 23 24 25 26 27 28

Gynostemium and ovary (G Undisturbed (control) Undisturbed (control) Pollinated Pollinated Emasculated Emasculated NAA on stigmag NAA on stigma

29 30 31 32 33 34 35 36

Gynostemium only (G) Undisturbed (control) Undisturbed (control) Pollinated Pollinated Emasculated Emasculated NAA on stigmag NAA on stigma

14.8 14.0

Str Str

c1 C1

6.0 6.4

CUr Cur Str Str Cur

0

25

8.7 10.0 13.0 13.0 10.0 9-7 13.3 13-7

cur Str Str

4.5 4-4 4.6 5.3 4.3 4.0 4-7 4.2

0 25 0 25 0 25 0 25

9.7 9.3 12.0 11.0 8.7 9.0 10-8 11.0

CUr CUr Str SCur Cur Cur SCur SCur

0 25

+ 0) 0

25 0

25 0

25

0

0

c1 c1 0 0

c1 c1

0 0 c1

c1 0 0 c1 c1

Ovary only (0)

40

No NAA on cutk No NAA on cut NAA on cut f l ~k NAA on cut

41 42

Pollen on cut Pollen on cut

37 38 39

0 25

0 25 0 25

4.53 4.13 5.03 5.18 5.17 4-83

continued

TABLE 41-continued

Number

43

44 45 46

Treatment Labellum only (L) No NAA between calli No NAA between calli NAA between callis NAA between calli

NAA in culture medium, 4 3 mi

Gynostemium Swelling,

mma

0

25 0 25

Measured at the lower edge of the stigma. straight. C C l,closed; 0, open. d Measured with calipers at the ovary midsection. e OR, orange-red; R, red; Y, yellow; YBr, yellow-brown. f N W ,not wilted; SW,slightly wilted; VSW, very slightly wilted. fl NAA was applied at 25 pg/flower or floral part. NAA was applied between the calli. 4 Cut resulting from removal of the gynostemium. Cut resulting from removal of the labellum. Cut resulting from removal of all other floral segments. a

b Cur, curved; SCur, slightly curved; Str,

Curvatureb

Stigmac

Ovary diameter,

Labellum

mmd

caW YJ3r YBr YBr YBr

Perianth or labellum wiltingf

Nw

Nw NW

Nw

TABLE 42 Nitrogen Content in Orchid Flowers Following Pollination (Abstracted From Gessner, 1948; Hsiang, 1951b). See text, p . 604.

% of Species

Cattleya Iabiata

Coelogyne cristata Coelogyne cristata var hololeuca Cymbidium tracyanum cv Doris

Treatment

Flower segment

Time after treatment

Control Control Pollinated Pollinated NAA treated NAA treated Pollinated Pollinated Pollinated Pollinated Pollinated Pollinated Pollinated Pollinated Pollinated Pollinated

perianth gynostemium perianth gynostemium perianth gynostemium gynostemium ovary perianth gynostemium ovary perianth gynostemium ovary perianth whole flower

167 hours 167 hours 167 hours 167 hours 167 hours 167 hours 10 days 10 days 10 days 10 days 10 days 10 days 16 days 16 days 16 days 16 days

Control

Initial levels

mg N/ flower 7-5 2 2.2 7-8 1.8 5.3

100 100 63 161 45 177

188 117 24-66 165 218 2-41 160 155 43-100 113

2.2 1-71 0.24-0.47 6.24

600

J. ARDITTI

problem may have arisen from the decline in the study of whole plants or entire organs. An important advantage offered by orchid flowers is that a whole organ can be studied in situ unaffected by treatments or influences other than those which are a part of their normal life cycle (i.e., the flowers are not battered). Even in vitro the extraneous influences are minimal. Yet another advantage of orchid flowers is that the fast process (which is probably senescence) can be initiated at will (Arditti and Knauft, 1969). This renders the flowers suitable for studies of the earliest stages of senescence. (b) Swelling and straightening of the gynostemium. One of the most noticeable effects of pollination is the swelling of gynostemia (Curtis, 1943; Dolcher, 1961, 1967; Duncan and Schubert, 1943; Fitting, 1909a; Hubert and Maton, 1939; Morita, 1918; for reviews see Arditti, 1969, 1976a, b). This has been reported for a number of species including Coelogyne speciosa, Odontoglossum, Cyrtopodium punctatum, Phalaenopsis amabilis, Cymbidium, Arachnanthe sulingi, Aerides odoraturn (Figs 81, 82). In Cymbidium the swelling starts on the second day after pollination and usually reaches a maximum in seven days (Fig. 82C). During that time the gynostemium also straightens (Fig. 82B). Removal of the rostellum has no effect on either the swelling or the straightening (Table 40; Arditti and Flick, 1974). Auxins can also induce straightening and swelling of the column (Tables 40, 41 ;Arditti and Knauft, 1969; Hubert and Maton, 1939). Excised gynostemia also swell if pollinated or when NAA is applied to the stigma. The same is true for gynostemia attached to ovaries with all other floral segments removed. Gynostemia on flowers minus labella also swell when NAA is applied to the cut resulting from removal of the lip. However, when NAA is supplied through the base there is no swelling (Table 41 ; Arditti and Flick, 1976). Experiments with sections of gynostemia have shown that the upper parts react more rapidly and extensively than basal portions (Dolcher, 1961b). The swelling of the gynostemium has been described as “. . . a characteristic response to ethylene” (Burg and Dijkman, 1967). However, ethylene (10 pl l-l) treatments of Cymbidium flowers for up to 4-5 days did not induce swelling (Arditti et al., 1973). This fact points to auxin as the factor which initiates the swelling. The following reports lend support to this view. First, the swelling is due to an increase in cell size (i.e., cell enlargement), not cell number (i.e., cell division) at least in Odontoglossum hybrids, Cymbidium tracyanum, Cyrtopodium punctatum and Vanda tricolor var suavis (Figs 8 1, 82; Hubert and Maton, 1939). Second, pollination and NAA applications bring about increases in fresh and dry weight of gynostemia (Hsiang, 1951a). Third, osmotic pressure increases after pollination (Hsiang, 1951a). Fourth, cut discs of pollinated columns take up more water than those from untreated ones (Hsiang, 1951a). Fifth, pollinated columns have a greater water holding capacity (Hsiang, 1951a). The obvious conclusion from these facts is that the swelling is due to in-

ASPECTS OF ORCHID PHYSIOLOGY

601

creased water uptake (brought about by the increased osmotic concentration which is the probable result of the higher dry matter content) coupled with auxin induced wall softening (for a more detailed discussion see Arditti er al., 1971b; Arditti and Knauft, 1969). ABA, GA, and kinetin do not inhibit or reverse the effects of auxin (Arditti et al., 1971a, b). The same is true for actinomycin D, but cycloheximide and ethionine reduce the swelling and straightening (Arditti and Knauft, 1969 and unpublished results from my laboratory). This may be taken as an indication that swelling and straightening do not require the synthesis of RNA, but are dependent to some extent on new proteins. (c) Stigmatic closure (Figs 79A, 81A, B, E, F; Tables 40,41) is a very common post-pollination phenomenon in orchids (which probably serves to protect pollen deposited on the stigma). In general, stigmatic closure is caused in a similar fashion and by the same factors which bring about stigmatic swelling i.e., pollination and auxin treatments. The only exceptions are full or partial closing of the stigma induced by: 1. 10 or 100 pg GA, per flower. 2. 0.01, 0.05 or 0-1 pg ABA plus 10 pg GA, per flower. 3. 1 pg GA, plus 10 pg kinetin per flower. 4. 0.01, 0-05 or 0.1 pg ABA plus 10 pg kinetin per flower. 5. 100 pg kinetin per flower. These exceptions are difficult to explain except possibly in terms of increased synthesis or sparing of auxin. Stigmatic surfaces of Epipactis atropurpurea contain leucoplasts with concentric thylakoids, abundant stroma and stroma-containing vesicles which include at least one starch grain and lipid globules (Pais, 1973-1974). If the same is true for other orchids it is possible to speculate that gibberellins could initiate synthesis of a-amylase which hydrolyses the starch grains, thereby increasing glucose concentration in the cells. This would increase osmotic concentration and subsequently water influx which can cause closure. NAA induced stigmatic closure is inhibited by cycloheximide (unpublished results from my laboratory), but not by actinomycin D (Arditti and Knauft, 1969). As with swelling and straightening of the column this is an indication that closure does not require RNA synthesis but is probably dependent on new proteins. (d) Anthocyanins. Both production and destruction of anthocyanin have been reported to occur following pollination (Arditti and Knauft, 1969; Burg and Dijkman, 1967; Dijkman and Burg, 1970; Hsiang, 1951a; Lim et al., 1975; for reviews see Arditti, 1969, 1976a, b ; Withner, 1974). In some flowers (Cymbidium for example) destruction may follow production (Fig. 90) (Arditti et al., 1973). Anthocyanin production (Arditti e l al., 1975) or destruction (Burg and Dijkman, 1967; Dijkman and Burg, 1970) can also be induced by emasculation and ethylene, In addition, GA, and ABA treatments canalso

602

. I . ARDITTI

cause increased anthocyanin levels in Cymbidium gynostemia and labella (Arditti et al., 1971a, b). Kinetin applications induced only marginal increases in labella (Arditti et al., 1971b). When these hormones (ABA, auxin, cytokinins, ethylene, gibberellins) are applied in pairs they tend to reduce or inhibit each other’s effects (Arditti et al., 1971b, 1973). Applications of actinomycin D, cycloheximide, ethionine and puromycin after NAA inhibit anthocyanin production (Arditti and Knauft, 1969). This can be taken to indicate that anthocyanin production requires de novo synthesis of RNA and proteins. Ethionine inhibits anthocyanin synthesis while enhancing ethylene production. Therefore it appears that the effects of these inhibitors on anthocyanin levels are not due to their influence on ethylene production. (e) Chlorophyll. As already mentioned pollination induces chlorophyll synthesis in gynostemia or perianth segments (Tables 38,39; Fig. 77). In Angraecum flowers pollination induced increased chlorophyll levels in the ovaries but a reduction in the spurs. Actinomycin D and cycloheximide applications following pollination reduce the losses in chlorophyll content (Strauss, 1976). c f ) Nastic movements. Pollination, emasculation and NAA applications induce hyponasty of petals and sepals in Phalaenopsis and Angraecum, as well as downward bending of the pedicel in Phajus and sideways movement of the column in Mormodes (Figs 78, 79; Strauss, 1976; for reviews see Arditti 1969, 1976a, b). Hyponasty is less common than epinasty. Epinasty is induced by ethylene (Abeles, 1973) and the same is true for hyponasty (Strauss, 1976). In Angraecum blossoms hyponasty of petals and sepals is inhibited by cycloheximide (Strauss, 1976). ( g ) Ovaries. Ovaries of pollinated orchid flowers swell and elongate following pollination (Fig 83; Fitting 1909a; Heslop-Harrison, 1957; Hsiang, 1951a; Hubert and Maton, 1939), but less so than gynostemia. Both the fresh and dry weights of ovaries increase following pollination (Strauss, 1976 and unpublished results from my laboratory). However, the hydration value (FW-DW/DW) decreases. This is an indication that the increase of water content in ovaries is lower, in proportion, than that of dry matter. The swelling is due to increases in cell size rather than number (Fig. 83D). This suggests that the mechanism of swelling is the same as in gynostemia. When excised ovaries are inserted in auxin containing solutions they do not swell (Table 41 : 37, 38). The same is true for a number of reports regarding excised floral segments (Table 41 : 15-18,21,22,25-28). In other cases ovaries do swell under such conditions (Table 41: 2, 4, 6, 8, 10, 12-14, 19, 20, 24, 3942; Arditti and Flick, 1976). One possible explanation for these findings is that auxin (exogenous or from pollen)-mediated increased uptake of water (i.e., solution) from the medium carries auxin into the ovary and the hormone initiates swelling. Subsequent growth of ovaries is intermittent (Duncan and Curtis, 1942a, b ; 1943).

Epiderma I cel Is

Total number of cells

68

70

600

714

(a) c

E

r 5oc

/--O-O

Fig. 83. Effects of pollination and auxin on ovaries. A. Aruchnunthe sulingi. (a) unpollinated; (b) seven days after pollination. B. Rhynchostylis retusu. (a) unpollinated ; (b) seven days after pollination. C. Aerides odorutum. (a) unpollinated; (b) six days after pollination (Fitting, 1909a). D. Dendrobium bronkeurti. (a) unpollinated; (b) pollinated (Hubert and Maton, 1939). E. Increase in length of the ovaries of a Cymbidium hybrid under the influenceof a-naphthylacetic acid. Average of six ovaries (Hubert and Maton, 1939).

604

J. ARDITTI

(h) Hormone production. Pollination, emasculation and auxin application induce autocatalytic ethylene production by orchid flowers. The same is true for ethylene, wounding and possibly other hormones (for references, see section on ethylene). ( i ) Movement and mobilization of substrates. As already mentioned the fresh (FW) and dry (DW) weights of gynostemia increase following pollination. The same is generally true for NAA applications, but not for emasculation. These increases are due to (i) increased water uptake after pollination (Hsiang, 1951a), and (ii) increased influx of dry matter due to the creation of sinks as in citrus flowers (Goldschmidt and Huberman, 1974). The increases in FW and DW are accompanied by consistent losses from perianth segments in Angraecum (Strauss, 1976), Coelogyne and Cymbidium (Gessner, 1948; Hsiang, 1951a; and unpublished data from my laboratory). FW losses in perianth segments are due to increased transpiration (Hsiang, 1951a), but the decreases in DW are mostly the result of export of substances. Evidence for this is available from work with several species and a number of constituents. In unpollinated flowers of Phalaenopsis amabilis, Dendrobium nobile and Cattleya labiata, protein content remains balanced until “normal” death of the blossom; pollination accelerates degradation (Schumacher, 1931). Total nitrogen content per flower changes little or not at all following pollination of Coelogyne cristata and Cymbidium tracyanum (Table 42, p. 599; Gessner, 1948). After pollination of these species and Cattleya labiata (Hsiang, 1951b) the nitrogen content of perianth segments decreases and that of gynostemia and ovaries increases (Table 42). This suggests that increases can also be expected in perianth segments which turn green and persist on the capsule. The mechanisms which lead to these alterations in protein content are not clear. However, it is reasonable to assume that hydrolysis followed by export occurs in senescing segments (i.e., perianth) and mobilization occurs in the ones which persist (e.g., ovary and gynostemium). Even more interesting are the differences (if any) between these events and their control before and after pollination. Further, the nature of the nitrogenous substances in perianth segments, column and labella is not known. One can only speculate that in senescing segments hydrolysates (i.e., amino acids) may predominate whereas in those that remain, newly synthesized proteins could be a major fraction. In Cattleya labiata the content of reducing sugars in the perianth and gynostemium does not change 137.5 hours after pollination. The same is true for the perianth of Cymbidium lowianum after 222 hours (Table 43; Hsiang, 1951b). However, the content in gynostemia increases somewhat during that period and the same is true for total sugar in both Cymbidium and Cattleya (Table 43). After 20 days the levels of glucose and total sugar in the gynostemium increase considerably (Table 43 ; Gessner, 1948). Sucrose content of Cattleya gynostemia and perianth segments increases after pollina-

TABLE 43 Sugar and Starch Content in Orchid Flowers Following Pollination (Abstracted From Gessner, 1948: Hsiang, 19516) Content Glucose, mg/organ Species

Treatment

Floral segment

Initial

After 20 days

Reducing, mg g-l FW Initial

After 73.5 h

After 137.5 h

Sucrose, mg g-

After 222 h

Initial

After 73.5 h

~

Cattleya bowringiana Control Pollinated Control Pollinated Control Pollinated Cymbidium lowianum Contr o1 Pollinated NAA treated Control Pollinated NAA treated Control Pollinated Pollinated Cymbidium Pollinated tracyanum Pollinated Pollinated Pollinated

perianth perianth gynostemium gynostemium 6 whole flowers 6 whole flowers perianth perianth perianth gynostemium gynostemium gynostemium 6 whole flowers 6 whole flowers gynostemium ovary perianth whole flower labellum

76

141

84 78 146 146 52.1 53.4

109 44.5

24.3 28.8 26.7

20 8.2 48.6 49.7 41.9 46.9 68.9

30.2 11.28 22

0.94 1.83 7.6 5.4 3.0-5-4 45.9 11.8

24.0 10.8 46-8.3 71.5 8.6 continued

TABLE 43-continued Content Sucrose total, mg g-’ DW

Sucrose total, mg/organ Speciesa

After 137.5 h

After 222 h

lnitial

After 20 days

~

Cattleya bowringiana

Cymbidium lowianum

14 25 7 30 5.4 13-6

~~

~~~

90

52.6 17.5 9.8 15-7 20.3 11.8

After 137.5 h ~~

After 222 h ~~

Initial

After 15 days

~

~

98 103 153 176 57.5 67.1

129

54.5 40.1 48.6

03.7 03.3 Cymbidium trucyanum

After 733 h

Initial

Starch, mg/organ

66.1 59.5 57.6 67.2 80.6 1.31 2-16

12.3 7.8 4-8.6 69.3 17.1

32.2 255 6-2-15 109.7 16.0

8

4

4

1

a For reasons of space, the “Treatment” and “Floral segment” columns could not be repeated here. They correspond exactly with those given in the first part of the table on the previous page.

ASPECTS OF ORCHID PHYSIOLOGY

607

tion, but this is not the case in Cymbidium (Hsiang, 1951b). Starch levels in gynostemia and labella of Cymbidium decrease (Table 43; Gessner, 1948) but increase in ovaries of Habenaria, Cymbidium lowianum, Dendrobium nobile and D . thrysifolium (Seshagiriah, 1941; Hsiang, 1951b). Broadly interpreted, these observations suggest hydrolysis, export and/or utilization in some segments (perianth) coupled with accumulation in others (ovaries). As in the case of nitrogen content, this assumption is consistent with the hypothesis that substances from senescing organs are mobilized into those which undergo changes and persist. 32P phosphate taken up by intact racemes of Cymbidium accumulates mostly in gynostemia and ovaries. Much smaller amounts are found in perianth segments. Phosphorus content decreases with age at least in Arundina flowers (Lim et al., 1975). Pollination intensified the uptake and differences (Oertli and Kohl, 1960; Hsiang, 1951b). To determine the nature and rate of phosphate redistribution we applied 32P0, to dorsal sepals, columns and labella of Cymbidium cv Jungfrau (Fig. 84; Harrison and Arditti, 1976) flowers which were pollinated, auxin treated or left undisturbed (control), We then determined the radioactivity in ovaries, gynostemia, dorsal sepals, sepals, petals and labella 3, 8, 24, 48, 96 and 168 hours after treatment (Fig. 85). Our findings indicate that there is a continuing interchange of phosphate between floral segments. Pollination and NAA treatments increase transport into gynostemia and ovaries while reducing movement into perianth segments (Harrison and Arditti, 1976). Again the movement is from senescing segments to active, juvenile ones. Similarly, senescence increases efflux rates of ,WI-, soRb+ and 14C-metabolites from morning glory flower tissues (Hanson and Kende, 1975) and phosphate transport from “old” to “young”, from tip to base in corn leaves (Muller and Leopold, 1966). Movements of phosphate and water in orchid flowers are, of course, dependent on the vasculature and the following considerations (Harrison and Arditti, 1976) are therefore relevant: in Cymbidium, three vascular bundles from the inflorescence axis enter each flower (Swamy, 1948; Fig. 86). One forms the midrib trace of the bract, whereas another divides into three, reaching the two lateral petals and one lateral sepal (Fig. 86A). The third serves the median petal (labellum or lip), one lateral sepal and the dorsal sepal. Therefore, only two vascular bundles give rise to the six main traces of the ovary (Fig. 86A). Traces leading to three stigmas (GI, G2 and G3) separate first at the level of perianth insertion. They originate from traces which supply the dorsal and lateral sepals (Fig. 86B). Traces into the gynostemium (or column) branch from bundles serving petals (Fig. 86B, a,, a,), sepals (Fig. 86B, A,, A,, G,, G,), and the dorsal sepal (Fig. 86B, A,, G,). Thus, the perianth (Fig. 86C) is well interconnected with the column (Fig. 86B, a, b, c). Fusion occurs between all traces leading into the perianth (Fig. 86C). Altogether, every segment of a Cymbidium flower is connected to

608

. I . ARDITTI

application. (A) Flower of Fig. 84. Cymbidium flowers, their parts and sites of Cymbidium cv Jungfrau. x 0.35.(B) View of parts of Cymbidium cv flower parts showing sites of 32Papplication (stars). Explanation of symbols (these also include symbols for Fig. 86): A, anther; Al, median stamen (outer whorl of stamens) or traces leading to it; A,, and A,, lateral stamens (outer whorl of stamens) or traces leading to them; a,, a2. and a,, median stamens (inner whorl of stamens) or traces leading to them; Br, bract; Ca, calli; DS, dorsal (median) sepal (outer whorl of stigmas) or trace leading to it; GI, median stigma (whorl of stigmas) or trace leading to it; G, and G,, lateral stigmas (whorl of stigmas) or traces leading to them; Gy, gynostemium (column); L, labellurn (lip or median petal); LP, lateral petals or traces leading to them; LS, lateral sepals or traces leading to them; MP,traces leading to median petal (labellum or lip); 0, ovary; Pd, pedicel; St, stigma (Harrison and Arditti, 1976).

every other part by at least one vascular bundle (Fig. 86C). Hence, it seems likely that the vascular bundles provide an interconnecting network for the distribution of phosphate. Still, it is possible that some movement may also take place through parenchyma cells. Movement from dorsal sepals to labella is very limited ; transport in the reverse direction is slightly better. Both agree with previous reports concerning the vascular anatomy of Cymbidium flowers (Swamy, 1948). The trace leading to the labellum (MP) originates below that of the dorsal sepal (Fig. 86A). Phosphate movement from the dorsal sepal might be largely directed away from the labellum trace by the transpiration stream, or, transport through the phloem (unaffected by transpiration) may be minimal. Nevertheless, even with the interconnection of traces, the dorsal sepal and labellum are not in direct contact (Fig. 86B). This could explain the relatively low exchange between dorsal sepals and petals or sepals.

I

Lip base-

32 Sepal base- P applied to dorsal sepal

32

P applied lip

32

P applied to

100

100

---38 24 48

'"t

Column base-

NAA .n pol1inated.p ( a )

I

I

96

168 Hours

Hours 32 Ovary- P applied to lip

Ovarysepal

32

P applied to dorsal

Hours

loooL

Ovary-

32

P applied to stigma

loot

Hours

Fig. 85. Phosphate transport through Cymbidium cv Jungfrau floral parts. Organ analysed and the site of 32Papplication are listed at the top of each figure. (a) 32Papplied between calli on labella and content measured at lip bases. (b) 32Papplied to dorsal sepals and content measured at their bases. (c) 32Papplied to stigmas and content measured at gynostemium (column) bases. (d) applied to labella and content measured in ovaries. (f) 32Papplied to stigmas and content measured in ovaries. (9) 32Papplied between calli on labella and content measured in gynostemia. (h) 32Papplied to dorsal sepals and content measured in gynostemia. (i) applied to stigmas and content measured in petals. Measurements: amounts are expressed as picomoles (pmoles) of phosphate per mg dry weight at the intervals listed. Transport is expressed as concentrations calculated on the basis of radioactivity present and the initial 32P: 31Pratio. Explanation of symbols: solid line with dots, controls (c); broken line with triangles, NAA (n); slashed lines with squares, pollinated flowers (p). Range of coordinates: (c, d, f) from 0 to 1OOO; (a, b, e, g, h) from 0 to 100; (i) from 0 to 10 (Harrison and Arditti, 1976).

610

J. ARDITTI

m

A

C

Fig. 86. Vasculature of Cymbidium flower (reproduced from Swamy, 1948 with permission). (A) Differentiation of the six main traces of the ovary from vascular bundle of the inflorescence axis. (B) Vasculature of a monandrous orchid like Cymbidium; (a) transverse section of gynostemium (column) ;(b) drawing from wire-plasticine model of vasculature; (c) vascular diagram. (C) Vascular supply of perianth segments in Cymbidium. For explanation of symbols see caption to Fig. 84.

Only two traces connect the dorsal sepal and the gynostemium (Fig. 86B, A,, GI), and this is reflected in relatively low movement of phosphate in either direction. The somewhat better transport from stigmas to dorsal sepals in unpollinated flowers might be explained on the basis of evaporative water losses in the perianth segments. Pollination, which causes water influx into the gynostemium, reduces net transport into the dorsal sepal. Transport from labella to gynostemia is much greater than that into sepals and petals. Vasculature of the flower may again be the reason. The labellum trace is connected through the bundle from which it originates to traces leading into the dorsal sepal and a lateral sepal. Together, these provide four branch traces into the gynostemium (Fig. 86B, A, and G , plus either G , and A, or G, and A,). Traces leading into lateral petals do not originate from the same bundle as the labellum trace (Fig. 86A). A vascular contact is provided only by the fusion of traces (Fig. 86C). The labellum traces and the trace for one sepal originate from the same bundle (Fig. 86B). This plus fusion provide a limited connection which can be responsible, in part, for the low rates of transport.

ASPECTS OF ORCHID PHYSIOLOGY

61 1

Stigmas are connected with all parts of the perianth through traces A,, A,, a,, a,, GI, G, and G, of the gynostemium (Fig. 86B). Thus, the fusion between traces (Fig. 86B) may explain the movement of 32Pfrom the stigma to the perianth, gynostemium and ovary (Fig. 85). Another point to consider is that pollination initiates ovule development in orchid ovaries (Miiller, 1868 ; Poddubnaya-Arnoldi, 1964) thereby creating sinks for a number of substances including phosphate. This seems to be happening in Cymbidium flowers because pollination increases phosphate transport into ovaries. Furthermore, since all vascular traces lead to bundles which pass through the ovary (Fig. 86B) phosphate (32P)wherever applied, can move easily into that organ (Fig. 85). During the experiments described above no radioactivity could be detected in the culture medium. Radioactivity decreased drastically in serial sections below the ovary. This allows speculation as to where the transport of phosphate may be occurring. If phosphate moves into gynostemia along with water influx, transport probably occurs in the xylem. But, if transport takes place primarily or only in the xylem, leakage into the medium might be expected (due to diffusion from the cut ends of vessels which are immersed in the medium) unless tyloses are formed (as they are in cut flowers) and block outward flow. This suggests two possibilities. First, translocation out of flowers may occur only if they are attached to the plant as seems to be the case with apple leaves (Spencer and Titus, 1973). Second, it is possible that at least downward movement may be in the phloem. This assumption is supported by the fact that transport of 32P from stigmas is reminiscent, in principle, of IAA movement in Vundu flowers (see section on auxins; Burg and Dijkman, 1967; Dijkman and Burg, 1970). The lip or labellum in orchid blossoms is a modified median petal. This modification is morphological as well as physiological. It loses less fresh weight and dry weight than other petals, wilts more slowly and rolls inward forming a tube. Even when NAA causes dry weight losses in labella, there is still no fresh weight loss and water content remains high. The combination of a swollen gynostemium and the still turgid, often rolled, labellum could exclude would-be pollinators from already-pollinated flowers (Arditti et ul., 1971b; Gellert, 1923). ( j ) Respiration. Respiratory rates of orchid flowers decrease with age (Fig. 87A, B; Sheehan, 1952, 1954), but rise sharply with fruit set (Fig. 87B; Rosenstock, 1956). The increase starts within one hour after pollination or NAA application. Respiration in gynostemia reaches an initial peak 50 hours after pollination and a second one 170 hours later (Fig. 87C; Hsiang, 1951b). In the perianth respiration peaks at 50 hours, drops 20 hours later, reaches another peak after another 125 hours and then decreases to levels below those of the controls (Fig. 87C). A very large proportion of the measured increases in respiration may be

612

J. ARDITTI

o’200

-

0.175-

-Tight bud ---- Flower opening -.- Open 1-2days ..-..

Open 30days

._._.._.._..---0.050

-

-

0.025

OO o:

a0

$0

slo d o

1: Time ( h 1

A

B

Stage

1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 l ~ l L

0

50

100 150 Hours after treatment

200

C

Fig. 87. Respiration by orchid flowers. A. Respiration of Cattleya mossiae flowers in relation to age at 15°C (Sheehan, 1954). B. Respiration of Listera ovata (l), Platuntheru chlorantha (2), Neottia nidus avis (3), Epipactis latifolia (4), and Goodyera repens (5). Ordinate: respiration as mg C02 released h-l g-l fresh weight (Rosenstock, 1956). C. Change in respiratory rate (mm3 Oa g-l dry weight h-l) of Cymbidium lowianum as ratio to control following pollination or auxin treatment (Hsiang, 1951b).

centred in the placental tissues. The second peak is reminiscent of the climacteric rise reported for several fruits (Table 44;Avadhani et al., 1971). This observation is particularly interesting in view of a report that “there was no evidence that a climacteric occurred in the respiration of orchid [Cattleya mossiae] flowers between the tight bud stage and the fully mature flower.” (Sheehan, 1954). Respiration in Cymbidium lowianum columns increased by a factor of three, eight hours after pollination (Hsiang, 1951b). Gynostemia of Coelogyne mooreana and Cattleya bowringiana exhibited a two-fold increase in respira-

613

ASPECTS OF ORCHID PHYSIOLOGY

TABLE 44 Respiration Rates in Placentas of Vanda cv Miss Joaquim (From Data in Avadhani et al., 1971)

Respiration, oxygen uptake, p1 g-l fresh weight

Time Immediately after pollination Three weeks after pollination (when ovule primordia are organized) Five weeks after pollination Nine weeks after pollination (when 8celled embryo is formed) Twelve weeks after pollination Unspecified time period, later than 12 weeks

750

400 575 150 375 225

NU-P POL-C

0

I

50

I

too noun oftor lnatmont

I

I

150

200

Fig. 88. Change in catalase activity (ml0, g-' fresh weight min-l) in Cymbidiumlowianum as ratio to control following pollination or auxin treatment (Hsiang, 1951b).

tion. This information together with that regarding placental tissues of pollinated Vanda cv Miss Joaquim and 30 day old unpollinated Cattleya mossiae flowers is yet another indication of reduced activities in senescent organs and increased metabolism in those which become the centre of new developmental events. (k) Enzymes. Ageing, pollination and emasculation bring about changes in enzyme activities (Fig. 88) and complements (including isozymes) in orchid flowers (Table 45) (Hsiang, 1951b; Limet al., 1975;Tan and Hew, 1973;Tran

TABLE 45 Changes in Enzyme Complements in Orchid Flowers With Age or Following Pollination (Hsiang, 19516; Lim et al., 1975); Tan and Hew, 1973; Tran %nh Van and Trippi, 1970; Trippi and Tran Thanh Van, 1971) Enzyme Catalase Catalase Catalase Dehydrogenases Glu-6-P dehydrogenase

Floral segment

Orchid

Gynostemium, perianth Gynostemium, perianth Gynostemium, perianth

Cymbidium lowianum Cymbidium lowianum Cymbidium

Time after pollination 0-25 hours

Changes Increase

Reference Hsiang, 1951b Hsiang, 1951b

50-200 hours

Increase

Hsiang, 1951b T. T. Van and Trippi, 1970; Trippi and T. T. Van, 1971 T. T. Van and Trippi, 1970; Trippi and T. T. Van, 1971 T. T. Van and Trippi, 1970; Trippi and T. T.Van, 1971 Trippi and T. T. Van, 1971

lowianum

Corolla

Phalaenopsis amabilis

24-48 hours

Disappears

Glutamatedehydrogenase Corolla

Phalaenopsis amabilis

5 days

Malate dehydrogenase

Corolla

Phahenopsis amabilis

5 days

Peroxidase

Corolla

Phalaenopsis atnabilis

Peroxidase

Corolla

Phalaenopsis amabilis

Unopened to fully expanded flowers 5 days

Band 1 decreased more rapidly than Band 2 Band 1 decreased more rapidly than Band 2 Of 7 bands, 3 anodic, 4 cathodic, 5 appear Middle anodic band 2, and cathodic band 5 increase. Cathodic band 6 appears

Trippi and T. T. Van, 1971

Peroxidase

Corolla

Peroxidase

Flowers

Peroxidase

Flowers

Peroxidase

Flowers

Phosphatase, acid

Flowers

Polyphenol, oxidase

Corolla (emasculated flower) Corolla (emasculated flower) Gynostemium (emasculated flower) Gynostemium (emasculated flower) Gynostemium (emasculated flower) Corolla (pollinated flower) Corolla (pollinated flower) Corolla (pollinated flower) Corolla (pollinated flower) Corolla (pollinated flower)

Polyphenol oxidase Polyphenol oxidase Polyphenol oxidase Polyphenol oxidase Polyphenol oxidase Polyphenol oxidase Polyphenol oxidase Polyphenol oxidase Polyphenol oxidase

Phalaenopsis arnabilis Arundina graminifolia Arundina graminifolia Arundina graminifolia Arundina graminifolia Arachnis cv Maggie Oei Arachnis cv Maggie Oei Arachnis cv Maggie Oei Arachnis cv Maggie Oei Arachnis cv Maggie Oei Arachnis cv Maggie Oei Arachnis cv Maggie Oei Arachnis cv Maggie Oei Arachnis cv Maggie Oei Arachnis cv Maggie Oei

Ageing

Increase of peroxidase 3 isozymes

Trippi and T. T. Van, 1971 Lim et al., 1975 Lirn et al., 1975

1 day

4th band appears Activity increases 5 bands increase Decrease

1-28 days

Increase

Tan and Hew, 1973

1 day

Decrease

Tan and Hew, 1973

1-21 days

Increase

Tan and Hew, 1973

21-28 days

Decrease

Tan and Hew, 1973

1 day

Decrease

Tan and Hew, 1973

1-7 days

Increase

Tan and Hew, 1973

7-14 days

Plateau

Tan and Hew, 1973

14-21 days

Increase

Tan and Hew, 1973

21-28 days

Decrease

Tan and Hew, 1973

1 , 2 , 4 days after opening 6th day after opening 4th day 4th day

Lim et al., 1975 Lim et al., 1975 Tan and Hew, 1973

continued

TABLE 45-continued Enzyme Polyphenol oxidase Polyphenol oxidase Polyphenol oxidase

Floral segment Gynostemium (pollinated flower) Gynostemium (pollinated flower) Gynostemium (pollinated flower)

Orchid Arachnis cv

Maggie Oei Arachnis cv Maggie Oei Arachnis cv Maggie Oei

Time after pollination

Changes

Reference

1 day

Decrease

Tan and Hew, 1973

1-21 days

Increase

Tan and Hew, 1973

21-28 days

Decrease

Tan and Hew, 1973

ASPECTS OF ORCHID PHYSIOLOGY

617

Thanh Van and Trippi, 1970; Trippi and Tran Thanh Van, 1971). These changes indicate that senescence in the corolla and events in other segments originate from “. . . endogenous correlative effects inducing cytoplasmic alterations which act as gene regulators.”(Trippi and Tran Thanh Van, 1971). The increase in peroxidase activity correlates with senescence because this enzyme has been shown to be involved in ethylene synthesis (Lim et a f . , 1975), and, senescence of orchid flowers is affected by ethylene (see sections on senescence and ethylene). The increases of acid phosphatase in corollas can also be correlated with senescence (Lim et al., 1975). Flowers which senesce faster, like those of Cattfeya have a lower polyphenol oxidase (PPO) activity than those of longer lived ones (for example, Arachnis cv Maggie Oei). Levels of PPO in the corolla of A . cv Maggie Oei decrease with age (Tan and Hew, 1973.) After emasculation there were no changes in the corolla but a decrease followed by an increase in the gynostemium (Table 45). An increase was also noted following pollination (Table 45; Tan and Hew, 1973). Therefore it has been noted that “It is tempting to suggest that the changes in polyphenol oxidase of orchid flowers [pollinated and depollinated (emasculated)] might be correlated with the changes in ethylene production.” (Tan and Hew, 1973.) If so, the nature of the correlation is not clear. It is clear, however, that the changes in enzyme and isoenzyme levels, content and activities are correlated with the developmental events associated with pre- and post-pollination ageing and senescence. ( I ) R N A synthesis. Much of the evidence that RNA synthesis occurs following pollination of orchid flowers is circumstantial having been obtained from experiments involving the application of actinomycin D (Arditti and Knauft, 1969; Strauss, 1976). However, direct evidence from isolation of RNA is also available (Arditti and Flick, 1976b). Similar de n o w synthesis in Petunia hybrida (Solanaceae) styles has been reported and supports the findings with orchids. Purely theoretical considerations, based on the current state of knowledge also support the view that post-pollination, ageing, senescence and post-emasculation phenomena require de n o w synthesis of RNA. (m) Miscellaneous. Nectar production may cease or be initiated following pollination (Dodson, 1967; van der Pijl and Dodson, 1966; Wiefelputz, 1970; for reviews see Arditti et af., 1971b; Jeffrey et aE., 1970). Production of fragrances may also terminate following pollination, emasculation and NAA treatments (Strauss, 1976; van der Pijl and Dodson, 1966). The mechanisms which regulate these events are not clear. E. INDUCTION OF PHENOMENA

Post-pollination phenomena can be induced by a number of factors (Table 46). Pollination, of course, induces all of these phenomena in nature (Table 46). Emasculation or the mere dislodgement of pollinia can induce

TABLE 46 Post-pollinationPhenomena in Orchid Flowers as Aflected by Pollination, Emasculation and Plant Hormones Phenomena known to be induced by pollination only@

ABA, auxin, emasculation, ethylene, GA, or pollinationa

ABA, auxin, or pollination@

1. Greening of sepals, petals,

1. Senescence and death of

1. Hooding of dorsal

1. Changes in pedicel

labella and gynostemia 2. Cessation of scent and nectar production 3. Hydrolysis of storage and structural compounds 4. RNA production 5. Starch accumulation in ovaries 6. Changes in protein levels

perianth 2. Anthocyanin productionb 3. Anthocyanin destruction (fading)c 4. Changes in colour and shape of calli 5. Hyponasty 6. Ovule development 7. Changes in enzyme complements 8. Changes in UV reflection by flowers 9. Folding of labella

sepal 2. Yellowing of perianth segments

curvature 2. Closing of stigma 3. Swelling of gynostemia 4. Mobilization of substances from perianth into columns and ovaries 5. Loss of curvature by columns 6. Swelling of ovaries 7. Ethylene evolutiona 8. Altered patterns of phosphate movement 9. Changes in fresh and dry weight

Auxin, or pollination@

a All phenomena in the table are induced by pollination in one or more species. Those also brought about by other treatments or plant hormone are listed in separate columns. As we gain more information on the subject, phenomena may be moved from column to column. The listing of a phenomenon in the pollination only column indicates that no other information is available at present and not that other factors may not induce it. b In some orchids like Cymbidium, for example. C As in V d a , for example. a Also induced by ethylene, emasculation and wounding.

ASPECTS OF ORCHID PHYSIOLOGY

619

some but not all symptoms. Auxins can bring about many, but not all phenomena. Abscisic acid (ABA), gibberellins (GA3) and ethylene can each initiate fewer events. Interactions between hormones and other substances can affect post-pollination phenomena. For example, several auxin-induced phenomena (e.g., anthocyanin synthesis) can be inhibited or reduced in intensity by protein- or RNA-synthesis inhibitors, kinetin, GA, or ABA; others cannot, e.g., swelling of the column or stigmatic closure. ABA-induced anthocyanin synthesis can be inhibited or decreased by GAa, auxin or kinetin When anthocyanin production is initiated by GAB,it can be reduced by ABA kinetin or auxin (Arditti et al., 1971a, b). Since emasculation (like auxin and pollination) initiates ethylene evolution, a suggestion could be made that all post-pollination phenomena are initiated or controlled by the gas. However, work with Cymbidium flowers has shown that ethylene induces anthocyanin synthesis, but not other phenomena like swelling of the column (Table 47) (Arditti et al., 1973). Some agents (e.g., ethionine) may stimulate ethylene production while inhibiting anthocyanin synthesis. 1. Pollination

Pollen, dead or alive, can initiate all post-pollination phenomena in all orchid species (Table 46) (Curtis, 1943; Duncan and Schubert, 1943; Fitting, 1909b, 1910, 1921; Laibach, 1930; Morita, 1918; McClelland, 1919 among others; for reviews see Arditti 1969, 1976a, b; Withner, 1974). However, the effect of dead pollen may be less pronounced. A possible reason for this may be the exhaustion of auxin and/or other substances in dead pollinia and continuous synthesis in germinating pollen (Hubert and Maton, 1939). Reduced germination of Cyrtopodium punctatum pollen and complete inhibition of Phalaenopsis grains by naphthaleneacetic acid (Curtis and Duncan, 1947) could be taken to support this view. If one assumes that these observations are due to supraoptimal auxin levels it is possible that these concentrations could result when exogenous auxin (NAA) was added to that produced by the germinating pollen. 2. Emasculation Removal of pollinia or merely their dislodgement brings about the onset of several post-pollination phenomena in Cymbidium (Arditti and Knauft, 1969; Curtis, 1947; Knauft et al., 1970), Angraecum (Strauss, 1976) and other orchids (Table 46). The available evidence suggests that these effects are ethylene mediated. Folding of the labellum in Angraecum following emasculation starts eight hours later than in pollinated blossoms. However, after 60 hours folding in pollinated and emasculated flowers is equal. Hyponasty of Angraecum petals is more pronounced after emasculation than following pollination (Strauss, 1976).

TABLE 47 Efsects of 10 pl-I Ethylene on Cymbidium cv Samarkand Flowers One Week After Treatment With Ethylene and/or Auxin (Arditti et al., 1975) Condition of Treatment Untreated Untreated Pollinated Emasculated Lanolin NAA (25 pglflower) Ethylene (10 pl l-I) Ethylene (10 pl l-I) Ethylene (10 pl l-l) Ethylene (10 pl l-l) Ethylene (10 pl l-I) Ethylene (10 $1-l) Ethylene (10 pl 1- l) Ethylene (10 pl l-3 Ethylene (10 pl l-I) Ethylene (10 pl l-I) Ethylene (10 pI 1-3 plus NAA (25 pglflower)

Duration, hours

Column width, mma

0 164 164 164 164 164

Stigma

Column

Callib

open Open Closed Open open Closed open open

open open Open open Open Open open open

Curved CUNed Straight Curved Curved Straight Curved Curved Curved Curved CWVd Curved CUNed Curved Curved Curved

Y

16 21+ 64 78 111

10.3 10-6 14.3 11.0 11.0 16.3 10.8 10.5 10.5 10.0 11.5 11.0 10.5 108 10-5 11.0

164

15.0

Closed

Straight

*

1

Indicative of swelling, measured along the lower edge of the stigma. 0, orange; R,red; Y,yellow. NW, not wilted; SW, slightly wilted; VSW, very slightly wilted.

0 R

R YO OR

OR OR OR OR OR OR OR OR OR OR

OR

sepals and petalsC NW Nw

sw sw Nw sw vsw vsw vsw vsw vsw vsw sw sw sw sw sw

ASPECTS OF ORCHID PHYSIOLOGY

62 1

3. Auxin It was inevitable perhaps that the discovery of auxin would lead to a search for growth factors in orchid pollinia. Assays of pollinia and their extracts with A vena coleoptiles, Vicia faba stems, Coleus petioles, Bryonia dioica tendrils and Phaseolus multij7orus epicotyls demonstrated that they were a rich source of what was then referred to as “Wuchstoff” and assumed to be auxin or closely related to it (Laibach, 1930, 1932, 1933a, b; Laibach and Maschmann, 1933; Mai, 1934; Maschmann and Laibach, 1932).Confirmation that both Fitting’s Pollenhormon and the growth substance extracted by Laibach and Maschmann were auxin was obtained in experiments involving the application of known auxins to orchids (Hubert and Maton, 1939). These and subsequent experiments showed that auxin can mimic pollinia and initiate most post-pollination phenomena (Table 46) (Arditti et al., 1971a, b; Arditti and Knauft, 1969; Burg and Dijkman, 1967; Dijkman and Burg, 1970; Dolcher, 1961a, b, 1967; Heslop-Harrison, 1957; Hubert and Maton, 1939; Hsiang, 1951a, b; Zimmerman and Hitchcock, 1939; Strauss, 1976). Comparisons between several auxins have shown that a-naphthaleneacetic acid is the most effective (Dolcher, 1967; Hsiang, 1951b; Hubert and Maton, 1939) and all had more pronounced effects than indoleacetic acid (HeslopHarrison, 1957). This auxin can also induce embryo-sac formation (HeslopHarrison, 1957) and cause the seed capsule to show “. . . sign of parthenocapy . . .” (Zimmerman and Hitchcock, 1939). In Cymbidium flowers 0.01-250 pg NAA per flower bring about straightening of the column and raise anthocyanin levels of gynostemia and labella. Stigmatic closure is initiated by 0.05 pg per flower and the calli change from yellow to orange-red following application of 0.001 pg per blossom (Arditti et al., 1971b). In Vanilla 2,4-D or 8-naphthoxyacetic acid can cause formation of fruits. These develop faster than pollinated ones, but produce inferior perfume (Bouriquet, 1954). Stigmatic closure, loss of gynostemium curvature and changes in calli induced by NAA cannot be prevented by simultaneous application of kinetin, GA, or ABA. However, anthocyanin content is reduced to levels below those brought about by NAA alone. Kinetin may reduce the wilting brought about by NAA applications (Arditti et al., 1971a, b). Anthocyanin formation and wilting, but not swelling of the column and stigmatic closure can be inhibited by actinomycin D, cycloheximide, ethionine and puromycin (Arditti and Knauft, 1969). This suggests that anthocyanin synthesis and wilting may require de novo RNA and protein synthesis. Auxin transport in gynostemia is polar (i.e., from stigma to ovary) and NAA applied through the base may not reach the stigma and/or rostellum (Arditti and Flick, 1976a). Radioactivity from 14C-IAA has also been recovered from perianth segments and ovaries of Vanda and Angraecum (Fig. 89) (Burg and Dijkman, 1967; Dijkman and Burg, 1970; Hubert and Maton,

622

J. ARDITTI

U

3 HOURS

24 HOURS

Fig. 89. Spread of 14C-IAAfrom the stigmas of V. cv Petamboeran to the floral appendages. About 20 OOO cpm of carboxyl labelled 14C-IAA(8 mc mmol-’) at a concentration of 0.1 mM in 0.8 % agar was applied to the stigmas of each of several flowers, and after either 3, 6, or 24 hours the flowers were dissected as indicated (dotted lines). Thus the petals, sepals, and lip were excised and cut into outer and inner halves, and a 1 mm thick cross-section was cut from the base of the floral column. Values are total cpm radioactivity in the indicated tissues. The figures at six hours (not shown) were only slightly higher than those at three hours (Burg and Dijkman, 1967).

1939; Strauss, 1976). Application of 50 pg IAA per flower in addition to the low level of radioactive auxin increased 14C-IAA transport into perianth segments of Angraecurn whereas pollination suppressed it. However, added IAA reduced transport into ovaries (Strauss, 1976). Cycloheximide applied 2-6 hours after 14CIAA, reduced its transport into perianth segments but had no effect on movement into ovaries. If applied with the auxin, cycloheximide reduced transport into the ovary but after eight hours it enhanced movement (Strauss, 1976). Chromatograms of Angraecum and Cattleya stigma extracts following IAA applications indicate that the auxin is conjugated with aspartic acid to form indoleacetic acid aspartate. Cycloheximide treatments reduced the conjugate level sharply (Strauss, 1976). The induction of post-pollination phenomena in orchids by auxins (i.e., increased uptake of water, folding, wilting, cell enlargement and anthocyanin production) can all be explained in terms of the known effects of these hormones. Some of these effects are probably direct whereas others may be mediated by ethylene (Burg and Dijkman, 1967; Dijkman and Burg, 1970). 4. Ethylene What may have been one of the earliest observations on the effects of ethylene on orchids was made as a result of a misfortune: “During the prevalence of severe cold weather . . in Philadelphia . . gas [escaping) from the main pipes under the street . . made its way [into] the greenhouse

.

.

.

ASPECTS OF ORCHID PHYSIOLOGY

623

of B. A. Fahnestock . . . through the night and by the following morning. . .” damaged or destroyed a great many plants (Fahnestock, 1858). But it seemed “. . . anomalous that of the Orchidaceae . . . every individual should have exhibited total indifference to the gaseous influence.” Plants of Phajus (Bletia) tankervilliae, Paphiopedilum (Cypripedium) venustum and Paphiopedilum insigne and other species “. . . passed unscathed through the heat of the battle.” (Fahnestock, 1858). In view of our present knowledge at first glance it is indeed surprising that a mixture containing approximately 8-59 by volume “olefiant” gas (an old name for ethylene) and hydrocarbon vapours had no effect on orchids. However, on second thought, there may be no reason for surprise and no anomaly because there is no mention of flowers and orchid plants may not be very sensitive to ethylene. For example, Cattleya lueddemanniana and Cymbidium insigne showed no epinasty when exposed to ethylene (Crocker et al., 1932). Unlike the plants, orchid blossoms are very sensitive to ethylene. Concentrations as low as 0-002p1 1-1 for 24 hours or 0.1 pl I-’ for eight hours can damage the sepals in Cattleya flowers that are starting to open (Davidson, 1949). The wilting of sepal tips known as “dry-sepal” injury can be brought about by concentrations ranging from 0.1 pl 1-1 for 16 hours to 0.01 pl 1-’ for 48 hours (Davidson, 1949). Dry-sepal injury which is induced by ethylene (Jester, 1952; Saylor, 1954) has become a major cause of considerable losses by flower producers in air polluted areas (Anonymous, 1960; Boyd and Millar, 1965; Bracey, 1963; Clayton and Platt, 1967; Cottrell, 1968; Hetherington, 1970; Hindawi, 1970; Kendrick et al., 1956; Middleton et al., 1956; Thompson, 1963). Another problem caused by ethylene is damage to packaged blossoms (Akamine, 1976; Fischer, 1949, 1950; Lindner, 1946). Brominated charcoal (Akamine and Sakamoto, 195l), potassium permanganate-impregnated paper, cloth and other substances as well as packaging the flowers under nitrogen or carbon dioxide have all been used in efforts to prevent ethylene damage to Vanda flowers, but success has not been notable. More recent experiments with silver nitrate treatment of Phalaenopsis (Halperin and Halevy, 1973/1974) and Cattleya (Beyer, 1976) and hypobaric storage of Vanda (Burg, 1973) appear more promising. Ethylene oxide, an antagonist of ethylene metabolism (Lieberman et al., 1964) which can prolong the life of other ethylene-sensitive flowers has apparently not been tested with orchids. A possible analogue, ethylene carbonate, has had no effects on Cattleya flowers to date (unpublished results from my laboratory). Rhizobitoxine, a compound isolated from Rhizobium japonicum (~-2-amino-4-(2’-amino‘3’ hydroxypropoxy)-trans-3-butenoic acid) is a potent inhibitor of ethylene effects (Owens et al., 1971). It has been applied to orchid flowers and seems to prolong their life (M. Lieberman, personal communication). In Cymbidium flowers ethylene (lop1 1-l) brought about an increase of

0.320.28 gj

::

-

0.240.20-

e 0.16‘i:

0 0

0.12-

-

0.08 0.04

-

Fig. 90. Effects of ethylene on anthocyanin content in Cymbidium flowers. (a) Anthocyanin levels in Cymbidium “Samarkand” columns (gynostemia, solid line) and lips (labella, broken line) followingexposure to lop1 1-l ethylene over extended periods of time. Inasmuch as there were no marked visual changes in flowers not subjected to ethylene, none were extracted until the end of the experiment. (b) Anthocyanin content in Cymbidium “Samarkand” columns (gynostemia, solid line) and lips (labella, broken line) one week after a 24 h exposure to 0.1, 1.0 and lop1 1-’ ethylene. (c) Anthocyanin concentration in Cymbidium “Samarkand” flowers: fresh, untreated, emasculated, pollinated, and following applications of NAA (25pg per flower) plus ethylene (lop1 1-l) or lanolin. Closed bars, gynostemia (columns); open bars, labella (lips) ; hatched bars, gynostemia corrected for swelling. In all cases, anthocyanins were extracted and measured one week after the start of the experiments.

ASPECTS OF ORCHID PHYSIOLOGY

625

anthocyanin content in labella and gynostemia (Fig. 90a, c). An initial increase was followed by a plateau, a second rise and a decrease (Fig. 90a, c). One week after exposure to ethylene flowers treated with 0.1 and 1.0 p1 1-l showed no appreciable increase in anthocyanin content (Fig. 90b, c). Pigment levels increased one week after a 24 hour exposure to 10 pI 1-l ethylene (Fig. 90b, c). Ethylene production in orchid flowers is autocatalytic (Burg and Dijkman, 1967; Dijkman and Burg, 1970). For example, fresh Cymbidium flowers exposed to 0 . 5 ~ 1I-' ethylene for only two hours start to produce ". . . abnormal endogenous . . ." amounts of the gas (Davidson, 1971). This is biologically important since pollination, emasculation, the mere dislodgement of pollinia and auxin also induce ethylene formation. In turn, this gas causes the evolution of increasing amounts of ethylene and the onset of postpollination phenomena such as senescence and anthocyanin production or destruction. Cattleya flowers start to produce ethylene within four hours after pollination (Davidson, 1971). Ethylene evolution by Cymbidium blossoms becomes noticeable within 10-12 hours after pollination, emasculation or NAA treatment (unpublished results from my laboratory). In Van& cv Miss Joaquim ethylene evolution starts 15 hours after emasculation and after 32 hours reaches a level of 3442 pl kg-1 h-' (Table 48). This production is 8.5 times greater than that of the pink passion fruit and one of ". . . the highest reported for any plant material." (Akamine, 1963). Ethylene production by flowers of Vanda cv Rose Marie and Vanda cv Petamboeran starts sooner after pollination and NAA application than following emasculation (Fig. 91A, B). Actinomycin D, cycloheximide and puromycin when applied with or following NAA reduce but do not greatly delay the onset of ethylene evolution. This suggests that initial ethylene production may depend on preexisting proteins and RNA but subsequent evolution requires de n o w synthesis. Other post-pollination phenomena are affected similarly by these inhibitors (Arditti and Knauft, 1969; Straws, 1976) thereby lending support to this view. The same is true of a report that pollination induces RNA synthesis in gynostemia (Arditti and Flick, 1976b). Ethionine when applied one hour before or with NAA brings about greatly increased ethylene evolution. Even when applied alone this analogue of methionine initiates considerable production of the gas. This suggests that in Cymbidium flowers ethionine may substitute for methionine and act as an ethylene precursor. An earlier report that ethylene may be formed from ethionine (Shimokawa and Kasai, 1967) tends to support this suggestion. However, this report has been questioned (Yang, 1974) and therefore my suggestion is speculative. Other possibilities are that ethionine merely activates the ethylene producing system perhaps by wounding the tissue because it is toxic.

626

J. ARDITTI

TABLE 48 Ethylene Production in Fading Vanda cv. Miss Joaquim Blossoms (Akamine, 1963) Time after removing pollinia (h)

0 12 15

22 25 28 30 32 34 36 46 49 a

Ethylene production

(4 kg h - 9 0 0 335.21 508.26 1016.53 1638.43 2582.02 3442.15 3359.00 285 1.24 1377.07 688.53

Degree of fading ( %I

0 5 30 75 85 90 95 97 97 99a

1 OOb 1OOC

Slightly glassy perianth. Glassy perianth and slimy peduncles. Very glassy perianth, slimy and darkened peduncles, and odoriferous.

In Vanda the gynostemium produces 10-8 nl g-l h-l ethylene, 150 minutes following IAA application (Burg and Dijkman, 1967) but the perianth does not evolve any. After 24 hours sepals and petals evolve 1.2 nl g-1 h-l ethylene, the labellum produces 81 nl g-' h-l and the gynostemium releases 50 nl g-l h-l. This points to the gynostemium as the earliest and major source of ethylene. The production site in the gynostemium may well be the rostellum (Fig. 92) which is a modified stigma. A number of observations support this suggestion. First, stigmas of other plants (Vaccinium angustifohm, the common lowbush blueberry) have been reported to produce ethylene following pollination (Hall and Forsyth, 1967; Forsyth and Hall, 1969). Second, the mere dislodgement of pollinia can bring about post-pollination changes of the type caused by ethylene (Curtis, 1943; Duncan and Schubert, 1947). Fig. 91. (A) Ethylene evolution by blossoms of V. cv Rose Marie after self-pollination or removal of pollinia. Control blossoms stayed fresh for about one week and produced no ethylene during that time. Fading of the lower petals first became evident after 8 to 12 hours in self-pollinated blooms, and after about 24 to 30 hours in emasculated flowers (Burg and Dijkman, 1967). (B) Ethylene evolution by blossoms of V. cv Petamboeran after pollination (with pollinia intact), application of 5 m IAA ~ in lanolin to the stigmas, or removal of the pollinia. The lower petals of blossoms which were pollinated or treated with IAA began to fade after 8 to 10 hours, those of emasculated flowers after about 35 hours, and controls after about 80 hours. Removal of the pollinia caused an initial transient production of ethylene, perhaps a wound response, which subsided within six hours (Burg and Dijkman, 1967).

V. Rose Marie

Hours

V. Petamboeran

Control I

9

e 60 Hours

Fig. 91

80

Fig. 92. Rostellum of Phalaenopsis. (a) Damage to rostellum 24 hours following detachment of the pollinia (Strauss, 1976). (b) Rostellum cells. CW, cell wall; M, mitochondria; V, vacuole (Strauss, 1976).

ASPECTS OF ORCHID PHYSIOLOGY

629

Such dislodgements only disturb the rostellum-viscidium interface (Fig. 92a) i.e., cause some, even if limited, wounding. Third, emasculation which wounds the rostellum induces considerable ethylene evolution (Burg and Dijkman, 1967; Dijkman and Burg, 1970) and other post-pollination phenomena which are mediated by the gas (Burg and Dijkman, 1967; Dijkman and Burg, 1970; Strauss, 1976; for short reviews see Arditti and Flick, 1974; Withner, 1974). Fourth, rostellum cells contain large numbers of mitochondria (Fig. 92b) which even if not a source of ethylene themselves may well provide the energy needed to produce it (Strauss, 1976). Fifth, surgical experiments involving the removal of gynostemium tips, rostella and stigmas (Fig. 93; Table 49) have shown that excision of the rostellum reduces

Fig. 93. Cymbidium gynostemia: Structure, parts and surgical treatments. (a) Intact gynostemium. x 2.1. (b) Gynostemium with anther cap removed, showing pollinia, rostellum, tip and viscidium. x 2.1. (c) Gynostemium, tip excised above the rostellum. x 1.05. (d) Self-pollination. x 1.05. (e) Emasculation. x 1.05. (f) Gynostemium following removal of the rostellum. x 1.05. (9) Gynostemium, tip excised below the rostellum. In some treatments pollen was placed in a drop of lanolin covering the cut surface. x 1.05. (h) Rostellum with pollinia still attached to it. x 4.2. (i) Rostellum following removal of the pollinia. x 4.2, a, anther cap; p , pollinia; r, rostellum; s, stigma (a cavity in Cymbidium and most orchids); t, tip of gynostemium above the rostellum; v, viscid disc or viscidium, which separates easily from the rostellum and attaches to pollinators.

TABLE 49 Effects of Pollination, Excision of the Rostellum, and Removal of the Cynostemium Tip on Post-pollinationPhenomena in Cymbidium Flowers (Arditti and Flick, 1974) Gynostemium Treatments First description

Second description

Time after first treatment,min.

None None Tip cut above None rostellum Tip cut at stigma None edge Rostellum excised None Pollinated None Pollinated 30 Tip cut above rostellum Pollinated Tip cut above 60 rostellum Pollinated 150 Tip cut above rostellum Pollinated 30 Tip cut at stigma edge Pollinated Tip cut at stigma edge 60 Pollinated Tip cut at stigma edge 150 Pollinated Rostellum excised 30 Pollinated Rostellum excised 60 Pollinated Rostellum excised 150 Tip cut at stigma Pollen placed on cut less than 1 edge edgef

1 9.8

Swelling, mm Day 7 2 4

9.9 10.0 10.8 +1-0 c c c c 9.8 0.0 c c c c

9.8 10.0 10.0

9.8

9.8

9-8 9-3 - 0 . 5

Stigmad Day 1 2 4

y y yor n n n s y y l o o n n s s

o o o o o o

c c c c

y y l o o

n n n s

9.8 9.8 10.3 10.0 f0.2 c c c c y y o o p n n s s o o o 9.8 11.5 13.1 14.2 +4*4 c c s s y o p p n s w w ; t c . t c c 9-8 11.0 12.7 13-0 +3.2 c c s c s c y l o o p n n s w + c + c c 9.8 11.8 12.8 13.1 +3*3 c c s c s c y l o o p

n n s w

fctcc

9.8 12.0 13-0 12-6 +2*8 c c s c s c y

n s w w

& c

9.8 9.5 9.7 9-8 9-8 10.3 9.8

10.0 9.4 9-7 11.0 11.3 11.2 10.8

9.8 9.5 9.0 8-5 9-0 9.0 12-3 14.8 13.0 14.3 12-3 14.3 9.8 10.0

a c, curved; sc, slightly curved; s, straight. b lo, light orange; 0, orange; p, purple; r, red; y, yellow.

n, no wilting; s, slightly wilted; w, wilting. c, closed; ;tC, threequarters closed; %, half closed; &, quarter closed; 0, open.

Td,total change during 7 days. f Pollen was placed

Tde

Perianth wiltingc Day 1 2 4 7

Curvaturea Callib Day Day 1 2 4 7 1 2 4 7

on the cut surface immediately after removal of the tip.

4.3 c c c c -1 .o c c c c

0

opp

yyoo p yyoyoyo -0-7 c c c c y y o o y o +5.0 c c c sc y y o o o p +4.5 c c sc sc y lo 0 op +40 c c sc sc y yo 0 op +02 c c c c y y o y o o p

n n n n n n n

n n s s n s n

s s w w s w w w w w w w s

s

c

o & c o $c c o o c

ASPECTS OF ORCHID PHYSIOLOGY

63 1

or delays anthocyanin production in Cymbidium (Arditti and Flick, 1974) which is an ethylene induced phenomenon (Arditti et al., 1973). However, the excision had no effects on phenomena which are mediated by ethylene (Table 49). Sixth, the rostellum is in close proximity to the stigma and can therefore be affected by pollen-borne or otherwise applied auxin(s) and other hormones or ethylene evolved by it. This does not, of course, eliminate the possibility that the stigma itself can also produce ethylene. Despite all this evidence, final proof that the rostellum is a major source of the gas is still lacking since there are no reports of direct measurements of ethylene evolution by this organ. Such measurements would be easy to make but unfortunately no one has made them yet. It is not surprising, perhaps, that ethylene, the only gaseous plant hormone mediates several post-pollination phenomena. This is because as a gas it diffuses readily and reaches all flower segments much faster than hormones which must be transported within the plant (such as auxin). Another reason is the fact that ethylene evolution can be induced quickly by a number of substances as well as by mechanical means. 5 . Cytokinins Post-pollination phenomena which might be regulated to a large extent by cytokinins may include: (i) ovule formation (for reviews see PoddubnayaArnoldi, 1964; Poddubnaya-Arnoldi and Selezneva, 1957; Veyret, 1974; Wirth and Withner, 1959) even in cases where NAA could play an important role (Dolcher, 1967; Heslop-Harrison, 1957); (ii) protein metabolism (Schumacher, 1931); (iii) fruit growth (Duncan and Curtis, 1942a, b, 1943); (iv) mobilization of substrates and creation of sinks (Gessner, 1948; Goldsmith and Huberman, 1974; Harrison and Arditti, 1972, 1976; Oertli and Kohl, 1960; Seshagiriah, 1941); (v) greening of perianth segments and gynostemia; (vi) regulation of senescence (Arditti, 1969). Some developing fruits and seeds are known to contain cytokinins. It is possible that they may act as post-pollination sources of these hormones in orchid flowers. In addition it is possible that orchid pollinia may contain cytokinins (unpublished results by G. Hayes in my laboratory). Exogenous kinetin, 0.1-10pg per flower, does not induce post-pollination phenomena in Cymbidium flowers. At 1OOpg per flower kinetin causes slight stigmatic closure and at 1 and 0.1 pg per flower it brings about a limited increase in anthocyanin levels. In combination with NAA, kinetin does not prevent stigmatic closure, swelling and straightening of the column, and wilting or changes in the calli. When applied with GA, and ABA, kinetin generally reduces the intensities of their effects. However, when 1Opg kinetin per flower are combined with 1 pg GA, per flower or 0.01, 0.05 or 0.1 pg ABA per flower stigmas close without swelling or straightening of columns (Arditti et al., 1971a, b). These observations are difficult to explain except in terms of increased auxin levels either

632

J. ARDITTI

due to increased production, reduced destruction or both. On the whole, though, the effects of kinetin on Cymbidium flowers are in line with the known influence of cytokinins on other systems.

6. Gibberellins To the extent that gibberellins may be required by pollinated orchid flowers, they are probably supplied by the young fruits and seeds (known sources of GA,) and/or the pollen (according to unpublished preliminary assays by Dr. M. S. Strauss and M. Ma, orchid pollinia may contain gibberellins). Exogenous GA, when applied alone at 0-001-1 pg per flower does not bring about straightening or swelling of the column and stigmatic closure. At 10 and 100 pg per flower GA, causes slight swelling and stigmatic closure. Anthocyanin levels in gynostemia and labella increase following applications of 0~001-1OOpg GA, per flower. In combinations with NAA, GA, reduced anthocyanin levels, but did not affect other post-pollination phenomena. Combinations of ABA and GA, have effects which are similar to those of ABA alone, except that anthocyanin levels are reduced. Flowers treated with GA, plus kinetin wilted slightly in most cases but columns did not swell and retained their curvature; calli did not develop colour and anthocyanin content was generally equal to that of flowers given only kinetin (Arditti et al., 1971a, b). 7. Abscisic Acid ABA, 250 and 500 ppm sprays, inhibit the development of new growths on Cymbidium plants (Brewer et al., 1969). When applied to flowers, ABA, 0*001-1 pg per flower, induces some, but not all, post-pollination symptoms. The hormone raises anthocyanin levels in labella, petals, sepals and gynostemia; initiates wilting; brings about folding by the dorsal sepal which forms a hood (“hooding”); and induces calli to develop colour and lose turgidity. However, it does not cause straightening and swelling of columns or stigmatic closure. ABA cannot inhibit most NAA induced post-pollination phenomena, but it does lower anthocyanin levels (Arditti et al., 1971a).

8. Interaction Between Hormones It appears reasonable to assume that several hormones participate in postpollination phenomena of orchid flowers and in subsequent development and if so, it is obvious that the system is self-sustaining. An exogenous supply of hormones can provide some indications regarding the role played by each. However, the system is very complex with numerous phenomena occurring in close physical proximity and ordered succession. Therefore, the effects of exogenously applied hormones may not be simple to explain since they may be interacting with regulators already present in the orchid flowers. In any case, the available data from orchid flowers indicate that all observed effects

ASPECTS OF ORCHID PHYSIOLOGY

633

are due to the treatments, but may represent effective concentrations other than those applied.

9. Summation Regardless of their nature, the post-pollination phenomena in orchids serve three basic functions: (i) To protect the pollinia, ensure a close contact between them and the stigmatic surface and provide a favourable environment for pollen germination and tube growth. This is accomplished by the swelling of the column and stigmatic closure. (ii) To recycle substances from senescing organs into those that become the centre of new activities. Production and/or activation of enzymes as well as transport and/or mobilization of substrates play important roles in this function. (iii) To render the pollinated flowers no longer attractive to pollinators thereby conserving “pollinator power” and increasing the likelihood that unpollinated flowers will be visited by vectors (Arditti and Fisch, 1977). This is important because only a small proportion of orchid flowers is pollinated as a rule. The effects of cessation of scent and nectar production are obvious: pollinators are no longer attracted to the flower. Folding of the perianth and movement of the entire flower and visible colour changes also serve the same function in an obvious manner. In addition, the visible colour changes also alter the UV reflectance image of the flower which also renders it less attractive to pollinators (Fig. 94; Kugler, 1966; Thien, 1971). This is especially true for flowers where the UV image is an important attracting and orienting feature. In some orchids the labellar surface markings (Figs 95, 96) and consistency may also play important roles in orienting and attracting pollinators. Alteration of these features (as for example the trichomes of Bifrenaria harrisoniae, the calli on Cymbidium and labellar surface on other orchids), all of which are attractants (Porsch, 1908) may serve to discourage visits by vectors to flowers which have already been pollinated. The more rapid onset of postpollination phenomena in pollinated blossoms than in emasculated ones is an indication of the extremely “fine tuning” of the system. Pollinated flowers have completed their function, but emasculated ones have not. Their pollen has been removed by a vector (which may deposit it on another flower) but they are still unpollinated. Consequently a second vector, should it carry pollinia, may pollinate such a flower. However, the vector can not obtain pollen from it and if the vector carries no pollinia the visit will be “wasted”. Hence, it would appear that conservation of energy and survival of the species would favour removal of emasculated blossoms, albeit at a slower rate than pollinated ones because the “extra” time may allow pollination. This is indeed the situation even if at first glance (on reading) the hypothesis would appear to be teleological to some extent. In any case, this seems to be the most likely explanation.

634

I. ARDITTI

Fig. 94. Epidendrurn cochleaturn flower image. Photographs taken without (left) and with (right) a UV filter (Thien, 1971, photograph courtesy of Dr Leonard B. Thieu). Inset, UV pattern of Orchis laxifolia (Kugler, 1966).

VI. TISSUE CULTURE Orchids may well be the very first flowering plants of commercial value to be propagated in vitro both through seeds and tissue cultures. Symbiotic (i.e., dixenic) seed germination was the first procedure to be developed (Bernard, 1909a). The second was asymbiotic germination (Knudson, 1921, 1922, 1946). Aseptic culture of flower stalk cuttings (each with only one axillary bud) of Phalaenopsis was the third method (Rotor, 1949). Fourth, perhaps the most dramatic, was the development of Cymbidium shoot tip (meristem) cultures as a means of clonal propagation (Morel, 1960, 1974). This procedure, based on work with Tropaeolum and Lupinus (Ball, 1946) revolutionized the orchid industry and pointed to the usefulness of tissue culture for fast clonal propagation of other plants (Murashige, 1974).

ASPECTS OF ORCHID PHYSIOLOGY

635

Fig. 95. Single epidermal cell on the upper side of the labellum of the flower ofPolystachiu fullax Kraenzlin. The cell surface shows a rather complicated polycentric cuticular fold pattern. x 2100 (Courtesy of Dr W. Barthlott).

Since the initial report, nearly twenty years ago (Morel, 1960) that Cymbidium can be propagated by shoot tip culture (a process now called mericloning and which produces mericlones) methods have been developed for nearly 40 genera (for reviews see Arditti 1977a; Morel, 1974; Rao, 1977). These methods differ considerably not only as they might be applied to separate genera and species, but also in relation to the parts of a plant. Thus, media which are suitable for shoot tips may not support growth of root or leaf tips, and media which are suitable for leaf tipsof one speciesmay not apply to those of another (Arditti, 1977a). Consequently, as this is being written, development of tissue culture procedure for orchids is mostly an empirical science (Vajrabhaya, 1976; Vajrabhaya and Vajrabhaya, 1976). My initial intent was to review tissue culture at length in this chapter. However, my resolve to do so weakened as this chapter increased in length and I finally decided to present only this short account and refer the reader to recently published reviews (Arditti, 1977a, c; Morel, 1974; Murashige, 1974; Rao, 1977). The new information which has accumulated since these reviews were published does not justify a new survey.

Fig. 96. (a) Labelium of Ophrys bertolonii Moretti. c. x 21. (b) Detail from (a) showing that the omega-sign on the labellurn differs not only in colour, but also in the form of the papillate epidermal cells. x 206 (Courtesy Dr W. Barthlott).

ASPECTS OF ORCHID PHYSIOLOGY

637

VII. EPILOGUE Through the years orchids have been considered to have magic properties, medicinal value or magical powers (Emboden, 1974). They have been the subject of expensive searches which even cost human lives. At one time only the rich could own, grow, give and/or wear them. At present they are within the reach of many as plants for the hobby trade or cut flowers and important factors in several economies. Unfortunately, however, orchids have been largely neglected by scientists (Dressler and Williams, 1970) despite offering a number of advantages (Arditti, 1971~).And, only now when all orchids are endangered or threatened are steps being taken to preserve them (for a review see Withner, 1977). The Orchidaceae is ". . . a vast family, so vast that it is beyond imagination . . ." (Withner, 1977). With a staggering 20 000-30 000, orchids exceed in number the flora of many regions. For example, they are 4-6 times the flora of England which has counts of 5000 species (Withner, 1977). In size orchids range from the few millimetres (pseudobulbs are 1-1-5 mm in diameter) and grams of the dimunitive Bulbophyllum minutissimum (Nicholls, 1969) to the several metres (three or more) and tons of the gigantic Grammatophyllum speciosum (Grant, 1895; Holttum, 1957). They may be soft, delicate and herbaceous or hard and nearly (but not) woody. Their pollination ranges from natural selfing to pseudocopulation in one and the same or different species. Carbon fixation patterns include C,, C , and CAM. The chemical diversity and adaptive features of the orchids are enormous. They can be found in all climatic regions and most niches populated by flowering plants (except under water). And this list of superlatives (especially when compiled by an orchidologist) can go on and on, perhaps ad nauseum. Its most surprising feature is that it can be compiled from studies of relatively few species by a limited number of investigators. Even the most liberal of orchidologists would probably estimate the total number of workers since Theophrastus to be less than the count of those who are currently working with E. coli. This review has not discussed all aspects of orchid physiology. Several areas remain untouched. However, my hope is that it contains enough of the seductive features and mystique of the orchids to generate if not a passion at least sufficientcuriosity for these remarkable plants. ACKNOWLEDGEMENTS I thank Dr P. N. Avadhani, University of Singapore; Dr R. Ernst, University of California, Irvine; Dr G . Hadley, University of Aberdeen, Scotland ; Dr A. Stoessl, Canada, Department of Agriculture, London, Ontario ; J. Michaud and Dr. M. S. Strauss, University of California, Irvine for reading

638

J. ARDITTI

and commenting on the manuscript; Dr C. R. Harrison, Saddleback College, Mission Viejo, California for allowing me to reproduce parts of his doctoral dissertation; Dr W. Barthlott, University of Heidelberg; Dr C. H. Dodson, Marie Selby Botanical Gardens, Sarasota, Florida; Dr B. Kullenberg, Uppsala University, Sweden; Dr L. B. Thien, Tulane University, New Orleans; the editors of Scient$c American for providing or allowing the use of photographs and line drawings; and K. Thurston for photographic darkroom work. Our work in orchid physiology and development has been supported by grants from the American Orchid Society; Far Best Corporation; Malihini Orchid Society, San Jose, Ca.; Elvenia J. Slosson Fund, University of California; Mrs Emma D. Menninger ; National Science Foundation; Office of Naval Research; Orchid Digest Corporation; Orchid Society of San Francisco; Population Council ; Public Health Service ; Stanley Smith Horticultural Trust; Textilana Corporation ; University of California fellowships, scholarships and research funds for several students ; UCI Industrial Associates; USDA; and the South Bay Orchid Society.

REFERENCES Abeles, F. B. (1973). “Ethylene in Plant Biology”. Academic Press, New York. Adams, M. S. (1970). Bull. Torrey bot. Club 97, 219-224. Adams, P. and Nightingale, C. (1976). Planting time. In “Antigrav” (P. Strick, Ed.), pp. 219-224, Taplinger Publishing Co., New York. Akamine, E. K. (1963). Science, 140, 1217-1218. Akamine, E. K. (1976). UortScience 11, 125-127. Akamine, E. K. and Sakamoto, H. T. (1951). Am. Orch. SOC.Bull. 20, 149-152. Alvarez, M. R. (1968). Am. J. Bot. 55, 1036-1041. Alvarez, M. R. (1969). Exp. CeIl Res. 57, 179-184. Alvarez, M. R. (1970). Exp. Cell Res. 61, 191-198. Alvarez, M. R. and King, D. 0. (1969). Am. J . Bot. 56, 180-186. Alvarez, M. R. and Reyniers, J. P. (1970). Exp. Cell Res. 61, 326-332. Ames, 0. (1938). Leafl. bot. Mus. Harvard Univ. 6, 145-183. Ames, 0. (1942). Am. Orch. SOC.Bull. 11, 102-106. Ames, 0. (1947). Am. Orch. SOC.Bull. 16, 442-443. Ames, 0. (1948). “Orchids in Retrospect”. Bot. Mus. Harvard Univ. Andersen, J. (1863). J. Horf. and Cottage Gardener. 4 (NS), 287. Anonymous. (1864). Jahresber. Agrikchem. 7, 100. Anonymous. (1890). J. Orchideis (Bruxelles) 1, 267-269. Anonymous. (1894). Orch. Rev. 2, 338-340. Anonymous. (1899). Orch. Rev. 7, 165-166. Anonymous. (1922). Orch. Rev. 30, 78-81. Anonymous. (1960). Ethylene. In “Technical Report of California Standards for Ambient Air Quality and Motor Vehicle Exhaust”, State of California, Dep. of Public Health, Berkeley. Arditti, J. (1966a). Am. Orch. SOC.Bull. 35, 175-182.

ASPECTS OF ORCHID PHYSIOLOGY

639

Arditti, J. (1966b). Orch. Dig. 30, 88-90. Arditti, J. (1966~).Orch. Rev. 74, 298-317. Arditti, J. (1966d). Sci. Am. 214, 70-78. Arditti, J. (1966e). Cymbidium SOC.News 20, 10-1 I . Arditti, J. (1967a). Bot. Rev. 33, 1-97. Arditti, J. (1967b). Am. J. Bot. 54, 291-298. Arditti, J. (1967~).Orch. Rev. 75, 197-199. Arditti, J. (1967d). Orch. Rev. 75, 253-256. Arditti, J. (1968a). Am. Orch. SOC.Bull. 37, 112-116. Arditti, J. (1968b). Orch. Rev. 76, 191-194. Arditti, J. (1969). Aust. Orch. Rev. 34, 155-158. Arditti, J. (1971a). Am. J. Bot. 58, 480-481. Arditti, J. (1971b). Am. Orch. SOC.Bull. 40, 211-214. Arditti, J. (1971~).“What is New in Plant Physiology”. Vol. 3, No. 7, pp. 1-4. Botany Department, University of Florida, Gainesville. Arditti, J. (1972a). Am. Orch. SOC. Bull. 41, 899-904. Arditti, J. (1972b). Die Orchidee 23, 243-247. Arditti, J. (1972~).Orquidea, Mexico 2, 299-236. Arditti, J. (1972d). Am. Orch. Soc. Bull. 41, 298-300. Arditti, J. (1975a). Orch. Rev. 83, 127-128. Arditti, J. (197513). Orch. Rev. 83, 187-188. Arditti, J. (1976a). Orch. Rev. 84, 261-268. Arditti, J. (1976b). Proc. Eighth World Orch. Con$, Frankfurt, pp. 528-533. Arditti, J. (1977a). Clonal propagation of orchids by means of tissue culture-a manual. In “Orchid Biology Reviews and Perspectives” (J. Arditti, Ed.), pp. 203-293. Cornell University Press, Ithaca. Arditti, J. (1977b). PI. Sci. Bull. 23, 38. Arditti, J. (1977~).Orch. Rev. 85, 102-103. Arditti, J. and Dueker, J. (1968). Am. Orch. SOC.Bull. 37, 862-866. Arditti, J. and Ernst, R. (1974). Proc. Seventh World Orch. Cot.., Medellin, pp. 299-307. Arditti, J. and Fisch, M. H. (1977). Anthocyanins of the Orchidaceae: Distribution, heredity, functions, synthesis and localization. I n “Orchid Biology Reviews and Perspectives” (J. Arditti, Ed.), pp. 117-1 55. Cornell University Press, Ithaca. Arditti, J. and Flick, B. H. (1974). Am. J. Bot. 61, 643-651. Arditti, J. and Flick, B. H. (1976a). Am. J. Bot. 63, 201-211. Arditti, J. and Flick, B. H. (1976b). Am. J. Bot. (Suppl.) 63, 37. Arditti, J., Flick, B. H., Ehmann, A. and Fisch, M. H. (1975). Am. J. Bot. 62, 738-742. Arditti, J., Flick, B. H. and Jeffrey, D. C. (1971a). New Phytol. 70,333-341. Arditti, J. and Harrison, C. (1971). Orch. Dig. 35, 21-23. Arditti, J. and Harrison, C. R. (1977). Vitamin requirements and metabolism in orchids. In “Orchid Biology Reviews and Perspectives” (J. Arditti, Ed.), pp. 157-1 75. Cornell University Press, Ithaca. Arditti, J., Ernst, R., Fisch, M. H. and Flick, B. H. (1972a). J. Chem. SOC.Chem. Comm. 1972, 1217-1218. Arditti, J., Healey, P. L. and Ernst, R. (1972b). Am. Orch. Soe. Bufl. 41, 503-510. Arditti, J., Healey, P. L. and Ernst, R. (1974). Die Orchidee 25, 289-294. Arditti, J., Hogan, N. M. and Chadwick, A. V. (1973). Am. J. Bot. 60,883-888. Arditti, J., Jeffrey, D. C. andFlick, B. H. (1971b). NewPhytol. 70,1125-1141. Arditti, J. and Knauft, R. L. (1969). Am. J. Bot. 56, 620-628.

640

J. ARDITTI

Arens, P. (1923). Ann. Jard. Bot., Buitenzorg 32, 103-124. Avadhani, P. N. (1963). Univ. Sinpapore Sci. J. 2, 6-8. Avadhani, P. N. and Goh, C. J. (1974). J. Singapore Nut. Acad. Sci. 4, 1-3. Avadhani, P. N., Rao, A. N. and Ong, P. Y. (1971). Fifty-eighth Indian Sci. Congr., Bangalore, pp. 8-10. Baker, P. G. (1968). Orch. Rev. 76, 199-201. Ball, E. A. (1946). Am. J. Bot. 33, 301-318. Barthlott, W. (1976). Proc. Eighth World Orch. Con& Frankfurt, pp. 444-445. Baskin, S. I. and Blis, C. A. (1969). Phytochem. 8, 1139-1145. Bateman, J. (1843). “The Orchidaceae of Mexico and Guatemala”. Johnson Reprint Corp., New York, 1973. Beau, C. (1914). Riviera Scientifique 1, 49. Beaumee, J. (1927). De Tropische Natuur 16, 54-55. Beer, J. G. (1863). “Beitrage zur Mophologie und Biologie der Familie Orchideen”. Druck und Verlag von Carl Gerold‘s Sohn, Wien. Beijerink, M. W. (1907). Bot. Zbl. 104, 89. Bendrat, M. (1929). Plunta 7, 508-584. Benike, L. A. (1910). Pri. Isp. Kharkovskum Univ. 43, 1-58. Benzing, D. H. (1973). Quart. Rev. Biol. 48, 277-290. Bernard, N. (1899). C.R. Lebd. SPanc. Acad. Sci., Paris 128, 1253-1255. Bernard, N. (1909a). Ann. Sci. Nut. Bot. ser. 9 9, 1-196. Bernard, N. (1909b). Bull. h s t . Pusteur 7, 369-386. Bernard, N. (1911). Ann. Sci. Nut. Bot. ser. 9 14, 221-234. Beyer, E. M., Jr. (1976). PI. Physiol. 58, 268-271. Bivins, J. L. (1968). Am. Orch. SOC.Bull. 37, 385-387. Bivins, J. L. (1970). Am. Orch. SOC.Bull. 39, 1005-1006. Blakeman, J. P., Mokahel, M. A., and Hadley, G. (1976). New Phytol. 77,697-704. Blowers, J. W. (1966). Card. Chron. 159, 293. Blowers, J. and Arditti, J. (1970). Orch. Rev. 78, 248-249. Boller, A., Corrodi, H., Gaumann, E. H., Hardegger, E., Kern, H., WinterhalterWild, N. (1957). Helv. Chim. Acta 40, 1062-1066. Borriss, H. (1967). Die Orchidee 18, 396-406. Borriss, H. (1969). Proc. Second European Orch. Cong., pp. 74-78. Borriss, H. (1971). Biol. Rundschuu 9, 177-180. Borris, H., Jeschke, E. M., and Bartsch, G. (1971). Biol. Rundschau 9, 177-180. Bose, T. K. and Mukhopadhyay, T. P. (1977). Orch. Rev. 85,245-247. Bouriquet, G. (1954). Culture. I n “Le Vanillier et la Vanille dans le Monde” (G. Bouriquet, Ed.), pp. 429-458. Editions Paul Lechevalier, Paris. Boyd, H. and Millar, A. A. (1965). Aust. Orch. Rev. 30, 143-145. Boyd, J. (1969). “The Pollinators of Eden”. Dell Publ. Co., New York. Bracey, S. (1963). Orch. SOC.So. Cal. Rev. 5, 9-10. Brewer, K., Gradowski, C. and Meyer, M. (1969). Am. Orch. SOC.Bull. 38,591-592. Brieger, F. G. (1957). Am. Orch. SOC.Bull. 26, 546-550. Brieger, F. G., Maatsch, R. and Senghas, K. (1977). “Die Orchideen, Rudolph Schlechter”. 3rd ed. Verlag Paul Parey, Berlin, W. Germany. Brongniart, A. (1831). Ann. Sci. Nut. ser. 1 24, 113-130. Brown, R. (1831). Phil. Mag. 10 (NS), 4 3 7 4 1 . Brown, R. (1833). Trans. Linn. SOC.London 16, 685-745. Brown, R. (1834). Allgemeine Bot. Ztng. 2, 17-20. Bulliard, P. (1802). In “Dictionnaire Elementaire de Botanique”. 2nd ed., pp. 56-57, Fuchs, Paris.

ASPECTS OF ORCHID PHYSIOLOGY

64 1

Burg, S. P. (1973). Hort. Sci. 8, 202-205. Burg, S. P. and Dijkrnan, M. J. (1967). PI. Physiol. 42, 1648-1650. Burgeff, H. (1909). “Die Wurzelpilze der Orchideen ihre Kultur und ihr Leben in der Pfianze”. Gustav Fischer Verlag, Jena. Burgeff, H. (191 1). “Die Anzucht tropischer Orchideen auch Samen”. Gustav Fischer Verlag, Jena. Burgeff, H. (1932). “Saprophytismus und Symbiose”. Gustav Fischer Verlag, Jena. Burgeff, H. (1934). Ber. dt. bot. Ges. 52, 384-390. Burgeff, H. (1 936). “Samenkeirnung der Orchideen”. Gustav Fischer Verlag, Jena. Burgeff, H. (1959). Mycorrhiza of orchids. I n “The Orchids” (C. L. Withner, Ed.), pp. 361-395. Ronald Press, New York. Burges, A. (1936). New Phytol. 35, 117-131. Burges, A. (1939). New Phytol. 38, 273-283. Burkill, I. H. (1917). Gdns. Bull. Staits Settlements, ser. 3 1, 400-405. Burstrom, H. (1948). Physiol. Plant. 1, 209-215. Capesius, I. (1976). FEBS 68, 255-258. Capesius, I., Bierweiler, B., Bachrnan, K. and Rucker, W. (1975). Biochim. Biophys. Acta 395, 67-73. Cappelletti, C. (1933). Bull. SOC.Ital. Biol. Sper. 8, 288-291. Cappelletti, C. (1947). Lovori di Botanica 8, 57-76. Casamajor, R. and Went, F. W. (1953). Am. Orch. SOC.Bull. 22, 126. Chatin, J. (1874-1875). Rev. Sci. Nat. 3, 236-240. Chautard, J. (1864). C. R. hebd. SPanc. Acad. Sci. Paris 58, 639-641. Chennaveeraiah, M. S. and Patil, S . J. (1973). Proc. Sixtieth Indian Sci. Congr., Part 111, 410-41 1. Churchill, M. E., Ball, E. A. and Arditti, J. (1972). Am. Orch. SOC.Bull. 41,726-730. Clarke, A. C. (1974). The reluctant orchid. I n “The Nine Billion Names of God”, pp. 184-194. Signet Books. New American Library, New York. Clayton, G. D. and Platt, T. S. (1967). J. Am. Hyg. SOC.28, 151-160. Clifford, H. T. and Smith, W. K. (1969). Phytomorphology 19,133-139. Collett, A. (1971). J. Roy. Hort. SOC.96, 358-361. Cook, M.T. (1926). J. Dept. Agr. Puerto Rico 10, 5-9. Coomans de Ruiter, L. (1930). Trop. Natuur 19, 116-119. Cortesi, F.(1912). Atti. SOC.Ital. Prog. Sci., Rome 5, 860-864. Costantin, J. and Magrou, J. (1922). Annls Sci. Nat. ser. loe, T. IV, I-XXXIV. Coster, C. (1925). Ann. Jard. Bot., Buitenzorg 35, 125-162. Cottrell, G.G. (1968). Florist and Nursery Exchange 150, 5-8. Coutinho, L. M. (1963). Bol. Fac. Fil. Ciknc Letr. Univ. Sao Paolo 288, Bot. 20, 83-98. Coutinho, L. M. (1964). Untersuchungen uber die lage des lichtkompensationspunkte einigen pflanzen zu verschiedenen tageszeiten mit besonderer beriicksichtigung des “de Saussure-Effektes” bei sukkulenten. I n “Be itrage zur Physologie” (K. Kreeb, Ed.), pp. 1-8. Verlag Eugen Ulrner, Stuttgart. Coutinho, L. M. (1965). Bol. Fac. Fil. Ciknc Letr. Univ. Sao Paolo 294, Bot. 21, 397-408. Coutinho, L. M. (1969). Bol. Fac. Fil. Ciknc Letr. Univ. Sao Paolo 331, Bot. 24, 77-102. Coutinho, L. M. (1970). Ciknc. Cult. 22, 364-368. Coutinho, L. M. and Schrage, C. A. F. (1970). Ann. Acad. B r a d Ciknc. 42, 843849. Crafts, C. B. and Miller, C. 0. (1974). PI. Physiol. 54, 586-583.

642

J. ARDITTI

Crocker, W., Zimmerman, P. W. and Hitchcock, A. E. (1932). Contrib. Boyce Thompson Znst. 4, 177-218. Cruger, H. (1851). Bot. Ztng. 9, 73-74, 79-80. Curtis, J. T. (1937). Proc. SOC. Exp. Biol., Med. 36, 4 3 4 . Curtis, J. T. (1943). Am. Orch. SOC.Bull. 11, 258-260. Curtis, J. T. (1946). Ecology 27, 264-266. Curtis, J. T. (1947). Am. Orch. SOC.Bull. 16, 654-660. Curtis, J. T. (1954). Bull. Torrey bot. Club 81, 340-352. Curtis, J. T. and Duncan, R. E. (1947). Am. Orch. SOC.Bull. 16, 616-619. Czapek, F. (1909). Sitzber. Math. Naturwiss KI. Akad. Wiss. Vienna 118,1555-1580. Darwin, C. (1862). “The Fertilization of Orchids by Insects” (1st ed. cited by Van der Pijl and Dodson), John Murray, London. Darwin, C. (1904). “The Fertilization of Orchids by Insects” (7th Impression of 2nd ed.), John Murray, London. Davidson, 0. W. (1949). Proc. Am. SOC.Hort. Sci. 53, 440-446. Davidson, 0. W. (1967). Am. Orch. SOC.Bull. 36, 811. Davidson, 0. W. (1971). Orch. Dig. 35, 198. Davidson, 0. W. (1977). Am. Orch. SOC.Bull. 46, 798. De Bruijne, E. and Debergh, P. (1974). Mededelingen Fakulteit Landbouw-wetenschappen, Ghent 39,210-215. De Lehaire, J. H. (1910). Bull. SOC.Royale Bot. Belgique 47, 45-52. De Vries, J. T. (1953). J . Annls. Bogoriensis 1, 61-76. De Wit, H. C. D. (1959). “Rumphius Memorial Volume” (H. C. D. De Wit, Ed.), Greshaff’s Rumphius Fonds, Baarn, Holland. De Wit, H. C. D. (1977). Orchids in Rumphius’ Herbarium Amboinense. Zn “Orchid Biology, Reviews and Perspectives” (J. Arditti, Ed.), pp. 47-94, Cornell University Press, Ithaca. Dijkman, M. J. and Burg, S. P. (1970). Am. Orch. SOC.Bull. 39, 799-804. Dodson, C. (1962). Ann. Mo. Bot. Gdn 49, 35-36. Dodson, C. H. (1967). Atlas do Simpdsio s6bre a Biota Amaz6nica (Zool.) 5 , 1-72. Dodson, C. H. (1975). Coevolution of orchids and bees. In “Coevolution of Animals and Plants” (L. E. Gilbert and P. H. Raven, Eds), pp. 91-98, University of Texas Press, Austin. Dodson, C. H. and Frymire, G. P. (1961). Mo. Bot. Gdn Bull. 49, 133-152. Dodson, C. H. and Hills, H. G. (1966). Am. Orch. SOC. Bull. 35, 720-825. Dodson, C. H., Dressler, R. L., Hills, H. G., Adams, R. M. and Williams, N. H. (1969). Science 164, 1243-1249. Dolcher, T. (1961a). Nuevo Giorn. Bot. Ztal. N. S. 68, 213-215. Dolcher, T. (1961b). Nuevo Giorn. Bot. Ztal. N. S. 68, 216-219. Dolcher, T. (1967). Nuevo Giorn. Bot. Ztal. 101, 41-57. Dorr, I. and Kollman, R. (1969). Planta 89, 372-375. Downie, D. G. (1943). Trans. and Proc. Bot. SOC.Edinburgh 33, 380-382. Downie, D. G. (1949). Trans. and Proc. Bot. SOC.Edinburgh 35, 120-125. Dressler, R. L. and Williams, N. H. (1970). Am. J. Bot. 54, 754. Dueker, J. and Arditti, J. (1968). PI. Physiol. 43, 130-132. Duncan, R. E. and Curtis, J. T. (1942a). Bull. Torrey bot. Club 69, 167-183. Duncan, R. E. and Curtis, J. T. (1942b). Bull. Torrey bot. Club 69,353-359. Duncan, R. E. and Curtis, J. T. (1943). Bull. Torrey bot. Club 70, 104-119. Duncan, R. E. and Schubert, C. K. (1943). Am. Orch. SOC.Bull. 11, 1-3. Duncan, R. E. and Schubert, C. K. (1947). Am. Orch. SOC. Bull. 16, 208-210. Dunsterville, G. C. K. and Dunsterville, E. (1967). Am. Orch. SOC. Bull. 36,790-797.

ASPECTS OF ORCHID PHYSIOLOGY

643

Dykus, A. M. and Knudson, L. (1957). Bot. Gaz. 119, 78-87. Edwards, G. E. (1974). What’s New in PI. Physiol. 6 (6), 1-4. Emboden, W. A. (1974). “Bizarre Plants, Magical, Monstrous, Mythical”. Macmillan, New York. Erickson, L. C. (1957a). Am. Orch. SOC.Bull. 26, 401-402. Erickson, L. C. (1957b). Am. Orch. SOC.Bull. 26, 560-563. Ernst, R. (1967a). Am. Orch. SOC.Bull. 36, 1068-1073. Ernst, R. (1967b). Am. Orch. SOC.Bull. 36, 694-704. Ernst, R. (1974). Am. Orch. SOC.Bull. 43, 35-38. Ernst, R. (1975). Am. Orch. SOC.Bull. 44, 12-18. Ernst, R. (1976). Proc. Eighth World Orchid Conf., Frankfurt, pp. 379-383. Ernst, R., Arditti, J. and Healey, P. L. (1970). Am. Orch. SOC.Bull. 39, 599-605, 691-700. Ernst, R., Arditti, J. and Healey, P. L. (1971a). Am. J. Bot. 58, 827-835. Ernst, R., Arditti, J. and Healey, P. L. (1971b). New Phytol. 70, 457-475. Esser, G. (1973). Die Orchidee 24, 12-14. Fahnestock, G. W. (1958). Proc. Acad. Nut. Sci. Phila. 18, 118-134. Fast, G. (1964). Die Orchidee 15, 23-29. Fast, G. (1967). Die Orchidee 18, 445-452. Fast, G. (1971). Die Orchidee 22, 189-192. Fast, G. (1976). Proc. Eighth World Orchid Conf., Frankfurt, pp. 359-363. Fehlandt, P. R. (1960). Orch. Dig. 25, 391-392. Fisch, M. and Arditti, J. (1972). Am. J. Bot. 59, 672. Fisch, M. H., Flick, B. H. and Arditti, J. (1973a). Phytochem. 12, 437-441. Fisch, M. H., Ernst, R., Flick, B. H., Arditti, J., Barton, D. H. R.,Magnus, P. D. and Menzies, 1. D. (1973b). J. C.S. Chern. Comm. 1973, 530. Fisch, M. H., Shechter, Y. and Arditti, J. (1972). Am. Orch. SOC.Bull. 41, 605607. Fischer, C., Jr. (1949) N , Y. State Flower Industries Bull. 49, 9. Fischer, C. W. (1950). N. Y. State Flower Growers Bull. 61, 1, 1. Fitting, H. (1909a). Zeitschr. fur Bot. 1, 1-86. Fitting, H. (1909b). Biol. Centralblutt 29, 193-205, 225-239. Fitting, H. (1910). Zeitschr. fur Bot. 2, 225-267. Fitting, H. (1921). Naturwissenschaften 9, 1-9. Forsyth, F. R. and Hall, 1. V. (1969). Le Naturalisfe Canadien 96, 257-259. Frackowiak, E. (1933). Am. Orch. SOC.Bull. 2, 34. Franklin, L. W. (1967). Orch. Rev. 75, 213-215. Frei, J. K. (1973a). Am. Orch. SOC.Bull. 42, 307-314. Frei, J. K. (1973b). Am. Orch. SOC.Bull. 42, 701-708. Frei, J. K. (1976). The ecology of epiphytic orchids in relation to their substrates. In “First Symposium on the Scientific Aspects of Orchids” (H. H. Szrnant and J. Wernple, Eds), pp. 46-62. Chern. Dept. University of Detroit. Frei, J. K. and Dodson, C. H. (1972). Bull. Torrey bot. Club 99, 301-307. Freson, R. (1969). Bull. SOC.Royale Bot. Belgique 102, 205-209. Fuchs, A. and Ziegenspeck, N. (1924). Bot. Arch. 6, 193-206. Furrnan, T. E. (1959). Proc. Ninth Int. Bot. Congr. Toronto 2, 1271. Cailhofer, M. and Thaler, I. (1975). Phyton 17, 159-165. Gallaud, I. (1904). C. R. Acad. Sci., Paris 138, 513-515. Garay, L. A. (1960). Bot. Mus. Leaflets, Harvard Univ. 19, 57-87. Gaumann, E. (1960). C. R. Acad. Sci., Paris 250, 19441947. Gaumann, E. (1963-1964). Phytopath. Z. 49,211-232.

644

J . ARDITTI

Gaumann, E., Braun, R. and Bazzigher, G. (1950). Phytopath. Z . 17, 3663. Gaumann, E. and Hohl, H. R. (1960). Phytopath. Z . 38,93-104. Gaumann, E. and Jaag, 0. (1945). Experientia 1, 21-22. Gaumann, E. and Kern, H. (1959a). Phytoputh. 2. 23, 347-356. Gaumann, E. and Kern, H. (1959b). Phytopath. 2. 36, 1-26. Gaumann, E., Niiesch, J. and Rimpau, R. H. (1960). Phytopath. Z . 38, 274-308. Gaumann, E., Muller, E., Niiesch, J. and Rimpau, R. H. (1961). Phytopath. Z . 41, 89-96. Gellert, M. (1923). Feddes Repertorium Beih. 25, 1-66. Gessner, F. (1948). Biol. Zentralblatt 67, 457-477. Gessner, F. (1956). Der wasserhaushalt der Epiphyten un Lianen. In “Handbuch den Pflanzenphysiologie” (W. Ruhland and 0. Stocker, Eds), p. 915. Springer Verlag, Berlin. Godfery, M. J. (1925). J. Bot., Lond. 63, 33-40. Goh, C. J. (1971). Int. Symp. Morphogenesis in PI. Cell, Tissue and Organ Cult. VIII. 4, 65-66. Goh, C. J. (1973). Malayan Orch. Rev. 11, 16-18. Goh, C. J. (1975). Ann. Bot. 39, 931-934. Goh, C. J. (1976a). Ann. Bot. 40, 645-646. Goh, C. J. (1976b). PI. Physiol. 57 (Ann. Meeting Suppl.), 64. Goh, C. J. (1977a). Ann. Bot. 41, 1061-1063. Goh, C. J. (1977b). Ann. Bot. 41, 1065-1067. Goh, C. J. (1977~).Rep. Second ASEAN Orch. Sem., Jukarta, Indonesia. Goh, C. J. (1977d). Ann. Bot. 41, 763-769. Goh, C. J., Avadhani, P. N., Lob, C. S., Havegraaf, C. and Arditti, J. (1977). New Phytol. 78, 365-372. Goh, C. J. and Seetoh, H. C. (1973). Ann. Bot. 37, 113-119. Goh, C. J. and Wan, H. Y. (1973). The role of auxins in the flowering of a tropical orchid hybrid Vanda Miss Joaquim. I n “Plant Growth Substances 1973” (Y. Sumiki, Ed.), pp. 945-952. Horokawa Publishing Co., Tokyo. Goh, C. J. and Yang, A. L. (1977). PI. Physiol. 59 (Suppl.), 121. Goldschmidt, E. E. and Huberman, M. (1974). J. Exp. Bot. 25, 534-541. Gosse, P. H. (1863). Gardener N. S . 4, 217. Grant, B. (1895). “The Orchids of Burma”. Hanthawaddy Press, Rangoon. Gray, R. W., Guggisberg, A., Segebarth, K. P., Hesse, M. and Schmid, H. (1976). Helv. Chem. Acta 59, 645-649. Gregg, K. B. (1973). Studies on the control of sex expression in the genera Cycnoches and Catasetum, subtribe Catasetinae, Orchidaceae. Ph.D. Dissertation, University of Miami, Coral Gables, Fla. Gregg, K. B. (1975). Selbyana 1, 101-113. Gregg, K. B. (1979). Selbyana 2 (in press). Griffon, E. (1898). C. R. hebd. Skance Acad. Sci., Paris 127,973-976. Griffon, E. (1899). Ann. Sci. nut. (bot.) ser. 18, 10, 71-76. Gruignard, L. (1886). Ann. Sci. nat. (bot.)ser. 7 , 4, 202-240. Haber, W. (1952). Die Orchidee 3, 14-18. Haberlandt, G. (1928). “Physiological Plant Anatomy”. MacMillan, London. Hadley, G. (1968). Malayan Scientist 4, 23-27. Hadley, G. (1969). New Phytol. 68, 933-939. Hadley, G. (1970a). New Phytol. 69, 549-555. Hadley, G. (1970b). New Phytol. 69, 1015-1023. Hadley, G. (1975). Organization and structure of orchid mycorrhiza. I n “Endo-

ASPECTS OF ORCHID PHYSIOLOGY

645

mycorrhizas” (F. E. Sanders, B. Mosse and P. B. Tinker, Eds), pp. 335-351. Academic Press, London. Hadley, G. (1977a). Abstr. Second Int. Mycol. Congr. Vol. A-L, 258. Hadley, G. (1977b). Abstr. Second Int. Mycol. Congr. Vol. A-L, 259. Hadley, G. and Harvais, G. (1968). New Phytol. 67, 441-445. Hadley, G., Johnson, R. P. C. and John, D. A. (1971). Planta. 100, 191-199. Hadley, G. and Perombelon, M. (1963). Nature 1337. Hadley, G . and Purves, S. (1974). New Phytol. 73,475-482. Hadley, G. and Williamson, B. (1971). New Phytol. 70, 4 4 4 5 5 . Hadley, G. and Williamson, B. ( 1 972). New Phytol. 71, 1 1 11-1 1 1 8. Hager, H. (1957). Proc. Second World Orch. Con$ 130-132. Halevy, A. H. (1975). Proc. Third MPP Meeting, Ege Univ., Bornova, Izmir, Turkey 84-88. Hall, I. V. and Forsyth, F. R. (1967). Can. J. Bot. 45, 1163-1166. Halperin, M. and Halevy, A. (1973-1974). Ann. Rep. Dept. Ornament. Hort. Hebrew Univ. 58-59. Halperin, M. and Halevy, A. (1974-1975). Ann. Rep. Dept. Ornament. Hort. Hebrew Univ. 49-52. Halpin, J . E. and Farrar, M. D. (1965). Am. Orch. SOC.Bulf. 34,416425. Hamilton, R. M. (1977). “When Does It Flower?“ R. M. Hamilton, Publisher, Richmond, Canada. Hampton, J. (1955). Orch. Dig. 19, 53-58. Hanke, M. and Dollwet, H. H. A. (1976). Sci. Biof. J. 2, 227-230. Hanson, A. D. and Kende, H. (1975). PI. Physiof. 55, 663-669. Harbeck, M. (1963). Die Orchidee 14, 58-65. Harbeck, M. (1964). Die Orchidee 15, 57-60. Harbeck, M. (1968). Die Orchidee 19, 1-5. Hardegger, E., Biland, H. R. and Corrodi, H. (1963a). Helv. Chim. Acta 45, 1354-1 360. Hardegger, E., Rigassi, N., Seres, J., Egli, C., Miiller, P. and Fitzi, K. 0. (1963b). Helv. Chim. Acta 46,2543-2551. Hardegger, E., Schellenbaum, M. and Corrodi, H. (1963~).Helv. Chim. Acta 46, 1171-1180. Harley, J. L. (1969). “The Biology of Mycorrhiza” (2nd ed.). Leonard Hill, London. Harrison, C. R. (1972). Am. J. Bot. 59, 672. Harrison, C. R. (1973). “Physiology and Ultrastructure of Cattfeya aurantiaca (Orchidaceae) Germination” Ph.D. Dissertation, Univ. of California, Irvine. Harrison, C. R. (1977). Bot. Gar. 138, 41-45. Harrison, C. R. and Arditti, J. (1970). Orch. Dig. 34, 199-204. Harrison, C . R. and Arditti, J. (1972). Am. J. Bot. 59 (Suppl.), 669. Harrison, C. R. and Arditti, J. (1976). Am. J. Bot. 63, 911-918. Harrison, C. R. and Arditti, J. (1978). Bot. Gaz. 139, 180-189. Harvais, G. (1972). Can. J. Bot. 50, 1223-1279. Harvais, G. (1973). Can. J. Bot. 51, 327-332. Harvais, G. (1974). Can. J. Bot. 52, 451-460. Harvais, G. and Hadley, G. (1967a). New Phytof. 66, 205-215. Harvais, G. and Hadley, G. (1967b). New Phytol. 66, 217-230. Harvais, G. and Pekkala, D. (1975). Can. J. Bot. 53, 156-163. Harvais, G. and Raitsakas, A. (1975). Can. J. Bot. 53, 144-155. Hayes, A. B. (1968). Am. Orch. SOC.Bulf. 37, 705-707.

646

J. ARDITTI

Hayes, A. B. (1969). Mycopafh. Mycol. Appl. 38, 139-144. Healey, P. L., Ernst, R. and Arditti, J. (1971). New Phytol. 70,477-482. Hele, M. P. (1953). Biochem. J. 55, 857-863. Henrici, M. and Senn, G. (1925). Ber. Schweiz. Bot. Ges. 34, 110-141. Heslop-Harrison, J. (1957). Bot. Notiser 110, 28-49. Hetherington, E. E. (1970). Am. Orch. SOC.Bull. 19, 537-539. Hew, C. S. (1976). Proc. Eighth World Orchid ConJ, Frankfurt pp. 426-430. Hijner, J. A. and Arditti, J. (1973). Am. J. Bot. 60, 829-835. Hildebrand, F. (1863a). Ann. Mag. Nut. Hist. Ser. 3, 12, 169-174. Hildebrand, F. (1863b). Bot. Ztng. 21, 329-334, 338-345. Hildebrand, F. (1865). Bot. Ztng. 23, 245-249. Hildebrand, F. (1868). Bot. Ztng. 26, 113-116. Hills, H. G., Williams, N. H. and Dodson, C. H. (1968). Am. Orch. Soc. Bull. 37, 967-971. Hindawi, 1. J. (1970). Ethylene injures vegetation in urban areas. In “Air Pollution Injury to Vegetation”, pp. 32. National Air Pollution Control Administration Publication No. AP-71, U.S. Dep. Health, Educ. and Welfare, Public Health Serv., Environment. Health Serv., Raleigh, N.C. His, Ch. (1807). J. Phys., Chim., Hist. Nut., Arts 65, 241-264. Hoehne, F. C. (1949). “Iconografia de Orchidaceas do Brasil”. Secretaria de Agricultura, Silo Paulo, Brasil. Holdson, J. and Laurie, A. (1949). Proc. Am. Soc. Hort. Sci. 51, 379-380. Hollander, S. (1932). “Ernahrung-physiologische Untersuchungen an Wurzelpilzen saprophytisch lebender Orchideen” Dissertation, Univ. of Wiirzburg (cited by Burgeff, 1936). Holttum, R. E. (1953). Symp. Soc. Exp. Biol. 7, 159-173. Holttum, R. E. (1957). “A Revised Flora of Malaya, I, Orchids of Malaya” (2nd ed.), Government Printing Office, Singapore. Holttum, R. E. (1969). “Plant Life in Malaya”, Longman, London. Homes, J. (1973). J. Microscopie 17, 66a. Homes, J., Freson, R. and Vermylen-Guillaume, M. (1973). Bull. SOC.Royale Bot. Belgique 106, 123-129. Homks, J., Freson, R., Vermylen-Guillaume, M. and Michel, M. (1971a). Ninetysixth Congr. Nut. SOC.Savantes, Toulouse 4, 93-105. Homes, J., Freson, R., Vermylen-Guillaume, M. and Michel, M. (1971b). Ninetysixth Congr. Nut. SOC.Savantes, Toulouse 4, 86. Homes, J. and Vanseveren-Van Espen, N. (1972). J. Microscopie 14, Coll. Annuel. SOC.Francaise Microscopie Electronique, Nantes, p. 55a. Homes, J. and Vanseveren-Van Espen, N. (1973a). Bull. SOC.Royale Bot. Belgiqirr 106, 89-106. Homes, J . and VansCveren-Van Espen, N. (1973b). Bull. SOC.Royale Bot. Belgiqire 106, 117-121. Hsiang, T.-H. T. (195la). PI. Physiol. 26, 441-455. Hsiang, T.-H. T. (1951b). PI. Physiol. 26, 708-721. Huber, B. (1921a). Akad. Wiss. Wien, Math. Naturwiss. KI. Anzeiger 58, 134-135. Huber, B. (1921b). Sitzber. Akad. Wiss. Wien, Math. Naturwiss. KI. Anzeiger 130, 307-328. Hubert, B. and Maton, J. (1939). Biol. Jaarboek 6, 244-285. Hurst, C. C. (1898). J. Roy. Hort. SOC.21, 442-486. Ichihashi, S. and Yamashita, M. (1977). J. Japan SOC.Hort. Sci. 45, 407-413. Israel, H. W. (1963). Am. Orch. SOC.Bull. 32, 441-443.

ASPECTS OF ORCHID PHYSIOLOGY

647

Inthuwong, 0. and Vachrapai, T. (1964). Kasetsart J. 4, 1-6. Ito, I. (1965). Japan Orch. SOC.Bull. 11, 4-15. Janke, H. (1915). Orchis 9, 3-4. Jeffrey, D. C. and Arditti, J. (1968). Orch. Rev. 76, 315-316. Jeffrey, D. C. and Arditti, J. (1969). Am. Orch. SOC.Bull. 38, 866-868. Jeffrey, D. C., Arditti, J. and Koopowitz, H. (1970). New Phytol. 69, 187-195. Jester, R. A. (1952). “A Study of Dried Sepal Injury to Orchid Flowers”. M. S. Thesis, Rutgers University, New Brunswick. Jeyanayaghy, S. and Rao, A. N. (1966). Bull. Torrey bot. Club 93, 97-103. Johanson, D. R. (1974). Acta Phytogeogr. Suecica 59, Uppsala. Johanson, D. R. (1977). Am. Orch. SOC.Bull. 46, 703-707. Johnson, E. and Laurie, A. (1943). Proc. Am. SOC.Hort. Sci. 42,607-608. Johnson, E. and Laurie, A. (1945). Proc. Am. SOC.Hort. Sci. 46, 388. Jones, D. L. (1 972). Victorian Naturalist 89, 260-263. Jurgen-Schrenk, W. (1973). Die Orchidee 24, 108-1 11. Kaewbanrung, M. (1967). Orch. SOC.Thailand Bull. 1, 18-22. Kamemoto, H. and Sagarik, R. (1965). Florida Orchidist 8, 161. Kendrick, J. B., Jr, Darley, E. F., Middleton, J. T. and Paulus, A. 0. (1956). CaliJ Agric. 10, 9-15. Kerr, A. D. (1972). Am. Orch. SOC.Bull. 41, 208-211. Khan, I. (1964). “Preliminary investigations into the Crassulacean acid metabolism of Vandu cv Miss Joaquim” B.Sc., Hons. Dissertation, Botany Dept., University of Singapore. Klass, C. S. (1964). “The Extraction and Identification of Free Auxin from Orchid Pollinia”. M.A. Thesis, Brooklyn College, New York. Knauft, R. L. and Arditti, J. (1969). New Phytol. 68, 617-661. Knauft, R. L., Arditti, J. and Flick, B. (1970). Die Orchidee 21, 132-135. Knudson, L. (1916). Cornell Univ. Agr. Exp. Stu. Mem. 6, 9-75. Knudson, L. (1921). Bull. Real, Sa. Espan’ola Hist. Nut. 21, 25C260. Knudson, L. (1922). Bot. Gaz. 73, 1-25. Knudson, L. (1929). Proc. Znt. Congr. PI. Sci. 2, 1183-1189. Knudson, L. (1946). Am. Orch. SOC.Bull. 15, 214-217. Knudson, L. (1951). Bot. Gaz. 112, 528-532. Koch, L. (1973). Gartenbauwissenscha~t.38, 419-426. Koch, L. and Schulz, D. (1975). Die Orchidee 26, 27-30. Koch, U. (1972). Zierpflanzenbau 4, 131-132. Koopowitz, H. (1964). Am. Orch. SOC.Bull. 33, 1026-1028. Kosugi, K. (1952). J. Hort. Assoc. Japan 21, 179-182. Kosugi, K., Yokoi, M. and Maekawa, S. (1973). Tech. Bull. Far. Hort. Chiba Univ. 21,25-30. Kosugi, K., Yokoi, M. and Yokobori, H. (1973). Tech. Bull. Fac. Hort. Chiba Univ. 21, 31-35. Kristen, U. (1965). “Untersuchungen iiber den Zusammengang zwischen dem C02Gaswechsel und der gleichzeitigen Luftwegigkeit an sukkulented mit un ohne De Saussure-Effect.” Diplomarbeit, Univ. Hamburg. Krizck, D. T. and Lawson, R. H. (1974). Am. Orch. SOC.Bull. 43, 503-510. Kugler, H. (1966). Ber. dt. bot. Ges. 79, 57-70. Kuijper, J. (1931). Trop. Natuur 20, 90-94. Kuijper, J. (1933). Rec. Tvav. bot. nPerZ. 30, 1-22. Kukulczanka, K. and Sobieozczariski, J. (1974). Proc. Nineteenth Znt. hort. Congr., Warsaw lB, 852.

648

J. ARDlTTl

Kukulczanka, K. and Twarda-Prcdota, B. (1973). Acta SOC. Bot. Poloniae 42, 281-294. Kullenberg, B. (1961). “Studies in Ophrys Pollination”. Almquist and Wiksells, Boktryckeri, Uppsala. Kullenberg, B. (1973). Zoon. Suppl. 1, 9-14. Kullenberg, B. and Bergstrlim, G. (1976). Bot. Notiser 129, 11-19. Kusano, E. (1911). J. Coll. Agr. Tokyo 4, 1-66. Kusumoto, M. and Furukawa, J. (1977). J. Japan. SOC.hort. Sci. 45,421-426. LaGarde, R. V. (1929). Ann. Missouri Bot. Card. 16,499-514. Laibach, F . (1930). Planta 9, 341-387. Laibach, F. (1932). Ber. dt. bot. Ges. 50, 383-390. Laibach, F. (1933a). Ber. dt. bot. Ges. 51, 336-340. Laibach, F. (1933b). Ber. dt. bot. Ges. 51, 386-392. Laibach, F. and Maschmann, E. (1933). Jb. wiss. Bot. 78, 399-430. Lebedev, A. F. (1948). Am. Rev. Soviet Med. 5, 15-27. Lee, Y. T. (1970). “Carbon dioxide fixation in orchid leaves”. B.Sc. Hons Dissertation, Botany Dept, University of Singapore. Leimbach, G. (1911). Deutsche Bot. Monatschrift 22, 114-118, 138-142, 155158.

Leopold, A. C. (1964). “Plant Growth and Development”. McGraw-Hill Book Co., New York. Leopold, A. C. (1975). Bio Science 25, 659-662. Lerman, J. C., Deleens, E., Nato, A. and Moyse, A. (1974). PI. Physiol. 53,581-584. Lerman, J . C. and Queiroz, 0. (1974). Science 183, 1207-1209. Letcher, R. M. and Nhamo, L. R. (1973). J. Chem. SOC.12, Perkin Trancations 1, 1263-1 265. Lewis, D. H. (1973). Biol. Rev. 48, 261-278. Lieberman, M., Asen, S. and Mapson, L. W. (1964). Nature 204, 765-758. Lim, S. L., Chin, T. Y. and Hew, C. S. (1975). Proc. Seminar Singapore Inst. Biol. and Singapore Nac. Acad. Sci., pp. 18-26. Lindley, J. (1830). “An Introduction to the Natural System of Botany”, Longman, London. Lindner, R. E. (1946). Hawaii Agr. Exp. Sta. Prog. Notes 49, 1-5. Lindt, 0. (1885). Bot. Z. 43, 825-839. Lugo-Lugo, H. (1955). Am. J. Bot. 42,679-684. Magnus, P. (1887). VerhandlungenBot. Vereins Proviz Brandenburg 28, IV. Magnus, P. (1890). Deutsche Bot. Monatschrifi 8, 97-98. Magnus, P. (1891). Deutsche Bot. Monatschr# 9, 49-51. Magrou, J. (1924). Ann. Sci. nat. (bot.) Ser. 10, Vol. VI, 256-276. Magrou, J. (1936). Third Znt. Congr. Comp. Path., 73-80. Magrou, J . (1938). Ann. Inst. Pasteur 60, 565400. Magrou, J. and Mariat, F. (1945). Ann. Inst. Pasteur 71, 49. Magrou, J., Mariat, F. and Rose, H. (1949). C. R. hebd. SPanc. Acad. Sci. Paris 229, 665-688. Mai, G. (1934). Jb. wiss. Bot. 79, 681-713. Malguth, R. (1902). “Biologische Eigentiimlichkeiten der Friichte epiphytischer Orchideen”. Inaugural Dissertation, Univ. Breslau. Mariat, F. (1944). Rev. Hort. 29, 68-69. Mariat, F. (1948). Rev. Gen. Bot. 55, 229-243. Mariat, F. (1952). Rev. Gen. Bot. 59, 324-377. Mariat, F. (1954). Action de vitamines sur les germinations d‘orchidees. In “Le

ASPECTS OF ORCHID PHYSIOLOGY

649

Vanillier et la Vanille dans le Monde” (G. Bouriquet, Ed.), pp. 4281-4283, Paul Lechavalier, Paris. Martin, C. and Thimann, K. V. (1972). PI. Physiol. 49, 64-71. Maschmann, E. and Laibach, F. (1932). Biochem. Ztschr. 255,446-452. Massart, J. (1895). Bull. SOC.Royale Bot. Belgique 34, 151-343. Matsumoto, K. (1966). Cymb. Soc. News 20, 11-14. McClelland, T. B. (1919). J. Agr. Res. 16, 245-251. McDade, E. (1947). Orch. Dig. 11, 141-146. McIntyre, D. K., Veitch, G. J. and Wrigley, J. W. (1971). Mimeogr. Canberra Bot. Gdns. (also Aust. PI. 6, No. 50, March 1972). McIntyre, D. K., Veitch, G. J. and Wrigley, J. W. (1972a). Mimeogr. Canberra Bot. Gdns. McIntyre, D. K., Veitch, G. J. and Wrigley, J. W. (1972b). Am. Orch. SOC. Bull. 41, 1093-1097. McWilliarns, E. L. (1970). Bot. Cuz. 131, 2850290. Mead, J. W. and Bulard, C. (1975) New Phytol. 74, 33-40. Medawar, P. B . (1957).“TheUniquenessof thehdividual”. BasicBooks, New York. Mejstfik, V. (1970). Biol. Plantarurn 12, 105-109. Menke, W. (1940). Naturwissenschaften 28, 31. Menke, W. and Schmid, G. H. (1976). PI. Physiol. 57, 736-719. Menke, W., and Wolfersdorf, B. (1968). PIantu 78, 134-143. Middleton, J. A., Crafts, A. S., Brewer, R. F. and Taylor, 0. C. (1956). Calif. Agr. 10, 9-10. Milborrow, B. V. (1974) Ann. Rev. PI. Physiol. 25, 259-307. Mitra, G. C. (1970). Proc. Seventh Indian Sci. Congr., 269-270. Mitra, G. C. (1971). IndianJ. Exp. Biol. 9, 79-85. Miura, G. and Hall, R. H. (1973). PI. Physiol. 51, 563-569. Miwa, A. (1937). Orch. Rev. 45, 146-152. Miyazaki, S. (1968). Agr. Bull. Saga Univ. 43, 101-116. Miyazaki, S. and Nagamatsu, T. (1965). Agr. Bull. Saga Prefectorial Univ. 21, 131-1 49. Molisch, H. (1905). Bot. Ztng. 63, 131-144. Moller, A. (1921). “Fritz Muller, Werke, Briefe, Leben” (3rd ed.). (A. Moller, Ed.), Gustav Fischer Verlag, Jena. Montfort, C. (1940). Ber. dt. bot. Ces. 58, 4148. Montfort, C. and Kiisters, G. (1940). Bot. Arch. 40,571-633. Montgomery, J. and Laurie, A. (1944). Ohio Exp. Stu. Bimonthly Bull. 29, 48-55. Morawiecka, B., Kubicz, A., Kukulczanka, K., Koch, A., and Markefka, E. (1973). Acta SOC. Bot. Poloniue 42, 133-141. Morel, G. (1960). Am. Orch. SOC.Bull. 29,495497. Morel, G. M. (1974). Clonal multiplication of orchids. In “The Orchids: Scientific Studies” (C. L. Withner, Ed.), pp. 169-222, Wiley-Interscience, New York. Morita, K. (1918). Bot. Mag., Tokyo 32, 39-52. Mowe, B. L. (1973). Singupore J. Pri. Ind. 1, 20-30. Mukherjee, T. P., Roy, S. and Bose, T. K. (1973). Orch. Rev. 81, 322-323. Mukherjee, T. P., Roy, S. and Bose, T. K. (1974). Indian Hort. 18, 21-24. Muller, F. (1868). Bot. Ztng. 26, 630-631. Muller, F. (1 869). Bot. Ztng. 27, 224-226. Miiller, H. (1883). “The Fertilization of Flowers” (English translation by DArcy W. Thompson), MacMillan and Co., London. Miiller, K. (1966). Sci. J. May, 3-7. Miiller, K. and Leopold, A. C. (1966). Pluntu 68, 167-185.

650

J. ARDITTI

Muller, R. (1953). Beitr. Biol. Pflanzen 30, 1-32. Murashige, T. (1974). A. Rev. PI. Physiol. 25, 135-166. Murashige, T., Kamemoto, T. H. and Sheehen, T. J. (1967). Proc. Am. SOC.Hort. Sci. 91, 672-679. Nagl, W. (1972). Cytobios. 5, 145-154. Nagl, W., Hendon, J. and Rucker, W. (1972). Cell Diff. 1, 229-237. Nagl, W. and Riicker, W. (1972). Z. PI. Physiol. 67, 120-134. Nagl, W. and Rucker, W. (1974). Cytobios. 10, 137-144. Nagl, W. and Rucker, W. (1976). Nucleic Acid Res. 3 , 2033-2039. Nakamura, S. J. (1962). Reich. F. Z. Bor., 50, 487-497. Nakamura, S. J. (1976). Bot. Mag. Tokyo 89, 21 1-218. Nakamura, S. J., Uchida, T. and Hamada, M. (1975). Bot. Mag. Tokyo 88,103-109. Neales, T. F. and Hew ,C. S. (1975). P/antu 123, 303-306. Nicholls, W. H. (1969). “Orchids of Australia”. Thomas Nelson (Australia), Melbourne. Nicolai, W. (1914). Mollers Deutsche Gartner-Zeitung 29, 253-254. Nieuwdorp, P. J. (1972). Acta bot. neerl. 21, 128-144. Nishikawa, T. and Ui, T. (1976). Trans. Mycol. SOC.Japan 17, 77-84. Nishimura, G., Kosugi, K. and Furukawa, J. (1972). J. Japan SOC.hort. Sci. 41, 297-399. Nishimura, G., Kosugi, K. and Furukawa, J. (1976). Orch. Rev. 84, 175-179. Nobkcourt, P. (1923). C. R. hebd. SPanc. Acud. Sci. Paris 177, 1055-1057. Nobecourt, P. (1938). Assoc. Diplomes Microbiol. Fac. Pharm. Nancy Bull., 10-17, 9-32. Noggle, G. R. and Wynd, F. L. (1943). Bot. Gar. 104,455459. Nuerenbergk, E. L. (1961a). Planta 56, 28-70. Nuerenbergk, E. L. (1961b). “Kunzlicht und Pflanzenkultur”. B. V. Verlagsgesellschaft, Miinchen. Nuerenbergk, E. L. (1963). Proc. Fourth World Orch. Con$, Singapore, 158-169. Nuesch, J. (1963). Symp. SOC.Gen. Microbiol. 13, 335-343. Oertli, J. J. and Kohl, H. C. (1960). Gartenbuuwissenscha~t.25, 107-114. O’Neill, M. W. (1958). Am. Orch. SOC.Bull. 27, 537-540. Ordin, L. and Bonner, J. (1957). PI. Physiol. 32,212-215. Owens, L. D., Lieberman, M., and Kunishi, A. (1971). PI. Physiol. 48, 1-4. Pages, P. D. (1971). SERCA Bull. No. 2, College Laguna, Philippines. Pais, M. S. S. (1973-1974). Rev. Cien. Nut. Fac. Cien. Univ. Lisbon ser. C 17, 437-444. Payne, J. H. (1964). Bull. Pac. Orch. SOC.Hawaii 22, 7-8. Payne, J. H. (1968). Orch. SOC.So. Cul. Rev. 7, 9. Peklo, J. (1906). Flora 96, 260-275. Pegg, G. F. (1973). Tr. Br. Mycol. SOC.61, 277-286. Pkrombelon, M. and Hadley, G. (1965). New Phytol. 64, 144-151. Pfitzer, E. (1882). “Grundzuge einer Vergleichenden Morphologie der Orchideen”. Carl Winter’s Universitatsbuchhandlung,Heidelberg. Pierik, R. L. M. and Steegmans, H. H. M. (1972). Z. PJI Physiol. 68, 228-234. Poddubnaya-Amoldi, V. A. (1960). Phytomorphology 10, 185-198. Poddubnaya-Amoldi, V. A. (1964). “Obshtaya Ernbriologia Pokritosemenyikh Rastenii” (General Embryology of Angiosperms), Pub. House Nauka, Moscow. Poddubnaya-Amoldi, V. A. and Selezneva, V. A. (1957). “Orhidei i h Koultoura” (Orchids and their Culture), U.S.S.R. Acad. Sci. Moscow.

ASPECTS OF ORCHID PHYSIOLOGY

651

Poddubnaya-Arnoldi, V. A. and Zinger, N. V. (1961). Application of histochemical technique to the study of embryonic processes in some orchids. I n “Recent Advances of Botany”, Sect. 8, 71 1-714, Univ. of Toronto Press. Pohl, F. (1927). Jb. wiss. Bot. 66, 556-577. Pollard, G. E. (1973). Orquidea (MPx.)3, 187-190. Porsch, 0. (1 908). Mitteilungen Naturwiss. Verein Steiermark 45. Powell, K. B., and Arditti, J. (1975). Mycopathologiu 53, 163-167. Poyanne, A. (191 7). Bull. SOC.Hist. Nut. Afv. N. 8, 1 . Prilieux, E. (1874). Ann. Sci. Nut. Bot. Bio. Veg. ser. 5, 19, 19, 108-118. Purves, S. and Hadley, G. (1975). Measurement of carbon compounds between the partners in orchid mycorrhiza. I n “Endomycorrhizas” (F. E. Sanders, B. Mosse and P. B. Tinker, Eds), pp. 175-194, Academic Press, London. Purves, S. and Hadley, G. (1976). New Phytol. 77, 689-696. Quednow, K. G. (1930). Bot. Arch. 30, 51-108. Quisumbing, E. (1968). Arunetu J. Agr. 15, 195-212. Raghavan, V. (1964). Bot. Cm. 125, 260-267. Raghavan, V. (1976). “Experimental Embryogenesis in Vascular Plants”. Academic Press, New York. Raghavan, V. and Torrey, J. G. (1964). Am, J. Bot. 51,264-274. Raghavendra, A. S. and Das, V. S. R. (1976). New Phytol. 76, 301-305. Ramsbottom, J. (1922a). Charlesworth and Co. Catalog, pp. ii-xvii. Ramsbottom, J. (1922b). Orch. Rev. 30, 197-202. Ramsbottom, J. (1929). Proc. Int. Congr. PI. Sci. 2, 1676-1687. Rangaswamy, N. S. (1968). Phytomorphol. 17,477-487. Rao, A. N. (1967). Histogenesis and organogenesis in orchid seedlings. I n “Seminar on Plant Cell, Tissue and Organ Cultures, New Delhi, India” (B. M. Johri, Ed.), pp. 71-74. Rao, A. N. (1977). Tissue culture in the orchid industry. I n “Applied and Fundamental Aspects of Plant Cell, Tissue and Organ Culture” (J. Reinert and Y. P. S. Bajaj, Eds), pp. 44-69, 207-248, Springer Verlag, Berlin. Rao, A. N. and Avadhani, P. N. (1963). Proc. Fourth World Orch. ConJ, Singapore, 194-202. Rauh, W., Barthlott, W. and Ehler, N. (1975). Bot. Jb. Syst. 96, 353-374. Resvoll, T. R. (191 1). Biol. Arbejaer., 159-165. Reznik, H. (1958). Planta 51, 694-704. Reznik, H., Lichtenthaler, H. K. and Peveling, E. (1969). Plantu 86, 353-359. Ricardo, M. J., Jr and Alvarez, M. R. (1971). Am. J. Bot. 58, 229-238. Ringstrom, S. (1968). Am. Orch. SOC.Bull. 37, 512. Riviere, A. (1866). Jard. Fr. N. S . 12, 268-291. Riviere, A. (1872). Bull. SOC.Bot. France 19, 20-22. Rosenstock, G. (1956). Z. Bot. 44, 77-87. Rossner, F. (1923). Bot. Arch. 3, 61-128. Rotor, G. (1949). Am. Orch. SOC.Bull. 28, 738-739. Rotor, G. B., Jr (1952). Cornell Univ. Agr. Exp. Stu. Bull. 885. Rotor, G. B., Jr (1959). The photoperiodic and temperature response of orchids. I n “The Orchids, a Scientific Survey” (C. L. Withner, Ed.), pp. 397417, Ronald Press, New York. Rubenstein, R., Hunter, D., McGowan, R. E. and Withner, C. L. (1976). Western Regional Meeting Am. SOC.PI. Physiol., 10.

652

J. ARDITTI

Rucker, W. (1974). Z. Pfr Physiol. 72, 338-351. Rucker, W. (1975). Z. Pfr Physiol. 76, 229-237. Ruinen, J. (1953). Ann. Bogor. 1, 101-157. Rutgers, A. A. L. and Went, F. A. F. C. (1915). Ann. Jard. Bot. Buitenzorg 16, 129-160. Salisbury, R. A. (1804). Trans. Linn. SOC.London 7, 29-32. Sanford, W. W. (1971). Bot. J. Linn. SOC.64, 163-181. Sanford, W. W. (1974). The ecology of orchids. In “The Orchids, Scientific Studies” (C. L. Withner, Ed.), pp. 1-100, Wiley-Interscience, New York. Sano, Y., Kataoka, K. and Kosugi, K. (1961). J. Japan. SOC.horr. Sci. 30, 178-182. Savina, G.I. (1974). Fertilization in orchids. In “Fertilization in Higher Plants” (H. F. Linskens, Ed.), pp. 197-204, American Elsevier Co., New York. Sawa, Y. (1969). Res. Reps. Kcchi Univ. Agr. Sci. 18, 37-40. Saylor, J. C., Jr (1954). “A Study of Ethylene-induced Injury to Orchid Flowers”. M. S. Thesis. Rutgers Univ., New Brunswick. Schlechter, R. (1977). “Die Orchideen, Third Edition” (F. G. Brieger, R. Maatsch and K. Senghas, Eds), Vol. 2. Verlag Paul Parey, Berlin. Schmid, G. H., Jankowicz, M. and Menke, W. (1976). J. Microscope et Biol. Cell. 26,25-28. Schopfer, W. H. (1943). “Plants and Vitamins”, pp. 250-253, Chronica Botanica, Waltham. Schultes, R. E. and Pease, A. S. (1963). “Generic Names of Orchids, Their Origin and Meaning”. Academic Press, New York. Schumacher, W. (1931). Jb. wiss. Bot. 75,581-608. Schweizer, D. and Nagl, W. (1976). Exp. Cell Res. 98, 411423. Seifriz, W. (1923). Am. J. Bot. 10, 32-37. Senn, F. (1927). Verhandl.Naturforsch. Ges. Basel. 38, 516526. Seshagiriah, K. N. (1941). Curr. Sci. 10, 30-32. Shechter, Y. and Arditti, J. (1972). Orch. Rev. 80, 147, 152. Shechter, Y. and Arditti, J. (1973). Orch. Rev. 81, 241-251. Sheehan, T. J. (1952). “A Study of Respiration and Storage of Flowers of the Family Orchidaceae”. Doctoral Dissertation, Cornell University. Sheehan, T. J. (1954). Am. Orch. SOC.Bull. 23, 241-246. Sheehan, T. J., Murashige, T. and Kamemoto, H. (1965). Am. Orch. SOC.Bull. 34, 228-232. Shimokawa, K. and Kasai, Z. (1967). Science 156, 1362-1363. Sideris, C. P. (1950). Bull. Pacific Orch. SOC.Hawaii 8, 337-339. Smith, D., Muscatine, L. and Lewis, D. (1969). Biol. Rev. 4, 17-90. Smith, J. J. (1926). Ann. Jard. Bot. Buitenzorg 35, 55-70. Smith, J. J. (1927). Orch. Rev. 35, 13. Smith, S. E. (1966). New Phytol. 65, 488-499. Smith, S. E. (1967). New Phytol. 66, 371-378. Smith, S. E. (1973). New Phytol. 72,497-499. Smith, S . E. (1974). CRC Crit. Rev. Microbiol. 3, 275-313. Smyth, R. (1969). The Orchudian 3, 61. Spencer, P. W. and Titus, J. S. (1973). PI. Physiol. 51, 89-92. Sprau, F. (1937). Jb. wiss. Bot. 85, 151-168. Sprengel, C. H. (1793). “Das Entdekte Geheimnis der Natur im Bau und in der Befruchtung der Blumen”. Friedrich Vieweg der reltern Berlin (Reprint, 1972, by J. Cramer, Lehre; Wheldon & Wesley Ltd, Codicote and Stechert-Hafner, New York).

ASPECTS OF ORCHID PHYSIOLOGY

653

Stephen, R. C. and Fung, K. K. (1971a). Can. J. Bof. 49,407410. Stephen, R. C. and Fung, K. K. (1971b). Can. J. Bot. 49, 411415. Stephens, G. C. and North, B. B. (1974). Biology. John Wiley & Sons, Inc., New York. Storey, W. B., Kirch, H. W., Pierce, D. J. and Kamemoto, H. (1947). Rep. Univ. Hawaii Agr. Exp. Sta., June, 1946, 82-86. Stoessl, A., Rock, G. L. and Fisch, M. H. (1974). Chem. and Ind. 17, 703-704. Stoutamire, W. (1974). Terrestrial orchid seedlings. In “The Orchids, Scientific Studies” (C. L. Withner, Ed.), Wiley-Interscience, New York. Strauss, M. S. (1976). “The Physiology of Pollination Induced Phenomena in the Orchidaceae”. Doctoral Dissertation, Univ., California, Irvine. Strauss, M. S. and Arditti, J. (1973). Am. J. Bot. 60, 29 (Abstr.). Strauss, M. S. and Koopowitz, H. (1973). Amer. Orch. SOC.Bull. 42, 495-502. Strauss, M. S. and Reisinger, D. M. (1976). Am. Orch. SOC.Bull. 45, 722-723. Strullu, D. G. (1974). Coll. SOC.Fr. Phytopath., 339-340. Strullu, D. G. and Gourret, J. P. (1974). J. Microscopie 20, 285-294. Sunding, P. (1963). BIythia 21, 110-1 15. Summerhayes, V. S. (1968). “Wild Orchids of Britain”. Collins, London. Svihla, R. D. and Osterman, E. (1943). Science 98, 23-24. Swamy, B. G. L. (1948). Lea.. bot. Mus. Harvurd Univ. 13, 61-95. Szarek, S. R., and Ting, I. P. (1977). Photosynthetica 11, 330-342. Tan, T. N. and Hew, C. S. (1973). J. Singapore natn. Acad. Sci. 3, 282-296. Thien, L. B. (1971). Am. Orch. SOC.Bull. 40, 877-880. Thimann, K. V. (1956). Am. J. Bot. 43,241-250. Thompson, C. R. (1963). Orch. SOC.So. Calif. Rev. 5, No. 5, p. 14. Thompson, P. A. (1974a). J. Royal Hort. SOC.99, I 1 7-122. Thompson, P. A. (1974b). Orch. Rev. 82, 179-183. Thompson, P. A. (1977). “Orchids from Seed”. Roy. Bot. Gardens, Kew. Tienken, H. G. (1947). Am. Orch. Soc. Bull. 35, 11 1. Tokunaga, Y. and Nakagawa, T. (1974). Trans. Mycol. SOC.Japan 15, 121-133. Torikata, H., Sawa, Y. and Sisa, M. (1965). J. Japan. SOC.hort. Sci. 34, 63-70. Tran Thanh Van, M. (1970). Proc. Int. Hort. Conf., Tel Aviv, 35-36. Tran Thanh Van, M. (1974). Am. Orch. SOC.Bull. 43, 699-707. Tran Thanh Van, M. and Trippi, V. (1970). Reps. Proc. Int. Hort. Con/ 1, 3. Treub, M. (1887). SOC.Royale Bot. Belgique 26, 182-186. Treviranus, L. C. (1827). Untersuchungenuber die Natur des Menschen der Thiere und der Pganzen 2, 189-229. Trippi, V. S. and Tran Thanh Van, M. (1971). PI. Physiol. 48, 506-508. Tsuchiya, I. (1935). Orch. Rev. 43, 330-337. Ueda, H. and Torikata, H. (1968). J. Japan. Soc. hort. Sci. 37, 240-248. Ueda, H. and Torikata, H. (1969). J. Japan. SOC.hort. Sci. 38, 188-193. Ueda, H. and Torikata, H. (1972a). Am. Orch. SUC.Bull. 41, 322-327. Ueda, H. and Torikata, H. (1972b). Znt. SOC.Hort. Third Symp. Sub-Trop., Trop. Hort. Inst., Hort. Res (ICAR), Bangalore, India. Ueda, H. and Torikata, H. (1974). J. Japan. SOC.hort. Sci. 43,281-285. Urech, J., Fechtig, B., Niiesch, J. and Visher, E. (1963). Helv. Chim. Acta 46, 2758-2766. Urmston, J. W. (1949). Am. Orch. SOC.Bull. 18, 652-655. Vacin, E. F. (1952). Am. Orch. SOC.Bull. 21, 601-813. Vacin, E. F. and Went, F. W. (1949). Bot. Gaz. 110, 605-613. Vajrabhaya, M. (1976). Fuc. Sci. Chulalongkorn Univ. 1, 105-1 15.

654

J. ARDIlTI

Vajrabhaya, M., and Vajrabhaya, T. (1976). Chulalongkorn Res. J. 3, 109-127. Van der Donk, J, A. W. M. (1974). Molec. Gen. Genet. 131, 1-8. Van der Pijl, L. and Dodson, C. H. (1966). Orchid Flowers, Their Pollination and Evolution. Univ. of Miami Press, Coral Gables. Vanseveren, N. (1969). Bull. SOC.Royale Bot. Belgique 102,211-220. Vanseveren-Van Espen, N. (1971). Bull. SOC.Royale Bot. Belgique 104, 2945. Vanseveren-Van Espen, N. (1973). Bull. SOC.Royale Bot. Belgique 106, 107-1 15. Vanseveren-Van Espen, N. and Courtez-Geerink, D. (1974). Bull. SOC.Royale Bot, Belgique 107, 259-270. Varner, J. E. (1965). Death. Zn “Plant Biochemistry” (J. Bonner and J. E. Varner, Eds), pp. 867-892, Academic Press, New York. Veitch, G. J. and McIntyre, D. K. (1972). Mimeogr. Canberra Bot. Gdns. Veitch, H. J. (1887). J. Linn. SOC.(Bot.) 24, 395406. Veitch, H. J. and Sons (1894). “A Manual of Orchidaceous Plants”. Reprint 1962 by A. Asher & Co., Amsterdam. Vermeulen, P. (1947). “Studies on Dactylorchis” Dissertation. Univ. of Amsterdam. Veyret, Y. (1969). Trav. Lab. “La Jaysinia” 3, 89-98. Veyret, Y. (1974). Development of the embryo and the young seedling stages of orchids. I n “The Orchids, Scientific Studies’’ (C. L. Withner, Ed.,) pp. 223265, Wiley-Interscience, New York. Von Hofsten, A. (1973). Zoon Suppl. 1, 93-96. Voth, W. (1976). Proc. Eighth World Orch. Conf., Frankfurt, 351-358. Wachter, J. (1799-1801). Arch. Bot. 2, 209-215. Wan, A. S. C., Axel, R. T. and Nicholas, H. J. (1971). Phytochem. 10, 22672269. Warburg, 0. (1886-1888). Untersuchungen Bof. Znst. Tiibingen 2, 53-150. Warcup, J. H. (1971). New Phytol. 70, 41-46. Warcup, J. H. (1973). New Phytol. 72, 387-392. Warcup, J. H. (1975). Factors affecting symbiotic germination of orchid seed. In “Endomycorrhizas” (F. E. Sanders, B. Mosse and P. B. Tinker, Eds), pp. 87-104, Academic Press, London. Warcup, J. H. and Talbot, P. H. B. (1967). New Phytol. 66, 631-641. Warcup, J. H. and Talbot, P. H. B. (1962). Trans. Br. Mycol. SOC.45, 495-518. Warcup, J. H. and Talbot, P. H. B. (1963). Trans. Br. Mycol. SOC.46,465-472. Warcup, J. H. and Talbot, P. H. B. (1965). Trans. Br. Mycol. SOC.48, 249-259. Warcup, J. H. and Talbot, P. H. B. (1971). New Phytol. 70, 35-40. Ward, E. W. B., Unwin, C. H. and Stoessl, A. (1975a). Can. J. Bot. 53, 964-971. Ward, E. W. B., Unwin, C. H. and Stoessl, A. (1975b). Phytopath. 65, 632-633. Webber, F. (1920). Ber. dt. bot. Ges. 38, 233-242. Wells, H. G. (1960). The flowering of the strange orchid. I n “The Short Stories of H. G. Wells”, pp. 203-21 1. Ernest Benn Ltd, London, 18th Impression. Went, F. A. F. C. (1898). Ann. Jard. Bot. Buitenzorg (Suppl. II), 73-77. Went, F. A. F. C. (1917). Naturwissenschaften 5, 72-76. Went, F. A. F. C. and Rutgers, A. A. L. (1915). Afdeeling Natuurkunde 24,513-517. Went, F. W. (1930). Trop. Natuur 19, 119. Went, F. W. (1940). Ann. Jard. Bot. Buitenzorg 50, 1-198. Went, F. W. (1946). Orch. Dig. 10, 86-89. Went, F. W. (1951). Cymbidium SOC.News 6, 10-12. Werkmeister, P. (1970a). Die Orchidee 21, 126-131. Werkmeister, P. (1970b). Ber. dt. Bot. Ges. 83, 19-26. Werkmeister, P. (1971). Bot. Gaz. 132, 346-350.

ASPECTS OF ORCHID PHYSIOLOGY

655

Wheeler, L. C. and Ramos, L. J. (1965). Mimeogr., Dep. Biol. Sci., Univ. So. Cal., Los Angeles. Wiefelputz, E. (1970). Die Orchidee (Special Issue). Wiesmeyer, H. and Hofsten, A. V. ( 1 976). Electron microscopy of orchid seedlings, In “First Symposium on the Scientific Aspects of Orchids” (H. H. Szmant and J. Wemple, Eds), pp. 1 6 2 6 , Chem. Dep., Univ. Detroit. Wiesner, J. (1865). Jb. wiss. Bot. 8, 575-594. Wiesner, J. (1871). Bot. Ztng. 29, 619. Wiesner, J. (1 874). Flora 57, 73-77. Wiesner, J. (1897). Sitzb. Math.-Naturwiss. KI. Acad. wiss. Vienn 56, 77-98. Wildhaber, 0. J. (1972). J. Ber. Naturwiss. Verein Whippertal25, 61-66. Wildhaber, 0 .J. (1974). Die Orchidee 25, 225-230. Williams, B. S. (1877). “Orchid Grower’s Manual”, Victoria and Paradise Nurseries, London. Williamson, B. (1970). Planta 92, 347-354. Williamson, B. (1971). Absfr. First Int. Mycol. Conf, Exeter VII, 2A. Williamson, B. (1973). Plunta 112, 149-158. Williamson, B. and Hadley, G. (1969). Nature 222, 582-583. Williamson, B. and Hadley, G. (1970). Phytopathology 60, 1092-1096. Wirth, M. and Withner, C. L. (1959). Embryology and development in the orchidaceae. In “The Orchids, A Scientific Survey” (C. L. Withner, Ed.), pp. 155188, Ronald Press, New York. Withner, C. L. (1951). Am. Orch. SOC.Bull. 20, 276-278. Withner, C . L. (1959). Orchid physiology. I n “The Orchid, A Scientific Survey” (C. L. Withner, Ed.), pp. 315-360, Ronald Press, New York. Withner, C. L. (1974). Developments in orchid physiology. I n “The Orchids, Scientific Studies”, pp. 129-168, Wiley-Interscience, New York. Withner, C. L. (1977). Threatened and endangered species of orchids. I n “Extinction is Forever” (G. T. Prance and T. S. Elias, Eds), New York Botanical Garden, Bronx. Withner, C. L., Nelson, P. K. and Wejksnora, P. J. (1974). The anatomy of orchids. I n “The Orchids, Scientific Studies” (C. L. Withner, Ed), pp. 267-374, WileyInterscience, New York. Wolff, H. (1927). Jb. W ~ S S Bot. . 66, 1-34. Wolff, H. (1933). Jb. wiss. Bot. 77, 657-684. Wong, S. C. and Hew, C. S. (1973). J , Singapore Acad. Sci. 3, 150-157. Wong, S. C. and Hew, C. S. (1975). Am. Orch. SOC.Bull. 44,902-906. Wright, D. (1967). Orch. Rev. 75, 120-122. Wrigley, J. W. (1973). Mimeogr. Canberra Bot. Gdns. Wrigley, J. (1976). Proc. Eighth World Orch. Conf. Frankfurt, 397-399. Wrobel-Stermiriska, W. (1961). Bid. Ogorod. Bot. 5, 163-165. Wynd, F. L. (1933). Ann. Missouri Bot. Card. 20, 223-237. Yang, S . F. (1974). Recent Advances in Phytochem. 7, 131-164. Ziegler, A., Sheehan, T. and Poole, R. (1967). Am. Orch. SOC.Bull. 36, 185-202. Zimmerman, P. W. and Hitchcock, A. E. (1939). Contr. Boyce Thompson Znst. PI. Res. 10, 481-508. Zimmermann, W. (1932). Jb. wiss. Bot. 3, 393-506. Zotkiewicz, R. (1961). B i d . Ogorod. Bot. 5, 102-104.

This Page Intentionally Left Blank

AUTHOR INDEX The numbers in italics indicate the pages on which names are mentioned in the reference lists A

Abeles, F. B., 602, 638 Abdel Kader, A. B., 84, 109 Abreu, I., 20 100 Adams, D. H., 345, 348, 349, 359, 365, 370, 380,417 Adams, M. S., 522, 529, 638 Adams, P., 566, 638 Adams, R. M., 522, 555, 564, 642 Adams, S. N., 158,219 Agapova, L. P., 227, 288, 289, 299, 332 Akamine, E. K., 623, 625, 626, 638 Akazawa, T., 55, 96, 110, 114 Akerlund, H. E., 11, 100 Albertsson, P. A., 1 1 , 28, 100, 108, 116 Aleksandrovskaya, V. A., 122, 222 Allen, M.B., 54, 100, 116, 118, 125, 126, 127, 138, 144, 145, 149, 219 Allum, A. I., 282, 331 Alvarez, M.R., 438, 487, 488, 638, 6.51 Amelunxen, F., 19, 100 Ames, O., 424, 508, 533, 652, 575, 638 Amin, J. V., 120, 221 Amoore, J. E., 288, 328 Anagnostakis, S. L., 347, 348, 417 Andersen, J., 568, 638 Anderson, J. M., 3, 82, 100 Andersson, B., 11, 28, 100, 108 Andreeva, I. N., 227, 278, 279, 288, 289, 332 Anonymous, 423, 568, 623, 638 Apostolakos, P., 18, 104 Appelqvist, L. A., 91, 100 Arber, A., 281, 308, 328 Arditti, J., 422, 423, 424, 425, 438, 440, 441, 442, 443, 446, 447, 448, 449, 450, 451,455,456,458,461, 462,464,466, 467,468, 469, 471, 474, 477, 481, 482, 483, 489,490, 491, 494, 495, 496, 503, 505, 506, 507, 508, 510, 511, 512, 516,

519, 521, 522, 523, 524, 527, 528, 529, 530, 531, 532, 533, 534, 543,544, 545, 546, 549, 550, 551, 552, 553, 554, 555, 558, 559, 560, 566, 568, 569, 570, 572, 573, 574, 575, 576, 577, 580, 581, 582, 583, 585, 586,588, 589, 592, 594, 595, 596, 600, 601, 602, 607, 608, 609, 611, 617, 619, 620, 621, 625, 629, 630, 631, 632, 633, 635, 637, 638, 639, 640, 641, 642, 643, 644, 64.5, 646, 647, 651, 6.52, 653 Arens, P., 550, 640 Armstrong, J. J., 22, 114 Armstrong, W., 244, 250, 255, 257, 260, 266, 272, 275, 277, 281, 282, 283, 284, 285, 287, 288, 290, 294, 297, 305, 307, 308, 309, 31 1, 3 12, 320, 328, 330, 331 Arnon, D. I., 54, 100,116, 118, 125, 126, 127, 138, 144, 145, 149, 157, 168, 219, 221 Arntzen, C. J., 3, 5, 14, 100 Asahi, T., 70, 100 Asen, S., 623, 648 Ashby, W. C., 124, 138, 141, 190, 219 Astafurova, T. P., 280, 328 Astier, T., 81, 107 Atchison, B. A., 21, 116 Atkins, C. A., 70, 104 Attoe, 0. J., 120, 223 Avadhani, P. N., 447,461,464,465, 471, 519, 521, 522, 523, 527, 528, 529, 532, 533, 573, 612, 613, 640, 644, 6.51 Avrameas, S. 9, I00 Axel, R. T., 51 1, 654 B Backman, K., 488, 641 Bahl, J., 34, 38, 41, 42, 43, 44, 100 Bailey, A., 6, 101 Bain, J. M., 18, 21, 102

657

658

AUTHOR INDEX

Baker, D. A., 237,330 Baker, P. G., 546, 640 Baldry, C. W., 27, 54, 57, 60, 71, 100, 101,102 Ball, E. A., 532, 634, 640, 641 Ballantine, J. E. M., 17, 29, 108 Bamberger, E. S., 58, 100 Banath, C. L., 288, 328 Bange, G. G. J., 260, 261, 328 Bannister, T. T., 311, 330 Barash, H., 185, 207, 221 Barber, D. A., 226,272,296, 328, 332 Barber, D. J., 72, 73, 100, 108 Barber, J., 27, 32, 66, 110 Barrett, D. K., 348, 359, 365, 373, 417 Barnett, H. L., 392, 418 Barsukov, L. I., 84, I01 Barthlott, W., 438, 640, 651 Barton, D. H. R., 510, 512, 643 Barton, R., 4, 100 Bartsch, G., 503, 640 Baskin, S. I., 552, 640 Bassham, J. A., 27, 53, 54, 59, 65, 70, 100,106,111 Bateman, J., 640 Baumeister, W., 121, 219 Bauza, J., 387, 418 Bazzigher, G., 510, 644 Beadle, N. C. W., 124, 138, 141, 190, 219 Bear, F. E., 128,223 Beardsell, D. V., 279, 305, 307, 331 Beau, C., 489, 640 BeaumCe, J., 550, 640 Becker, W. M., 21, 104 Bedbrook, J. R., 74, 102 Beer, J. G., 424, 427, 428, 429, 438, 640 Beevers, H., 91, 94, 104, 115, 204, 220 Beijerind, M. W., 505, 640 Belobrodskaya, L. K., 67, I01 Belsky, M. M., 148, 186, 223 Bendrat, M., 519, 521, 524, 526, 527, 529, 640 Benecke, W., 219 Benike, L. A., 489, 640 Benne, E. J., 121, 127, 158,220 Benson, A. A., 11, 29, 32, 34, 35, 37, 39, 41, 42, 43, 44, 45, 47, 53, 56, 67, 69, 79, 86, 87, 95, 100, 102, 103, 104, 106, 108 Benzing, D. H., 528, 640 Bergelson, L. D., 84, 101

Berger, K. C., 120, 223 Berger, S., 56, 64, 114 Bergfield, R., 12, 107 Bergstrom, G., 522, 566, 648 Bernard, N., 424, 447, 489, 490, 496, 508, 519, 634, 640 Berry, J. A., 166, 223 Berry, L. J., 284, 328 Bershein, B. I., 122, 205, 222 Bertrams, M., 89, 93, 101, 105 Bertrand, G., 127, 219 Berzborn, R., 7, 111 Beyer, E. N., Jr, 523, 640 Biale, J. B., 43, 113 Bickel, H., 48, 101 Bierweiler, B., 488, 641 Bill, K. Ya., 67, 101 Biland, H. R., 510, 516, 645 Billecocq, A., 7, 9, 26, 43, I01 Bird, I. F., 59, 101 Birner, H., 119, 219 Bisalputra, T., 4, 5, 6, 17, 18, 101 Bismuth, A., 67, 69, 102 Bivins, J. L., 544, 640 Black, C. C., 166, 197, 199, 207, 220 Black, R. F., 138, 149, 159,219 Blaich, R., 346, 359, 373, 418 Blair, E. G., 52, 74, 78, 101, 103 Blakeman, J. P., 501, 503, 640 Blank, M., 67, I01 Blee, E., 87, 101 Blis, C. A., 552, 640 Bloch, K., 89, 111 Block, M., 48, 102 Blobel, G., 75, 77, 78, 79, 101, 102, 103 Blowers, J., 489, 640 Blowers, J. W., 489, 640 Boag, T. S., 200, 202, 208, 209,219 Boardman, N. K., 3, 52, 95, I01 Boasson, R., 12, 101 Boatman, D. J., 282, 328 Boeke, J. E., 281, 328 de Boer-Bolt, B. M., 281, 331 Bogorad, L., 21,74,95,96,101, 102,106 Bohnert, H. J., 95, 106 Boidin, J., 353, 379,417 Boisard, J., 48, 101, 109 Bollen, W. B., 384, 385, 419 Boller, A., 510, 640 Bolton, P., 93, I01 Bomsel, J. L., 65, I01 Bonner, J., 447, 650

659

AUTHOR INDEX

Bonner, W. D., 26, 35, 56, 103, 111 Borriss, H., 442, 464, 465,467, 471, 503, 519, 523, 526, 528, 532, 640 Bonner, W. D., 26, 35, 56, 103, 111 Borriss, H., 442, 464, 465, 467, 471, 503, 519, 523, 526, 528, 532, 640 Bose, T. K., 443, 460,465,467,473,477, 452, 456, 640, 649 Bostwick, C. D., 144,219 Bottomley, W., 52, 74, 101 Bouck, G. B., 18, 101 Boulter, D., 280, 328 Bourdu, R., 20, 101 Bouriqvet, G., 621, 640 Bouyovcos, G. J., 231, 328 Bove, C., 81, 107 Bove, J. M., 81, 107 Bowes, G., 65, 101 Bowling, D. J. F., 237, 328 Bos, G. F., 128, 129, 132, 219 Boyd, H., 623, 640 Boyd, J., 566, 640 Boyle, J. E., 33, 114 Boynton, W. P., 233, 328 Bracey, S., 623, 640 Bracker, C. E., 44, 106, 205, 221 Bradbeer, J. W., 21, 74, 101, 109 Brattain, W. H., 233, 328 Braun, R., 510, 644 Brefeld, O., 350, 417 Bretscher, M. S., 44, 73, 78, 84, I01 Brewer, K., 544, 632, 640 Brewer, R. F., 623, 649 Briantais, J. M., 3, 5, 14, 100 Brieger, F. G., 542, 543, 545, 546, 640 Brierley, G., 56, 110 Briggs, W. R., 48, 101, 108, 109 Brongniart, A., 568, 640 Broaks, J. L., 92, I01 Brown, C. L., 234, 280, 281, 305, 307, 308, 309, 330 Brown, H., 360, 328 Brown, L. R., 144, 219 Brown, R., 226,328, 569, 640 Brown, R. H., 166, 197, 220 Brown, R. M., Jr, 95, 113 Brown, W. V., 159, 223 Brownell, P. F., 118, 125, 126, 127, 128, 129, 131, 134, 135, 136, 137, 139, 140, 142, 143, 144, 145, 146, 150, 151, 152, 153, 154, 156, 157, 159, 161, 162, 166, 167, 168, 169, 170, 172, 173, 174, 175,

176, 177, 180, 182, 186, 187, 188, 189, 190, 191, 193, 194, 195, 196, 197, 198, 199, 202, 203, 212, 214, 219 Browning, G., 48, 101 Broyer, T. C., 121, 124, 134, 143, 144, 149,219,221, 222, 237, 239 Brulfert, J., 168, 219 Bruins, H. R., 233, 328 Bryant, M. P., 148, 219 Bucke, C. 27, 54, I01 Buckingham, E., 226, 328 Bulard, C., 422, 443, 447, 649 Buller, A. H. R., 350, 376, 417 Bulliard, P., 568, 640 Bunning, J., 70, 112 Burg, S. P., 542, 573, 600, 601, 611, 621, 622, 623, 625, 626, 629,641,642 Burgeff, J., 423, 425, 438, 462, 489, 490, 491, 496, 497, 503, 505, 507, 519, 641 Burgeff, H., 424,425,446,450,489, 490, 496, 505, 641 Burghoffer, C., 48, 102 Burges, A., 477, 505, 508, 641 Burnett, J. H., 349, 351, 372, 376, 377, 378, 417 Burkill, I. H., 550, 641 Burkhard, C., 21, 112 Burstrom, H., 447, 641 Burton, A. C., 26, 101 Butcher, A. C., 348, 417 Butler, G. M., 413, 418 Butler, R. D., 4, 102 Butow, R. A., 77, 107 Butz, R. G., 215, 220 Buvat, R., 4, 102 Bystrom, B. G., 95, 114

C Caldwell, M., 166, 223 Calvin, M., 53, 100, 102 Camefort, H., 18, 102 Campbell, A. H., 344, 345, 348, 418 Campbell, P. N., 75, 77, 102 Campbell, W. H., 199, 207, 220 Canny, M. J., 234, 328 Canvin, D. T., 70,91,104,116,204,220 Capesius, I., 488, 641 Capindale, J. B., 54, 116 Cappelletti, C., 446, 469, 641 Carde, J. P., 19, 102

660

AUTHOR INDEX

Carlton, A. B., 124, 134, 143, 144, 149, 219, 221 Carr, N. G., 16, 116 Carrayol, E., 48, 102 Casamajor, R., 534, 641 Cartwright, K. St G., 345, 406, 418 Caten, C. E., 347, 374, 377, 418 Catesson, A. M., 14, 102 Catlin, P. B., 279, 331 Cavalier-Smith, T., 96, 102 Chadwick, A. V., 573, 583, 600, 602, 619, 631, 639 Champigny, M. L., 39, 67, 69, 102 Chap, H. J., 84, 102 Chapman, E. A., 18,21, 39, 102, 111 Chapman, S., 233, 328 CharriCre-Ladreix, Y., 19, 102 Chatin, J., 519, 520, 525, 641 Chautard, J., 424, 641 Cheesbrough, T. M., 84,102 Chen, J. L., 52, 102 Chen, T. M., 166, 197,220 Cheng, K. H., 214, 220 Chennaveeraiah, M. S., 461, 464, 641 Chen-She, S. H., 59, 102 Chkriff, A., 92, 102 Cheung, Y. S., 72, I10 Chevallier, D., 73, 102 Childs, T. W., 345, 348, 380, 418 Chin, T. Y., 581,589,601,607,613,614, 615, 616, 617, 648 Chirkova, T. V., 280,328 Chollet, R., 16, 17,22,102, 159, 166,223 Chon, C. J., 65, 66, 73, 105, 111 Van Deenen, L. L. M., 44, 84, 87, 93, 102, 105,110 Chon, C. J. A., 59, 60,108 Chu, H., 148, 219 Chua, N., 79,102 Chua, N. H., 78,103 Churchill, M. E., 532, 641 Christensen, E. L., 56. 103 Clarke, A. C., 566, 641 Clarkson, D. T., 239, 331 Claus, G., 95, 106 Clayton, G. D., 623, 641 Clifford, H. T., 429,437,438,641 Cobb, A. H., 34, 49, 51, 67, 74, 80, 102 Cockburn, A., 233,330 Cockburn, W., 27, 54,57, 71,100,102 Coen, D. M., 74, I02 Coggins, C. W., 14, 114

Cohen, S. S.,95, 102 Coleman, R. A., 48, 108 Collander, R., 239, 328 Collet, A., 442, 641 Collin, C., 28, 108 Colman, B., 214, 220, 221 Connor, D. J., 121, 220 Conway, V. M., 226,328 Cook, M. T., 495, 641 Cooke, R. J., 48, 49, 67, 102 Cosmans de Ruiter, L., 550, 641 Coombs, J., 3, 5, 60, 102 Cornelius, M. J., 59, I01 Corrodi, H., 510, 516, 640, 645 Cortesi, F., 505, 641 Costantin, J., 424, 641 Coster, C., 550, 641 Costes, C., 48, 102 Cottrell, G. C., 623, 641 Coult, D. A., 280, 294, 296, 297, 328 Courtez-Geerink, D., 451, 654 Coutinho, L. M., 519, 522, 523, 524, 525, 526, 527, 528,641 Coutts, M. P., 305, 308, 309, 310, 311, 328,329 Cowling, T. G., 233, 328 Cowling, E. B., 408, 418, 419 Crafts, A. S.,237, 329, 623, 649 Crafts, C. B., 462, 461 Craig, A. S., 18, 114 Craig, I. W., 16, 116 Craig, S., 166, 197, 221 Cran, D. G., 12, 20, 102 Crank, J., 243, 329 Crawford, R. M. M., 238,280,307,329, 330 Cresti, M., 20, 111 Crocker, W., 623, 642 Croft, J. H., 346, 347, 418 Crofts, A. R., 27, 115 Crossland, C. J., 118, 125, 126, 159, 162, 166, 167, 168, 169, 172, 197, 199, 203, 219 Croteau, R., 19, 108 Crotty, W. J., 19, 20, 21, 102 Cruger, H.,568, 642 Currie, J. A., 226, 238, 259, 313, 315, 326, 329 Curtis, J. T., 442, 444, 445, 446, 474, 497, 534, 600, 602, 619, 626, 631, 642 Czapek, F., 519, 532, 642

66 1

AUTHOR INDEX

D Dalling, M. J., 70, 103 Dallner, G., 84, 110 Van Dam, L., 231, 311, 331 Damon, S. C., 120, 220 Damsz, B., 12, 103 Danielli, J. F., 3, 103 Darley, E. F., 623, 647 Darwin, C., 422, 534, 567, 642 Das, D. K., 281, 329 Das, V. S. R., 70, 111, 206, 223, 525, 527, 651 Davidson, 0. W., 519, 544, 546, 623, 625, 642 Davies, D. D., 280, 329 Davis, G. J., 17, 108 Davson, H., 3, 103 Dawson, R. M. C., 84, 103 Day, D. A., 26, 103 Day, P., 34, 35, 44, 67, 69, 111 Day, S. C., 391, 418 De Bary, A., 413, 418 Debergh, P., 443, 446, 642 De Bruijne, E., 443, 446, 642 de Duve, C., 34, 103 Deleens, E., 528, 648 De Lehaire, J . H., 438, 642 De Lubac, M., 34, 37, 44, 82, 114 Dennis, C., 402, 419 De Pierre, J. W., 84, I03 Devaux, P., 86, 103, I12 Devor, K. A., 82, 83, 103 De Vries, J. T., 546, 642 De Wit, H. C. D., 424, 642 Diano, M., 48, 102 Dieckert, J. W., 129, 137, 223 Diers, L., 12, 20, 113 Dijkman, M . J., 572, 573, 600, 601, 61 1, 621, 622, 625, 626, 629, 641,642 Dilly, R. A., 66, 103 Dittrich, P., 199, 207, 220 Dobberstein, B., 78, 103 Dodson, C., 546, 555, 642 Dodson, C. H., 469, 471, 534, 552, 555, 558, 559, 560, 561, 563, 564, 566, 567, 6 17, 642, 643, 646, 654 Dohler, G., 170, 220 Dolcher, T., 600, 621, 631, 642 Dolcher, T., 600, 621, 631, 542 Dollwet, H. H. A., 506, 645 Dolzmann, P., 21, 103

Donaldson, R. P., 43, I03 Doonan, S., 79, 109 Dorne, A. M., 77, 108 Dorph-Petersen, K., 119, 220 Dorr, I., 530, 642 Douce, R., 3, 7, 21, 26, 27, 28,29, 32, 34, 35, 37, 38, 39,41,42,43,44,45,47,48, 49,50, 51, 52, 53, 56,62, 66, 67, 69,70, 71,72, 75,79, 82, 84, 86, 87, 89,92,93, 96, 101, 102, 103, 105, 106, 107, 108, 110, I l l , 113 Downie, D. G., 462, 468, 469, 563, 506, 642 Downton, W. J. S., 17, 101, 123, 166, 220, 221, 223 Drapeau, G. R., 184, 220 Dressler, R. L., 423, 552, 556, 567, 637, 642, 643 Droop, M. R., 147, 220 Dubacq, J. P., 43, 92, 102, 103 Ducet, G., 48, 102 Dueker, J., 519, 524, 529, 530, 531, 549, 552, 568, 639, 642 Dumas, C., 18, 103 Duncan, E. R., 578, 589, 600, 602, 619, 626, 631, 642 Dunsterville, E., 534, 642 Dunsterville, G. C . K., 534, 642 Dupont, G., 81, 107 Dupont, J., 41, 43, 44, 109 Dyer, A. F., 20, 102 Dykus, A. M., 523, 525, 526, 527, 643

E Eaglesham, A. R., 52, 103 Ebert, M., 226, 272, 328, 329 Echlin, P., 16, 103 Edwards, G. E., 55, 103, 106, 107, 166, 197, 220, 529, 643 Effer, W. R., 280, 329 Egli, C., 510, 645 Ehler, N., 438, 651 Ehmann, A., 423,510,511,601,620,639 Ehrlich, B. A,, 58, 100 Eisenstadt, J. M., 52, 105 Ellis, R. J., 45, 49, 51, 52, 53, 74, 78, SO, 101, 103, 106, 113 101, 103, 106, 113 El-Sheikh, A. M., 121, 122, 220 Emboden, W. A., 637, 643 Emerson, R., 144, 220

662

AUTHOR INDEX

Fisher, J. D., 205, 220 Fitting, H., 534, 569, 589, 590, 592, 600, 602, 619, 643 Fitzgerald, M. P., 27,29,54,62, 108,115 Fitzi, K. O., 510, 645 Flavell, R., 95, 104 Flick, B., 534, 619, 647 Flick, B. H., 423,447,495,510,511, 512, 516, 519, 583, 594, 595, 596, 598, 599, 600, 601, 602, 611, 617, 620, 621, 625, 629, 630, 631, 632,639 Fliege, R., 12, 59, 63, 104, 105 Flowers, T. J., 119, 220 Floyd, G. L., 34, 49, 51, 115 Fliigge, U. I., 37, 49, 51, 59, 62, 104 Fock, H., 22, 112 Fogg, G. E., 206, 214, 220 Forde, B. J., 17, 108 Forde, J., 43, 104 Forger, 111, J., 54, 100 Forrester, M. L., 284, 286, 329 Forsyth, F. R., 626, 643, 645 Fourier, J. B., 243, 329 Frackowiak, E., 521, 533, 643 Francke, B., 34,38,41,42,43,44,100 Franki, R. 1. B., 52, 95, I01 Franklin, L. W., 543, 546, 643 F Fraser, A. I., 311, 329 Frederick, S. E., 21, 22, 104, 110 Fahnestock, G. W., 623, 643 Frei, J. K., 469, 471, 474, 643 Freson, R., 450,451,472,477,484, 643, Falck, O., 350, 418 646 Falck, R., 350, 418 Falk, H., 34, 38, 43, 44, 49, 51, 79, 93, Frey-Wyssling, A., 12, 110 Friedlander, M., 4, 107 103, 108 Farrar, M. D., 472, 477, 645 Frymire, G. P., 552, 642 Fast, G., 442, 443, 471, 473, 474, 477, Fuchs, A., 505, 643 Fulton, J. M., 238, 281, 329 643 Fung, K. K., 468, 491, 494, 505, 653 Faure, M., 7, 26, 43, I01 Fay, P., 206, 220 Furukawa, J., 443,470, 546, 648, 650 Furman, T. E., 495, 643 Fechtig, B., 510, 516, 517, 653 Fehlandt, P. R., 643, 643 G Feierabend, J., 78, 103 Ferrari, R. A., 87, 104 Gailhofer, M., 462, 643 Filner, P., 177, 220 Findlay, W. P. K., 345, 376, 385, 406, Galatis, B., 18, 104 Gallagher, E. A., 49, 111 418 Fineran, B. A., 6, 104 Gallaud, I., 400, 643 Fisch, M., 423, 643 Galston, A. W., 67, 68, 112 Fisch, M. H., 447, 495, 510, 511, 512, Gambrel], R. P., 281, 329 516, 518, 519, 601, 620, 639, 643, 653 Garay, L. A., 422, 643 Fischer, C., Jr, 623, 643 Gardiner, J. B. H., 311, 329 Fischer, C. W., 623, 643 Garnier, M., 81, 107

Engelbrecht, A. H. P., 6, 115 Engler, R. M., 226, 329 Epstein, E., 123, 220 Epstein, S., 159, 166, 223 Erdmann, R., 119, 222 Erickson, A. E., 238, 272, 281, 318, 329, 330 Erickson, L. C., 519, 523, 529, 530, 643 Ernst, R., 440, 447, 448, 449, 450, 451, 455, 462, 466, 468, 469, 471, 473, 475, 476, 477, 481, 482, 483, 503, 505, 506, 507, 510, 511, 512, 639,643,646 Ernster, L., 84, 163 Esau, K., 4, 103 Escombe, F., 260, 328 Esser, G., 519, 643 Esser, K., 346, 359, 373, 378, 418 Etheridge, D. E., 385, 387, 396, 402, 418, 419 Etherington, J. R., 226, 282, 329 Evans, A., 29, 48,49,103 Evans, E. J., 349,417 Evans, H. J., 122, 123,220, 222 Evans, N. T. S., 226, 272, 328,329 Everson, R. G., 66, 103 Eyster, C., 144, 220

663

AUTHOR INDEX

Garrett, S. D., 334, 336, 405, 412, 418 Gaumann, E., 510, 512, 513, 516, 640, 643,644

Gayler, K. R., 33, 34, 104 Gaynard, T. J., 265, 276, 277, 285, 287, 296, 328, 329 Gee, R., 205, 206, 222 Geller, G., 63, 66, 67, 69, 105, 115 Gellert, M., 61 1, 644 Gessner, F., 532, 599, 604, 605, 606, 607, 631, 644 Gibbs, J. N., 403, 418 Gibbs, M., 29, 54, 58, 60, 13, 100, 105, 107, 109, I l l , 114 Gibbs, S. P., 18, 95, 104 Giles, K. L., 33, 74 Gill, C. J., 305, 307, 329 Gimmler, H., 72, 104 Glaser, T. E., 387, 418 Godfery, M. J., 566, 644 Goh, C. J., 461, 519, 521, 522, 523, 521, 528, 529, 532, 533, 534, 542, 544, 546, 640, 644

Goldschmidt, E. E., 92, 106, 604, 631, 644

Goldstein, S., 148, 186, 223 Gonzalez, E., 94, 104 Good, M. H., 413,418 Goodchild, D. J., 18, 111 Goodenough, U. W., 22,114 Goodman, D., 238, 272, 315, 326, 329 Goodwin, P. B., 239, 329 Goodwin, T. W., 19, 46, 48, 104, 112 Gordon, M., 67, 68, 112 Gorharn, P. R., 234, 331 Gosse, P. H., 568, 644 Gossett, D., 121, 221 Gourret, J. P., 501, 653 Cove, D. W., 18, 21, 102 Grable, A. R., 231, 329 Gracen, N. E., 166, 197, 220 Gradowski, C., 544, 632, 640 Graham, D., 55, I10 Grainjer, J., 413, 418 Granick, S., 3, 104 Grant, B., 422, 637, 644 Grant, B. R., 33, 34, 70, 104 Gray, J. C., 74, 104 Gray, R. W., 510, 644 Green, M. S., 226, 282, 329 Greene, R. W., 95, 114 Greenway, H., 279, 280, 330

Greenwood, A. D., 3, 5, 28, 102, 104 Greenwood, D. J., 227, 229, 238, 253, 255, 257, 259, 270, 272, 275, 288, 315, 326,329, 332 Gregg, K. B., 543, 544, 551, 644 Gregory, R. P. F., 54, 104 Griffin, D. M., 257, 329 Griffith, N. T., 392, 418 Griffon, E., 519, 520, 525, 644 Grindle, M., 347, 418 Gronau, G. Z., 19, 100 Gronegress, J., 14, 104 Gruber, P. J., 21, 104 Gruignard, L., 568, 644 Guarnierri, M., 82, 104 Guerrier, D., 168, 319 Guggisberg, A., 510, 644 Guidotti, G., 73, 104 Guillermond, A., 74, 104 Guillot-Salomon, T., 43, 56, 82, 84, 103, 104,114

Gunning, B. E. S., 3, 4, 5, 12, 13, 15, 18, 21,48, 104, 239,329 Gurr, M. I., 92, 104 Gutierrez, M., 166, 197, 220

H Haber, W., 543, 544, 644 Haberlandt, G., 532, 644 Hackenbrock, C. R., 23, 25, 104 Hackett, D. P., 263, 284, 286, 332 Hadley, G., 442,450,459, 460,461,462, 463,464,465, 487, 489, 490, 491, 494, 497, 499, 500, 501, 502, 503, 505, 501, 640, 644, 645, 650, 651, 655

Hageman, R. H., 65, 70, 101, 103, 112, 123, 222 Hager, H., 534, 645 Halevy, A., 546, 623, 645 Halevy, A. H., 546, 645 Hall, C. W., 86, 110 Hall, D. O., 27, 53, 54, 104, I05 Hall, I. V., 626, 643, 645 Hall, J. L., 237, 330 Hall, R. H., 462, 649 Hall, W. T., 95, 105 Hallier, U. W., 58, 105 Halperin, M., 546, 623, 645 Halpern, Y.S., 185, 207, 221 Halpin, J. E., 472, 477, 645

664

AUTHOR INDEX

Hamada, M., 477, 478,650 Hamilton, R. M., 534, 645 Hampp, R., 49, 70, 71, 105, 112 Hampton, J., 545, 645 Hanke, M., 506, 645 Hansen, D., 205, 220, 222 Hanson, A. D., 607, 645 Hansson, G., 205, 220 Harbeck, M., 442, 645 Hardegger, E., 510, 516, 640, 645 Harder, R., 386, 418 Harmer, P. M., 121, 127, 158, 220 Harley, J. J., 489, 495, 496, 505, 645 Harper, J. E., 381, 418 Harris, E. H., 52, 105 Harris, R. V., 92, 105 Harrison, A., 6, 115 Harrison, G . R., 438, 443,450, 453, 454, 456, 457, 458, 466, 468, 477, 478, 479, 480, 484, 486, 495, 505, 506, 528, 573, 607, 608, 609, 631, 639, 645 Hartley, M. R., 52, 74, I03 Hartt, C. E., 119, 120, 220 Hartwell, B. L., 120, 220 Harvais, G., 422,442,443,447,460,462, 464, 465, 467, 468, 469, 470, 473, 474, 477, 491, 494, 495, 497, 498, 505, 645 Harwood, J. L., 93, 101 Hashimoto, H., 34, 38, 43, 47, 105 Hatch, M. D., 125, 158, 159, 166, 197, 199, 201, 206, 220, 221, 223 Hattori, A., 214, 222 Hatzopoulou, C., 18, 104 Hauser, I]., 34, I05 Hailssinan, K., 21, 105 Havegraaf, C., 521, 522, 523, 527, 532, 533, 644 Haverkate, F., 44, 105 Havir, E. A., 54, 105 Hawke, J. C., 91, 105 Hawker, J. S., 123, 221 Hayes, A. B., 451, 460, 462, 572, 575, 571, 645, 646 Haystead, A,, 206, 223 Healey, P. L., 440, 447, 448, 449, 450, 451, 455, 469, 481, 482, 483, 503, 639, 642, 646 Healey, M. T., 288, 304, 305, 308, 330 Heber, U., 3, 12, 27, 32, 48, 54, 57, 58, 59, 60, 62, 63, 64, 65, 66, 69, 70, 72, 73, 104, 105, 107, 111, 112 Heinrich, G . , 19, I05

Heinz, E., 87, 89, 91, 93, 101, 105, 106, 107,113 Heitz, E., 3, 4, 105 Heldt, H. W., 3, 12, 32, 37, 48, 49, 51, 54, 55, 56, 57, 58, 59, 60, 62, 63, 64, 65, 66, 67, 69, 73, 81, 104, 105, 108, I I I , 112,115 Hele, M. P., 447, 646 Hellebust, J. A., 214, 221 Hellriegel, H., 119, 221 Henden, J., 488, 650 Henningson, B., 386, 387, 388, 394, 395, 397,402, 405, 418 Henrici, M., 519, 646 Henshaw, G. G., 280, 328 Herold, A., 54, 59, 60, 73, 105, 106, 115 Herrmann, R. G . , 95, 106 Heslop-Harrison, J., 602, 621, 631, 646 Hesse, M., 510, 644 Hetherington, E. E., 623, 646 Hew, C. S., 519, 521, 522, 523, 524, 525, 526, 527, 528, 529, 532, 549, 581, 589, 591, 601, 607, 613, 614, 615, 616, 617, 646, 648, 650, 653,655 Hewitt, E. J., 118, 121, 126, 134, 142, 154, 169, 221 Highfield, P. E., 52, 78, 103, 106 Hijner, J. A., 468,469,494,495, 505,646 Hildebrand, F., 569, 570, 646 Hill, R., 54, 106, 115 Hiller, R. G., 65, I06 Hilliard, J. H., 17, 106 Hills, H. G., 552, 555, 567, 642, 646 Hind, G . , 66, 106 Hindawi, I. J., 623, 646 Hintikka, V., 407, 409, 418 Hippmann, H., 91, 106 His, Ch., 424, 646 Hitchcock, A. E., 621, 623, 642, 655 Hitchcock, C., 82, 106 Hoagland, D. R., 138, 221 Hoarau, A., 66, 109 Hodges, T. K., 205, 206, 221, 222 Hodges, R. K., 205, 206, 220, 222 Hodson, R. C., 70, 112 Hoehne, F. C., 429, 430, 646 Hofsten, A. V., 441, 655 Hogan, N. M., 573, 583, 600, 602, 619, 631, 639 Hohl, H. R., 510, 513, 644 Holborow, K., 66, 108 Holdson, J., 544, 646

665

AUTHOR INDEX

Holland, R., 91, 112 Hollander, S., 505, 646 Hollis, J. P., 282, 331 Holttum, R. E., 422, 533, 542, 546, 550, 551, 637, 646 Hokum, J. A. M., 200, 207, 208, 221 Holtz, R. B., 29, 32, 34, 35, 37, 39, 41, 42, 43, 44, 45, 56, 67, 69, 86, 103 Homes, J., 451, 472, 477, 484, 646 Honda, S. I., 4, 21, 26, 106, 116 Hongladarum, T., 4, 21, 26, 106, 116 Hoober, J. K., 80, 106 Hook, D. D., 234, 280, 281, 305, 307, 308, 309, 330 Hoshina, G . , 26, 106 Howard, R. J., 33, 34, 104 Hsiang, T-H, T., 583,599,600,601,602, 604, 605, 606, 607, 611, 612, 613, 614, 616, 621, 646 Huber, B., 505, 646 Huber, C. T., 23, 106 Huber, D. J., 14, 106 Huber, S. C., 55, 65, 103, 106 Huberrnan, M., 604, 631, 644 Hubert, B., 569, 583, 592, 594, 600, 602, 603, 619, 621, 646 Hubert, E. E., 344, 345, 418 Huck, M. G., 288, 299, 330 Hudson, H. J., 334, 410, 418 Hudson, M. A., 58, 105 Hulme, M. A., 384, 418 Humpherson, P. C., 4, 112 Humphries, W. J., 232, 233, 330 Hunter, D., 519, 521, 523, 524, 526, 528, 529, 651 Hurkman, W. J., 44, 106 Hurst, C. C., 568, 646 Huselbos, T. J. M., 87, 115 Hutton, J. T., 134, 221

I Ichihashi, S., 442, 443, 446 Ikediugwu, F. E. O., 394, 401, 402, 419 Ingle, R. K., 214, 221 Inthuwong, O., 546, 647 Israel, H. W., 13, 106, 462, 646 Ito, I., 441, 647 Izawa, G., 123, 222 Izawa, S., 66. I06

J Jaag, O., 510, 644 Jackman, M. E., 126, 172, 186, 187, 188, 190, 191, 193, 194, 195, 196, 198, 203, 219 Jackson, W. A., 215, 220 Jacobson, A. B., 21, 95, 106 Jacobson, B. S., 92, 106, 107 Jacobson, L., 141, 144, 192, 221 Jagendorf, A. T., 52,65,74,79,106,110, 112, 114 James, A. T., 44, 92, 104, 105, 106 Janacek, K., 44, 107 Janke, H., 441, 647 Jankowicz, M., 520, 522, 652 Jat, R. L., 281, 329 Jaworski, J. G., 92, 106 Jeffrey, D. C., 552, 555, 583, 601, 602, 611, 617, 619, 621, 631, 632, 639, 647 Jeffrey, S. W., 47, 106 Jennings, D. H., 118, 205, 221 Jennings, R. C., 81, 106 Jensen, C. R., 249, 272, 278, 299, 330 Jensen, R. G., 27, 54, 59, 100, 106 Jeschke, E. M., 503, 640 Jeschke, W. D., 131, 221 Jester, R. A., 623, 647 Jeyanayaghy, S., 542, 647 Jinks, J. L., 346, 347, 377 Joham, H. E., 118, 120, 121, 122, 128, 130, 131, 134, 137, 138, 139, 142, 190, 221, 222, 224 Johanson, D. R., 495, 647 Johanson, L., 121, 221 Johansson, G., 1 1 , 116 John, C. D., 279, 280, 330 John, D. A., 501, 645 Johnson, C. M., 124, 134, 143, 144, 149, 219, 221 Johnson, E., 544, 647 Johnson, R. M., 82, 104 Johnson, R. P. C., 501, 645 Johnson, W. H., 81,109 Jokela, A. T., I I , 100 Jones, A. G., 127, 221 Jones, D. L., 580, 647 Jones, L. W., 214, 221 Jones, M. H., 17, 108 Jope, C., 21, 116 Joy, K. W., 45, 49. 51. 52, 70, 106. 112

666

AUTHOR INDEX

Joyard, J., 3,7,26,27,28,29, 32, 34, 35, 39,44,45,47,48, 52,53,56,62,66,67, 69,82,87,89,92,93,96, 102,103,105, 106, 107, 110, 113 Jiirgen-Schrenk, W., 576, 647

K Kaarik, A. A., 335, 387, 407, 413, 419 Kader, J. C., 43, 84, 103, 107 Kaewbanrung, M., 443, 647 Kagan-Zur, V., 4, 107 Kagawa, T., 166, 197,221 Kahane, M. M., 185, 207, 221 Kamemoto, H., 443, 542, 543, 544, 545, 546,647, 650, 652, 653 Kanai, R., 55, 107 Kandler, O., 73, 107 Kannangara, C. G., 92,106,107 Kanwisher, J. W., 238, 283, 294, 331 Karpilov, Y.S., 67, I01 Karlsson, J., 205, 221 Kasai, Z., 625, 652 Kasemir, H., 12, 107 Kataoka, K., 542, 652 Kates, M., 82, 83, 109 Kavanagh, J., 80, 111 Kawashima, N., 74, 107 Kaye, G. W. C., 233, 330 Keen, N. T., 394,419 Keenan, T. W., 205, 221 Kekwick, R. G., 74, 104 Kekwick, R. G. O., 91,115 Kellems, R. E., 77, 107 Kelly, G. J., 58, 107 Kemper, W. D., 314, 330 Kende, H., 607, 645 Kendrick, J. B., Jr, 623, 647 Kendrick, R. E., 48, 49, 67, 102 Kennedy, E. P., 84, 86,112 Kennedy, R. A., 171, 221 Kenyon, C. N., 95,107 Kern, H., 510, 640, 644 Kerr, A. D., 568, 569, 589, 647 Keyani, E., 77, 107 Key, C., 121, 220 Keys, A. J., 59, I01 Keys, C. R., 387, 396,419 Khan, I., 527, 647 Khan, M. U., 87,116 Khazova, I. V., 200, 328

Kim, M., 48,111 King, D. O., 487, 638 Kirch, H. W., 443, 653 Kirk, J. T. O., 3, 12, 74, 107, 279, 330 Kirk, M., 59, 73, 95, 96, 100 Kirk, M. R., 59, 60, 105 Kirk, P. R., 73, 107 Klasova, S., 19, 113 Klass, C. S., 569, 570, 647 Kleinig, H., 87, 91, 107, 108 Klingenberg, M., 56, 111 Knauft, R. L., 519, 521, 523, 528, 532, 534, 583, 589, 600,601, 602, 617, 619, 621, 625, 639, 647 Knudson, L., 425, 441, 447, 456, 477, 478, 497, 519, 523, 525, 526, 527, 634, 643, 647 Koch, A., 487, 649 Koch, L., 437, 443, 647 Koch, V., 443, 647 Kodama, T., 148, 185, 221 Kohl, H. C., 607, 631, 650 Kohler, E., 349, 419 Kollman, R., 503, 642 Konigs, G., 91, 107 Koopowitz, H., 542, 555, 557, 574, 575, 583, 617, 647, 653 Kordan, H. A., 279,330 Korhonen, K., 407, 418 Kormanik, P. P., 280, 307, 308, 330 Kornberg, R. G., 84, 107 Kortschak, H. P., 18, 107 Kosugi, K., 542, 545, 546,647,650,652 Kotyk, A., 44, 107 Koves, E., 298, 331 Kowallik, K. V., 95, 106 Kozlova, G. I., 227, 278, 279, 288, 289, 299, 332 Kramer, D., 171, 224 Kramer, P. J., 311, 330 Kraminer, A., 59, 60, 73, 105 Kraminer, H., 66, 72, 104 Kratz, W. A., 118, 144, 221 Krause, G. H., 12, 39, 54, 62, 65, 66, 105, 107 Krinsky, N. I., 48, 107 Kristen, U., 519, 521, 647 Kristensen, K. J., 227,250,257,314,330 Krizck, D. T., 533, 647 Krotkov, G., 284, 286, 329 Kubicz, A., 487, 649

AUTHOR INDEX

Kugler, H., 633, 634, 647 Kuhner, R., 379, 419 Kuijper, J., 550, 647 Kukulczanka, K., 469, 481, 487, 647, 648, 649 Kullenberg, B., 534, 552, 563, 566, 647 Kunishi, A., 623, 650 Kursanov, A. L., 21, 107 Kusano, E., 519, 647 Kushizaki, M., 121, 221 Kusters, G., 519, 520, 649 Kusumota, M., 443, 470, 647 Kylin, A., 205, 206, 220, 221, 222

Laborde, J. A., 14, 107 Laby, T. H., 233, 330 Laetsch, W. M., 2, 6, 8, 12, 13, 17, 18, 28, 29, 34, 35, 37, 45, 46, 49, 51, 67, 101, 107, 113, 114 Laflkhe, D., 81, 107 La Garde, R. V., 447, 648 Laibach, F., 569, 619, 621, 648, 649 Laing, H. E., 226 Lance, A., 4,41,43, I07 Lance, C., 26, 44, 84, 103, I09 Lang, W., 4,107 Lang, N. J., 16, 95, 108 Larson, S., 58, 114 Larsson, C., 28, 108 Laties, G. G., 21, 26, 106 Laughlin, W. M., 121, 220 LaulhBre, J. P., 17, 108 Laurie, A., 534, 544, 646, 647, 649 Lawson, R. H., 533, 647 Lebedev, A. F., 520, 648 Ledbetter, M. C., 19, 20, 21, I02 Lee, Y. T., 522, 648 Leech, R. M., 25, 26, 27, 28, 29, 35, 39, 42, 43, 44, 45, 54, 73, 14, 82, 91, 104 105, 106, 108, 112 LCger, J. C., 380, 419 Lehner, K., 12, 63, 64, 105, 108 Lehr, J. J., 119, 120, 122, 222 Leimbach, M., 568, 648 Lembi, C. A., 20, 86, 109 Lemon, E. R., 227, 250, 256, 257, 272, 314, 318, 327, 330 Lenard, J., 84, 112 Lendzian, K., 70, I 1 1

667

Leonard, R. T., 205, 206, 221,222 Leopold, A. C., 450, 587, 589, 607, 648, 649 Lerbs, S., 77, 108 Lerman, J. C., 528, 648 Letcher, R. M., 510, 648 Letey, J., 249, 250, 266, 272, 275, 283, 284, 288, 292, 299, 300, 302, 321, 330, 331, 332 Leung, S. P. K., 87, 93, 116 Levi, C., 60, 111 Levi, M. P., 408, 419 Levine, B. A., 34, 105 Levine, G., 54, 100 Levine, R. P., 22, 114 Lewis, C. M., 144, 220 Lewis, D., 503, 652 Lewis, D. H., 59, 102, 106, 507, 648 Lewis, L. N., 14, 114 Leyton, L., 307, 330 Lichenthaler, H. K., 28, 108, 520, 651 Lieberman, M., 623, 648, 650 Liedvogel, B., 34, 38, 43, 44, 49, 51, 87, 91, 93, 107, 108 Lilley, R. McC., 27, 29, 54, 59, 62, 66, 108, 115

Lim, S. L., 581, 589, 601, 607, 613, 614, 615, 616, 617, 648 Lin, C. H., 77, 114 Lindley, J., 533, 648 Lindner, R. E., 623, 648 Lindt, O., 519, 520, 648 Lips, S. H., 4, 107 Loh, C. S., 521, 522, 523, 527, 532, 533, 644 Loiseaux, S., 18, 108 Lopez-Real, J. M., 344, 345, 419 Loomis, W. D., 19,108 Loveys, B. R., 49, I08 Lucanus, B., 119, 219 Lucy, J. A., 23, 108 Ludwig, G. D., 48, 110 Luginger, C., 73, 107 LUgO-LUgO, H., 446, 648 Lundegardh, H., 192, 222 Lunney, C. A., 17, 108 Liittge, U., 190, 222 Luxmoore, R. J., 249,266, 275,278, 283, 284, 288, 292, 299, 330 Lyons, J. M., 39,108 Lyttleton, J. W., 17, 29, 74, 95, 108

668

AUTHOR INDEX

M Maatsch, R., 542, 543, 545, 640 Mache, R., 18, 92, 102, 108 Mackender, R. O., 25, 26, 27, 35, 38, 39, 42, 43, 44, 45, 108 Mackenzie, J. M., 48, 108, 109 MacLeod, R.A., 147,148,184,185,207, 220,222 MacManmon, M., 280, 330 Mae Kawa, S., 545, 647 Magnus, P., 519, 648 Magnus, P. D., 510, 512, 569, 643 Magrou, J., 424, 446, 469, 508, 510, 641, 648 Mai, G., 569, 621, 648 Malaise-Lagae, F., 74, 110 Malgoth, R., 568, 648 Mancha, M., 89, 113 Mangenot, M. F., 387, 419 Mannella, C. A., 26, 35, 56, 103 Mapson, L. W., 623, 648 Mansfield, T. A., 260, 262, 330 Margulies, M. M., 52, 79, 109 Margulis, L., 95, 109 Mariat, F., 446, 468, 469, 648 Markefka, E., 487, 649 MarmC, D., 48, 101, 109, I l l Maronde, D., 59, 60, 73, 10.5 Marra, E., 79, 109 Marschner, H., 118, 119, 123, 221, 222 Marshall, M. O., 82, 83, 109 Martin, C., 590, 591, 592, 649 Maschmann, E., 569, 621, 648, 649 Mason, T. G . , 235 Mason, T. L., 75, 330 Massart, J., 534, 550, 649 Mathieu, Y., 27, 58, 109 Maton, J., 569, 583, 592, 594, 600, 602, 603, 619, 621, 646 Matsumoto, K., 568, 644, Mayer, F., 21, 109 Mazelis, M., 280, 330 Mazliak, P., 43, 84, 92, 93, 109, 114 McCarty, R. E., 41, 43, I08 McClelland, T. B., 619, 649 McCollum, R. E., 123, 222 McConnell, H. M., 84, 86, 103, 107, I12 McDade, E., 534, 540, 550, 649 McGowan, R. E., 519, 521, 523, 524, 526, 528, 529, 6.51

McLachlan, J., 147, 222 McIntyre, D. K., 442, 507, 649, 654 McIntyre, D. S., 250, 273, 320, 321, 330 McLachlan, T. L., 13, 108 McLaren, J. S., 72, 73, 108 McLean, J. D., 3, 108 McManus, T. T., 91, 110 McWilliams, E. L., 519, 522, 524, 525, 526, 527, 528, 529, 532, 649 Mead, J. W., 442, 443, 447, 649 Medawar, 587 Meidner, H., 260, 262, 330 Mejstrik, V., 503, 649 Mendiola-Morgenthaler, L. R., 34, 49, 51, 52, 53, 74, 109, 11.5 Menke, W., 2, 12, 109, 519, 520, 522, 525, 649, 6.52 Menzies, I. D., 5 10, 5 12, 643 Mercer, F. V., 21, I l l Meredith, D. S., 403, 419 Mereschkowsky, C., 95, 109 Merrill, W., 408, 419 Merritt, W. D., 20, 86, 109 Meyer, M., 544, 632, 640 Michaels, A., 79, 109 Michel, M., 451, 472, 646 Middleton, J. A,, 623, 649 Middleton, J. T., 623, 647 Miflin, B. J., 70, 109 Miginiac-Maslow, M., 66, 109 Mikulska, E., 12, 103 Milborrow, B. V., 49,109,591, 592,649 Mill, D. W., 191, 192, 195 Millar, A. A., 566, 640 Miller, A. G . , 214, 220 Miller, C. O., 462, 641 Miller, G. W., 123, 222 Miller, K. J., 23, 25, 104, 109 Miller, K. R., 6, 109 Millington, R. J., 227, 233, 330 Milovancev, M., 12, 63, 65, 10.5, 11.5 Mitra, G. C., 442, 443, 446, 472, 649 Mivra, C., 462, 649 Miwa, A., 519, 521, 533, 649 Miyazaki, S., 443, 460, 465, 471, 481, 649 Mohapatra, S. C., 81, 109 Mohr, H., 12,107 Mokahel, M. A., 501, 503, 640 Molisch, H., 226, 330, 520, 649 Mollenhauer, H. H., 18, 76, 110, I l l

669

AUTHOR INDEX

Moller, A., 534, 569, 649 Moneger, R., 34, 38, 41, 42, 43, 44,

Muscatine, L., 503, 652 Myers, J., 118, 144, 221

100 Montasir, A. H., 120, 121, 222 Monteith, N. H., 288, 328 Montes, G., 21, 109 Montfort, C., 519, 520, 649 Montgomery, H. A. C., 233, 330 Montgomery, J., 534, 649 Moor, H., 6, 109, 110 Moore, F. D., 48, I09 Moore, T. S., 84, I02 Morawiecka, B., 487, 649 Moreau, F., 26, 35, 41, 43, 44, 48, 109 Morel, G. M., 481, 634, 635, 649 Morgenthaler, J. J., 29, 34, 49, 51, 52, 53, 74, I09 Morita, K., 568, 569, 594, 600, 619, 649 Morre, D. J., 20,44, 86,94,95,106, 109, I10 Morriset, C., 278, 299, 331 Morrison, D. L., 412, 419 Morrison, M., 23, I06 Mosbach, A. M., 59, 66, 108, I11 Moscolov, I. V., 122, 222 Moser, H., 159, 222 Mouches, C., 81, I07 Mounce, I., 348, 419 Mourioux, G., 70, 71, 72, I10 Mowe, B. L., 471, 649 Moyse, A., 528, 648 Mudd, J. B., 82, 83, 86, 87, 91, 93, I03 110, I l l Miihlethaler, K., 3, 6, 12, I09, IIO Mukherjee, T. P., 443, 460, 465, 467, 473, 477, 649 Mukhopadhyay, T. P., 542, 546, 640 Muller, E., 644 Muller, F., 534, 569, 61 1, 649 Miiller, H., 568, 649 Miiller, K., 508, 607, 649 Muller, P., 510, 645 Muller, R., 569, 650 Mullison, E., 120, 222 Mullison, W. R., 120, 222 Munnecke, D. E., 394, 419 Murakami, M., 34, 47, 65, I14 Murakami, S., 32, 38, 54, 105, I10 Murashige, T., 542, 543, 544, 546, 634, 635, 650, 652 Murphy, D. J., 54, 73, 82, 91, 108

N

Nagamatsu, T., 443, 460, 465, 477, 649 Nagl, W., 488, 650, 652 Nakagawa, T., 490,497, 653 Nakamura, S. J., 477, 478, 650 Nakamura, Y., 91, 110 Nakatani, H. Y., 27, 32, 66, 106, 110 Nascimento, K. H., 280, 329 Nato, A., 528, 648 Neales, T. F., 519, 521, 522, 523, 524, 526, 527, 528, 529, 650 Nedukha, E. M., 81, 110 Nelson, C. D., 234, 284,286,329,331 Nelson, J. I., 413, 418 Nelson, P. K., 532, 655 Neuburger, M., 7, 35,44,45,67, IIO Neufeld, E. F., 86, I10 Neumann, J., 65, 110 Neupert, W., 48, I10 Newcomb, E. H., 21, 22,104, 110 Newman, D. W., 14, I06 Nharno, L. R., 510,648 Nicholas, D. J. D., 126, 127, 144, 145, 146, 170, 172, 173, 174, 175, 176, 177, 180, 182, 212, 214, 219 Nicholas, H. J., 51 1, 654 Nicholls, W. H., 422, 637, 650 Nichols, B. W., 44,82, 93, 95, 106, 110, I12 Nicholson, G. L., 6, 7, 11, I13 Nicolai, W., 489, 650 Nicolson, J., 66, 100 Niewdorp, P. J., 501, 650 Nightingale, C., 566, 638 Nilsson, 0. S., 84, 110 Nishida, K., 26, 106 Nishikawa, T., 490, 497, 650 Nishimura, G., 546, 650 Nishimura, M., 55, 96, 110,I14 Nitsos, R. E., 123, 222 Nivison, H., 77, 114 Nobbe, F., 119,222 Nobecourt, P., 508, 510, 650 Nobel, P. S., 55, 72, 110, 115, 236, 239, 260, 331 Noggle, G. R., 468, 650 Norris, W. E., Jr, 284, 328 North, B. B., 556, 561, 653

670

AUTHOR INDEX

Northcote, D. H., 18, 19, 110, 116 Nuerenbergk, E. L., 519, 521, 522, 523, 524, 525, 526, 527, 528, 529, 532, 534, 650 Nuesch, J., 508, 510, 511, 512, 513, 516, 517, 518, 519, 644, 653 Nurit, F., 18, 81, 108, 110

0 OBrien, R. L., 56, 110 Odoro, K. A., 394, 419 Oertli, J. J., 607, 631, 650 Ogren, W. L., 22, 65, 101, 102 Ohad, I., 4, 13, 75, 81, 106, I10 Ohki, K., 120, 223 Ohmori, K., 214, 222 Ohmori, M., 43, I10 Oji, Y., 123, 222 Okajima, H., 293, 331 ONeill, M. W., 546, 650 Ong, P. Y., 573, 612, 613, 640 Ongun, A., 86, 87,111 Onofrey, W., 147, 184, 222 Oppermann, A., 386, 419 Orci, L., 74, 110 Ordin, L., 192, 221, 447 Osmond, C. B., 18, 111, 158, 166, 171, 190, 201, 206, 220, 222 Osterhout, W. J. V., 119, 146, 222 Ostermann, E., 441, 653 Ota, Y., 282, 332 Oursel, A., 92, 102 Owens, L. D., 623, 650 Ozanne, P. G., 149, 222

P Pacini, E., 20, 111 Packer L., 32, 110 Pages, P. D., 447,460,461,470,471,650 Pais, M. S. S., 601, 650 Palade, G. E., 4, 13, 79, 102, 110 Palladina, T. A., 122, 205, 222 Pallaghy, C. K., 190, 222 Pallas, E. J., 18, 111 Paolillo, D. J., 16, 17, 102 Paramonova, N. V., 21,107, 111 Parenti, F., 52, 109 Park, R. B., 14, 111 Park, D., 408, 419 Parsons, D. F., 56, 111

Partington, M., 349, 377, 417 Patil, K. D., 280, 329 Patil, S. J., 461, 464, 641 Patrick, W. H., 281, 329 Patrick, W. H., Jr, 226, 329 Patterson, R., 74, I12 Paulas, A. O., 623, 647 Payne, J. H., 552, 650 Peasson, H. W., 206, 223 Pease, A. S., 422, 552, 652 Peavey, D. G., 60, 73, 111, 114 Peel, A. J., 234, 331 Pekkala, D., 468, 469, 495, 505, 645 Peklo, J., 519, 650 Pegg, G. F., 466, 650 Pember, F. R., 120, 220 Penman, H. L., 226,249,331 Perkins, H. J., 234, 331 Perombelon, M., 491, 645 Perrelet, A., 74, I10 Peveling, E., 520, 651 Pfaff, E., 56, 111 Pfaller, A., 48, 110 Pfeffer, W., 124, 222 Wtzer, E., 424, 650 Pfluger, R., 66, 111 Phillipson, J. J., 308, 309, 310, 31 1, 329 Phillis, E., 235, 330 Pierce, D. J., 443, 653 Pierik, R. L. M., 463, 650 Pineau, B., 49, 50, 51, 52, 75, 111 Pitts, G., 282, 331 Platt, T. S., 623, 641 Platt-Aloia, K. A., 12, 111 Plaut, W., 74, 95, 112 Plaut, Z., 70, I l l Pleunneke, R. H., 118, 128, 130, 131, 134, 137, 138, 139, 142, 190, 222 Poddubnaya-Arnoldi, V. A., 456, 462, 506, 569, 611, 631, 650 Pohl, F., 568, 651 Poincelot, R. P., 34, 35, 38, 39, 42, 44, 45, 49, 66, 67, 69, 111 Pollard, G. E., 474, 651 Pomerleau, R., 387, 419 Ponnamperuma, F. N., 281, 331 Poole, R., 447, 655 Porsch, O., 633, 651 Porter, J. W., 48, 111 Porter, H. K., 73, 108 Portis, A. R., 65, 66, 111 Possingham, J. V., 12, 21, 102, 111

AUTHOR INDEX

Potrykus, I., 4, 107 Powell, K. B., 491, 494, 505, 651 Poyanne, A., 552, 562, 566. 651 Poyton, R. D., 80, 111 Pradet, A., 65, 101 Pratt, D., 147, 222 Pratt, L. H., 48, 108 Preiss, J., 60, 111 Pressman, B. C., 69, 111 Preston, J. F., 52, 105 Price, C. A., 29, 35, 52, 109, 111 Price, I., 12, 101 Price, J., 18, 107 Priestley, D. A., 4, 12, 18, 22, 23, 25, 34, 35, 42, 47, 49, 51, 111 Prilieux, E., 519, 651 Probine, M. C., 6, 113 Purczeld, P., 65, 69, 70, 105, 111 Purves, S., 503, 505, 507, 645, 651

Q Quagliariello, E., 79, 109 Quail, P. H., 48, 49, 111 Quebedeaux, B., 159, 166, 223 Quednow, K. G., 447, 651 Queiroz, O., 168, 219, 528, 648 Quereshi, F. A., 235, 236, 331 Quisumbing, E., 534, 651 Qureshi, A. A., 48, I l l Qureshi, N., 48, 111

R Raciborski, M. M., 226, 331 Racusen, D., 49, I11 Radunz, A., 7, I I I Raghavan, V., 443, 446, 447, 651 Raghavendra, A. S., 206, 323, 525, 527, 651

Rains, D. W., I 1 8, 223 Raison, J. K., 39, 111 Raitsakas, A., 491, 494, 505, 645 Ramos, L. J., 443, 655 Ramsbottom, J., 424, 425, 489, 506, 651 Rangaswamy, N. S., 484, 651 Ransom, S. L., 280, 329 Rao, A. N., 447, 461,464,465, 471,484, 542, 573, 612, 613, 635, 640, 647, 651 Raper, J. R., 346, 365, 370, 378, 419

67 1

Rapley, L., 37, 55, 56, 59, 63, 64, 105 Rathnam, C. K. M., 70, 111 Rauh, W., 438, 651 Raven, P. H., 95, 111 Rayner, A. D. M., 338, 341, 343, 345, 350, 35 1, 355, 359, 364, 369, 377, 379, 386, 387, 396, 399, 400, 403, 405, 409, 414, 419, 420 Rayner, R. W., 354, 419 Read, D. J., 312,331 Redfern, D. B., 412, 419 Reed, H. S., 226, 331 Reisinger, D. M., 458, 459, 653 Renkonen, O., 89, 111 Resvoll, T. R., 568, 651 Reyniers, J. P., 488, 638 Reznik, H., 519, 520, 525, 651 Ricard, J. L., 384, 385, 419 Ricardo, M. J., Jr, 487, 651 Rice, H. V., 48, I01 Rich, A., 74, 102 Ridley, S. M., 74, 112 Rientis, K. G., 27, 29, 54, 62, 108 Rigassi, N., 510, 645 Riley, W. S., 311, 330 Rimpau, R. H., 512, 516, 644 Ringstrom, S., 651 Ris, H., 74, 95, 112 Rishbeth, J., 401, 403, 412, 414, 419 Risley, E. B., 6, 115 Ritenour, G. L., 70, 112 Ritt, E., 56, 111 Riviere, A., 568, 651 Robards, A. W., 4, 112, 239, 329, 331 Robertson, J. D., 3, 6, 112 Robertson, R. N., 192, 223 Robinson, I. M., 148, 219 Robinson, J. M., 29, 109 Robinson, M. P., 92, 104 Robinson, S. P., 59, 64, 65, 112 Rock, G. L., 510, 653 Rogers, L. J., 19, 112 Rollins, J. B., 314, 330 Roos, G., 28, 108 Rosado-Alberio, J. T., 17, 18, 112 Rose, H., 446, 648 Rosen, W. G., 12, 13 Rosenberg, L. L., 54, 116 Rosenberg, T., 73, 112 Rosenstock, G., 611, 651 Rosinski, J., 12, 13, 112 Rossner, F., 568, 575, 651

672

AUTHOR INDEX

Rosso, S. W., 14, 112 Roth, L. F., 345,348,349,359,365,370, 417 Rothman, J. E., 84, 86, 112 Rotor, G., 634, 651 Rotor, G. B., Jr, 543, 544, 545, 546, 651 Roughan, P. G., 91,92, 112 Roughton, F. J. W., 61, 101 Rousseau, L. Z., 307, 330 Roux, S. J., 48, 112 Rowe, R. N., 279, 280, 305, 307,331 Roy, H., 74,112 Roy, S., 443,460,465,467,473,417,649 Rubenstein, R., 519, 521, 523, 524, 526, 528, 529, 651 Rucker, W., 446,463,482,641,650,652 Ruigrok, Th. J. C., 86, 112 Ruinen, J., 495, 652 Rumsby, M. G., 91, 105 Rutgers, A. A. L., 550, 652, 654 RypaEek, V., 386,388,397,402,407,419

S Sabatini, D., 75, I01 Sabnis, D. D., 67, 68, I12 Saccone, C., 79, 109 Safford, R., 93,112 Sagarik, R., 545, 647 Sager, R., 74, 112 Sakamoto, H. T., 623,638 Salisbury, J., 49, 51, 115 Salisbury, J., 34, 115 Salisbury, R. A., 424, 652 Sanderson, P. L., 282, 309, 312, 331 Sane, P. V., 3, 14, I l l , 112 Sanford, W. W., 438,447,469,474,533, 534, 535, 536, 537, 538, 539, 540, 541, 542, 545, 546, 547, 548, 652 Sano, Y.,542, 652 Santarius, K. A., 27, 39, 58, 62, 64, 105, 107,l I2 Santos, A., 20, 100 Sarafis, V., 33, 74, 104 Sauer, F., 37, 55, 64, IOS Saunders, P. F., 48, 101 Savina, G. I., 423 Sawa, Y., 447,459,464,470, 652, 653 Saylor, J. C. Jr, 623, 652 Scandella, C. J., 86, 112 Scanu, A. M., 44, 112 Schafer, E., 48, 111

Schafer, G., 66, 72, 73, 104, 112 Schantz, R., 87, 101 Schatz, G., 75, 109 Schellenbaum, M., 510,516, 645 Scherphof, G. L., 86, 112 Schiff, J., 70, 112 Schilling, N., 73, 107 Schimper, A. F. W., 95, 112 Schlecter, R., 534, 652 Schmid, G. H.. 519. 520, 522, 525, 649, 652 Schmid, H., 510, 644 Schmidt, L., 121, 219 Schnarrenberger, C., 21, 22,43, 103, 112 Schneider, M. M., 49, 112 Schnepf, E., 18, 19, 95, 96, 113 Scholander, P. F., 231, 311, 331 Scholander, S. I., 231, 311, 331 Scholte, H. R., 23, 113 Schopfer, W. H., 466,652 Schotz, F., 12, 20, 113 Schrader-Reichhardt, 78, 103 Schrage, C. A. F., 519, 528, 641 Schreiner, O., 226, 331 Schroeder, J., 119, 222 Schubert. C. K., 578, 589, 600, 619, 626, 642 Schultes, R. E., 422, 552, 652 Schulz, D., 437, 647 Schultz, G., 48, I01 Schumacher, W., 604, 63 1, 652 Schweizer, W., 488, 652 Schwenn, J. D., 70, 113 Schwertner, H. A., 43, 113 Sculthorpe, C. D., 281, 331 Seetoh, H. C., 542, 644 Segebarth, K. P., 510, 644 Seifriz, W., 550, 652 Selezneva, V. A., 569, 631, 650 Sellami, A., 64, 65, 101, I13 Senghas, K., 542, 543, 545, 640 Senn, F., 519, 652 Senn, G., 519, 646 Seres, J., 510, 645 Seshagiriah, K. N., 607, 631, 652 Sexton, R., 237, 330 Shacher-Hill, B., 190, 222 Shah, S. P. J., 19, 112 Sharovbeem, H. H., 120, 121, 222 Shechter, Y., 423,424,425, 510,643,652 Sheehan, T. J., 447, 542, 543, 544, 545, 546, 611, 612, 650, 652, 655

AUTHOR INDEX

Shepard, D. C., 48, I09 Shields, J. K., 384, 418 Shigo, A. L., 385, 423, 420 Shimokawa, K., 625, 652 Shine, W. E., 89, I13 Shiryayev, A. I., 23, l I 3 Shomer-Ilan, A., 171, 223 Shumway, L. K., 17, 25, II3, 115 Siddell, S . G., 52, I I3 Sideris, C . P., 477, 652 Sidrak, G. H., 120, 121, 222 Siebertz, H. P., 87, 93, I13 Siefermann-Harms, D., 47, 48, I13 Siegenthaler, P. A., 148, 186, 223 Siekevitz, P., 4, 13, 14, 79, 102, I I O , I13 Sifton, H. B., 281, 331 Signol, M., 43, 82, 84, I03, 114 Silayeva, A. M., 21, 23, 113 Silverthorne, J., 52, 78, I03 Simon, E. W., 4, I02 Simonis, W., 147, 223 Simpson, E. E., 87, 116 Sims, J. J., 394, 419 Singer, S. J., 6, 7, 11, 73 Singh, J., 129, 137, 223 Sisa, M., 447, 459, 470, 653 Sistrom, W. R., 148, 223 Sitte, P., 14, 34, 38,43,44,48,49, 51,93, 108, I13 Sjolund, R. D., 13, 14, 113 Skou, J. C., 205, 223 Slabas, A. R., 54, I15 Slack, C. R., 66, 91, 92, 103, 112, 125, 158, 159, 201, 220, 221, 223 Smirnov, B. P., 91, 113 Smith, B. N., 159, 166, 223 Smith, D., 503, 652 Smith, D. C., 33, I14 Smith, F. A., 206, 222 Smith, H., 29, 48, 49, 103, I13 Smith, J. J., 545, 550, 652 Smith, K. A., 272, 315, 331 Smith, M. E., 403, 418 Smith, M. K., 175, 177, 178, 179, 181, 183, 210, 212, 214, 223 Smith, S. E., 450, 466, 489, 490, 491, 492, 493, 494, 495, 496, 497, 503, 505, 506, 507, 652 Smith, T. A., 126, 169, 221 Smith, W. K . , 429, 437, 438, 641 Smyth, R., 560, 652 Sobieozczanski, J., 469, 647

673

Sorger, G. J., 122, 220 Spanner, D. C., 235, 236, 331 Spanswick, R. M., 239, 331 Spencer, D., 27, 52, 74, IOI, 113 Spencer, P. W., 61 I , 652 Spolsky, C., 96, 114 Sprav, F., 490, 652 Sprengel, C. H., 568, 652 Sprey, B., 2, 6, 8, 13, 18, 28, 29, 34, 35, 37, 45, 46, 49, 51, 67, 107, 108, II3 Spurr, A. R., 14, I07 Staehelin, L. A., 6, I09, 113 Stanier, R . Y . , 16, 95, 107, 113 Stankovic, Z. S., 59, 60, 65, 73, 105, 114 Steegmans, H. H. M., 463, 650 Steenbjerg, F., 119, 220 Steer, M. W., 3, 4, 5, 12, 13, 15, 18, 21, 43, 48, I04 Steinberg, R. A., 169, 223 Stelter, W., 131, 221 Stephen, R. C., 468, 491, 494, 505, 653 Stephens, G. C., 556, 561, 653 Stetler, D. A., 13, I14 Steup, M . , 60, 73, I I I , 114 Steward, F. C., 13, I06 Stewart, W. D. P., 206, 214, 220, 221, 223 Stillwell, M. A., 385, 420 Stocking, C. R., 3, 17, 18, 25, 58, 62, 70, 77,112,114, I15 Stoessl, A., 510, 511, 513, 514, 515, 516, 517, 518, 653, 654 Stoeckenius, W., 56, 114 Stokes, D. M., 58, I14 Stoll, U., 84, 109 Stolzy, L. H., 249, 250, 266, 272, 275, 278, 283, 284, 288, 292, 299, 300, 302, 321, 330, 331, 332 Storey, W. B., 443, 653 Stout,P. R., 118,124,134,143,144,149, 157, 168, 219, 221 Stoutamire, W., 442,473,474, 476, 477, 653 Strauss, M. S., 458, 459, 557, 574, 575, 583, 602, 604, 617, 619, 622, 625, 628, 629, 653 Strotmann, H., 54, 56, 64, 65, 110, 114 Strugger, S., 3, I14 Strullu, D. G., 466, 501, 653 Struntz, G. M., 385, 420 Stumpf, P. K., 89, 91, 92, ZOO, 101, 106, 107, II3, 114

674

AUTHOR INDEX

Sukuki, M., 282, 332 Sullivan, M. S., 226, 331 Summerhayes, V. S., 424, 653 Sunding, P., 552, 653 Svrzycki, S. J., 22, 114 Svibla, R. D., 441, 653 Swader, J. A., 70, 114 Swamy, B. G. L., 607, 608, 610, 653 Swift, H., 21, 95, 106 Swift, M. J., 345, 413, 419, 420 Szarek, S. R., 522, 526, 528, 653 Szarkowski, J. W., 6, 110 Szymanek, J., 122, 223

T

Takabe, T., 96, 114 Talbot, P. H. B., 492, 493, 654 Tan, T. N., 591,613,614, 615, 616,617, 653 Taniguchi, S., 148, 185, 221 Tao, K. L. J., 79, 114 Tarocinski, E., 387, 418 Taylor, A. O., 18, 114 Taylor, 0. C., 623, 649 Taylor, D. L., 95, 114 Teal, J. M., 238, 283, 294, 331 Telfer, A., 66, 100 Ternynck, T., 9,100 Thaler, I., 462, 643 Theden, G., 406 Thien, L. B., 573, 575, 633, 634, 653 Thimann, K. V., 447, 590, 591, 592,649, 653 Thom, N. S., 233, 330 Thompson, C. R., 623, 653 Thompson, P. A., 443, 446, 477, 653 Thomson, W. W., 3, 4, 5, 12, 14, 111, 114,115 Tienken, H. G., 443, 653 Tiffany, H. L., 79,109 They-Bassett, R. A. E., 3, 12, 74, 95, 96, 107,279, 330 Ting, I. P., 168, 171, 223, 522, 527, 528, 653 Tischer, R. G., 144, 219 Titus, J. S., 611, 652 Todd, N. K., 345, 350, 351, 355, 359, 364,369,377,380,419,420 Tokunaga, Y.,490, 497, 653

Tolbert, N. E., 22, 43, 70, 103, 114 Torikata, H., 443, 447, 459, 463, 468, 470, 472, 473, 474,653 Torrey, J. G., 443,446,447, 651 Toth, S. J., 128, 223 Townsend, B. B., 413, 420 Trippi, V., 614, 616, 617, 653 Tran Thanh Van, M., 543,544,545,546, . 591, 614, 615, 617,653 Trebst, A., 70, 113 TrCcul, A., 25, 114 Treggi, G., 122, 223, 224 Tregunna, E. B., 17, 101, 166, 220, 223 Trelease, R. N., 21, 104 TrCmoli&res,A., 84, 92, 93, 102, 114 Trench, R. K., 33,95, 114 Treub, M., 534, 550, 653 Treviranus, L. C., 568, 653 Trippi, V. S., 591,614,615,616,617,653 Troke, P. F., 119, 220 Troughton, J. H., 18, 111 Truog, E., 120,223 Tsuchiya, I., 519, 521, 533, 653 Tullin, V., 119, 223 Tuquet, C., 29, 34, 37, 44, 82, 114 Turner, J. S., 192, 223 Turton, A. G., 134, 223 Twarda-Predota, B., 481, 648 Tyner, E. H., 123,222 Turee, M. T., 239, 331 Tzagoloff, A., 75, 114

U Uchida, T., 477, 478, 650 Ueda, H., 443, 447, 463, 468, 472, 473, 474, 653 Ui, T., 490, 497, 650 Ullrich, H., 21, 103 Ullrich, W., 235, 331 Ullrich-Eberius, C. I., 147, 223 Ulrich, A., 120, 121, 122, 220, 223 Unt, H., 27, 113 Unwin, C. H., 510, 511, 523, 514, 515, 516, 517, 518, 654 Urbach, W., 147,223 Urech, J., 510, 516, 517, 653 Urmston, J. W., 543, 544, 653 Uscuplic, M., 348, 359, 365, 373, 417 Uzzel, T., 96, 114

675

AUTHOR INDEX

V Vachrapai, T., 546, 647 Vacin, E. F., 443, 477, 534, 653 Vajrabhaya, M., 635, 653, 654 Vajrabhaya, T., 635, 654 Valanne, N., 18, 114 Vallance, K. B., 297, 328 Vamos, R., 298, 331 Van Bavel, C. M. M., 226,331 Van Besouw, A., 87, 115 Van der Donk, J. A. W. M., 544, 654 Van der Heide, H., 281, 331 Van der Pijl, L., 534, 552, 555, 558, 559, 560, 561, 563, 564, 567, 617, 654 Van Eijk, M., 119, 121,223 Van Gundy, S. D., 249, 278, 330 Van Hummel, H. C., 87, 114, 115 Van Niel, C. B., 95, 113 Van Raalte, M. H., 226, 281, 331, 332 Vanseveren, N., 654 Vanseveren-Van Espen, N., 451, 452, 454, 484, 646, 654 Van Vliet, H. H. D. M., 87, 93, I10 Van Zaone, D., 86, 112 Varner, J. E., 592, 654 Vartapetian, B. B., 227, 278, 279, 288, 289, 299, 332 Vasalli, P., 74, 110 Vasconcelos, A. C., 34, 49, 51, 52, 53, 109, 115 Veitch, G. J., 442, 507, 649, 654 Veitch, H. J., 424, 568, 654 Vennesland, B., 280, 330 Verboon, J. G., 56, 111 Vermeulen, P., 468, 654 Vermylen-Guillaume, M., 451, 472, 477, 484, 646 Vernon, L. P., 66, 103 Vesk, M., 21, 111 Veyret, Y., 438, 442, 631, 654 Vick, B., 91, I15 Vidaver, G. A., 73, 115 Vigil, E. L., 21, 115 Vignias, P. V., 64, 115 Visher, E., 510, 516, 517, 653 Vishniac, H. S., 148, 185, 223 Vogel, A. I., 128, 223 Vogell, W., 56, 111 Von Fick, A., 243, 332 Von Hofsten, A., 501, 654

Von Wettstein, D., 4, 12, 91, 100, 115 Vothe, W., 442, 447, 467, 471, 473, 474, 477, 654

W Wachter, J., 568, 654 Waisel, Y., 119, 171, 223 Walker, D. A., 3, 27, 29, 32, 48, 54, 57, 58, 59, 60, 62, 65, 66, 71, 73, 100, 101, 102, 105, 106, 108, 114, 115 Wallace, A., 128, 223 Walsby, A. E., 206, 220 Walsh, A., 124, 128, 129, 132, 219, 223 Wan, A. S. C., 511, 654 Wan, H. Y., 546, 644 Wang, C., 72, 110 Wang, C. T., 55, 115 Warburg, O., 521, 522, 523, 524, 525, 526, 527, 528, 529, 654 Warcup, J. H., 424, 489, 490, 491, 492, 493, 496, 654 Ward, A. K., 127, 177,181,182,214,223 Ward, E. W. B., 510, 511, 513, 514, 515, 516, 517, 518, 654 Watson, G. R., 87, 93, I16 Weaire, P. J., 91, 115 Weast, R. C., 232, 233, 332 Webb, M. J., 200, 208,223 Webber, F., 519, 520, 654 Webster, J., 381, 394, 401, 402, 418, 419 Weier, T., 17, 113 Weier, T. E., 3, 4, 6, 12, 13, 14, 17, 18, 25, 112, 113, 114,115 Wejksnora,'P. J., 532, 655 Welkie, G. W., 166, 223 Wellburn, A. R., 29, 49, 51, 67, 74, 80, 102, 111, 115 Wellburn, F. A. M., 29, 34, 49, 115 Wells, H. G., 566, 654 Went, F. A. F. C., 534, 550, 652, 654 Went, F. W., 443, 477, 495, 519, 521, 534, 544, 641, 653, 654 Werdan, K., 12, 59, 63, 65, 66, 67, 69, 104, 105, I15 Werkmeister, P., 472, 474, 476, 654 West, K. R., 190, 222 West, S. H., 17, 106 Westrin, H., 11, 100, 116 Wetmore, R. H., 234, 281, 305, 306, 330 Wetzel, R. G., 127, 177, 181, 182, 214, 223

676

AUTHOR INDEX

Whateley, F. R., 54, 100 Whatley, J. M., 12, 18, 54, 116 Wheeler, L. C., 443, 655 Whitenberg, D. C., 121, 224 Whitfield, P. R., 52, 74, 101 Whittingham, C. P., 59, I01 Whitton, B. A., 16, 95, 108, 116 Wiefelputz, E., 574, 617, 655 Wiegand, C. L., 227, 257, 330 Wieler, A., 3, 116 Wiesmeyer, H., 441, 655 Wiesner, J., 519, 520, 530, 655 Wilbrandt, W., 73, 112 Wildes, R. A., 123, 220 Wildhaber, 0. J., 438, 655 Wildman, S. G., 4, 21, 26, 52, 74, 95, I0I,102,106,107,116 Willert, von, D. J., 171, 224 Willfarth, H., 119, 221 Williams, B. S., 555, 655 Williams, G. C., 377, 420 Williams, J. P., 28, 87, 93, 104, 116 Williams, M. C., 127, 134, 138, 166, 190, 203, 224 Williams, N. H., 423, 552, 555, 567, 637, 642, 646 Williams, R. J. P., 34, 105 Williams, W. T., 296, 332 Williamson, B., 487, 489, 497, 502, 503, 645, 655 Winter, K., 171, 224 Winterhalter-Wild, N., 510, 640 Wintermans, J. F. G. M., 87, 114, 115 Wirth, M., 631, 655 Wirtz, K. W. A., 84, 86, 112, 116 Wiskich, J. T., 26, 59, 64, 65, 103, 112 Withner, C. L., 438, 441, 442, 443, 444, 445, 447, 448, 450, 458, 462, 466, 468, 469, 477, 503, 505, 519, 521, 523, 524, 526, 528, 529, 532, 533, 601, 619, 629, 631, 637, 651, 655 Wojtezak, L., 56, 116 Wolfersdorf, B., 520, 649 Wolff, H., 505, 655 Wollgiehn, R., 77, 108 Wolk, C. P., 206, 224 Wong, S. C., 519, 522, 524, 525, 526, 527, 529, 532, 655 Wood, F. A., 385,420

Wood, J. G., 124, 125, 126, 134, 135, 144, 149, 150, 153, 166, 203, 219, 224 Wood, J. T., 229, 332 Woodgate-Jones, P., 396, 399, 420 Wooding, F. B. P., 18, 19, 110, 116 Woolhouse, H. W., 22, 34, 42, 47, Ill, 116 Woolley, J. T., 118, 127, 128, 130, 131, 134, 135, 138, 139, 141, 142, 143, 149, 158, 159, 166, 190, 191, 222,224 Wright, E. J., 244, 250, 255, 266, 272, 320,328 Wright, D., 519, 521, 533, 655 Wright, S. J., 34, 104 Wrigley, J., 442, 507, 655 Wrigley, J. W., 442, 507, 655 Wrobel-Stermiliska, W., 534, 655 Wybenga, J. M., 118, 120, 122,222,224 Wynd, F. L., 442,443, 468, 650, 655

Y Yamada, M., 43,91,110 Yamada, N., 282,332 Yamashita, M., 442, 443, 446, 646 Yang, A. L., 542, 644 Yang, S. F., 625, 655 Yasuda, T., 121,221 Yeo, A. R., 119, 220 Yguerabide, J., 48, 112 Yocum, C. S., 263, 284, 286,332 Yokobori, H., 546, 647 Yokoi, M., 545, 546, 647 Yoshida, T., 282, 332 Yu, P. T., 300, 302, 332

Z Zaluska, H., 56, 116 Ziegenspeck, N., 505, 643 Ziegler, A., 447, 655 Ziegler, H., 49, 105, 112 Ziegler, I., 70, 71, 105 Zilkey, B. F., 91, 116 Zimmerman, P. W., 621,623,642,655 Zimmelman, W., 552, 655 Zinger, N. V., 456, 462, 506, 650 Zovkiewicz, R., 534, 655 Zwaal, R. F. A., 84,102

SUBJECT INDEX trees, 305-313 wetland plants, 278-298 Abscisic acid, and post-pollination models phenomena in orchids, 584, 601, 619, calibration, 270 632 programming, 271-272 Acanthophippium bicolor, seed morphothe basic unit, 267-270 logy, 427 wetland soil sink, 270-271 A. sylhetense, seed morphology, 437 Aerenchyma, and aeration, 189-297 Aceras anthropophorum Aerides, flowering period, 427, 542 A . odoratum carbon fixation, 522 phytoalexins, 512 post-pollination phenomena, 593,600, thylakolds, 520 603 Achromobacter, effect of sodium on seed morphology, 426 amino acid uptake, 185 Agrostaphyllum sp., seed morphology, Acremonium, colonization of wood, 416 427 Acrolaene punctata, seed morphology, Amaranthus gangeticus, chloroplast427 adenosine triphosphatase, 67 Acropera citrina, seed morphology, 426 A . paniculatus, adenosine triphosphaA . loddigesii tase, 206 carbon fixation, 522 A. retroflexus, chloroplast-adenosine triphosphatase, 67 seed morphology, 426 A . tricolor, effect of sodium on growth, A. luteola, seed morphology, 426 162, 164, 167, 197 A . maculata, seed morphology, 426 Acyltransferases in chloroplast envelope, Amino acid transport, in chloroplast, 89 71-73 Aminoisobutyric acid, effect of sodium Adenosine triphosphatase on uptake, 184, 185 activity in chloroplast envelopes, 6769 Ammonium sulphamate, effect on fungal colonization, 411, 414-415 sodium requirement, 205-207 Adenosine triphosphate, transport from Anabaena cylindrica culture medium, 144 chloroplasts, 58 effect of sodium on Adenylate kinase, effect of sodium, 199, 200 acetylene reduction, 181 Adenylate translocator in chloroplast carbon metabolism, 181-183 glycolate release, 183, 214 envelope, 56, 6 4 7 4 nitrate reductase, 126, 170, 174, 175 Aeluropus litoralis, effect of sodium on nitrogen metabolism, 172-181 carbon fixation, 171 role of sodium, 211-215 Aerangi.r, flowering period, 535, 542 Aeration sodium requirement, 125-127, 138, 145, 146, 168 in unsaturated soil, 3 13-324 in wetland condition Anacamptis pyramidalis, synthesis of phytoalexins, 512, 513, 516, 517 non-wetland plants, 298-305 677 A

678

SUBJECT INDEX

Anacystis montana, thylakold structure,

16 A . nidulans, sodium requirement, 144,

168 Anaectochylus setaceus, seed morpho-

logy, 427 Ancistrorhynchus, flowering, 535, 536,

538,450, 547 Andreana squalidu, orchid pollination,

564 Aneura pinguis, chloroplast membrane

structure, 4

Armillaria mellea

biological control agents, 396 competitive ability in culture, 389, 390, 394, 396 in nature, 400, 401 decay of sapwood, 410, 412, 413 endophyte of orchid, 491 Arundina

post-pollination phenomena, 580, 581, 607 promeristem development, 484 stomatal rhythm, 521

Angraecurn

A. bambirsifolia

auxin transport, 621 flowering, 535, 540, 541, 541, 569 post-pollination phenomena, 574, 575, 602, 604, 619, 622 A. bilobum, seed morphology, 425

culture, 442 seed germination, 472 A. chinensis, vitamin production, 468, 494 A. gramini~olia

carbon fixation, 522, 528 pollination, 555, 557 post-pollination phenomena, 570, 582 Anguloa ruckeri, seed morphology, 437 Ankistrodesmus braunii, phosphate uptake, 114 Ansellia, flowering period, 536, 541, 547 Anthocyanins effect of NAA, 621 effect of pollination in orchids, 601, 602, 619 Apatunia senilis, seed morphology, 427 Aplectrum hyemale, carbon fixation, 522 Arabinose, transport in chloroplast, 73 Arachnis cv. Maggie Oei carbon fixation, 522 flowering period, 542 post-pollination, 615, 616

carbon fixation, 522, 528, 529, 533 post-pollination phenomena, 615 Ascocentrum ampullaceum, carbon fixation, 522, 528 Aster tripolium, effect of sodium on growth, 121, 161, 166 Atomic absorption measurement of sodium, 128-133 Atriplex albicans, effect of sodium on growth, 161, 165 A. angustifolia, effect of sodium on growth, 161, 165 A. glabriuscula, effect of sodium on growth, 161, 165 A. hastata, effect of sodium on growth, 121 A. inflafa, effect of sodium on growth, 124, 138, 141, 160, 164, 190 A. leptocarpa, effect of sodium on growth, 160, 164 A . lindleyi, effect of sodium on growth, 16&165

Aranda

A. nummularia

carbon fixation, 532, 549 stomatal rhythm, 521 A . cv. Deborah carbon fixation, 522 flowering period, 542 A . cv. Wendy Scott carbon fixation, 522, 528 flowering period, 542 Aranthera cv. James Storie, carbon fixation, 522 Archinis, stomatal rhythm, 521

effect of sodium on enzyme activity, 123 growth, 124, 138, 141, 160, 164, 190 nitrogen fraction, 197, 198 physiology 186-188, 203, 204 respiration, 191, 193-195 A. paludosa, effect of sodium on growth, 160, 164 A . quinii, effect of sodium on growth, 150, 164

A. eburneum

pollination, 552 post-pollination phenomena, 570, 582 A . sesquipedale

SUBJECT INDEX

A. semibaccata, effect of sodium on

679

Bijirenaria harrisoniae, post-pollination growth, 160, 164 phenomena, 581 A . semilunalaris, effect of sodium on Biofin, effect on orchids in culture, 468 growth, 160, 165 Bjerkandera adjusta A. spongiosa an tagonisrn decarboxylation system, 197 between synthesized dikaryons, 365 effect of sodium on growth, 160, 165 intraspecific, 351, 354, 356 A . tripolium, sodium content of seeds, interspecific, 398 143 ammonium sulphamate treatment, A. vesicaria 414 sodium competitive ability in culture, 388, content, 124, 142, 143, 191 389, 390, 391, 393, 396, 403 effect at low concentrations, 149di-mon matings, 369 157 fungal replacement, 400, 401 effect on yield, 151, 152, 165, 188sporophore morphology, 361 wood colonization, 337,357,416, 383, 190 408 requirement, 125, 126, 135, 138, 168 Auxins B. fumosa, competitive ability in culture, in orchids, 569, 570, 602, 619, 621,622 389, 391, 392, 396 orchid growth, 458, 459-462 Bletia sheperdii, seed morphology, 426 Avena sativa, chloroplast envelope, 34, Bletilla 51,75 flowering period, 542 post-pollination phenomena, 578 Avicennia marina effect of sodium on growth, 121 seed germination, 424, 442 salt tolerance and adenosine triphos- Blue-green algae, sodium requirement, 146146 phatase, 205 A . nitida, aeration, 311 Bolusiella, flowering period, 538 Botryris, wood colonization, 416 B. cinerea, competitive ability in culture, 389, 396 B Brassavola cordata, seed morphology, 421 Bacteria, sodium requirement, 148 B. nodosa, flowering period, 542 Bacteriodes succinogenes, sodium re- B. perrinii, carbon fixation, 522 Brassia, pollination, 553 quirement, 148 Bangia fusco-purpurea, chloroplast en- B. cowanii, seed morphology, 426 B. verrucosa, flowering period, 542 velope structure, 6 Barkeria melanocaulon, seed morpho- Brassica oleracea, effect of sodium on growth, 161, 166 logy, 425 Bean, response of mitochondria to Brassocattleya, seed germination, 472 Brassolaelia, post-pollination phenoanoxia, 278 mena, 583, 585 Benzyl adenine, effect on orchids in Brassolaeliocattleya cv. Maunalani, carculture, 463 bon fixation, 522, 528 N-Benzyladenosine, effect on orchids in Bromheadia culture, 463 promeristem development, 484 Beta vulgaris stomata1 rhythm, 521 adenosine triphosphatases, 205 effect of sodium on growth, 158, 160, B. alticola, flowering period, 542, 551 B. finlaysoniana 164 carbon fixation, 522, 529, 551 effect of sodium and potassium on flowering period, 542 enzymes, 123 Bryophyllum tubiforum, effect of sodium sodium content of seeds, 142, 143

680

SUBJECT INDEX

Bryophyllum tubiporum-contd. on carboxylation, 200, 202, 208 on growth, 125, 159, 168, 169, 170, 197, 199, 209 Bulbophyllum flowering period, 536, 537, 538, 539, 540, 541 post-pollination phenomena, 580 B. distans, flowering period, 547 B. gibbosum, carbon fixation, 522, 528 B. lobii, flowering period, 542 Buller phenomenon, in wood-decaying Basidiomycetes, 365-372

C

C, photosynthetic pathway, and sodium requirement, 126, 169-171 C4 dicarboxylic acid photosynthetic pathway species and sodium effecton growth, 157-159, 160, 162, 163-166, 169 effect on physiology and metabolism, 186-207 requirement, 125, 126 role, 207-212 chloroplast membranes ATPase activity, 67-68 structure, 16, 17, 18, 97 protoplast preparation, 55 Cabbage, sodium content of seed, 142, 143 Calanthe amethustina, seed morpho~ogy, 426 C . bicolor, seed morphology, 426 C. crispa v. purpurea, seed morphology, 427 C . discolor isolation of fungus, 490 longevity of flowers, 569 C . forbesii, seed morphology, 427 C. harrissonii, seed morphology, 427 C . lobata, seed morphology, 425 C . loddigesii, seed morphology, 427 c . rose% 1ongevitY of flowering, 542 C . tigrina, seed morphology, 426 C . trianaei, seed morphology, 425 c. CV. Veitchii, longevity of flowers, 542 C. veratrifolia, seed morphology, 426

C . vestita carbon fixation, 522, 528 longevity of flowers, 543 C . viridi purpurea, seed morphology, 427 ~ongevityof flowers, 569 Calopogon, promeristem development, 484 Calothrix scopulorum, effect of sodium on release of nitrogen, 214 Calypso bulbosa 470 seed morphology, 434 Calyptrochilum, flowering period, 540 Campylocentriim fasciola, carbon fixation, 530 C . pachyrrhizum, carbon fixation, 530 Candido lbobtica, effect of PhYtoalexin on growth, 516, 518 c*effect marina of potassium on growth, 149 effect of sodium on growth, 148, 149 Carbon dioxide transport, across chloroplast envelope, 66-70 Carotenoids, in chloroplast envelopes, 45-48 Catalese, in orchids following pollination, 614 Catasetum longevity of flowers, 543, 551, 552 pollination, 558, 563 post-pollination phenomenon, 574

c*528

fimbriatumy

5239

Catenularia, colonization of interaction zones, 405,416 Cattleya carbon fixation, 530 C. aurantiaca carbohydrate metabolism, 450, 453, 454,456,457 chlorophyll level, 478, 479 development, 484-486 effect of hormones, 459, 463 seed ultrastructure, 438, 439, 441 C. autumnalis, carbon fixation, 523 C . bicolor, carbon fixation, 523 C. cv. Bow Bells, carbon fixation, 523, 528 C. bo wringiana carbohydrate content, 605, 606 carbon fixation, 528

SUBJECT INDEX

leaf anatomy, 532 longevity of flowers, 543 respiration, 612 C . cv. Dupreana, longevity of flowers, 543 C. cv. Enid endogenous auxins, 569 longevity of flowers, 543 C. forbesii, carbon fixation, 523 C. gigas, carbon fixation, 523 C. intermedia, carbon fixation, 523 C. labiata

carbohydrate content, 604 carbon fixation, 523, 528 longevity of flowers, 543 nitrogen content, 599, 604 C. loddigesii, carbon fixation, 523 C. lueddemanniana, effect of ethylene, 623 C. mendelii, longevity of flowers, 543 C. mossiae, carbon fixation, 523 C . percivaliana, longevity of flowers, 543 C. schroederae, longevity of flowers, 544 C. skinneri

carbon fixation, 523 longevity of flowers, 544 C. trianae

carbon fixation, 523 longevity of flowers, 544 symbiotic specificity, 496 C. walkeriana, carbon fixation, 523 C. warneri, carbon fixation, 523 C. warscewiczii, flowering period, 544 effect of vitamins in culture, 468 germination, 472 longevity of flowers, 568, 569 post-pollination phenomena, 572, 622 resupination, 554 stornatal rhythm, 521 Centris geminata, orchid pollination, 561 Cephafanthera pallens, seed morphology, 437 Ceratobasidium cornigerum

ethylene production, 506 fungal-protocorm interaction, 499 orchid endophyte, 490, 491 symbiotic specificity, 496 C . obscirrum

fungal-protocorm interaction, 499

68 1

symbiotic specificity, 496 C. sp. indet, fungal-protocorm interaction, 499 C . sphaerosporum, symbiotic specificity, 496 C. sterigmaticus, orchid endophyte, 492 Chamaengis, flowering period, 535, 536, 539, 547 Charcoal, effect on orchid growth in culture, 475, 476 Charmorchis alpha, phytoalexin production, 512 Chauliodon, flowering period, 540 Chenoppdium, sodium requirement, 158, 160, 164 Chlamydomonas reinhardtii, chloroplast membranes, 13, 81 Chloris barbata, effect of sodium on growth, 162, 163

c. gayana

decarboxylation system, 197 response to sodium, 162, 163, 171 Chroococcus, sodium requirement, 144 Chlorogloea fritschii, thylakoi’d-plasmalemma connections, 16 Chlorophyll, and sodium requirement, 145 Chloroplast envelope chemical composition peptides, 49-53 pigments, 44-49 polar lipids, 37-44 enzymes lipid synthesis, 81-95 metabolite transport, 54-74 protein transport, 74-81 isolation principles, 25-32 procedure, 32-37 origin of membranes, 95 relationships with other cell membranes, 12-25 structure, 3-1 1 Chondrorhynca discolor x Lycasfe aromatica, effect of auxin in culture, 459 Chrondrostereum purpureum

coloured zone formation, 395, 397 competitive ability, 388, 389, 390, 391, 392, 396 wood colonization, 386, 400, 401 Chrysanthemum segetum, chloroplast membrane structure, 4

682

SUBJECT INDEX

Cladosporium cucumerinum, sterol activity, 511 colonization of interaction zones, 405, 416 Coelia alba, seed morphology, 426 Coeloglossum viride effect of hormones in culture, 459, 463 phytoalexin production, 512 protocorm-fungal interaction, 499500 Coelogyne cristata carbon fixation, 523, 528 nitrogen content, 599, 604 C . massangeana, carbon fixation, 523, 529 C. mayeriana, carbon fixation, 524, 528, 529 C . mooreana, respiration, 612 C . rochussenii, carbon fixation, 524, 528, 529 C . speciosa, longevity of flowers, 569 post-pollination phenomena, 593, 604 Coriolus sterquilinus, and the “unit mycelium”, 350 C . versicolor antagonism between synthesized dikaryons, 364-365 di-mon matings, 365-372 genetics of intraspecific antagonism, 359-372 intraspecific antagonism, 356, 360, 361, 362 population genetics, 352, 353, 354 vegetative characteristics, 339 vegetative incompatibility, 349, 35035 1 Coprinus comatus, vegetative incompatibility, 349 Corallorhiza innata carbon fixation, 520 endophyte of orchid, 490 seed morphology, 425 C . maculata, interaction with rhizoctonia, 498 C . trifida, ultrastructure, 501 Corticum catonii, thamine production, 469 Coryanthes rodriguezii, pollination, 560

C. speciosa, pollination, 560 seed morphology, 435 Corycium crispum, seed morphology, 426 C. orobanchoides, seed morphology, 426 Corymborkis, flowering period, 537 Coryne sarcoides competitive ability in culture, 398, 396 effect of ammonium sulphamate, 414 replacement of fungi in stumps, 400 Cotton, low sodium culture, 137, 139, 142 Crassulacean acid metabolism and sodium requirement, 125, 126, 159, 168, 170, 171 effect of sodium on physiology and metabolism, 186-207 in orchids, 521-529 role of sodium, 207-212 Crepidotus variabilis, wood colonization, 416 Cryptopodium paranaensis, carbon fixation, 524 C. punctatum, post-pollination phenomena, 593, 594 Cryptosporiopsis, interspecific competition, 385 Cucurbita pep0 aeration in seedlings, 226 phytochrome content of plastid membranes, 49 Cyanophyceae, as progenitors of eukaryotic chloroplasts, 95-97 Cycnocher flowering period, 544, 551, 552 post-pollination phenomena, 574 Cymbidium acid phosphatase production, 487 carbon fixation, 528, 530, 568 culture, 443, 446, 470, 481, 634, 635 DNA synthesis, 488 effect of hormones, 459,462, 624,625, 629, 631 longevity of flowers, 568 pollination, 553 post-pollination phenomena, 572,573, 583, 584, 595, 596, 598, 600-611, 619, 630 C. canaliculatum, protocorm-fungal interaction, 499-500

683

SUBJECT INDEX

C. cv. Chelsea, carbon fixation, 534, 549 C. chinense, carbon fixation, 524 C. finlaysonianum post-pollination phenomena, 594 symbiotic specificity, 496 C. goeringii germination, 472 nitrogen metabolism, 447 C. cv. IndependenceDay, carbon fixation 524, 531, 549 C. cv. In Memorium, effect of cytokinin, 463 C.insigne, effect of hormones, 463, 623 C.kanran, effect of hormones, 459, 464 C. lowianirm carbohydrates, 604-606, 607 enzymes, 614 floral respiration, 612, 613 C . lowianum cv. “Yorktown”, carbon fixation, 524 C . madidum, effect of auxin, 459 C. odontorrhizon, seed morphology, 426 C. pumilum, nitrogen metabolism, 447 C. cv. Samarkand, effect of ethylene, 620 C.cv. “Sicily Grandee”, flowering period 544 C . sinense carbon fixation, 524, 529 seed morphology, 427 C. tracyanum carbohydrates, 605, 606 post-pollination phenomena, 594, 599, 604 C . virescens effect of hormones, 459, 464 flowering period, 544, 569 germination, 443 Cynodon dactylon decarboxylation system, 197 effect of sodium on growth, 162, 163 Cynorkir fastigiata, seed morphology, 433 Cypripedium auxin in seeds, 462, 506 longevity of flowers, 569 post-pollination phenomena, 576, 578 C . aucale, carbon fixation, 524, 528 C. barbatum, seed morphology, 425 C. calceolus

culture, 442, 464, 465 germination, 473 C . reginae culture, 464, 467, 470 germination, 473 Cyrtorchis arcuata flowering period, 535 seed morphology, 436 C. hamata, flowering period, 535, 539, 541, 547 Cytokinins, and orchid seedlings, 462, 463, 464, 465 Cytoplasmic streaming, lateral movement of respiratory gases, 239-242

D Dactylorhiza, vitamin production, 468 D . incarnate, symbiotic specifity, 496 D. purpurella carbon transport, 504, 505 culture, 442, 459, 460, 464, 470 protocorm-fungal interaction, 499500 symbiotic specifity, 496 ultrastructure of mycorrhiza, 501, 504 vitamin production, 494, 495 Daedalea confragosa coloured zone formation, 395, 397 competitive ability in culture, 389, 390, 394, 396 pseudosclerotial plate formation, 395, 397 D . quercina coloured zone formation, 395, 397 competitive ability in culture, 354, 389,390,391,396 Dendrobium cv. Anne Marie, flowering period, 544 D. antennatum, seed morphology, 437 D. appecdiculatum, longevity of flowers, 568 D . bronkearti, effects of pollination on ovaries, 630 D. cornatum, flowering period, 545 D . cretaceum, seed morphology, 427 D. crumenatum, flowering period, 545, 550, 551, 568, 569

684

SUBJECT INDEX

Dendrobium--cont d . D . dicuphum, symbiotic specificity, 496 D . draconis, flowering period, 545 D . findlayanum, flowering period, 545 D . formosum, flowering period, 545 D. infundibulum, flowering period, 545 D . cv. Jaquelyn Thomas, flowering period, 545 D . cv. Lady Fay, flowering period, 545 D. cv. Merlin, flowering period, 545 D. nobile effect of hormones in culture, 460,465 flowering period, 545 symbiotic specificity, 496 D . phalaenopsis flowering period, 545 nitrogen metabolism, 446 D . plicatile, seed morphology, 427 D . scabrilingue, flowering period, 545 D . superbum, longevity of flowers, 569 D . taurinum, carbon fixation, 524 D. thrysifolium, post-pollination phenomena, 607 D . cv. Thwaitesiae, flowering period, 545 Dendrophylax funalis, carbon fixation, 530 Diaphananthe, floweringperiod, 535,538, 541, 547 Dicarboxylate translocator in chloroplast membrane, 56, 62-64 Dicrypta bauerii, seed morphology, 421 D . glaucescens, seed morphology, 426 Didymoplexis pallens, flowering period, 431 Diffusion definition, 242 diffusivities of atmospheric gases, 232, 233 Fick's Law, 243-244 movement of respiratory gases in plants, 227, 228 movement of respiratory gases in the gas-space, 228-234, 239 Ohm's law and the diffusion analogy, 247-248 planar diffusion, 244-247, 253-256 pore-space resistance, 248-249 radial diffusion, 250-251, 256-260 Dihydroxyacetone phosphate, transport in chloroplasts, 54, 57, 58, 59, 60, 62 Disa cernua, seed morphology, 427 D . cornuta, seed morphology, 427

D . pulchella, seed morphology, 427 D. tenella, seed morphology, 426 Diuris pedunculata, symbiotic specificity, 496 Dryopteris borreri, chloroplast envelope, 20 Dunaliella tertiolecta, sodium requirement, 147

E Echinochloa utilis effect of sodium on enzymes, 170, 177, 200,201 effect of sodium on growth, 162, 163, 167 sodium uptake, 190, 197 Electrophoresis of chloroplast proteins, 49, 50, 51 Eleusine coracana, nitrate reductase in chloroplast envelope, 70 E. indica, effect of sodium on growth, 162, 163 Encheiridion, flowering period, 535 Encyclia atropurpurea, carbon fixation, 524, 528 E. flabellifera, carbon fixation, 524 E. odoratissima, carbon fixation, 524 Endothia parasitica, heterokaryon incompatibility, 347, 379 Epidendrum effect of vitamins in culture, 468 post-pollination phenomena, 574 vitamin production, 468 E. alatum, carbon fixation, 524, 528 E. cinnabarinum, seed morphology, 426 E. cochleatum, seed morphology, 426 E. crassifolium, seed morphology, 421 E. ellipticum, carbon fixation, 524 E. jloribundum, carbon fixation, 525 E. lancifolium, seed morphology, 426 E. nocturnum, protocorm-fungal interaction, 499, 500 E. obrienianum, protocorm-fungal interaction, 499, 500 E. obrienii, interaction with rhizoctonia, 498 E. papillosum, seed morphology, 421 E. radicans

SUBJECT INDEX

68 5

carbon fixation, 525, 528 galactolipid synthesis, 87 flowering, 545 isolation of chloroplast envelopes, 34 protocorm-fungal interaction, 499,500 protein synthesis in chloroplasts, 53 seed morphology, 437 Euglossa dudsoni, orchid pollination, 559 E. ramosum, seed morphology, 427 E. schomburgkii, carbon fixation, 525 E. superba, orchid pollination, 560 E. stamfortianum, seed morphology, E. viridissima, orchid pollination, 559 425 Eulaema cingulata, orchid pollination, E. xanthinum, carbon fixation, 525, 558 53 I Eulophia cuculata, flowering period, 545 Epipactis E. euglossa, flowering period, 541 post-pollination phenomena, 574 E. gracilis, flowering period, 536 E. atropurpurea, stigmatic closure, 601 E. horsfallii, flowering period, 540 E. erecta, longevity of flowers, 569 E. keithii, carbon fixation, 525, 528, E. falcata, longevity of flowers, 569 529 E. papillosa, longevity of flowers, 569 E. quatiniana, flowering period, 537 E. rubiginosa, carbon fixation, 520 E. streptopetala, seed morphology, 426 E. thunbergii, longevity of flowers, 569 Eulophidium, flowering period, 541 Epipogum aphyllum, endophyte of Eurychone, flowering period, 535 orchid, 490 Exomis E. nutans, isolation of hymenomycetous sodium requirement, 158 fungus, 490 E. axyrioides Epistephium parvi’orum, seed morphoeffect of sodium on growth, 161,165 logy, 426 sodium in seeds, 143 Eria flowering period, 545 F post-pollination phenomena, 578 Eriophoriim angustijolium aerenchyma characteristics, 291, 292, Fatty acids in chloroplast envelope, 39, 42, 82, 92 296 effect of aeration on occlusion of in thylakoi’d, 39, 42, 92 Ferricytochrome c, binding to chlorostomata, 265 gas-space characteristics, 230, 291 plast envelopes, 7, 10 photosynthesis and aeration, 298 Filuloporia vaillantii, moisture levels and Escherichia coli K-12, sodium-depengrowth, 407 dent glutamate transport, 185 Fistulina hepatica, competitive ability in Ethionine, effect on orchid flowers, 625 culture, 389, 390, 396 Flame photometer, measurement of Ethylene sodium, 128, 129, 131 effect on flowering, 551, 583, 584, Flavonoids 622-63 1 effect on seed germination, 477, 506 suppression of fungal growth, 410 post-pollination phenomena in transport in endoplasrnic reticulum, orchids, 600, 604, 619, 624, 625 19 Eucera longicornis, orchid pollination, Fomes cajanderi di-mon matings, 370 565 E. nigriabris, orchid pollination, 563 intraspecific antagonism, 348, 349 Freeze-etching of chloroplast envelope, E. oraniensis, orchid pollination, 565 Euglena, chloroplast envelope structure, 6-8, 13, 18 6 Fructose bisphosphatase and stromal magnesium ion concentration, 66 E. gracilis electrophoresis of chloroplast pro- Funariu hygrGmetrica, chloroplast envelope, 22, 81 teins, 51

686

SUBJECT INDEX

Fungal community structure factors influencing fungal growth host resistance, 410 interactions with other organisms,

competitive ability in culture, 389,

390,396 pseudosclerotial plate formation, 393,

398 zone-line formation, 345 Gas-space system characteristics, 228,229,230 lateral movement of respiratory gases,

413,414

method of entry, 410413 moisture and aeration, 406,407 nitrogen content, 408 239 nutrition, 404-406 longitudinal diffusion of gases, 228,229 pH, 409 longitudinal mass-flow movement, temperature, 407-408 229,231 toxic substances, 410 Genyorchis, flowering period, 536 wood anatomy, 409-410 Gibberellins interspecific antagonism in post-pollination phenomena, 584, interactions and their significance,

601,619,632

384-388 interactions in nature, 399-403 laboratory studies, 388-399 succession and community development, 380-384 intraspecific antagonism genetics, 359-372 physiology, 372-373 unit mycelium concept, 349-352,

in orchid seedlings, 465,466 Gleotulasnella calospora, orchid endophyte, 490 Glomerella cingulata, effect of orchinol,

517 Glossodia minor, seed morphology, 425 Glucose transport in chloroplasts, 73-74 Glutamate transport in chloroplasts, 56,

62

376

schema for fungal community develop- Glyceraldehyde-3-phosphate, transport in chloroplasts, 56,69 ment, 415-417 Fusarium oxysporum, effect of phyto- sn-Glycerol-3-phosphate synthesis, 91 Glycolate, release in sodium deficient alexin, 512,517 algal cells, 214,215 F. solani, infection and phytoalexins, 517 Glycolipids, in chloroplast envelope,

38-41,43,81 G

Galactolipids, origin in the chloroplast,

86-95 Galactose, effect on orchid seedlings,

440,447,449,450 Galactosyltransferase, activity in chloroplast envelope, 37 Galeola hydra, isolation of hymenomycetous fungus, 490 G . javanica, isolation of hymenomycetous fungus, 490 G . septentrionalis, isolation of hymenomycetous fungus, 490 seed germination, 443,477,478 Camoplexis orobanchoides, seed morphology, 427 Ganoderma adspersum

Gomesa crispa, carbon fixation, 525 Gongora armeniaca, orchid pollination,

559 G . bufonia, seed morphology, 426 G. horichiana, orchid pollination, 559 G . maculata v. pallida, seed morphology,

426 Goodyera discolor, seed morphology, 426 G. repens carbon fixation, 520 effect of hormones in culture, 460,464 floral respiration, 612 protocorm-fungal interaction, 499500 seed morphology, 426,437 G . repens var. ophioides, interaction with rhizoctonias, 498 G. semipellucida, seed morphology, 427

SUBJECT INDEX

Gorytes campestris, orchid pollination, 563 Gossypium hirsutum, concentration of sodium in leaves, 191 Grammatophyllum speciosum, seed morphology, 437 Graphorkis lurida flowering period, 536, 547 seed morphology, 434 Gymnadenia albida, phytoalexin production, 512 G. conopea phytoalexin production, 512 seed morphology, 425, 437 G. cucirllata, longevity of flowers, 569 G. longifolia, seed morphology, 425 G. odoratissima, phytoalexin production, 512

H Habenaria englerana flowering period, 535 H. hispidula, seed morphology, 426 H. macrandra, flowering period, 539 H . obtusata, interaction with rhizoctonias, 498 H. platyphylla, carbon fixation, 525 H. psycodes, interaction with rhizoctonias, 498 H. tridentata, seed morphology, 426 H. discolor x H . rubrovenia,seedmorphology, 437 Haematorchis altissima, seed morphology, 426 Halogeton glomeratus sodium and growth, 127, 134, 138, 165, 190 sodium uptake, 190 Harrisela porrecta, carbon fixation by roots, 530 Helianthus, oxygen movement in protoxylem, 237 Hemiselmis virescens, sodium tolerance, 147 Heterobasidion annosum bioloeical control. 403 colonization of living wood, 412

687

competitive ability in culture, 389, 391 395, 396 decay of sapwood, 410 hyphal interference, 394 interspecific amount in nature, 400, 403 moisture level and growth, 407 Heteroporus biennis, interspecific antagonism in nature, 400 trans-d3-Hexadecanoic acid, in chloroplasts, 44 Hexosamine, in chloroplast envelope, 52 Hexose monophosphate, uptake by chloroplasts, 59 Himantoglossum hircinum, seed morphology, 425 Hircinol isolation, 510 specificity, 512, 516, 517 structure, 51 1 Hirschioporus abietinus, moisture level and wood colonization, 407 Hordeum invagination of inner membrane of plastid envelope, 80 H . distichon, sodium and potassium in roots, 131-132 H . vulgare cv. Pallidium, effect of sodium on growth, 158, 160 Huntleya violacea, seed morphology, 427 Hymenochaete rubiginosa competitive ability in culture, 390, 396 hyphal interference, 394 Hypholoma capnoides moisture level and wood colonization, 407 H . fasciculare competition in culture, 388, 390, 391, 396, 399 formation of zone lines, 341 interspecific antagonism, 384 mycelial cord formation, 39CL391 vegetative characteristics, 339 Hypoxy Ion multiforme effect of ASM treatment, 407 formation of zone lines, 345, 355, 357, 406 interspecific antagonism, 384, 390 mode of nutrition. 404 H . serpens, effect of ASM treatment, 414

688

SUBJECT INDEX

I Indoleacetic acid, effect on orchids in culture, 459-461 Indolebutyric acid and peroxidase activity in orchids, 487 effect on orchids in culture, 460 transport in orchid flowers, 621-622 Inositol, effect on orchids in culture, 448, 450, 460 Zrispseudacorus,effect on metabolism in anaerobic conditions, 280 Isochilus linearis, seed morphology, 427

K Kalanchoe blossfeldiana, CAM metabolism, 168 a-Ketogiutarate, transport across chloroplast envelope, 62, 63 Kinetin effect on orchids in culture, 463, 464 in post-pollination phenomena, 619, 631, 632 Knudson medium, in culture of orchids, 478-481, 484, 485, 486 Kochia sodium requirement, 158 K. childsii effect of sodium on growth, 162, 165, 167 effect of sodium on respiration, 195, 197 effect of sodium on phosphoenolpyruvate carboxylase, 200 sodium content of seeds, 143 sodium concentration in leaves, 190, 191, 192 sodium requirement, 197 K. pyramidata, effect of sodium on growth, 161 Kyllinga brevifolia, effect of sodium on growth, 162, 163

L Laelia albida flowering period, 545

L. anceps, seed morphology, 427 L. cinnabarina, carbon fixation, 525 L. crispa, carbon fixation, 525 L. Jlava, carbon fixation, 525 L. galeottiana, seed morphology, 427 L. lobata, symbiotic specificity, 496 L. millerii, carbon fixation, 525 L. perrinii, carbon fixation, 525 L. perrinii v. major, seed morphology, 427 L. purpurata, flowering period, 545 L. xanthina, carbon fixation, 525 Laetiporus sulphureus competitive ability in culture, 389, 390 growth in acid medium, 409 Lanium avicula, carbon fixation, 525 Leaf resistances boundary layer resistance, 260-261 gas-space resistance, 262-263 leaf-wall resistance, 261-262 ventilation of higher plants, 260-263 Leptotes bicolor, seed morphology, 426 Leucine, transport across chloroplast envelope, 72 Limnodorum abortivum carbon dioxide fixation, 520, 525 chlorophyll content, 520 seed morphology, 435, 437 L. verecundum, seed germination, 424 Limonum vulgare, arrangement of aerenchyma, 291 2-Linolenic acid, occurrence in chloroplasts, 42, 82, 84 Liparis, flowering period, 535, 537, 540 Lipid composition of chloroplast envelope membranes, 38-44 Listera, post-pollination phenomena, 578 Listrostachys, flowering period, 539 Lodgepole Pine, aeration under different conditions, 309, 310, 311, 312 Longevity of orchid flowers, 568, 569 Loroglossol antifungal activity, 516 effect on germination of Monilia fructicola, 5 14, 5 18 isolation, 510 structure, 512 Loroglossum hircinum phytoalexin production, 510, 512 L. longibracteatum, phytoalexin production, 512

SUBJECT INDEX

689

Luisia teretifolia, seed morphology, 426 Microcoelia, flowering period, 540, 541, Lycaste harrissonii, seed morphology, 547 427 Miltonia Lycopersicum exulentum culture, 451 relationship between endoplasmic flowering period, 545 reticulum and plastids, 20 M . anceps, flowering period, 545 L. peruvianum, relationship between M. ioezlii, flowering period, 546 endoplasmic reticulum and plastids, M . spectabilis, flowering period, 546 20 M . spectabilis var. moreliana, effect of L. esculentum cv. “Grosse Lisse” auxins, in culture, 460 effect of sodium on growth, 158, 161 Mitochondrion sodium content of seeds, 143 adenine nucleotide translocation, 64 L. esculentum cv. “Marglobe” and chloroplast membrane conconcentration of sodium in leaves, tinuity, 21, 22 191 bursting in low osmolarity medium, effect of sodium on growth, 166 26 sodium content of seeds, 142, 143 carotenoid content, 48 lipid composition, 41, 43 morphological response to changing M osmolarity, 56 phosphate transport, 60 Malate, transport across chloroplast phospholipid synthesis, 84 envelope, 56, 62, 63 Monilia fructicola, effect of photoalexins Malate dehydrogenase on spore germination, 512, 514, 517 and decarboxylate translocator of Monochrysis lutheri, sodium tolerance, chloroplast envelope, 62 147 in orchid flowers, following pollina- Mormodes ation, 614 flowering period, 546, 55 1 Malaxia lilifolia, seed morphology, 427 post-pollination phenomena, 580, 581 Malaxis, flowering period, 535, 537, 541 M . buccinator, seed morphology, 425 Mannose, transport across the chloro- M. pardina, seed morphology, 425 plast envelope, 73 M. viridiflora, seed morphology, 425 Marasmius coniatus var. didymoplexis. Morphactins, effect on growth of culendophyte of orchid, 490 t ures of Cymbidium, 48 l M . oreades, vegetative incompatibility, Mycelial cord formation and nutrient level, 391 349 Margarinomyces, colonization of inter- Mycena galericulata, competitive ability action zones in wood, 405 in culture, 389, 396 Maxillaria arornatica, carbon fixation, Mycoparasitism, 402 525 M . crocea, seed morphology, 425 Megaclinium, longevity of flowers, 569 N Melanomma pulvispyrius in wood decay, 337 Nannochloris oculata, sodium tolerance, mode of nutrition, 404 147 Membrane structure Napht haleneacetic acid lipid-globular protein mosaic model, 6 effect on orchids in culture, 458, 459, unit membrane, 3-4, 6 461 post-pollination effects on orchids, Menadenium, post-pollination pheno584, 592, 596-599, 600,601,602,603, mena, 576, 587, 588 Mesembryanthenum crystallinurn, induc607, 609, 611, 617, 619, 620, 621, 625, tion of CAM habit by sodium, 171 631, 632

690

SUBJECT INDEX

Narcissus poeticus, chromoplast structure, 15 N . pseudonarcissus chloroplast envelope, 51 chromoplast envelope isolation, 34 electrophoresis of polypeptides of galactolipid synthesis of chromoplast envelope, 57 lipid composition of chloroplast envelope, 38 Nastic movements, in orchid flowers, 602 Neottia aestivalis, seed morphology, 426 N. nidusavis carbon dioxide fixation, 520, 525 respiration by flowers, 612 seed germination, 424 symbiosis, 489 N . orchioides, seed morphology, 427 N. pubescens, seed morphology, 427 N . speciosa, seed morphology, 426 N . vitalis, seed morphology, 427 Nervilia, flowering period, 537 Neutron activation analysis of sodium, 129, 130 Niacin, effect on orchids in culture, 467, 468, 469 Nicotiana tabacum, chloroplast envelope pores, 81 NAD-glyceraldehyde-3-phosphate dehydrogenase in chloroplast envelope, 58 NADP-glyceraldehyde-3-phosphate d e hydrogenase in chloroplast envelope, 58 Nigritella angustqolia, seed morphology, 425 N. nigra, phytoalexin production, 512 Nitrate reductase activity and sodium requirement, 126 Nitrogen content of orchid flowers after pollination, 599 Non-wetland plants adaptability, 299-300 analogue data, 300-302 critical oxygen pressures, 298-299 oxygen transport in pea, 302-305 Nuphar advenum, photosynthesis and aeration, 297 Nvssa aauatica. effect on metabolism in -anaerobic conditions, 280 N. sylvatica, effect on metabolism of

anaerobic conditions, 280

0 Octomeria lanc$?lia, seed morphology, 425 Odontoglossum culture, 45 1, 468 longevity of flowers, 569 0. bictonense, flowering period, 546 0. citrosmum, flowering period, 546 0. grande, effect of auxin in culture, 460 0.pescatorei, seed morphology, 425 0. pulvinatum, seed morphology, 426 0. schlieperianum, effect of auxin in culture, 460 0.sphacelatum, seed morphology, 425 post-pollination phenomena, 576,593, 600 Oenothera hookeri, chloroplast envelope structure, 12 plastid membranes, 20 Oliveonia pauxilla, symbiotic specificity, 496 Oncidium longevity of flowers, 569 0. flexuosum, carbon fixation, 526, 529 0. lanceanum, carbon fixation, 526, 528 0. planilabre, pollination, 561, 566 0. pumilum, carbon fixation, 526 0. sphacelatum carbon fixation, 526 flowering period, 546 0. splendidum, flowering period, 546 Ophrys self-pollination, 566 0. apifera, pseudocopulation, 565 0.arachnvormis, pseudocopulation, 564 0. arachnites, phytoalexin production, 512 0.bertolonii, labellum ultrastructure, 636 0. funerea, seed morphology, 427 0. fusca, pseudocopulation, 562, 565 0. insectgera pseudocopulation, 563 ultrastructure, 501 0. lutea, pseudocopulation, 562 0. speculum, pseudocopulation, 562 0. sphecodes provincialis, .pseudo-

69 1

SUBJECT INDEX

copulation, 565 self-pollination, 566 Orchidaceae carbon fixation C3 photosynthesis, 522-527, 529 C, photosynthesis, 529, 530 Crassulacean acid metabolism, 521529 fixation by different plant organs, 530-532 history, 519-521 photorespiration, 532 stomata1 rhythms, 521 flowers history, 534 introduction, 534-552 pollination, 552-566 post-pollination phenomena, 566617 induction of post-pollination phenomena abscisic acid, 632 auxin, 621-622 cytokinins, 631-632 emasculation, 619 ethylene, 622-631 gibberellins, 632 hormone interactions, 623, 633 pollination, 619 phytoalexins action spectrum and activity, 512517 biological role, 517-5 19 chemistry production and distribution, 510-512 history, 508-510 seeds asymbiotic germination, 4 4 4 8 9 external morphology, 438 history, 423437 longevity, 441 structure and ultrastructure, 43844 1 symbiotic germination, 489-506 tissue culture, 634-635 Orchis nitrogen metabolism, 447 0. acuminata, seed morphology, 426 0. aristata, symbiotic specificity, 497 0. bifolia, carbon fixation, 520 0. brevicornu, seed morphology, 425

0 . coriophora phytoalexin production, 512 seed morphology, 425 0 .fragrans, seed morphology, 425 0. lactifolia carbon fixation, 520 phytoalexin production, 512 seed morphology, 425 UV flower image, 634 0 . longicornu, seed morphology, 425 0. maculata phytoalexin production, 512 seed morphology, 425 0 . mascula carbon fixation, 520 phytoalexin production, 512 0. militaris, phytoalexin production, 510-513, 517, 519 0. morio carbon fixation, 520 phytoalexin production, 508, 509, 512 seed germination, 424 symbiotic specificity, 496 Oudemansiella radicata competitive ability in culture, 389, 391, 396 effect of ammonium sulphamate, 414 intraspecific antagonism, 348 pseudosclerotial plate formation, 393 Oxaloacetate, transport in chloroplast envelope, 56, 62, 63 Oxygen source atmospheric diffusion, 260-265 photosynthesis, 263-265

P Panicum maximum, effect of sodium on growth, 162, 163 P . milioides, effect of sodium on growth, 159, 162, 163 Pantothenic acid, effect on orchids in culture, 467, 468 Paphiopedilum carbon fixation, 528, 529 culture, 468, 471 flowering period, 546, 568, 569 germination, 473, 475 P . barbatum, carbon fixation, 526, 528 P . crirtisii, seed morphology, 425

692

SUBJECT INDEX

Paphiopedilum-contd. P . insigne carbon fixation, 526 effect of ethylene, 623 flowering period, 546 P . cv. Mildred Hunter, carbon fixation, 526 P . parishii, seed morphology, 425 P . specierum, seed morphology, 425 P . venustum anatomy, 532 carbon fixation, 526, 528 effect of ethylene, 623 P . villosum carbon fixation, 526, 529 pollination, 554 Paxillus panuoides, humidity and growth, 406 Pelexia adnata, seed morphology, 427 Peristeria elata, longevity of flowers, 569 Peroxidase, in orchid flowers, 614, 615 Peroxysomes, association with chloroplasts, 21, 22, 23 Pettandra virginica, photosynthesis and aeration, 197 Petunia hybrida, RNA synthesis in flowers, 617 Phaeodactylon tricornutum, sodium tolerance, 147 Phaeolus schweinitzii, intraspecific antagonism, 348, 359, 373 Phaeoplasts, relationship with endoplasmic reticulum, 18, 19 Phajus post-pollination phenomena, 574, 575 P . albus, seed morphology, 425 P . bicolor, seed morphology, 426 P . grandifolius, seed morphology, 426 P . maculatus, seed morphology, 426 P . tankervilliae effect of ethylene, 623 flowering period, 546 post-pollination phenomena, 583,585 P . wallichii, seed morphology, 427 Phalaenopsis carbon fixation,. 526,. 531 extracellular enzyme production, 451, 455 germination, 473, 475 longevity of flowers, 546, 568, 569 metabolism, 446, 450

post-pollination phenomena, 569,572, 573, 586, 587, 619 P . amabilis carbon fixation, 526, 528 flowering period, 546, 569 post-pollination phenomena, 593, 600, 614, 615 P. esmeralda, carbon fixation, 528 P. schilleriana carbon fixation, 526, 528 flowering period, 546 P. violacea, flowering period, 569 Phallus impudicus antagonism, 398, 399, 400 effect of ammonium sulphamate, 414, 415 mycelial cord production, 41 1, 412 vegetative characteristics, 339, 341 Phanerochaete velutina effect of ammonium sulphamate, 414, 415 culture, 396, 402 mycelial cord formation, 391, 392 wood colonization, 409, 412, 416 Phaseobis vulgaris carotenoid content of chloroplast envelopes, 47 chloroplast envelope isolation, 35 polypeptides of chloroplast envelope, 51 RuBPCase synthesis, 74 Phenanthrenes, antifungal activity, 51 1, 514, 515 Phialophora, colonization of wood, 405, 416 Phlebia gigantea, colonization of wood, 402,403,407 P. merismoides antagonism, 356, 386, 388-390, 396, 398, 399, 400, 416 mating behaviour, 379 zone line formation, 345, 395, 396 Pisum sativum chloroplast envelope, 51, 57, 67, 68 oxygen transport, 302-305, 308 root growth and oxygen pressure, 288 Plastoglobuli, in chloroplast stroma, 28, 29, 32, 34, 35, 46, 41 Platanthera post-pollination phenomena, 578 P . bifolia culture, 461, 464

693

SUBJECT INDEX

phytoalexin production, 512 P. chlorantha, floral respiration, 612 P . yatabei, longevity of flowers, 569 Pleltrothalis ophiocephalus, carbon fixation, 526 P. racemiflora, seed morphology, 432 P. sessiliporum, seed morphology, 427 Pleurotus ostreatus competitive ability in culture, 389, 391, 396 zone formation, 395, 396 Poa pratensis, effect of sodium on growth, 159, 162, 163 Podangis, flowering period, 536 Polyphenol oxidase, in orchids, following pollination, 615, 617 Polystachya, flowering period, 535, 536, 537, 538, 539, 540, 541, 547 P. cultriformis, flowering period, 546 Portulacagrandijora, effect of sodium on growth, 162, 166, 197 P. oleracea, effect of sodium on carbon fixation, 171 Prasophyllum, post-pollination phenomena, 580 Promenaea stapelioides, seed morphology, 426 Protein synthesis by chloroplast in vitro, 52, 53 Pseudomonas stutzeri, effect of sodium, on growth, 148, 185 Pseudosclerotial plate formation, 364, 388, 389, 393, 398 Pseudotrametes gibbosa competitive ability in culture, 389, 390, 391, 394, 395, 396, 398, 399, 403 mycoparasitism, 383, 392, 396 Pteris vittata, chloroplast envelope, 20 Pterostylis nutans, symbiotic specificity, 496 Pterygodium caffrum, seed morphology, 426 P. catholicum, seed morphology, 427 P. inversum, seed morphology, 427 P. voluere, seed morphology, 427 Pumpkin, response of mitochondria to anoxia, 278 Pyridoxine, effect on orchids in culture, 467, 468 Pyrola rotundifolia, seed morphology, 425

Pyruvate orthophosphate dikinase, effect of sodium, 199, 200 Pythium ultimum, effect of orchinol on germination and growth, 517

R Renanthera imschootiana, flowering period, 546 Respiration and the ventilation process in higher plants, 251-260 in orchid flowers, 61 1-613 Rhinocladiella spp. colonization of interaction zones, 405, 406,416 effect of ASM, 414 Rhizobitoxine, and longevity of orchid flowers, 623 Rhizoctonia niacin production in culture, 469 thiamine production, 469 R. goodyerae repentis, isolated from orchids, 491 R. languinosa, isolated from orchids, 490 R . repens isolated from orchids, 490 nitrogen source, 491 phytoalexins, 509, 51 1 R. solani fungal-protocorm interactions, 500 isolated from orchids, 490, 493 Rhizomorphs, and the colonization of wood, 410, 412, 413 Rhodopseudomonas sodium requirement, 148 R. spheroides, sodium requirement, 148 Rhynchostylis post-pollination phenomena, 580, 581 R . gigantea, flowering period, 546 R. retusa effects of pollination and auxin on ovaries, 603 flowering period, 546 longevity of flowers, 569 Riboflavin, effect on orchids in culture, 467, 468 Ribulose bisphosphate carboxylase effect of magnesium ions, 66

694

SUBJECT INDEX

Ribulose bisphosphate carboxylasecontd. transport across chloroplast envelope, 74-75, 78, 79 Ribulose 1,S-diphosphate carboxylase, in cultured orchid seedlings, 479, 480, 487 Rice aerenchymatous structure, 292, 293, 294, 295 effect on mitochondria of anoxic culture, 278, 279 gas-space characteristics, 230 lacuna production in roots, 290 root growth in relation to oxygen pressure, 288 Root aeration in the unsaturated soil, 3 1 3-324

Scirpus validus, photosynthesis and aeration, 297 Selenipedium schlimmii, seed morphology, 427 Senescence of orchid flowers, 587-600 Serapias lingua, phytoalexin production, 512 S. neglecta, phytoalexin production, 5 12 S. orientalis culture, 467, 470 seed germination and development, 473 S . parviflora, culture, 470 S . vomeracea, phytoalexin production, 512 Serpula himantoides, moisture level and wood colonization, 407 S. lacrymans mycelial cord formation, 413 temperature requirement, 408 Sodium, and plant metabolism S metabolic and physiological effects of low sodium Sarcanthus rostratus, seed morphology, Anabaena cylindrica, 172-1 83 426 C4 and CAM plants, 184-207 Sarcochilus luniferus, carbon fixation, responses to low sodium 530 Atriplex vesicaria, 149-1 57 Satyrium C4 pathway species, 157-159, 163caproic acid content, 423 167 pollination, 552 CAM species, 159-168 S. bicallosum, seed morphology, 427 lower plants, 144-149 S. carneum, seed morphology, 427 schemes for the role of sodium, 207S. nepalense, seed morphology, 427 212 Saxifraga sarmentosa, oxygen move- Solanum tuberosum, endoplasmic reticument in phloem, 235 lum, 20 Scaphyglotis vestita, seed morphology, Sophronitis cerna, carbon fixation, 527 425 Spartina townsendii, gas-space characSchizophyllum commune, pioneer wood teristics, 291 colonization, 46 Spathoglottis S. cytalidium promeristem development, 484 biological control of wood decay, 384, stomata1 rhythm, 521 385,404 S . plicata wood colonization, 416 carbon fixation, 527-529, 533 Schombocattleya, carbon fixation, 526, hormones in culture, 461, 464 529 protocorm-fungal interactions, 499, Schomburgkia 500 carbon fixation, 526 Spinacia oleracea post-pollination phenomena, 576, 577 chloroplast envelope S. crispa. carbon fixation, 527, 528 ATPase activity, 67 S . tibicinis, post-pollination phenomena, lipid composition, 38-40, 44 583, 585 phosphate translocator, 57 S. undulata, seed morphology, 437 pigment composition, 47

695

SUBJECT INDEX

preparation, 29-37 proteins, 29, 50, 51 RuBPCase transport, 79 structure, 5, 6, 24 sulphate transport, 71, 72 phosphatidylcholine synthesis, 82 Spiranthes australis, longevity of flowers, 569 S . cerna, symbiotic specificity, 496 S . sinensis, symbiotic specificity, 496 S . speciosa, carbon fixation, 527, 528, 529 Stanhopea longevity of flowers, 569 post-pollination phenomena, 576 S . aurea, seed morphology, 425 S. insignis, seed morphology, 426 S. oculata, seed morphology, 426, 437 S. tigrina var. superba, seed morphology, 425 S. violacea, seed morphology, 421 S. warczewitzii, seed morphology, 425 Starch synthesis, control by phosphate translocator, 59-60 Stelar tissue, movement of respiratory gases in phloem, 234-235 in xylem, 236-238 Stereumgausapatum,competitive ability, 389, 396 S. hirsutum an tagonism interspecific, 386, 400, 401 intraspecific, 354, 355, 356 colonization of wood, 410, 416 competitive ability in culture, 389, 390, 396, 398, 399 hyphal interference, 394 population structure, 352, 353, 357, 358, 359 zone-line formation, 345, 355, 360, 395, 397 Sturmia loeselii, seed morphology, 425 Sulphate, transport across chloroplast envelope, 70, 71, 72 Sulpholipid, location in chloroplast envelope, 7, 9, 43 Symbiosis theory of chloroplast evolution, 95-97 Synergism and the ventilation of higher plants, 251-260

T Taeniophyllum promeristem development, 484 T. filiforme, carbon fixation, 532 T. zollingen, carbon fixation, 530 Tainia penangiana, carbon fixation, 527, 528, 529 Tannins, restriction of fungal growth, 410 Tetragamestus modestus, seed morphology, 425 Thanatephorus cucumeris fungal-protocorm interaction, 499 infection of Dactylorhiza purpurella, 502 specificity, 491, 496 T. orchidicola, fungal-protocorm interaction, 499 T. sterigmaticus fungal-protocorm interaction, 499 specificity, 496 Thelymitra aristata, symbiotic specificity, 496 T. grandiflora, symbiotic specificity, 496 T. ixioides, seed morphology, 426 Thiamine, effect of orchids in culture, 467,468,469 Thielaviopsis basicola, effect of orchi nol, 517 Thraustochytrium globosum, sodium requirement, 148 T. roseum effect of sodium on physiology, 186 sodium requirement, 148 Thrixspermum, flowering period, 546 Thunia marshalliana, carbon fixation, 527 Thylakold association with chloroplast envelope, 12-18 plastoglobuli, 28, 29 carotenoid composition, 45, 47 lipid composition, 39, 42, 43 polypeptide composition, 49, 50 Tissue culture of orchids, 634635 Tomato chlorine requirement, 124, 149 chromoplast structure, 15 low sodium culture, 136-139, 141142, 143 mitochondria1 response to anoxia, 278

696

SUBJECT INDEX

Tomato-contd. silicon requirement, 136 sodium requirement, 130, 136, 138, 149, 159 Transport of respiratory gases lateral transport, 239-242 longitudinal transport through gas-space system, 228-234 through stele, 234-238 Trichoderma biological control of wood decay, 384, 404 colonization of wood, 416 Trichopilia albida, seed morphology, 426 Tricholomopsisplatyphylla, wood colonization, 412 Tridactyle, flowering period, 536, 537, 538. 541 Trifolium repens cv. 'Palestine', effect of sodium on growth, 161, 166 Triphora pendula, seed morphology, 426 Triticum sativum isolation of etioplast envelopes, 34 lipid composition of chloroplast envelopes, 38, 44 Tulasnella allantospora, symbiotic specificity, 496, 497 T. asymmetrica, symbiotic specificity, 496 T. calaspora ethylene production, 506 fungal-protocorm interaction, 500 nitrogen source, 491 symbiotic specificity, 496 vitamin requirement, 494 T. cruciata fungal-protocorm interaction, 500 specificity, 496 T. violea, symbiotic specificity, 496 Typha latifolia, photosynthesis and aeration, 297

U UDP-galactose incorporation into galactolipids, 86 synthesis, 91 Unit membrane, 3-4, 6

V Vanada carbon fixative, 527 culture, 446, 471 development, 487 effect of auxin in culture, 461 effect of cytokinin in culture, 464 gibberellic acid, 465 peroxidase activity, 487 Vanda DNA synthesis, 487, 488 effect of ethylene, 623 flowering period, 546, 569 resupination, 554 V. coerulea, seed morphology, 427 V. tesselata, carbon fixation, 527 V. tricolor, leaf anatomy, 532 V. suavis, carbon dioxide fixation, 531 Vanilla effect of p-naphthoxyacetic acid on flowers, 621 seed morphology, 425 V. aromatica, carbon fixation, 527 V.fragrans, carbon fixation, 527, 528 V. planifolia carbon fixation, 527 culture, 442, 446 seed morphology, 426, 432 Venturia inaequalis, effect of orchinol, 517 Verticillum dahliae, effect of orchinol, 517 Viciafaba galactolipid content content of mitochondria, 43 isolation of envelope enriched fraction, 25, 26, 35 structure of proplastid in root tip cell, 13 synthesis of amino acids, 72-73 Vitamins effect on orchids in culture, 466-469 orchid endophyte requirements, 494

W Wetland plants in the wetland condition aerenchyma and aeration, 289-297 critical oxygen pressure, 284-288

697

SUBJECT INDEX

oxygen pressure and root growth, 288-289 photosynthesis and aeration, 297-298 radial oxygen loss and phytotoxins, 28 1-284 responses to anoxia, 278-281 Wood decay analysis, 337-345 hyphal interactions, 339-345

x Xanthopan morganii praedicta, orchid pollination, 557 Xerotus javanicus, orchid endophyte, 490 Xylaria hypoxylon effect of ASM, 414, 41 5 X . polymorpha interspecificamount in nature, 400,401 zone line-formation in decaying wood, 344 Xylobium squalens, seed morphology, 427

Xylose, transport across chloroplast envelope, 73

Z Zea mays chloroplast structure, 16, 38 effect of sodium, 171 etiochloroplast membrane structure, 14 Zeuxine elongata, flowering period, 537 Zone lines in decaying wood, 341, 343-345 pseudosclerotial plate formation, 344-345 resulting from a single mycelium, 344 resulting from mycelial interaction, 345 Zygopetalum flowering period, 546 post-pollination phenomena, 572, 578 Z . intermedium, seed morphology, 426 Z. mackaii, seed morphology, 426

This Page Intentionally Left Blank