Advances in Ecological Research, Volume 23

Advances in

ECOLOGICAL RESEARCH VOLUME 23

This Page Intentionally Left Blank

Advances in

ECOLOGICAL RESEARCH Edited by

M. BEGON Department of Zoology, University of Liverpool, Liverpool, L69 3BX, UK

A. H. FITTER Department of Biology, University of York, York, YO1 SDD, UK

VOLUME 23

ACADEMIC PRESS Harcourt Brace Jovanovich, Publishers London San Diego New York Boston Sydney Tokyo Toronto

ACADEMIC PRESS LTD 24/28 Oval Road London NWI 7DX United States Edition published by

ACADEMIC PRESS INC. San Diego, CA 92101

Copyright 0 1992 by ACADEMIC PRESS LIMITED

AN 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 ecological research. Vol. 23 I . Ecology 1. Begon, Michael 574.5 ISBN&l2413923-5

This book is printed on acid-free paper Typeset by Latimer Trend & Company Ltd, Plymouth Printed in Great Britain by T. J. Press (Padstow) Ltd, Padstow, Cornwall.

Contributors to Volume 23 A. D. Q. AGNEW, Department of Biological Sciences, University College of Wales, Aherystwyth SY23 3DA, UK. J. BASTOW WILSON, Botany Department. University of Otago, PO Box 56, Dunedin, New Zealand. M . L. CIPOLLINI, Department of Biological Sciences, Rutgers University, New Brunswick, NJ 08855-1059, USA. R. M . M. CRAWFORD, Department of Biology and Preclinical Medicine, Sir Harold Mitchell Building, The University, St Andrews. Fife KY16 9AL, UK. H. LAMBERS, Department of Plant Ecology and Evolutionary Biology, PO Box 800.84, NL-3508 T B Utrecht. The Netherlands. J . LUSSENHOP, Department of Biological Sciences, University of Illinois at Chicago, Box 4348, Chicago, I L 60680, USA. H. POORTER, Department of Plant Ecology and Evolutionary Biology, PO Box 800.84, NL-3508 T B Utrecht, The Netherlands. E. W. STILES, Department of Biological Sciences, Rutgers University, New Brunswick, N J 08855-1059. USA.

This Page Intentionally Left Blank

Preface The contributions to this volume are linked by their concern with topics or questions that have suffered a degree of neglect. Nobody can now seriously doubt the crucial importance of biotic interactions, hidden from the sight of most ecologists, within the soil, nor of the increasing necessity of conscious management of the soil biota, within agricultural and forest soils at least. Interactions between micro-organisms and micro-arthropods are central to many, if not most, soil processes. Lussenhop reviews what is known of the mechanisms through which these interactions occur, focusing separately on saprophytic systems and the rhizosphere, and ranging from simple grazing and dispersal, through the stimulation of microbial activity, to the potential regulation of pathogens. The conclusion, as so often, seems to be that the steps from description to useful quantification have yet to be taken. The co-evolutionary pressures connecting plants and their potential consumers are rarely if ever straightforward. Certainly, those addressed by Cipollini and Stiles, between fleshy fruits, their vertebrate dispersers and fruit-rot fungi are complex and subtle. In the past, studies of the secondary chemicals of fleshy fruits have concentrated on the almost certainly atypical, highly-selected cultivated species. By contrast, these authors evaluate selection pressures in a more general and natural context, generate a number of broad hypotheses, and then, using their own work with Ericaceous species as a springboard, show how these predictions may be given added specificity. When the distinction is drawn in introductory ecological texts between limiting and non-limiting resources, oxygen is sometimes advanced as a good example of the latter: crucially necessary, but always available in abundance to those aerobic organisms that require it. As Crawford shows, however, for plants at least, there are many situations where this view is quite simply wrong. Physiological and distributional data are combined to demonstrate that during the life cycle of most species of higher plants, there are critical periods when oxygen is a resource that is frequently limiting for germination, growth and survival. From the Arctic to the Tropics, the pattern of plant distribution frequently bears the imprint of oxygen as a limiting factor. It is no great surprise that plants growing on nutrient-poor soils have a lower growth rate than those on fertile soils. But even when grown under optimum conditions, species that naturally occur on nutrient-poor soils still have relatively low growth rates, as do those species (and ecotypes) charac-

... Vlll

PREFACE

teristic of shaded environments, dry habitats, saline conditions and other habitats intrinsically unsupportive of plant growth. Lambers and Poorter, therefore, ask two related questions. What are the physiological causes of these differing growth rates? And what are their ecological consequences? Their provisional answers are framed, perhaps not surprisingly, in the language of trade-offs, but the need remains for quantitative data that might fully support their contentions. Views of community (especially plant community) dynamics have been much influenced by Clements’ facilitation theory of succession and Watt’s theory of cyclic succession, both based on the idea of plants making their environments less suitable for themselves. The thrust of Wilson and Agnews’s argument, on the other hand, is that this has led to the comparative neglect of processes that do, broadly, just the opposite, where a community modifies the environment, making it more suitable for that community. They call these positive-feedback switches. Four types of switch and four effects of switches are distinguished, before their mediation by water, pH, soil-elements, light, temperature, wind, fire, allelopathy, microbes, termites and herbivores are reviewed. Many of the examples are speculative, but if community ecologists are persuaded to re-examine their perspectives, as seems likely, then such speculation will have been fruitful and constructive. Hence, the papers in this volume contribute to the series’ main aim: not to provide a vehicle for specialists to review topics of interest only to other specialists in the same field, but to allow ecologists in general to remain aware not only of the advances that are made, but of the lacunae that remain in a subject that grows every more diverse. M. Begon A. H. Fitter

Contents Contributors to Volume 23 . . . . . . . . . . . . . . . . . Preface . . . . . . . . . . . . . . . . . . . . . . .

V

vii

Mechanisms of Microarthropod-Microbial Interactions in Soil JOHN LUSSENHOP

I . Summary . . . . . . . . . . . . . . . . . . I1 . Introduction . . . . . . . . . . . . . . . . . 111. Historical and Biological Reasons for Interactions . . . . IV . Interactions in the Saprophytic System . . . . . . . . A . Competition . . . . . . . . . . . . . . . . B . Foraging . . . . . . . . . . . . . . . . . C . Microarthropods as Food of Bacteria and Fungi . . . D . Bacteria and Fungi as Food of Microarthropods . . . E . Fungal and Bacterial Response to Grazing . . . . . F. Microarthropod Digestion . . . . . . . . . . G . Microarthropod Excreta . . . . . . . . . . H . Dispersal . . . . . . . . . . . . . . . . I . Summary for the Saprophytic System . . . . . . V . Microarthropod-Microbial Interactions in the Rhizosphere A . Saprophyte-Pathogen-Microarthropod Interactions . B. Vesicular-Arbuscular Mycorrhizal-Microarthropod Interactions . . . . . . . . . . . . . . . C . Ectomycorrhizal-Microarthropod Interactions . . . D . Summary for Rhizosphere . . . . . . . . . . VI . Conclusions . . . . . . . . . . . . . . . . Acknowledgements . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . .

.

1

. .

2 3 6 6 7 8 8

. .

.

. . . . .

.

. . . . . .

.

.

.

.

.

.

. . . .

.

.

. . . .

.

.

. . . . .

. .

. .

. . . .

.

.

.

.

.

10

12 13 14 15

17 17

21 22 23 24 25 25

Relative Risks of Microbial Rot for Fleshy Fruits: Significance with Respect to Dispersal and Selection for Secondary Defense MARTIN L . CIPOLLINI and E D M U N D W . STILES

I . Summary . . . . . . . . . . . . . . . . . . I1 . Introduction . . . . . . . . . . . . . . . . A . Questions and Objectives . . . . . . . . . . B . Variations in Characteristics of Fleshy Fruits . . . C . lnterspecific Variation in Secondary Defense Chemistry 111. Fruit Rot and Effects on Dispersal . . . . . . . . .

ix

. . . . . .

.

.

. . . .

.

.

. . .

35 37 37 37 38 39

X

CONTENTS

A . Influence of Fruit Rot on Dispersal . . . . B. Factors that Affect Risk of Microbial Rot . . C . Natural Selection for Fruit Defenses . . . IV . General Hypotheses and Predictions . . . . . A . General Deterrent Nature of Fruit Rot . . . B . Microbe-specific Defenses . . . . . . . C . Interspecific Variation in Defense Effectiveness V. Predictions for Temperate Seed Dispersal Systems A . Temperate Fruiting Classes . . . . . . . B. Predictions for Temperature Species . . . . VI . Conclusions . . . . . . . . . . . . . Acknowledgements . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

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

. . . . .

39 43 48 56 56 56 57 58 58 60 67 68 68

Oxygen Availability as an Ecological Limit to Plant Distribution R . M . M . CRAWFORD I . Introduction . . . . . . . . . . . . . . . . . . I1. Plant Organs Liable to Oxygen Deprivation . . . . . . . A . The Hypoxic Seed . . . . . . . . . . . . . . . B. Underground Organs . . . . . . . . . . . . . . C. Above-ground Organs with Limited Access to Oxygen . . Ill . Plant Structure and Oxygen Supply . . . . . . . . . . A . Distribution and Function of Aerenchyma . . . . . . B . Mass Movement of Air in Aquatic Species . . . . . . IV . Symbiosis and Oxygen Supply . . . . . . . . . . . . A . Root Nodules . . . . . . . . . . . . . . . . B. Nitrogen Fixation in the Rhizosphere of Aquatic Plants . . C . Mycorrhizas . . . . . . . . . . . . . . . . . V . Consequences of Oxygen Deprivation for Survival and Metabolism A . Sensing Oxygen Deficiency in Plant Tissues . . . . . . B . Cellular Effects of Oxygen Deprivation . . . . . . . . C . Metabolic Adaptations to Anoxia . . . . . . . . . D . Causes and Prevention of Post-anoxic Injury . . . . . . E . Mineral Nutrition and Flooding Tolerance . . . . . . VI . Oxygen and Plant Competition . . . . . . . . . . . . VII . Consequences of Climatic Change for the Vegetation of Oxygen-deficient Habitats . . . . . . . . . . . . . . VIII . Conclusions . . . . . . . . . . . . . . . . . . Acknowledgements . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . .

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

93 97 97 110 I I8 120 120 123 126 126 129 129 133 134 141 144 150 152 158

. . .

171 171

.

161 171

Inherent Variation in Growth Rate Between Higher Plants: A Search for Physiological Causes and Ecological Consequences HANS LAMBERS and HENDRIK POORTER I . Summary . . . . . . . . . . . . . . . . . . . . I1 . Introduction . . . . . . . . . . . . . . . . . . .

188 189

CONTENTS

111. Growth Analyses .

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

IV . Net Assimilation Rate and Leaf Area Ratio . . . . . . . . V. Specific Leaf Area . . . . . . . . . . . . . . . . . A . Components of SLA . . . . . . . . . . . . . . . B. Plasticity in SLA . . . . . . . . . . . . . . . . VI . Biomass Allocation . . . . . . . . . . . . . . . . . A . Biomass Allocation at an Optimum Nutrient Supply . . . . B. Plasticity in Biomass Allocation . . . . . . . . . . . VII . Growth. Morphology and Nutrient Acquisition of Roots . . . . A . Root Growth and Nutrient Acquisition at an Optimum Nutrient Supply . . . . . . . . . . . . . . . . B. The Plasticity of Parameters Related to Root Growth and Nutrient Acquisition . . . . . . . . . . . . . . . C . Other Root Characteristics Related to Nutrient Acquisition . . D . Conclusions . . . . . . . . . . . . . . . . . . VIII . Chemical Composition . . . . . . . . . . . . . . . . A . Primary Compounds . . . . . . . . . . . . . . . B. Secondary Compounds . . . . . . . . . . . . . . C . Defence under Suboptimal Conditions . . . . . . . . . D . Effects of Chemical Defence on Growth Potential . . . . . E . The Construction Costs of Plant Material . . . . . . . . F . Conclusions . . . . . . . . . . . . . . . . . . IX . Photosynthesis . . . . . . . . . . . . . . . . . . A . Species-specific Variation in the Rate of Photosynthesis . . . B. Photosynthetic Nitrogen Use Efficiency . . . . . . . . C . Is There a Compromise between Photosynthetic Nitrogen Use Efficiency and Water Use Efficiency? . . . . . . . . . D . Photosynthesis under Suboptimal Conditions . . . . . . E . Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . X . Respiration A . Species-specific Variation in the Rate of Respiration . . . . B. Respiration at Suboptimal Nitrogen Supply or Quantum Flux Density . . . . . . . . . . . . . . . . . . . C . Conclusions . . . . . . . . . . . . . . . . . . XI . Exudation and Volatile Losses . . . . . . . . . . . . . A . The Quantitative and Qualitative Importance of Exudation . . B. The Quantitative and Qualitative Importance of Volatile Losses . C . Conclusions . . . . . . . . . . . . . . . . . . XI1. Other Differences between Fast- and Slow-growing Species . . . A . Hormonal Aspects . . . . . . . . . . . . . . . B. Miscellaneous Traits . . . . . . . . . . . . . . . XI11. An Integration of Various Physiological and Morphological Aspects . A . Carbon Budget . . . . . . . . . . . . . . . . . B . Interrelations . . . . . . . . . . . . . . . . . XIV . Species-specific Performance under Suboptimal Conditions . . .' xv . The Ecological Consequences of Variation in Potential Growth Rate A . What Ecological Advantage can be Conferred by a Plant's Growth Potential? . . . . . . . . . . . . . . . . B. Selection of Traits Associated with a Low SLA . . . . . . C . Selection for Other Traits Underlying R G R . . . . . . . D . Consequences of a High Growth Potential for Plant Performance in Specific Environments . . . . . . . . .

xi 191 194 196 196 198 199 199 200 200 200 202 205 205 206 206 207 210 211 212 214 214 214 215 219 221 222 222 222 225 226 227 227 228 229 229 229 231 23 I 231 233 234 235 235 236 237 238

xii

CONTENTS

E . A Low Growth Potential and Plant Performance in Adverse Environments. Other than Nutrient-poor Habitats . . . . F. Conclusions . . . . . . . . . . . . . . . . . XVI . Concluding Remarks and Perspectives . . . . . . . . . Acknowledgements . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . .

. . . .

239 239 239 242 242

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

264 265 265 267 267 268 270 271 271 271 275 276 278 280 280 281 281 282 283 286 288 290 290 290 292 292 292 292 294 296 296 297 300 301 301 303 303 305 306

Positive-feedback Switches in Plant Communities

J . BASTOW WILSON and ANDREW D . Q . AGNEW I . Summary . . . . . . . . . . . . . . . I1 . Introduction . . . . . . . . . . . . . . A . Switches . . . . . . . . . . . . . . B . Types of Switch . . . . . . . . . . . C . Boundaries . . . . . . . . . . . . . D . Vegetational Situations Produced by Switches . E . Agencies . . . . . . . . . . . . . . 111. Water-mediated Switches . . . . . . . . . . A . Concept . . . . . . . . . . . . . . B. Fog Precipitation . . . . . . . . . . . C . Infiltration . . . . . . . . . . . . . D . Sediment Entrapment: Salt Marsh Pans . . . E . Ombrogenous Bog Growth . . . . . . . . F. Snow Accumulation . . . . . . . . . . IV . pH-mediated Switches . . . . . . . . . . . V. Soil-element-mediated Switches . . . . . . . . A . NPK Increase . . . . . . . . . . . . B . NPK Decrease . . . . . . . . . . . . C . Heavy Metals . . . . . . . . . . . . D . Salt . . . . . . . . . . . . . . . . VI . Light-mediated Switches . . . . . . . . . . VII . Temperature-mediated Switches . . . . . . . . A . Concept . . . . . . . . . . . . . . B. Treeline . . . . . . . . . . . . . . C . Graminoid Tussocks . . . . . . . . . . VIII . Wind-mediated Switches . . . . . . . . . . A . Concept . . . . . . . . . . . . . . B. Soil Erosion and Trapping . . . . . . . . C . Wind Damage to Plants . . . . . . . . . IX . Fire-mediated Switches . . . . . . . . . . . A . Concept . . . . . . . . . . . . . . B . Australian Closed-forest/Savannah . . . . . C . African Closed-forest/Savannah . . . . . . D . Conclusion . . . . . . . . . . . . . X . Allelopathy-mediated Switches . . . . . . . . XI . Microbe-mediated Switches . . . . . . . . . A . Oldfield Succession and Nitrogen-fixing Microbes B. Forests and Mycorrhizas . . . . . . . . XI1 . Termite-mediated Switches . . . . . . . . .

. . . . . . . . . .

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

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

. . . . . . . . . .

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

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . .

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

...

CONTENTS

XI11. Herbivore-mediated Switches . . . A . Concept . . . . . . . . . B . Grass/Grass Boundary . . . . C . Grass/Woodland Boundary . . D . Grazing and Nitrogen Cycling .

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

308 308 310 311 311 312 312 313 321 321

Index . . . . . . . . . . . . . . . . . . . . . . . . Cumulative List of Titles . . . . . . . . . . . . . . . . .

331 353

E . Insects in Pine F. Conclusions . XIV . Discussion . . . Acknowledgements . . References . . . . .

. . . . . . . . . .

XI11

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

Forests . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

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

. . . . . .

This Page Intentionally Left Blank

Mechanisms of Microarthropod-Microbial Interactions in Soil JOHN LUSSENHOP

I. I1. I11. IV .

Summary . . . . . . . . . . . . . . . . . Introduction . . . . . . . . . . . . . . . . Historical and Biological Reasons for Interactions . . . Interactions in the Saprophytic System . . . . . . . A . Competition . . . . . . . . . . . . . . . B. Foraging . . . . . . . . . . . . . . . . C . Microarthropods as Food of Bacteria and Fungi . . D . Bacteria and Fungi as Food of Microarthropods . . E . Fungal and Bacterial Response to Grazing . . . . F. Microarthropod Digestion . . . . . . . . . . G . Microarthropod Excreta . . . . . . . . . . H . Dispersal . . . . . . . . . . . . . . . . I . Summary for the Saprophytic System . . . . . . V. Microarthropod-Microbial Interactions in the Rhizosphere A . Saprophyte-Pathogen-Microarthropod Interactions . B. Vesicular-Arbuscular Mycorrhizal-Microarthropod Interactions . . . . . . . . . . . . . . . C . Ectomycorrhizal-Microarthropod Interactions . . . D . Summary for Rhizosphere . . . . . . . . . . VI . Conclusions . . . . . . . . . . . . . . . . Acknowledgements . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . .

. . .

1

2 3 6 6 7 8 8 10

12 13 14 15 17

. . . . . . . . .

17

. . . . . .

21 22 23 24 25 25

. . . . . .

. . . . . .

I. SUMMARY Many aspects of the distribution. abundance. and activity of soil fungi and bacteria are controlled by microarthropods . In saprophytic successions. six mechanisms of interaction are important. Two control fungal distribution and abundance: (a) selective grazing of fungi by microarthropods. and (b) dispersal of fungal inoculum by microarthropods . Four additional mechanisms stimulate microbial activity: (a) direct supply of mineral nutrients in ADVANCES IN ECOLOGICAL RESEARCH VOL . 23

Copsright $: I Y Y Z Acudemic Press Limired All righrs of reproducrion in uny form rrsrrwd

ISBN s 1 2 4 1 3 9 2 3 - 5

1

2

J . LUSSENHOP

urine and feces, (b) stimulation of bacterial activity by microarthropod activity, (c) compensatory fungal growth due to periodic microarthropod grazing, and (d) release of fungi from competitive stasis due to microarthropod disruption of competing mycelial networks. Selective grazing can control species distribution and favor either efficient or inefficient decomposer fungi. Moderate grazing may stimulate microbial activity, thus augmenting either mineralization or immobilization of nutrients by micro-organisms. In the rhizosphere, the demonstrated mechanisms of interaction are dispersal and selective grazing. Microarthropods carry fungal propagules, including those of root pathogens, to root surfaces. Microarthropods also graze fungi on root surfaces, and they selectively consume saprophytic fungi. It has not been shown whether dispersal of pathogens to the rhizosphere is less important than preferential grazing o n pathogens. Vesicular-arbuscular mycorrhizal hyphae and germ tubes are also grazed preferentially, hence microarthropods are associated with fewer and less effective vesicular-arbuscular fungi. Ectomycorrhizal roots and their perennial networks in the soil may be physically and chemically protected from microarthropod grazing.

11. INTRODUCTION Interactions among fungi, bacteria, and invertebrates are central to many processes in soil ranging from decomposition to the functioning of the rhizosphere. The possible mechanisms of these interactions, including grazing, disturbance, and dispersal, have been little studied. It is important that these mechanisms be understood because in the future, conscious management of the soil biota in agricultural and forest soils will require knowledge of them. Grazing is the mechanism of interaction given most attention since Coleman et al. (1983) showed that mineral nitrogen and phosphorus levels in the rhizosphere were raised due to nutrients in excreta of bacterivorous nematodes and protozoa. But grazing, which is the consumption of parts of living organisms, is a complex phenomenon because of the modular nature of many grazed organisms, and because of the behavior of grazers. Grazers may be selective, and affect competition among grazed species. Grazers may cause disturbance that affects recovery of grazed species. Grazers may disperse propagules of the grazed species, and their excretions may control the rate and proportion of nutrient return to the grazed site. This complexity exists in grazed higher plant and algal systems, and it is likely that it exists in grazed microbial systems also. Seastedt (1984), for example, showed that while the presence of arthropods in litter bags increased mass loss by 23% on average, it had a smaller effect on mineral nutrient mineralization. Seastedt (1984) speculated that this smaller mineral nutrient effect might be due to arthropod

MECHANISMS OF MICROARTHROPOD-MICROBIAL INTERACTIONS IN SOIL

3

stimulation of microbial growth. Anderson ( I 988) suggested the importance of dispersal and control of fungal species composition. The present chapter focuses on these and additional mechanisms of interaction between fungi, bacteria, and microarthropods. Microarthropods (collembola, protura, mites, pauropods) form a distinct group when their mass is compared with that of macroarthropods (millipedes, isopods) and nematodes (Fig. 1). Since mass is correlated with density, resource use, movement, and reproductive rate (Peters, 1983), microarthropods should interact with micro-organisms as grazers and transporters, while macroarthropods are primarily comminutors of litter, and nematodes and protozoa are primarily bacterivores. The goal of this chapter is to describe and evaluate the importance of mechanisms by which micro-organisms and microarthropods interact in soil. It is a first step towards making quantitative predictions about functioning of saprophytic and rhizosphere foodwebs. This is a demanding goal for a biota whose natural history is poorly known. The present chapter builds on recent additions to the knowledge of natural history of soil fungi (Domsch et a[., 1980; Wicklow and Carroll, 1981; Cooke and Rayner, 1984; Rayner and Boddy, 1988), and microarthropods (Dindal, 1990; Norton, 1992).

111. HISTORICAL AND BIOLOGICAL REASONS FOR INTERACTIONS Soil micro-organisms and microarthropods have interacted since the Devonian when foodwebs developed in soil around the first terrestrial plants. Just as insect herbivores radiated in response to angiosperm evolution in the Cretaceous, soil microarthropods and fungi represent an earlier radiation in response to tracheophyte evolution. Microarthropod fecal pellets containing hyphae are known from the Silurian (Sherwood-Pike and Gray, 1985). By the Devonian, fossil oribatids (Shear et al., 1984), prostigmatids and collembola (Kevan et al., 1975) were present. In the Devonian, damaged Rhynia tissue (Kevan et al., 1975) suggests that arthropod herbivores had evolved, and the presence of branching septate hyphae in the secondary xylem of a fossil, arborescent, progymnosperm indicates that saprophytic fungi had evolved (Stubblefield and Taylor, 1988). Devonian terrestrial plants may have been aided in water and nutrient uptake by endomycorrhizal symbionts (Pirozynski and Malloch, 1975), although the first unquestioned fossil mycorrhizal arbuscules are from the Triassic (Stubblefield et a[., 1987). Protective tissue, spore ornamentation, and presence of cutin and suberin in Devonian fossils suggest a need for protection from herbivores as well as for water conservation. Later development of lignins, terpenoids and flavonoids in the Carboniferous is interpreted

4

J . LUSSENHOP

0

DIPLOPODA

0

CHI LO PO DA

0

ISOPODA

0

PSEUDOSCORPIONIDA

0

CO LLEMBOLA

0

ACARINA

0 0

PROTURA PAUROPODA

0

NEMAT0 DA

1 8

@

&

#

Fig. 1. Average live mass of soil invertebrate taxa shows three distinct groups when graphed on a logarithmic scale. Values are averages for species from many communities tabulated by Edwards (1967) and Peterson (1982).

MECHANISMS OF MICROARTHROPOD-MICROBIAL INTERACTIONS IN SOIL

5

as defense against pathogenic and decay fungi as well as from herbivorous insects; these compounds must have changed the chemical ecology of decomposers, including microarthropods (Swain and Cooper-Driver, 1981). The long history of co-occurrence of microarthropods with fungi and bacteria is one reason to expect interactions. Digestive and transport mutualisms are not likely to be as well-developed among soil as among terrestrial arthropods. For example, soil microarthropods appear to lack mycetomes-groups of specialized cells containing symbiotic micro-organisms-which aid digestion in many insects. Soil microarthropods may not need mycetomes since they ingest so many micro-organisms and microbial exoenzymes. Soil microarthropods also lack external cavities (mycangia) and internal sacs (sporothecae) for transport of fungal propagules. Soil microarthropods may not need these specialized transport structures as much as terrestrial arthropods, but they may have primitive forms of mycangia. The highly sculptured integument of 75% of the higher oribatid superfamilies may function as mycangia: figs 4 . 5 4 . 6 in Blackwell ( 1 984) show a Carahodes sp. with spores of the myxomycete Lycopalu epidendrum in cuticular depressions. On the other hand, well-developed reciprocal chemical interactions between microarthropods and micro-organisms should be expected. A suggestive example is the observation of Wicklow (1988) that detritivorous arthropods are more tolerant of mycotoxins than herbivorous arthropods. A second reason for expecting interactions between microarthropods and micro-organisms is the contrasting biology of the groups. Microarthropods, fungi, and many bacteria are heterotrophs, and so compete for similar resources, yet differences in their size, method of ingesting food, and population growth result in interactions. Soil bacteria are the smallest and most biochemically diverse of the three groups. They have the highest intrinsic rate of increase ( r ) , but due to limited carbon availability, Jenkinson and Ladd (198 1) estimated that the average cell divides once every 2.5 years in an English soil. Predominance of bacteria or fungi determines the invertebrate foodweb present. Bacterial biomass may sometimes be greater than fungal (Ingham et al., 1989), and bacterial activity may be briefly or locally greater than that of fungi. On a whole soil basis, Anderson and Domsch (1975) used selective inhibitors to show that fungal metabolic activity accounted for more than half the carbon mineralization in agricultural and forest soils. Hyphal growth is a central adaptation of fungi that determines much of their biology. The high surface to volume ratio of hyphae allows efficient utilization of the products of external enzymes, and hyphal networks allow translocation of nutrients to sites of active decomposition where growing hyphae are able to penetrate solid substrates. In contrast, bacteria occupy surfaces. It is the vulnerability of hyphal networks that is the basis for competitive interactions with invertebrates. Hyphae may form interwoven

6

J . LUSSENHOP

cords and rhizomorphs (Fig. 3C, p.19) as well as resting structures such as stroma, sclerotia, and pseudosclerotia: these are resistant to physical extremes as well as to animals. Because of their ability to grow at lower water potentials than bacteria, fungal-based foodwebs predominate in arid habitats (Whitford, 1989). Microarthropods are the right size to graze fungi and bacteria. Their mouthparts function by plucking and scraping (collembola, mites), shearing (mites), or piercing (prostigmatid mites and protura). They exhibit a range of life histories reflecting their strategy for utilizing micro-organisms or detritus. Microarthropod life histories include species with explosive reproductive rates (collembola with r as high as 0.15 (Gregoire-Wibo and Snider, 1977) to 0.3 (Rapoport and Aguirre, 1973)) allowing a numerical response to fungal growth. Other groups exhibit great population persistence-some oribatids with r close to zero (Cancela da Fonseca, 1980)-they are already present when fungal growth starts.

IV. INTERACTIONS IN THE SAPROPHYTIC SYSTEM A. Competition Among fungi, interspecific and even intraspecific competition between mutually antagonistic dicaryons is strong and slows decomposition. This is because saprophytic fungi gain access to resources by occupying volume with their hyphae. In the process hyphal networks compete by a number of mechanisms which have been summarized by Cooke and Rayner (1984). Effects on hyphae some distance away may be caused by chemicals; contact effects include parasitism, hyphal interference, hyphal fusion, or the production of dense zones of mycelia. As a result of these strong fungal competitive interactions, there is an inverse relationship between fungal species number and decomposition. Wicklow and Yocom (198 1) measured mass loss of rabbit feces by six species of coprophilous fungi growing singly or in combinations. They found that as species number increased, decomposition declined by 4.6%. The same result was obtained earlier by Norman (1930) who measured heat produced by Aspergillus, Trichoderma, and an actinomycete species growing singly and in combinations in a thermos. Similarly, in the field, Coates and Rayner (1985) compared decomposition of beech logs that were naturally colonized by saprophytic fungi, inoculated with four strong competitors, or recut to increase the number of colonizing species. Logs with the most species were least decomposed. Early in saprophytic succession microarthropods are strong competitors of fungi, and control fungal distribution and abundance by selective grazing

MECHANISMS OF MICROARTHROPOD-MICROBIAL INTERACTIONS IN SOIL

7

and carrying inoculum. In addition, microarthropods benefit from the resources not utilized by competing fungi. Due to competition, unoccupied zones of substrate are left between those occupied by fungi, and these may be used by microarthropods. There are two suggestive examples involving fly larvae. Boddy et a/. (1983) found that unoccupied zones of agar between competing basidiomycetes and ascomycetes were colonized by fungus gnat larvae (Bradysiu sp.). Using coprophilous fungal species combinations in rabbit feces, Lussenhop and Wicklow (1985) found that as the number of fungal species increased, the numbers of fungus gnat larvae (Lycoriella mali) increased. They interpreted this to mean that as numbers of fungal species increased, there was more unoccupied space between competing fungi available to the larvae. Since much of the fecal material was easily digestible, larval numbers increased as number of fungal species increased. In saprophytic successions, later-appearing fungi are stronger competitors of microarthropods than early species. In addition to sometimes presenting an impenetrable weft of mycelia (Binns, 1980), many late successional species are defended against herbivores. For example, the coprophilous fungus Chuetomium bostrycodes disperses spores relatively slowly from terminal hairs on perithecia. C . bostrycodes has lacerate terminal hairs on its perithecia that prevent grazing. Chaetomium species also produce chemical defenses (anthraquinones and chaetomin) (Wicklow, 1979).

B. Foraging Understanding the grazing interactions between microarthropods and fungi will require knowledge of the foraging behaviour of both groups. Fungi forage by varying growth patterns from diffuse, perennial networks to shortlived colonies, and by rhizomorphs. Dowson et al. (1988) offered baits to cord-forming basidiomycetes and demonstrated that Steccherinum fimbriatum switched growth pattern from slow-diffuse to fast-effuse exploration after contact with bait. Other basidiomycete species had longer-range foraging patterns. Foraging behavior of the hyphomycete Mortierellu isubelfina was shown to change in response to grazing by the collembolan Onychiurus armatus. Hedlund et al. (1991) showed that grazing caused slowgrowing appressed hyphae to switch to non-sporulating aerial hyphae. Foraging strategy and ability to respond to chemical cues are important for microarthropods. Streit and Reutimann et al. (1983) showed alternation between a searching and feeding mode of foraging using a surface-dwelling oribatid that was offered colonies of six different micro-organisms in petri dish experiments. Bengtsson et a/. (1988) found that the collembolan Onychiurus armatus was attracted to Mortierella isabellina and Penicillium spinulosum by odors.

8

J. LUSSENHOP

C. Microarthropods as Food of Bacteria and Fungi Microarthropods are surrounded by spores and conidia many of which can, if they lodge on the integument, germinate and grow into the animal, eventually killing it (Evans, 1988). These include specialized entomopathogenic species such as Acremonium sp., Beauveria bassiana, Conidiobolus coronatus, Metarrhizium anisopliae, and Verticillium lecanii (Domsch et a/., 1980; Keller and Zimmerman, 1989), as well as facultative pathogens such as Aspergillus frclvus and species of Fusarium. The impact of entomopathogenic fungi on natural populations of microarthropods is unknown. Purrini (1983) found only 0.7% of the collembola in European forests were infected with fungi; another 0.7% were infected with bacteria, and 2% with microsporidia. In the same study, Purrini and Bukva (1984) found that among oribatids, fungal and protozoan infections increased considerably in areas receiving high sulfur dioxide fallout. Microarthropods may be of great importance as vectors of the approximately 13 species of specialized entomopathogenic fungi that attack holometabolous insects. By using a Berlese funnel to force microarthropods to move through inoculated soil, Zimmerman and Bode (1983) showed that collembola and mites transport spores of Metarrhizium anisopliae. Adaptations to avoid touching the medium, and to prevent spore lodging, are important defenses against entomopathogenic fungi (Rawlins, 1984). This is because spores or conidia of entomopathogenic species germinate upon contact with arthropod cuticle. It may be no accident that 24% of collembola, and 40-56% of oribatids listed in Table 1 carried no inoculum. Further defensive adaptations include cuticular melanin which is believed to be toxic to fungi (Charnley, 1984), and fungitoxic secretions. The sex pheromone of the stored product mite, Caloglyphus polyphyllae is fungitoxic (Kuwahara et al., 1989). Collembola are flexible and are able to remove surface spores. The elaborate cuticular sculpture and setation of euedaphic collembola may be an adaptation to minimize contact with fungi.

D. Bacteria and Fungi as Food of Microarthropods A model for soil microarthropod grazing on saprophytic fungal colonies comes from observations of stream invertebrates grazing on leaves. Stream invertebrates selectively graze portions of leaves with fungal colonies (Barlocher, 1980; Arsuffi and Suberkropp, 1985). Similarly, isopods (Oniscus asellus) feed on pockets of mycelium of the leaf pathogen Rhytisma acerinum Fr. on maple leaves (Gunnarsson, 1987). It is likely that microarthropods selectively graze soil fungi in the same way. There is evidence that resource partitioning among microarthropods results in small species and juveniles grazing bacteria, and larger individuals grazing fungi (Bakonyi, 1989).

MECHANISMS OF MICROARTHROPOD-MICROBIAL

INTERACTIONS IN SOIL

9

Table 1 Numbers of fungal propagules carried by microarthropods Habitat/group Arctic and Subarctic Acari: Oribatida L. F, H horizons in aspen woodland Collembola Onychiurus subtenuis

L, F, H horizons in beech-maple woods Acari: Oribatida Collembola Diplopoda Coleoptera Staphylinidae

Number of fungal % Individuals species carried without inoculum 1.4

40

Reference Behan and Hill ( 1 978) Visser (1985)

2.4

ND"

Number of fungal genera carried 0.5 1.2 1.3

56 24 19

1 -4

14

Pherson and Beattie ( 1979)

N D= not determined.

Microarthropod grazing intensity is strong enough to control abundance and distribution of fungi. In early fall, ascospores of the saprophytic fungus Coniochaeta nepalica are briefly common in soil of the oak-birch forest in New York; Gochenaur (1987) recorded an 80% decline in frequency of C . nepalica spores during fall. Gochenaur (1987) placed ascospores of C. nepalica as well as Sordariajimicola in the A horizon on membrane filters and found that they disappeared at a rate of 60% per day. Since microarthropod fecal pellets accumulated on the filters at a similar rate, Gochenaur (1987) concluded that microarthropod feeding was responsible. In a second example, microarthropod feeding limited production of primary infective inoculum of two pathogens of black walnut (Juglans nigra): Mycosphaerella juglandis which causes mycosphaerella leaf spot, and Gnomonia leptostyla which causes walnut anthracnose. The primary inoculum of both fungi is produced by perithecia on fallen leaves. Kessler (1990) found that when perithecia-bearing walnut leaves fell into heavy leaf litter supporting microarthropod populations, the perithecia were eaten, primarily by collembola. Leaves falling into grassy areas with poor litter and low microarthropod populations were not subjected to intense grazing, and in these habitats walnut trees became infected the next year. Collembolan grazing controlled the vertical distribution of two perennial basidiomycetes in the litter of a 32-year-old Sitka spruce (Picea sitchensis) plantation in England. Newell (1984a,b) studied the two basidiomycetes that produced over 99% of the sporocarps at the site. Sporocarp depths showed

10

J. LUSSENHOP

that Marasmius androsaceus occurred naturally in the L horizon, and Mycena galopus in the F horizon. Newell (l984a) showed that the collemboIan Onychiurus latus preferred Marasmius androsaceus to Mycena galopus in laboratory feeding trials. When numbers of 0. l a m were experimentally increased in the field, density of M . androsaceus declined (Newell, 1984b). Because the competitively inferior fungus was the best decomposer, limitation of its distribution to the L horizon resulted in slower decomposition (Newell, 1984b). Some sclerotia are chemically protected (Wicklow, 1988) and collembola will graze on their conidial apparati but not on the sclerotia themselves (Aspergillusjavus; Lussenhop, personal observation). In other cases (Sclerotinia sclerotiorum) collembola apparently eat sclerotia in the field (Anas and Reeleder, 1987).

E. Fungal and Bacterial Response to Grazing 1 . Eflects of Grazing on Decomposition Microarthropod activity favors growth of bacteria, probably by mixing cells with fresh substrate. Even if microarthropods do not graze fungi, their activity may break hyphae just by walking through them (Lussenhop, personal observation; J. C . Moore, personal communication). Hanlon and Anderson (1 979) inoculated leached oak leaves with the basidiomycete Coriolus versicolor and added 0, 5, 10, or 20 collembola (Folsomia candida). Bacterial biomass exceeded fungal biomass when 10 or more collembola were present. The same results were obtained in a field experiment by Lussenhop et al. (1980). They found that beetle and/or fly larvae in cattle dung were associated with increases in bacteria and decreases in fungi even though mouthparts of the beetle and fly larvae made it impossible for them to ingest fungal hyphae. The possibility that invertebrate grazing stimulates fungal growth was suggested when the minor contribution of arthropods to soil respiration was recognized (Macfadyen, 1961). Such stimulation could occur as a result of what is called compensatory growth in studies of plant response to grazing (reviewed by Belsky, 1986). Compensatory growth is increased productivity or mass relative to a control due to grazing. Possible mechanisms include (a) fungal growth after senescent hyphae are grazed, and (b) regrowth after periodic grazing of actively growing mycelia. Periodic grazing is the mechanism associated with experiments showing compensatory response of fungi in Table 2. Bengtsson and Rundgren (1983) modeled what may happen in nature by alternating 2-day grazing bouts with 5-day growth periods; this increased fungal CO, output by about 5 % relative to ungrazed controls. In a more realistic physical setting, Bengtsson et al.

MECHANISMS OF MICROARTHROPOD-MICROBIAL

INTERACTIONS IN SOIL

II

Table 2 Compensatory growth of micro-organisms in response to microarthropod grazing

Fungal species Laboratory Soil dilution Botrytis cinerea Coriolus versicolor Mortierella isabellina Vert icillium bulbillosum Penicillium spinulosum Millipede faecal flora Field Soil dilution

a

Fob[somiaJimetaria F. candida F. candidu

~

-/+h -a

+

Reference Andren and Schnurer (1985) Hanlon (198 1a) Hanlon and Anderson (1979) Bengtsson and Rundgren (1983)

Ony ch iurus armatus 0. armatus

+

Bengtsson et al. (unpublished)

0 . armatus

+

Bengtsson et al. (unpublished)

0. quadr iocella tus

+"

Drift and Jansen (1977)

+ a.c

Addison and Parkinson (1978)

+ a.c

Addison and Parkinson ( 1 978)

Hypogastrura tullbergi Folsomia regularis

Soil dilution ~

Growth relative to Arthropod species controls

~~

Bacteria were present. Increase with fungi grown on high nutrient medium, otherwise decrease. In the less severe of the two field sites on Devon Island.

(unpublished) connected fungal colonies with tubing so that collembola could move from one colony to another: this resulted in periodic grazing and compensatory growth which increased CO, output by 4-5 times. Laboratory experiments listed in Table 2 as not showing compensatory growth had constant, relatively intense grazing by individuals belonging to species of Folsomia which tend to be larger than Onychiurus individuals. In addition, presence of bacteria may have affected the results in many of the experiments. The stimulating effect of microarthropods on fungal and bacterial growth effects nutrient transformation. The litter-inhabiting collembolan, Tomocerus minor, was associated with nitrogen immobilization in litter but with nitrogen mobilization in the fermentation layer by stimulating fungal growth in the different nutrient regimes (Verhoef et ul., 1989). Both T . minor and an isopod, Philosciu muscorum, in microcosms containing pine litter increased CO, output and exchangable phosphate, but only T . minor increased dehydrogenase, cellulase activity, and nitrate concentration, due to the collembolan's greater stimulation of microbial activity (Teuben and Roelofsma, 1990).

12

J. LUSSENHOP

2. Eflects of Grazing on Fungal Species Numbers Wicklow and Yocom (1982) showed that the number of species of coprophilous fungi on rabbit feces declined as the density of larvae of the sciarid fly, Lycoriella mali, increased. Whether the reduced species number was a result of grazing favoring competitive dominants, or of reversing the competitive superiority of competitive inferiors is not known. Collembolan grazing reversed the outcome of competition between two basidiomycetes studied by Newell (1 984a, b). In contrast, collembolan grazing favored a competitively superior fungus in Parkinson et al.’s (1979) study of two saprophytic fungi growing in aspen leaves when snow was melting. They isolated a competitively inferior, sterile, dark fungus that was grazed by the collembolan Onychiurus subtenius, and a competitively superior basidiomycete in whose presence in culture 0. subtenius died. They showed that collembolan grazing reinforced the competitive effects of the basidiomycete. Whittaker (1981) confirmed these interactions in the field.

F. Microarthropod Digestion Microarthropod habitats are rich in micro-organisms, microbial exoenzymes, and products of microbial degradation. For this reason ingestion of a variety of microbially conditioned materials and trituration of food materials may be the most important digestive adaptations of microarthropods. Mouthparts of oribatids (Phthiracarus sp.: Dinsdale, 1974b), and collembola (Tomocerus longicornis: Manton, 1977) function to minimize the size of food particles; this probably enhances activity of microbial and endogenous enzymes in the gut, and contributes in a minor way to comminution. Gut micro-organisms of microarthropods are derived from the microorganisms they ingest (Seniczak and Stefaniak; 1978; Haq and Konikkara, 1988). The particular microbial species present in the gut effect time to maturity and number of eggs laid in oribatids (Stefaniak and Seniczak, 1981). It is not surprising that collembola in culture can select hyphae with the highest nutrient content (Leonard, 1984; Amelsvoort and Usher, 1989), and that collembola produce more eggs when fed on fungi with higher nitrogen content (Booth and Anderson, 1979). Presence of fungi in microarthropod guts is associated with cellulases, while bacteria are associated with proteases, amylases, and chitinases (Stefaniak and Seniczak, 1981). Borkott and Insam (1990) presented evidence that chitinolytic bacteria and Folsomia candida have a mutualistic relationship. They fed F. candida microbially conditioned or unconditioned chitin, with and without antibiotics: the collembola gained the most mass on microbially conditioned chitin without antibiotics; numbers of chitinolytic bacteria were

MECHANISMS OF MICROARTHROPOD-MICROBIAL INTERACTIONS IN SOIL

13

greater in feces than in food. Although some microarthropods are believed to be bacterivores, two polyphagous collembola (Proisotoma minuta, Hypogastrura tullhergi) did not survive on diets of any of seven soil bacteria isolated from their habitat (Harasymek and Sinha, 1974). Microarthropods may facilitate the activity of fungal enzymes in their guts by maintaining basic pH, but this is poorly documented. If the basic gut pH ascribed to oribatids (Dinsdale, 1974a; Seniczak and Stefaniak, 1978) is general, then they may be similar to the sporocarp-inhabiting beetles studied by Martin (1987). Martin (1987) described a series of adaptations allowing arthropods to benefit from ingested fungal enzymes in digestion of plant structural carbohydrates. The adaptations range from (a) favoring activity of fungal enzymes by basic gut pH as illustrated by sporocarp-inhabiting beetles, to (b) transport and inoculation of wood-decomposing fungi as well as maintaining favorable gut conditions for their enzymes by siricid wood wasps and scolytid beetles, to (c) culture of fungi whose enzymes will be used in digestion by attine ants or harvesting termites. Microarthropods may have additional adaptations for benefiting from microbial enzymes. For example some collembola ingest clay to which bacterial enzymes are adsorbed. Kilbertus and Vannier (1981) showed that a cavernicolous collembolan lost mass without a dietary source of clay, and that clay was associated with bacterial cells in the gut. Touchot et al. (1983) later showed that dietary clay was important to the collembolan Folsomia candida, possibly because phenolic compounds were adsorbed to clay surfaces and did not inhibit bacterial activity.

G. Microarthropod Excreta The possibility that microarthropods return significant amounts of mineral nutrients in urine and feces was raised by Verhoef et al. ( 1 988). They fed the collembolan Tomocerus minor a diet of laboratory-grown hyphae, and estimated that 50% of the dietary nitrogen was released as urea, and that the nitrogen concentration of fecal pellets was 56% higher than the hyphae. In nature, the contribution of microarthropod urine to mineral pools may be significant but has not yet been quantified. Most experiments with soil arthropod fecal pellets were done with those of macroarthropods. Conclusions from experiments with millipede and isopod fecal pellets are that (a) bacterial activity is favored in fecal pellets due to the small size of particles (Webb, 1977; Hanlon, 1981b), and by gut conditions (Reyes and Tiedje, 1976; Anderson and Bignell, 1980), but that (b) fecal material does not decompose faster than uneaten material (Nicholson et al., 1966). These conclusions are supported by Grossbard’s ( 1 969) experiment showing that fecal pellets of oribatid mites fed I4Clabeled grass decomposed at the same rate as uningested grass.

14

J. LUSSENHOP

The possibility that microarthropod fecal pellets contribute to the formation of water-stable soil aggregates is of considerable interest. Tisdall and Oades (1982) pointed out that the smallest aggregates are formed from mineral particles held together by physical forces, but as smaller aggregates combine into larger, the importance of biological binding agents increases. In Tisdall and Oades’ (1 982) scheme, aggregates > 2000 pm are formed from aggregates between 20 and 250 pm in diameter and are held together by microbial- and plant-derived polysaccharides, as well as by fibrous plant roots and fungal hyphae, particularly those of vesicular-arbuscular mycorrhizal fungi. Since microarthropod fecal pellets are 30-90 pm in diameter (Rusek, 1975), it is not hard to imagine them forming nuclei of soil aggregates.

H. Dispersal Bacterial and fungal spores are dispersed through soil by physical mechanisms, but microarthropods modify natural distribution patterns by dispersing propagules from concentrations around sites of sporulation. In litter, fungal spores and bacteria are dispersed horizontally and vertically by the spreading pressure of monolayer-forming substances on aqueous films (Bandoni and Koske, 1974). Wettable surfaces of spores of some conidial fungi allow them to be moved a few millimeters by advancing water fronts (Hepple, 1960). Finally, hyphal growth, particularly along roots, is extensive enough to maintain propagules throughout soil. Microarthropod ingestion damages fungal spores, but the small fraction that survives is probably important (Table 3). Pherson (1980) speculated that some fungi are adapted for dispersal by microarthropods. He found that viable spores of Alternaria, Epicoccum, and Penicillium were most frequent in feces of litter microarthropods. When microarthropods were excluded from sterile leaf discs by 5-pm mesh bags, colonies of these three genera were significantly less frequent than other fungal genera compared with control disks in 500-pm mesh bags incubated in the F layer of a Michigan beechmaple forest. In their study of grazing selectivity, Moore et al. (1987) found that by sporulating quickly, Penicillium citrinum had more spores eaten and dispersed than other species in the study. Dispersal of microbial propagules by microarthropods appears to be passive, and thus a number of simple patterns exist:

(i) Microarthropods carry more propagules and species in litter than in mineral soil (Visser, 1985). (ii) Body-size is proportional to number of fungal genera carried (Table 1 : Pherson and Beattie, 1979). (iii) Aggregations of spores at sporulation sites are dispersed rapidly by microarthropods (Lussenhop and Wicklow, 1984). Visser et al. ( 1 981)

MECHANISMS OF MICROARTHROPOD-MICROBIAL INTERACTIONS IN SOIL

15

Table 3 Per cent survival of fungal propagules ingested by microarthropods and macroarthropods

Microarthropod species

Fungal species

Laboratory Collembola Entomobrya purpurascens Pseudosinella alba

Penicilliurn sp.

Onychiurus quadrocellatus

%Faecal pellets with viable propagules

Reference

2 (YOof spores) Cervek (1971)

11 species

Cladosporium sp.

2-25

Ponge and Charpentie (1981)

13

Drift (1965)

(control = 83) Acarina: Astigmata Rhizoglyphus echinopus Caloglyphus sp.

Vert icillium albo-atrum conidia microsclerotia Pythium myriotylum

Price (1976) 94 57-86 90

Field Collem bola Onychiurus subtenuis Aspen woodland Arctic soils Acarina: Oribatei microarthropods Oak-birch forest

50 13

8-30

Shew and Beute ( 1 979)

Visser (1985) Behan and Hill ( 1 978) Gochenaur (1987)

suggest that microarthropods bring fresh inoculum to sites they have grazed, with the overall effect of increasing nutrient immobilization.

I. Summary for the Saprophytic System Five mechanisms of interaction between microarthropods and micro-organisms occur in saprophytic systems. Two mechanisms affect distribution and abundance of fungi; three affect bacterial and fungal metabolic activity. Two mechanisms by which microarthropods affect fungal distribution and abundance are selective grazing and dispersal of fungal propagules: (i) Control of fungal species distribution by selective grazing is well supported by field observation and experiment (Parkinson et al., 1979; Whittaker, 1981; Newell, 1984a,b; Gochenaur, 1987; Kessler, 1990). However, generalizations as to effects of selective grazing cannot be

16

J . LUSSENHOP

drawn yet. If microarthropods always selectively grazed the competitively dominant fungus, decomposition would be slowed. But this only happened in Newell’s (1984a,b) study; in the study by Parkinson et a/. (1 979) the opposite occurred, and decomposition probably increased. (ii) Dispersal of fungal propagules seems particularly important early in saprophytic succession when it may increase the rate of decomposition. Cultural methods used to assess fungi associated with microarthropods lead to an underestimate of total fungal species numbers, and an overestimate of the importance of fast growing fungal species. Studies listed in Tables 1 and 3 have not quantified numbers of propagules carried by microarthropods, only numbers of different species or genera carried. A more appropriate cultural technique would be dilution plating of individual microarthropods on media that retard colony spread; this would give numbers of propagules carried per individual. Four mechanisms affect metabolic activity of micro-organisms: (i) Direct return of mineral nutrients in urine and feces has not been quantified, but is potentially an important mechanism stimulating microbial growth (Verhoef et a/., 1988). (ii) Bacterial growth is briefly stimulated by mixing and comminution of microarthropods. It is likely that interference with fungal growth indirectly benefits bacteria. The disturbance of cultivation favors bacteria in the same way though on a much larger scale (Hendrix et a/., 1986). (iii) Compensatory growth of fungi in response to episodic microarthropod grazing increases the decomposition rate above what it would be without grazing. Compensatory growth is likely to be important in the field, and is likely to be associated with nutrient immobilization by fungi. Compensatory growth has only been demonstrated in laboratory experiments designed to mimic episodic grazing; careful observation of grazing in situ is needed to substantiate this as an important mechanism. (iv) Decomposition rate is inversely proportional to fungal species number in the absence of microarthropods, because by competing for volume of substrate, fungi slow each other’s growth rates. Invertebrates thus increase decomposition rate by reducing competitive stasis among fungi. The effect increased decomposition rate by about 5% in laboratory studies.

In nature bouts of grazing would involve all of these mechanisms. Selective grazing by a microarthropod would add mineral nutrients, reduce fungal competition, stimulate bacterial growth and disperse fungal propagules (Fig.

MECHANISMS OF MICROARTHROPOD-MICROBIAL INTERACTIONS IN SOIL

17

2). Frequency of grazing bouts would determine whether compensatory growth occurred. The importance of each mechanism will change during saprophytic succession. With newly fallen leaves, introduction of fungal inoculum, selective grazing, and stimulation of bacteria by microarthropods are likely to be the most important mechanisms. As leaves age and enter the fermentation layer, compensatory growth response to microarthropod grazing and the mineral nutrients in excreta are likely to be important. Finally, as leaf fragments enter the humus layer, selective grazing and release of fungi from competitive stasis will be important. These mechanisms predict that microarthropods will stimulate fungal growth and magnify effects fungi have on limiting nutrients. In an immobilizing environment such as leaf litter, microarthropods should stimulate fungal growth which will reduce mineral nutrient concentrations of limiting nutrients. Lower in the soil horizons, in a mobilizing environment such as humus, microarthropods should stimulate fungal growth which will increase mineral nutrient concentration. An example is the decrease in ammonium-N in the L layer and the increase in ammonium-N in the F layer of a relatively lownutrient Pinus nigra forest caused by Tomocerus minor in microcosms (Verhoef et al., 1989). A second example is the increased inorganic-N concentration in response to reduced fungivores and reduced hyphal lengths in a Pinus contorta forest soil experimentally manipulated with biocides by Ingham et al. (1989). This pattern is not predictable, however, for even within the same study other variables affect the link between microarthropod stimulation of micro-organisms and nutrient mineralization. These additional variables include overgrazing, soil nutrient concentration, and numbers of bacteria relative to fungi.

V. MICROARTHROPOD-MICROBIAL INTERACTIONS IN THE RHIZOSPHERE Microarthropods interact with three groups of micro-organisms in the rhizosphere. These three groups-saprophytic and pathogenic bacteria and fungi, vesicular-arbuscular mycorrhizal fungi (VAM), and ectomycorrhizal fungi (ECM)-have distinct biologies and life histories, hence microarthropods interact differently with each (Fig. 3).

A. Saprophyte-Pathogen-Microarthropod Interactions Bacterial and fungal numbers are orders of magnitude higher around roots than in soil away from roots (Curl and Truelove, 1986). Microarthropod density is also higher around roots, though core sampling methods have

18

I. LUSSENHOP

FUNGI COLONIZE SUBSTRATE, OU TCO MPET E BACTERIA

FUNGAL GROWTH SLOWS DUE TO INTENSE COMPETITION AMONG FUNGI

GRAZING MICROARTHROPODS DESTROY COMPETING HYPHAL NETWORKS, ADD NUTRIENTS IN EXCRETA

BACTER AL POPULATIONS INCREASE

NEW FUNGAL SPECIES DISPERSED BY MICROARTHROPODS OUTCOMPETE BACTERIA

Fig. 2. Flow chart showing effects of microarthropod grazing on saprophytic micro-organisms.

MECHANISMS OF MICROARTHROPOD-MICROBIAL INTERACTIONS IN SOIL

19

A. NEMATODES

t PROTOZOA

t

BACTERIA

FUNGI

f

A \ EXUDATE

6.

DEAD CORTICAL CELLS

SLOUGHED

TISSUE

C.

Fig. 3. Three rhizosphere models based on Fogel (1991):A, Bacterial-based foodweb at root tip, and fungal-based foodweb along mature root; B, vesicular-arbuscular mycorrhiza; C, endomycorrhiza.

made this difficult to show. Curry and Ganley (1977) identified roots of pasture plants in soil cores and showed that grass roots were associated with higher microarthropod numbers, but could not distinguish rhizosphere from non-rhizosphere populations. Core samples collected in grid patterns between Picea abies trees showed highest collembola numbers in areas with the greatest density of fine, mycorrhizal roots (Poole, 1964). Wiggins et al. (1979) took 2.2- cm core samples next to and 20 cm away from tap roots of cotton plants in the field; they found statistically higher rhizosphere collembola densities, and a suggestion that the rhizosphere effect was greater in fertilized than in unfertilized soil. In pots, Wiggins et al. (1979) found an

20

J. LUSSENHOP

increase of 13 collembola for every gram of root mass, and that collembola aggregated around roots as the soil dried. Spatial relationships of microarthropods with roots can be directly observed in rhizotrons, minirhizotrons (Snider et al., 1990), borescopes, and behind glass plates sunk in the soil (Bohm, 1979). Microarthropod groups observed in a rhizotron in a mixed deciduous forest in Michigan were at least two orders of magnitude more dense on roots than in soil (Lussenhop et al., 1991). Microarthropods carry fungal and probably bacterial inoculum to roots. Collembola from Alabama cotton fields carried nine genera of fungi including Aspergillus, Fusarium, Verticillium (Wiggins and Curl, 1979). In laboratory experiments, collembola (Proisotoma minuta and Onychiurus encarpatus) transported fungal spores and bacteria through sterile soil to cotton seedling roots (Wiggins and Curl, 1979). Astigmatid mites (Rhizoglyphus sp.) carried Aspergillus J a w s to peanuts (Aucamp, 1969), and Verticillium alboatrum to bulbs (Price, 1976) and another astigmatid mite species (Caloglyphus micheali) transported Pythium myriotylum to peanuts (Shew and Beute, 1979). In an important and revealing part of their review, Beute and Benson (1979) showed that transport of pathogen inoculum to roots increases disease. Wounding of roots would still further increase disease, although it is not likely that microarthropods eat healthy tissue (Kooistra, 1964). Microarthropod grazing in the rhizosphere has a much more beneficial effect than dispersal of inoculum. This is because pathogenic fungi apparently lack antiherbivore defenses that saprophytic species have, and collembola prefer grazing pathogens (Curl et al., 1983; Lartey et al., 1989). Mankau and Mankau (1963) similarly found that the nematode Aphelenchus avenue had the strongest affect on pathogenic fungi. In petri dish experiments, Curl (1979) showed that both Proisotoma minuta and Onychiurus encarpatus preferred Rhizoctonia solani over Trichoderma harzianum. They ate the latter only when young. In pots containing R . solani-infested soil, presence of P . minuta and 0. encarpatus was associated with emergence of 58-83% more cotton seedlings, depending on collembolan density. Ulber (1983) obtained similar results, and in addition showed in pot experiments that sugar beet survival could be increased by 45% by adding Onychiurus j m a t u s to soil contaminated with Pythium ultimum 20 days before planting. The possibility that microbial-invertebrate interactions in the rhizosphere might contribute to nitrogen mineralization and its uptake by roots was demonstrated in a series of microcosm experiments performed by Coleman and his associates using bacteria, amoebae, nematodes, and blue grama grass (Bouteloua gracilis) (Coleman et al., 1978; Elliot et al., 1979). Mineral nitrogen was released from excreta of bacterivorous nematodes and amoebae, as well as from bacteria due to disturbance by the nematodes and amoebae. When Ingham et al. (1985) reported results of more extensive

MECHANISMS OF MICROARTHROPOD-MICROBIAL INTERACTIONS IN SOIL

21

microcosm experiments, they listed 17 studies associating nitrogen or phosphorus mineralization with amoebal or nematode bacterivory. In the field, Clarholm (1989) estimated that amoebal grazing contributed 1-17% of the nitrogen taken up by barley. Microarthropods have not been shown to increase mineralization in the rhizosphere as have bacterivorous protozoa and nematodes. This may be due to the association of microarthropods with the fungal-based food web that develops along older root segments behind the bacterial-based protozoan, nematode food web (Fig. 3A). Most microarthropods feed on suberized portions of roots, behind the nutrient-absorbing region (Lussenhop, personal observation of mixed deciduous forest tree and herb roots). Further, Wright and Coleman (1988) suggested that rhizosphere fungi are net mineralizers, and that fungivores decreased mineralization in microcosms they studied. Wright and Coleman (1988) used intact cores of field soil, applied factorial combinations of biocides to reduce densities of fungi, nematodes, arthropods, and mesofauna, and then planted Sorghum hicolor in the cores. Neither sorghum nutrient concentration or mass was raised by any invertebrate group, including microarthropods. Setala and Huhta (I99 1) increased mass and nitrogen concentration of birch (Betula pendula) seedlings by adding all groups of soil fauna to microcosms. They did not test microarthropods separately, and it is possible that microarthropods were not responsible for the improved seedling growth.

B. Vesicular-Arbuscular Mycorrhizal-Microarthropod Interactions As fungivores, microarthropods strongly affect all aspects of mycorrhizal growth and functioning, except dispersal. VAM spores and chlamydospores are too large to be dispersed by microarthropods; they are dispersed by wind (Warner et al., 1987), and macroarthropods (Rabatin and Rhodes, 1982). VAM spores, their germination tubes, and extramatrical hyphae are vulnerable to microarthropods. Collembola eat spores of some VAM species. Moore et al. (1985) showed that the collembolan Folsomia candida ate spores of Gigaspora margarita, but not spores of Gigasporafasciculatum or Glomus mosseae, in petri dish feeding trials. Mycorrhizas are established by germ tubes that grow from spores each time soil is moistened, and these germ tubes are susceptible to grazing (Koske, 1981). Grazing of germ tubes may be the reason Kaiser and Lussenhop (1991) found that F. candida reduced the number of infection sites if added to pots when soybeans (Glycine max) were planted, but not if the collembola were added 15 days after planting. Collembolan selectivity in grazing VAM hyphae was shown in Moore et al.’s (1985) petri dish experiments. Four species of collembola (F. candida,

22

J. LUSSENHOP

Onychiurus encarpatus, 0 .folsomi, and Proisotoma minuta) ate Gigaspora rosea hyphae. None ate G . mosseae hyphae, and only F. candida ate Glomus fasciculatum. Collembolan grazing of extramatrical hyphae reduces mycorrhizal benefits to plants, but the effect of collembolan grazing is least at intermediate densities of collembola. Warnock et al. (1982) demonstrated the response of collembolan populations to extramatrical hyphae. They added F. candida to pots with leeks (Allium porrum) and the mycobiont Glomus.fasiculatus. In the presence of collembola, mycorrhizal leeks weighed 50% less and contained 55% less phosphorus than mycorrhizal controls after 12 weeks. Collembola populations increased more in pots with mycorrhizal leeks and more individual collembola were observed with hyphae in their guts in these pots. Finlay (1985) grew leeks in pots with a range of densities of the collembolan Onychiurus ambulans. He found that collembola lowered the beneficial effect of mycorrhizal fungi, but that they had the least effect at intermediate densities. A similar compensatory response to collembola at intermediate densities was also observed by Harris and Boerner (1990) who added F. candida to pots containing Geranium robertianum and the endophyte Glomus fasicula tum . Collembola reduce the benefits of mycorrhizal infection to plants in the field. Finlay (1985) grew Trifolium pratense in field plots using chlorfenvinphos to reduce indigenous collembolan density, and benomyl to reduce infection by the mycobiont Glomus occultus. He found that reduced collembolan density was associated with the highest shoot mass and shoot phosphorus. By sampling four times during the experiment he showed that phosphorus accumulation per shoot mass was highest in treatments where collembolan density was lowest. Similarly, McGonigle and Fitter (1987) found that a two-thirds reduction in collembolan numbers was associated with higher phosphorus concentration in the grass Holcus lanatus. The impact of microarthropods on VAM is likely to be small in highly fertilized agricultural systems. But the literature just reviewed shows that microarthropods decrease benefits of VAM both in natural habitats where VAM may benefit members of plant populations locally or during brief periods (Fitter, 1986), and in low-input agriculture where soil phosphorus levels are low.

C. Ectomycorrhizal-Microarthropod Interactions The rhizosphere of short, ectomycorrhizal roots is controlled by the mycobiont. Ectomycorrhizal short roots are covered by the fungal mantle; they have no epidermis or root hairs, and have a reduced meristematic zone (Fig. 3C). Exudates and sloughed tissue from these short roots are fungal. Protection from herbivory may be very important for ECM fungi because

MECHANISMS OF MICROARTHROPOD-MICROBIAL INTERACTIONS IN SOIL

23

they can be perennial, and food reserves and phosphorus are stored in the mantle. Aphid and nematode feeding on ECM in the field was reviewed by Fogel (1988). There are no observations of microarthropods feeding on ECM in nature. In culture, protura feed on ECM fungi (Sturm, 1959). Shaw (1988) allowed the collembolan Onychiurus armatus to choose between agar plugs of 12 ECM fungi, and found a consistent hierarchy of preference. The least preferred ECM fungi were those with sporocarps toxic to 0. armatus. Other ECM fungi, such as Coenococcum geophilum, may be physically protected by thick-walled, knobby, melanic hyphae. Some ectomycorrhizas form mats of hyphae large enough to alter soil chemistry and biology. For example, hyphal density is 2.5 times greater in mats of Hysterangium setchellii than outside (Cromack et al., 1988) Calcium availability may be especially increased within mats. Many fungi produce oxalic acid as a waste product that forms crystals of calcium oxalate on hyphae and ectomycorrhizal mantles (Malajczuk and Cromack, 1982). Cromack et al. (1977) suggested that calcium oxalate is a source of calcium for soil biota, and that it may be broken down by micro-organisms in guts of arthropods including collembola (Sinella sp.) and oribatids (Pelopoidea sp.). Among microarthropods, calcium is especially important for oribatids; three groups of ptychoid oritabids harden their cuticle with calcium oxalate probably obtained from fungal hyphae (Norton and Behan-Pelletier, 1991). Comparing H . setchellii mat soil with adjacent soil, Cromack et al. (1988) found higher exchangeable calcium, organic nitrogen, and carbon as well as 3.2 times more oribatids. and 2.6 times more collembola.

D. Summary for Rhizosphere In the rhizosphere, microarthropods have their primary effect on microorganisms through dispersal and selective grazing. Microarthropods selectively graze pathogens in the rhizosphere. But they also move pathogens and saprophytes alike to root surfaces. In the future it will be important to know the net effect of these two activities. Beneficial bacteria such as plant-growth promoting rhizobacteria and rhizobia could be vectored to root surfaces as well as pathogens. Additional mechanisms of interaction between microarthropods and microorganisms and/or roots probably exist and would help explain Edwards and Lofty’s (1978) field experiment showing that the presence of arthropods (micro- and macroarthropods) leads to the production of greater root mass. They fumigated soil monoliths from fields cropped to cereals, planted barley in each, and added back natural densities of arthropods (microarthropods plus millipedes, insect larvae, etc.), and earthworms. Both earthworms and arthropods were associated with higher seed germination (relative to controls). In addition, arthropods were associated

24

J. LUSSENHOP

with the largest root mass: 1.2 times that associated with earthworms, 2.3 times fumigated controls, and only 0.8 times the root mass of unfumigated, ploughed soil.

VI. CONCLUSIONS Microarthropods control the distribution and abundance of fungi in soil, and they also stimulate microbial metabolic activity, thereby amplifying microbial immobilization or mineralization of nutrients. It is possible that microarthropods may be important as vectors of entomopathogenic fungi to holometabolous insects. Chemical interactions among the soil biota are probably extensive, yet are poorly known. In soils where fungi dominate there are six mechanisms of interaction with microarthropods. Litter microarthropod species selectively graze and disperse fungi. Deeper in the horizon, the same microarthropod species may stimulate bacterial activity; by grazing fungi they may control species occurrence, cause compensatory growth, and allow increased growth by disrupting competing hyphal networks and adding mineral nutrients in urine and feces. These stimulating effects on microbial growth will affect mineral nutrient concentrations in soil. If periods of plant nutrient uptake are synchronized with microbial immobilization or mobilization of nutrients, plant growth could be affected. The same microarthropod species may move inoculum to roots, and preferentially graze fungal pathogens. In soils dominated by bacterial foodwebs, e.g. agricultural soils, stimulation of bacterial activity and dispersal of bacteria by microarthropods are likely to be important, but there is much less information on microarthropod interactions with bacteria than with fungi. There is also little information on arid soils where prostigmatid mites may be important, or late in succession where oribatid mites may be important. A major obstacle to understanding how microarthropods and microorganisms interact is lack of spatiotemporal information. In the present chapter analogies with aquatic, coprophilous, and wood-decomposer systems were used to gain insight. But analogies are not sufficient for the saprophytic system and inappropriate for the rhizosphere, hence the need for new observation methods including direct observation with borescopes, minirhizotrons, and rhizotrons. None of the mechanisms reviewed is well quantified. In the future the effects of these mechanisms should be incorporated into regression and simulation models of soil microbial processes. Anderson er al. (1985) used temperature and arthropod density to predict nitrogen mineralization rate. Regression models might incorporate the mechanisms discussed in the present chapter to relate microarthropod density in the rhizosphere to the

MECHANISMS OF MICROARTHROPOD-MICROBIAL INTERACTIONSI N SOIL

25

number of rhizoplane pathogen colonies, or to VAM benefits to plants in the field. Simulation models of soil foodwebs might include responses to microarthropods. The simulation model of Hunt et a/. (1987) includes density-dependent control of microbial growth, thus roughly incorporating the retarding effect of fungal species number on decomposition. But the model omits selective grazing, dispersal of propagules, stimulation of bacteria by grazing and, compensatory growth. It is to be hoped that mechanisms just reviewed are quantified and included in future simulation models.

ACKNOWLEDGEMENTS I am most grateful to D.T. Wicklow, R. Fogel, and R.M. Miller for many years of stimulating interactions and for reviewing the manuscript. The advice and review by V. Behan-Pelletier, the review by H. A. Verhoef, and G. Bengtsson’s permission to cite unpublished research are much appreciated. I thank Helen Badawi and Gladys Odegaard of the UIC Science Library for their help.

REFERENCES Addison, J.A. and Parkinson, D. (1978). Influence of collembolan feeding activities on soil metabolism at a high arctic site. Oikos 30, 529-538. Amelsvoort, P.A.M.van and Usher, M.B. (1989). Egg production related to food quality in Folsomia candida (Collembola: Isotomidae): effects on life history strategies. Pedobiologia 33, 61-66. Anas, 0.and Reeleder, R.D. (1987). Recovery of fungi and arthropods from sclerotia of Sclerotinia sclerotiorum in Quebec muck soils. Phytopathology 77, 327-33 1. Anderson, J.M. ( 1 988). Spatiotemporal effects of invertebrates on soil processes. Biof. Fertil. Soils 6, 2 1 6 2 2 1 . Anderson, J.M. and Bignell, D.E. (1980). Bacteria in the food, gut contents, and faeces of the litter-feeding millipede Golrneris marginata (Villers). Soil Biol. Biochem. 12, 251-254. Anderson, J.M., Leonard, M.A., Ineson, P. and Huish, S. (1985). Faunal biomass: a key component of a general model of nitrogen mineralization. Soil Biol. Biochem. 17, 735-731. Anderson, J.P.E. and Domsch, K.H. (1975). Measurement of bacterial and fungal contributions to respiration of selected agricultural and forest soils. Can. J . Microbiof. 21, 314-322. Andren, 0. and Schnurer, J. (1985). Barley straw decomposition with varied levels of microbial grazing by Folsomiajmetaria (L.) (Collembola, Isotomidae). Oecologia 68, 57-62.

26

J. LUSSENHOP

Arsuffi, T.L. and Suberkropp, K. (1985). Selective feeding by stream caddis fly (Trichoptera) detritivores on leaves with fungal-colonized patches. Oikos 45, 5058. Aucamp, J.L. (1969). The role of mite vectors in the development of aflatoxin on groundnuts. J. Stored Prod. Res. 5, 245-249. Bakonyi, G. (1989). Effects of Folsomia candida (Collembola) on the microbial biomass in a grassland soil. Biol. Fertil. Soils. 7, 138-141. Bandoni, R.J. and Koske, R.E. (1974). Monolayers and microbial dispersal. Science 183, 1079-1081. Barlocher, F. (1980). Leaf-eating invertebrates as competitors of aquatic hyphomycetes. Oecologia 47, 303-306. Behan, V.M. and Hill, S.B. (1978). Feeding habits and spore dispersal of Oribatid mites in the North American arctic. Rev. Ecol. Biol. Sol 15, 497-516. Belsky, A.J. (1986). Does herbivory benefit plants? A review of the evidence. Am. Nut. 127, 870-892. Bengtsson, G. and Rundgren, S. (1983). Respiration and growth of a fungus, Mortierella isabellina, in response to grazing by Onychiurus armatus (Collembola). Soil Biol. Biochem. 15, 469-473. Bengtsson, G., Erlandsson, A. and Rundgren, S. (1988). Fungal odour attracts soil collembola. Soil Biol. Biochem. 20, 2530. Beute, M.K. and Benson, D.M. (1979). Relation of small soil fauna to plant disease. Ann. Rev. Phytopathol. 17, 485-502. Binns, E.S. (1980). Field and laboratory observations on the substrates of the mushroom fungus gnat Lycorriella auripila (Diptera:Sciaridae). Ann. Appl. Biol. 96, 143-152. Blackwell, M. (1984). Myxomycetes and their arthropod associates. In: Fungus-Insect Relationships (Ed. by Q. Wheeler and M. Blackwell), pp. 67-90. Columbia University Press, New York. Boddy, L., Coates, D. and Rayner, A.D.M. (1983). Attraction of fungus gnats to zones of intraspecific antagonism on agar plates. Trans. Br. Mycol. SOC.81, 149151. Bohm, W. (1979). Methoh of Studying Root Systems. Springer-Verlag, Berlin. Booth, R.G. and Anderson, J.M. (1979). The influence of fungal food quality on the growth and fecundity of Folsomia candida. Oecologia 38, 317-323. Borkott, H. and Insam, H. (1990). Symbiosis with bacteria enhances the use of chitin by the springtail, Folsomia candida (Collembola). Biol. Fertil. Soils 9, 126129. Cancela da Fonseca, J.P. (1980). Quelques approches analytiques en dynamique de populations e’daphiques (Acariens Oribates). Rev. Ecol. Biol. Sol. 17, 445-477. Cervek, S. (1971). Fertility of conidiospores of Penicillium sp. in the excrements of Entomobrya purpurascens (Packard) Collembola Entomobryidae. In: Organisms du Sol et Production Primaire (Proc. IV Int. CoII. Soil Zool.) Ann. Zoo/.-Ecol. anim. 71, 69-72. Charnley, A.K. (1984). Physiological aspects of destructive pathogenesis in insects by fungi: a speculative review. In: Invertebrate-Microbial Interactions (Ed. by J.M. Anderson, A.D.M. Rayner and D.W.H. Walton), pp. 229-270. Cambridge University Press, Cambridge. Clarholm, M. (1989). Effects of plant-bacterial-amoeba1 interactions on plant uptake of nitrogen under field conditions. Biol. Fertil. Soils 8, 373-378. Coates, D. and Rayner, A.D.M. (1985). Fungal population and community development in cut beech logs. 111. Spatial dynamics, interactions and strategies. New Phytol. 101, 183-198.

MECHANISMS OF MICROARTHROPOD-MICROBIAL

INTERACTIONS IN SOIL

27

Coleman, D.C., Cole, C.V., Hunt, H.W. and Klein, D.A. (1978). Trophic interactions in soils as they affect energy and nutrient dynamics. I. Introduction. Microb. Ecol. 4,345349. Coleman, D.C., Reid, C.P.P. and Cole, C.V. (1983). Biological strategies of nutrient cycling in soil systems. in: Advances in Ecological Research (Ed. by A. MacFadyen and E.D. Ford), pp. 1-55. Academic Press, New York. Cooke, R.C. and Rayner, A.D.M. (1984). Ecology of Saprotrophic Fungi. Longman, London. Cromack, K., Jr, Sollins, P., Todd, R.L. et al. (1977). Soil microorganism-arthropod interactions: fungi as major calcium and sodium sources. In: The Role of Arthropods in Forest Ecosystems (Ed. by W.J. Mattson), pp. 78-84. SpringerVerlag, New York. Cromack, K. Jr, Fichter, B.L., Moldenke, A.M., Entry, J.A. and Ingham, E.R. (1988). Interactions between soil animals and ectomycorrhizal fungal mats. Agr. Ecos. Environ. 24, 161-168. Curl, E.A. (1979). Effects of mycophagous collembola on Rhizoctonia solani and cotton-seedling disease. In: Soil-Borne Plant Pathogens (Ed. by B. Schippers and W. Gams), pp. 253-269. Academic Press, London. Curl, E.A. and Truelove, B. (1986). The Rhizosphere. Springer-Verlag, Berlin. Curl, E.A., Gudauskas, R.T., Harper, J.D. and Peterson, C.M. (1983). Effects of soil insects on populations and germination of fungal propagules. In: Ecology and Management of Soilborne Plant Pathogens (Ed. by C.A. Parker, A.D. Rovira, K.J. Moore and P.T.W. Wong), pp. 2&23. American Phytopathological Society, St Paul, MN. Curry, J.P. and Ganley, J. (1977). The arthropods associated with the roots of some common grass and weed species of pasture. Ecol. Bull. 25, 330-339. Dindal, D.L. (1990). Soil Biology Guide. Wiley, New York. Dinsdale, D. (1974a). The digestive activity of a phthiracarid mesenteron. J. Insect Physiol. 20, 2247-2260. Dinsdale, D. (1974b). Feeding activity of a phthiracarid mite (Arachnida:Acari). J. Zool. 174, 15-2 1 . Domsch, K.H., Gams, W. and Anderson, T.H. (1980). Compendium of Soil Fungi. Academic Press, London. Vol. 1, 859 pp., Vol. 2, 405 pp. Dowson, C.G., Rayner, A.D.M. and Boddy, L. (1988). Foraging patterns of Phallus impudicus, Phanerochaete laevis, and Steccherinum jimbriatum between discontinuous resource units in soil. FEMS Microbiol. Ecol. 53, 291-298. Drift, J. van der. (1965). The effects of animal activity in the litter layer. In: Experimental Pedology (Ed. by E.G. Hallsworth and D.V. Crawford), pp. 227235. Butterworths, London. Drift, J. van der and Jansen, E. (1977). Grazing of springtails on hyphal mats and its influence on fungal growth and respiration. In: Soil Organisms as Components of Ecosystems (Ed. by U. Lohm and T. Persson), Ecol. Bull. (Stockholm) 25, 203209. Edwards, C.A. (1967). Relationships between weights, volumes and numbers of soil animals. In: Progress in Soil Biology (Ed. by 0.Graff and J.E. Satchell), pp. 5 8 s 593. Friedrich, Braunschweig. Edwards, C. A. and Lofty, R. (1978). The influence of arthropods and earthworms upon root growth of direct drilled cereals. J. appl. Ecol. 15,789-795. Elliott, E.T., Coleman, D.C. and Cole, C.V. (1979). The influence of amoebae on uptake of nitrogen by plants in gontobiotic soil. In: The Soil-Root Interface (Ed. by L. Harley and R.S. Russell), pp. 221-229. Academic Press, London.

28

J . LUSSENHOP

Evans, H.E. (1988). Coevolution of entomogenous fungi and their insect hosts. In: Coevolution of’Futigi with Plants and Animals (Ed. by K.A. Pirozynski and D.L. Hawksworth), pp. 149-1 7 1. Academic Press, London. Finlay, R.D. (1985). Interactions between soil microarthropods and endomycorrhizal associations of higher plants. In: Ecological Interactions in Soil (Ed. by A.H. Fitter), pp. 319-331. Blackwell, Oxford. Fitter, A. H. (1986). Effect of benomyl on leaf phosphorus concentration in alpine grasslands: a test of mycorrhizal benefit. New Phytol. 103, 767-776. Fogel, R. (1988). Interactions among soil biota in coniferous ecosystems. Agr. Ecos. Environ. 24, 69-85. Fogel, R. (1991). Root system demography and production in forest ecosystems. In: Plant Root Growth: An Ecological Perspective (Ed. by D. Atkinson), pp. 89-101. Blackwell, Oxford. Gochenaur, S.E. (1987). Evidence suggests that grazing regulates ascospore density in soil. Mycologia 79, 445450. Gregoire-Wibo, C. and Snider, R.M. (1977). The intrinsic rate of natural increase: its interest to ecology and its application to various species of collembola. In: Soil Organisms as Components of Ecosystems (Ed. by U. Lohm and T. Person), Ecol. Bull. (Stockholm) 25, 4 4 2 4 8 . Grossbard, E. (1969). A visual record of the decomposition of I4Clabelled fragments of grasses and rye added to soil. J . Soil Sci. 20, 38-51, Gunnarsson, T. (1987). Selective feeding on a maple leaf by Oniscus asellus (Isopoda). Pedobiologia 30, 1 6 1- 165. Hanlon, R.D.G. (1981a). Influence of grazing by collembola on the activity of senescent fungal colonies grown on media of different nutrient concentrations. Oikos 36, 362-367. Hanlon, R.D.G. (1981 b). Some factors influencing microbial growth on soil animal faeces. 11. Bacterial and fungal growth on soil animal faeces. Pedobiofogia 21,264270. Hanlon, R.D.G. and Anderson, J.M. (1979). The effects of collembola grazing on microbial activity in decomposing leaf litter. Oecologia 38, 93-99. Haq, M.A. and Konikkara, I.D. (1988). Microbial associations in xylophagous oribatids. In: Progress in Acarology, Vol. 1 (Ed. by G. P. Channabasavanna and C. A. Viraktamath), pp. 469474. Oxford and IBH Pub. Co., New Delhi. Harasymek, L. and Sinha, R.N. (1974). Survival of springtails Hypogastrura tullbergi and Proisotoma minuta on fungal and bacterial diets. Environ. Entomol. 3, 965968. Harris, K.K. and Boerner, R.E.J. (1990). Effects of belowground grazing by collembola on growth, mycorrhizal infection, and P uptake of Geranium robertianum. Plant and Soil 129, 203-210. Hedlund, K., Boddy, L. and Preston, C.M. (1991). Mycelial responses of the soil fungus Mortierella isabellina to grazing by Onychiurus armatus (Collembola). Soil Biol. Biochem. 361-366. Hendrix, P.F., Parmelee, R.W., Crossley, D.A.J., Coleman, D.C., Odum, E.P. and Groffman, P.M. (1986). Detritus foodwebs in conventional and nontillage agroecosystems. Bioscience 36, 374380. Hepple, S. (1960). The movement of fungal spores in soil. Trans. Br. Mycol. Soc. 43, 73-79. Hunt, H.W., Coleman, D.C., Ingham, E.R. et al. (1987) The detrital food web in a shortgrass prairie. Biol. Fertil. Soils 3, 5 7 4 8 . Ingham, E.R., Coleman, D.C. and Moore, J.C. (1989). An analysis of food-web

MECHANISMS OF MICROARTHROPOD-MICROBIAL INTERACTIONS IN SOIL

29

structure and function in a shortgrass prairie, a mountain meadow, and a lodgepole pine forest. Biol. Fertil. Soils 8, 29-37. Ingham, R.E., Trofymow, J.A., Ingham, E.R. and Coleman, D.C. (1985). Interactions of bacteria, fungi, and their nematode grazers: effects on nutrient cycling and plant growth. Ecol. Monogr. 55, 119-140. Jenkinson, D.S. and Ladd, J.N. (1981). Microbial biomass in soil: measurement and turnover. In: Soil Biochemistry, Vol. 5 . (Ed. by E. A. Paul and J. N. Ladd), pp. 415471. Dekker, New York. Kaiser, P.A. and Lussenhop, J. (1991). Collembolan effects on establishment of vesicular-arbuscular mycorrhizae in Soybean (Glycine max). Soil Biol. Biochem. 23, 307-308. Keller, S. and Zimmerman, G. (1989). Mycopathogens of soil insects. In: InsectFungus Interactions (Ed. by N. Wilding, N.M. Collins, P.M. Hammond and J.F. Webber), pp. 239-270. Academic Press, New York. Kessler, K.J.J. (1990). Destruction of Gnomonia leptostylu perithecia on Juglans nigra leaves by microarthropods associated with Elueagnus umbellata litter. Mycologia 82, 387-390. Kevan, P.G., Chaloner, W.G. and Savile, D.B.O. (1975). Interrelationships of early terrestrial arthropods and plants. Palaeontology 18, 391417. Kilbertus, G. and Vannier, G. ( 1 981). Relations microflore-microfaune dans la grotte de Sainte-Catherine (Pyrenees arigeoises). 11. Le regime alimentaire de Tomocerus minor (Lubbock) et Tomocerus problematicus Cassagnau (Insectes Collemboles). Rev. Ecol. Biol. Sol. 18, 319-338. Koske, R.E. (1981). Multiple germination by spores of Gigasporugigantea. Trans. Br. Mycol. SOC.76, 328-330. Kooistra, G. (1964). Some data concerning the presence and behavior of springtails (Collembola) on grass and white clover. Neth. J . Plant Path. 70, 136141. Kuwahara, Y., Leal, W.S. and Suzuki, T. (1989). Antifungal activity of Culoglyphus polyphyllae sex pheromone and other mite exudates. Naturwissenschuften 76, 578579. Lartey, R.T., Curl, E.A., Peterson, C.M. and Harper, J.D. (1989). Mycophagous grazing and food preference of Proisotornu minuta (Collembo1a:Isotomidae) and Onychiurus encarpatus (Collembola: Onychiuridae). Env. Entomol. 18, 334-337. Leonard, M.A. (1984). Observations on the influence of culture conditions of the fungal feeding preferences of Folsomia candida (Collembola: Isotomidae). Pedobiologia 26, 361-367. Lussenhop, J. and Wicklow, D.T. (1984). Changes in spatial distribution of fungal propagules associated with invertebrate activity in soil. Soil Biol. Biochem. 16, 601404. Lussenhop, J. and Wicklow, D.T. (1985). Interaction of competing fungi with fly larvae. Microb. Ecol. 11, 175-182. Lussenhop, J., Kumar, R., Wicklow, D.T. and Lloyd, J. (1980). Insect effects on bacteria and fungi in cattle dung. Oikos 34, 5&58. Lussenhop, J., Fogel, R. and Pregitzer, K. (1991). A new dawn for soil biology: video analysis of root-soil-microbial-faunal interactions. In: Internutional Workshop on Modern Techniques in Soil Ecology (Ed. by D.A. Crossley, Jr), pp. 235-249. Elsevier, Amsterdam. Macfadyen, A. (1961). Metabolism of soil invertebrates in relation to soil fertility. Ann. appl. Biol. 49, 215-218. McGonigle, T.P. and Fitter, A.H. (1987). Evidence that collembola suppress plant

30

J . LUSSENHOP

benefit from vesicular-abruscular mycorrhizae (VAM) in the field. Proc. 8th North American Conf. on Mycorrhizae, p. 209. Gainesville, FL. Malajczuk, N. and Cromack, K. (1982). Accumulation of calcium-oxalate in the mantle of ectomycorrhizal roots of Pinus radiata and Eucalyptus marginata. New Phytol. 92, 527-53 1. Mankau, R. and Mankau, S.K. (1963). The role of mycophagous nematodes in the soil. 1. The relationships of Aphelenchus avenue to phytopathogenic soil fungi. In: Soil Organisms (Ed. by J. Doeksen and J. van der Drift), pp. 271-280. NorthHolland, Amsterdam. Manton, S.M. (1977). The Arthropoda. Clarendon, Oxford. Martin, M.M. (1987). Invertebrate-Microbial Interactions. Comstock Publishing Associates, Ithaca, NY. Moore, J.C., St John, T.V. and Coleman, D.C. (1985). Ingestion of vesiculararbuscular mycorrhizal hyphae and spores by soil microarthropods. Ecology 66, 1979-198 1. Moore, J.C., Ingham, E.R. and Coleman, D.C. (1987). Inter- and intraspecific feeding selectivity of Folsomia candida on fungi. Biol. Fertil. Soils 5, 6-12. Newell, K. (1984a). Interaction between two decomposer basidiomycetes and a collembolan under Sitka spruce: distribution, abundance, and selective grazing. Soil Biol. Biochem. 16, 227-233. Newell, K. (1984b). Interaction between two decomposer basidiomycetes and a collembolan under Sitka spruce: grazing and its potential effects on fungal distribution and litter decomposition. Soil Biol. Biochem. 16, 235-239. Nicholson, P.B., Bocock, K.L. and Heal, O.W. (1966). Studies of the decomposition of the facal pellets of a millipede (Glomeris marginata Villers). J . Ecol. 54,755-766. Norman, A.G. (1930). The biological decomposition of plant materials. 111. Physiological studies on some cellulose decomposing fungi. Ann. Appl. Biol. 17, 575-613. Norton, R. A. (1992). Evolutionary aspects of oribatid mite life histories and consequences for the origin of the astigmata. In: Life History and Reproductive Patterns in Mites (Ed. by M. A. Houck). Chapman and Hall, New York (in press). Norton, R.A. and Behan-Pelletier, V.M. (1991). Calcium carbonate and calcium oxalate as cuticular hardening agents in oribatid mites (Acari: Oribatida). Can. J . Zool. 69, 1 5 0 4 15 1 1. Parkinson, D., Visser, S . and Whittaker, J.B. (1979). Effects of collembolan grazing on fungal colonization of leaf litter. Soil Biol. Biochem. 11, 529-535. Peters, R.H. (1983). The Ecological Implications of Body Size. Cambridge University Press, Cambridge. Peterson, H. (1982). Structure and size of soil animal populations. Oikos 39,306-329. Pherson, D.A. (1980). Ecology of ArthropodlFungal Interactions in Decomposing Beech Leaf Litter. PhD thesis. Northwestern University. Pherson, D.A. and Beattie, A.J. (1979). Fungal loads of invertebrates in beech leaf litter. Rev. Ecol. Biol. Sol. 16, 325-335. Pirozynski, K.A. and Malloch, D.W. (1975). The origin of land plants; a matter of mycotrophism. BioSystems 6, 153-164. Ponge, J.F. and Charpentie, M.J. (1981). Etude des relations microfloremicrofaune: experiences sur Pseudosinella alba (Packard), Collembola mycophage. Rev. Ecol. Biol. Sol. 18, 291-303. Poole, T.B. (1964). A study of the distribution of soil collembola in three areas in a coniferous woodland. Pedobiologia 4, 3 5 4 2 . Price, D.W. (1976). Passage of Verticillium albo-atrum propagules through the alimentary canal of the bulb mite. Phytopathology 66, 46-50. Purrini, K . (1983). Comparison of pathogenic agents in collembola (Insecta, Aptery-

MECHANISMS OF MICROARTHROPOD-MICROBIAL INTERACTIONS IN SOIL

31

gota) from different forests in the Federal Republic of Germany, Austria, and Spain. Pedobiologia 25, 365-371. Purrini, K. and Bukva, V. (1984). Pathogenic agents of oribatid mites (Oribatei, Acarina): a gap in the research on population dynamics. In: Acarology VI, Vol. 2 (Ed. by D.A. Griffiths and C.E. Bowman), pp. 826837. Halsted Press, New York. Rabatin, S.C. and Rhodes, L.H. (1982). Acaulospora bireticulata inside oribatid mites. Mycologia 74, 859-86 1. Rapoport, E.H. and Aguirre, Y. (1973). Population analysis of Onychiurus yolandae, a parthenogetic collembolan insect with notes on possible prey tactics. Rev. Ecol. Biol. Sol. 10:341-358. Rawlins, J.E. (1984). Mycophagy in Lepidoptera. In: Insect-Fungus Relationships (Ed. by Q . Wheeler and M. Blackwell), pp. 382423. Columbia, New York. Rayner, A.D.M. and Boddy, L. (1988). Fungal communities in the decay of wood. Adv. Microb. Ecol. 10, 1 15-1 66. Reyes, V.G. and Tiedje, J.M. (1976). Ecology of the gut microbiota of Tracheoniscus rathkei (Crustacea, Isopoda). Pedobiologia 16, 67-74. Rusek, J. (1975). Die bodenbildende Funktion von Collembolen und Acarina. Pedobiologia 15, 299-308. Seastedt, T.R. (1984). The role of microarthropods in decomposition and mineralization processes. Ann. Rev. Entomol. 29, 2 5 4 6 . Seniczak, S. and Stefaniak, 0. (1978). The microflora of the alimentary canal of Oppia nitens (Acarina, Oribatei). Pedobiologia 18, 110-1 19. Setala, H. and Huhta, V. (1991). Soil fauna increase Betula pendula growth: laboratory experiments with coniferous forest floor. Ecology 72, 665-671. Shaw, P.J.A. (1988). A consistent hierarchy in the fungal feeding preferences of the collembola Onychiurus armatus. Pedobiologia 31, 179-1 88. Shear, W.A., Bonamo, P.M., Grierson, J.D., Rolfe, W.D.I., Smith, E.L. and Northon, R.A. (1984). Early land animals in North America: evidence from Devonian Age arthropods from Gilboa, New York. Science 224, 492494. Sherwood-Pike, N.A. and Gray, J. (1985). Silurian fungal remains: probable records of the class Ascomycetes. Lethaia 18, 1-20. Shew, H.D. and Beute, M.K. (1979). Evidence for the involvement of soilborne mites in Pythium pod rot of peanut. Phytopathology 69, 204-207. Snider, R.J., Snider, R. and Smucker, A.J.M. (1990). Collembolan populations and root dynamics in Michigan agroecosystems. In: Rhizoplane Dynamics (Ed. by J.E. Box and L.C. Hammond), pp. 168-191. Westview, Boulder, CO. Stefaniak, 0. and Seniczak, S. (1981). The effect of fungal diet on the development of Oppiu nitens (Acari, Oribatei) and on the microflora of its alimentary tract. Pedobiologia 21, 202-2 10. Streit, B. and Reutimann, P. (1983). Laboratory studies on feeding ecology of an oribatid mite from an alpine meadow ecosystem of Swiss National Park. In: New Trends in Soil Biology (Ed. by Ph. Lebrun, H.M. Andre, A. De Medts, C. GregoireWibo and G. Wauthy), pp. 614-61 5. Dieu-Brichart, Ottignies-Louvain-la-Neuve. Stubblefield, S.P. and Tayior, T.N. (1988). Recent advances in palaeomycology. New Phytol. 108, 3-25. Stubblefield, S.P., Taylor, T.N. and Trappe, J.M. (1987). Fossil mycorrhizae: a case for symbiosis. Science 237, 59-60. Sturm, H. (1959). Die nahrung der Proturan. Die Naturwissenschaften 46, 9 6 9 1 . Swain, T. and Cooper-Driver, G. (1981). Biochemical evolution in early land plants. In: Paleobotany, Paleoecology, and Evolution (Ed. by K.J. Niklas), pp. 103-134. Praeger, New York. Teuben, A. and Roelofsma, T.A.P.J. (1990). Dynamic interactions between func-

32

J . LUSSENHOP

tional groups of soil arthropods and microorganisms during decomposition of coniferous litter in microcosm experiments. Biol. Fertil. Soils 9, 145-151. Tisdall, J.M. and Oades, J.M. (1982). Organic matter and water-stable aggregates in soils. J . Soil Sci. 33, 141-163. Touchot, F., Kilbertus, G. and Vannier, G. (1983). R61e d’un collembole (Folsomia candida) au cours de la degradation des litieres de charme et de c h h e , en presence ou en absence d’argile. In: New Trends in Soil Biology (Ed. by Ph. Lebrun, H.M. Andre, A. De Medts, C. Grkgoire-Wibo and G. Wauthy), pp. 269-280. DieuBrichart, Ottignies-Louvain-la-Neuve. Ulber, B. (1983). Einfluss von Onychiurusfimatus Gisin (Collembola, Onychiuridae) und Folsomiafimetaria (L.) (Collembola, Isotomidae) auf Pythium ultimum Trow., einen Erreger des Wurzelbrandes der Zuckerube. In: New Trends in Soil Biology (Ed. by Ph. Lebrun, H.M. Andre, A. De Medts, C. GrCgoire-Wibo and G. Wauthy), pp. 261-268. Dieu-Brichart, Ottignies-Louvain-la-Neuve. Verhoef, H.A., Prast, J.E. and Verweij, R. A. (1988). Relative importance of fungi and algae in the diet and nitrogen nutrition of Orchesella cincta (L.) and Tomocerus minor (Lubbock) (Collembola). Funct. Ecol. 2, 195-201. Verhoef, H.A., Dorel, F. G. and Zoomer, H. R. (1989). Effects of nitrogen deposition on animal-mediated nitrogen mobilization in coniferous litter. Biol. Fertil. Soils 8, 255-259. Visser, S. (1985). Role of the soil invertebrates in determining the composition of soil microbial communities. In: Ecological Interactions in Soil(Ed. by A. H. Fitter), pp. 297-31 7. Blackwell, Oxford. Visser, S., Whittaker, J.B. and Parkinson, D. (1981). Effects of collembolan grazing on nutrient release and respiration of a leaf litter inhabiting fungus. Soil Biol. Biochem. 13, 215-218. Warner, N.J., Allen, M.F. and MacMahon, J.H. (1987). Dispersal agents of vesicular-arbuscular mycorrhizal fungi in a disturbed arid ecosystem. Mycologia 79, 721-730. Warnock, A.J., Fitter, A.H. and Usher, M.B. (1982). The influence of a springtail Folsomia candida (Insecta, Collembola) on the mycorrhizal association of Leek AIlium porrum and the vesicular-arbuscular mycorrhizal endophyte Glomus fasciculatus. New Phytol. 90,285-292. Webb, D.P. (1977). Regulation of deciduous forest litter decomposition by soil arthropod faeces. In: The Role of Arthropods in Forest Ecosystems (Ed. by D.A.Crossely), pp. 57-69. Springer-Verlag, New York. Whitford, W.G. (1989). Abiotic controls on the functional structure of soil food webs. Biol. Fertil. Soils 8, 1 4 . Whittaker, J.B. (1981). Feeding of Onychiurus subtenuis (Collembola) at snow melt in aspen litter in the Canadian Rocky Mountains. Oikos 36, 203-206. Wicklow, D.T. (1979). Hair ornamentation and predator defence in Chaetomium. Trans. Br. Mycol. SOC.72, 107-1 10. Wicklow, D.T. (1988). Metabolites in the coevolution of fungal chemical defence systems. In: Coevolution of Fungi with Plants and Animals (Ed. by K.A. Pirozynski and D.L. Hawksworth), pp. 173-201. Academic Press, London. Wicklow, D.T. and Carroll, G.C. (1981). The Fungal Community. Decker, New York. Wicklow, D.T. and Yocom, D.H. (1981). Fungal species numbers and decomposition of rabbit faeces. Trans. Br. Mycol. SOC.76, 29-32. Wicklow, D.T. and Yocom, D.H. (1982). Effect of larval grazing by Lycoriella mali (Diptera: Sciaridae) on the species abundance of coprophilus fungi. Trans. Br. M Y C O SOC. ~ . 78, 29-32.

MECHANISMS OF MICROARTHROPOD-MICROBIAL

INTERACTIONS IN SOIL

33

Wiggins, E.A. and Curl, E.A. (1979). Interactions of collembola and microflora of cotton rhizosphere. Phytopathology 69, 244-249. Wiggins, E.A., Curl, E.A. and Harper, J.D. (1979). Effects of soil fertility and cotton rhizosphere on populations of Collembola. Pedobiologia 19, 75-82. Wright, D.H. and Coleman, D.C. (1988). Soil faunal vs. fertilization effects on plant nutrition: results of a biocide experiment. Biol. Fertil. Soils 7 , 4 6 5 2 . Zimmermann, G . and Bode, E. (1983). Untersuchungen zur Verbreitung des insektenpathogenen Pilzes Metarhizium anisopliue (Fungi imperfecti, Moniliales) durch Bodenarthropoden. Pedobiologia 25, 65-7 1.

This Page Intentionally Left Blank

Relative Risks of Microbial Rot For Fleshy Fruits: Significance with Respect to Dispersal and Selection for Secondary Defense MARTIN L . CIPOLLINI* and EDMUND W . STILES

I . Summary . . . . . . . . . . . . . . . . . I1 . Introduction . . . . . . . . . . . . . . . . A . Questions and Objectives . . . . . . . . . . B . Variations in Characteristics of Fleshy Fruits . . . C . Interspecific Variation in Secondary Defense Chemistry 111. Fruit Rot and Effects on Dispersal . . . . . . . . . A . Influence of Fruit Rot on Dispersal . . . . . . . B . Factors that Affect Risk of Microbial Rot . . . . . C . Natural Selection for Fruit Defenses . . . . . . IV . General Hypotheses and Predictions . . . . . . . . A . General Deterrent Nature of Fruit Rot . . . . . . B . Microbe-specific Defenses . . . . . . . . . . C . Interspecific Variation in Defense Effectiveness . . . V. Predictions for Temperate Seed Dispersal Systems . . . A . Temperate Fruiting Classes . . . . . . . . . . B . Predictions for Temperature Species . . . . . . . VI . Conclusions . . . . . . . . . . . . . . . . Acknowledgements . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . . .

. . . .

. . . .

. . . . . .

. . .

. . . . . .

. . .

35 31 37 37 38 39 39 43 48 56 56 56 57 58 58 60 67 68 68

.

I SUMMARY Secondary plant chemicals are commonly thought to have evolved as feeding deterrents for herbivores. attractants for pollinators and seed-dispersal agents. and inhibitors of pathogens. and much evidence exists for such roles in various tissues of plants . Although a few hypotheses have been generated concerning ecological roles of secondary chemicals (other than pigments) in fleshy vertebrate-dispersed fruits. few empirical data refine these hypotheses

* Smithsonian Environmental Research Center. P.O. Box 28. Edgewater. MD 21037. ADVANCES IN ECOLOGICAL RESEARCH VOL. 23 lSBNCkl24l3923-5

Copyrighi 0 I992 Academic Press Limried A// righis of reproduction in an.v/orm reserved

35

36

M . L. CIPOLLINI AND E. W . STILES

or test their predictions. To date, virtually all data concerning secondary chemicals of fleshy fruits have come from studies of highly-selected cultivated species. In these species, secondary chemicals present in high concentrations in immature fruits diminish considerably during ripening, but patterns for wild species remain practically unexplored. Because wild plants bearing fleshy fruits benefit from the consumption of fruit by vertebrate seed-dispersal agents, but presumably d o not benefit from consumption by other organisms, an evolutionary conflict seems evident with respect to attraction of dispersers and defense against non-dispersers. Selection for specific secondary chemical patterns in ripe fleshy fruits may result from the need to provide palatable and non-toxic pulp for dispersers, while retaining defense against various non-disperser “frugivores”, including seed predators and microbial fruit-rot agents. Here we examine the specific case of fleshy fruits, their vertebrate dispersers, and fruit-rot fungi, and review the parameters necessary to evaluate selection pressures for secondary chemical defense. We arrive at three general hypotheses: (i) In addition to causing early drop, fruit rotted by fungi should be generally deterrent to frugivores, and thus antifungal defenses should be maintained in ripe fruit. However, considerable interspecific variation may exist in the effects of fungi upon dispersal, and thus fruit defense may vary considerably with respect to fungal species. (ii) Microbe-specific chemical agents, with little or no negative effects on frugivores, should be common for plants under strong selection to provide nutritious or otherwise palatable fruits as a means of attracting frugivores. (iii) The degree of antifungal activity present in ripe fruit may vary among plant species, dependent upon selection pressure for persistence. Within this latter hypothesis, we present two alternative models: (a) The removal-rate model, which states that fruit defenses should be low for plant species whose fruits are generally removed rapidly upon maturation, and (b) the relative-risk model, which states that fruit defenses should be allocated in proportion to the risk of microbial degradation resulting from other intrinsic and extrinsic variables, including time of ripening, ripening synchrony, pulp nutrient content and physical design, and environmental factors influencing patterns of microbial colonization. We define more specific predictions for temperate vertebrate-dispersed species of eastern North America, based upon our own work with Ericaceous species. We suggest that, despite evidence that plant-frugivorefungus interactions are generally complex (i.e. “diffuse”) in nature, broadscale patterns of ripe-fruit defense chemistry may reflect selective pressures

RELATIVE RISKS OF MICROBIAL ROT FOR FLESHY FRUITS

37

relating to seed dispersal. The hypotheses and predictions generated in this chapter provide a focus for elucidating the evolutionary significance of secondary chemicals of ripe fleshy fruits, and their potential effects in mediating plant-frugivore-fungus interactions.

11. INTRODUCTION A. Questions and Objectives This chapter concerns the defense of ripe fruit from microbial fruit-rot agents, and the degree to which characteristics relating to seed dispersal can be used to predict patterns of secondary chemical defense. We concentrate specifically on vertebrate-dispersed fruits of the temperate United States, and potential defense-attraction conflicts arising from the retention of defenses in ripe fruits. Few data are available on the importance of chemical defenses in increasing fruit persistence, or on the effects of these defenses on frugivorous animals, and only limited data exist concerning evolutionary patterns of plant-animal-microbe relationships for wild plants (Batra and Batra, 1985; Clay, 1988a, b; Pirozynski and Hawksworth, 1988; Barbosa et al., 1991). We first present a comprehensive review of the literature concerning fungal fruit rot as a factor in seed dispersal and as a selective pressure for fruit defense. We then use this information to generate general predictions concerning patterns of antifungal defense in ripe fruit with respect to fruit dispersal characteristics. These hypotheses and predictions are based primarily upon the “optimal defense” hypothesis that defenses are costly and should be allocated in direct relationship to fitness benefits accrued for particular plants and plant tissues (Rhoades, 1979, 1985).

B. Variation in Characteristics of Fleshy Fruits Following Ridley’s (1930) compendium on seed dispersal mechanisms, numerous researchers have recorded extensive variation in chemical and physical characteristics of fleshy fruits of vertebrate-dispersed plant species. Fruits vary in nutrient content, physical structure, color, size, seed number and size, seedlpulp ratio, water content, season of ripening, and ripening phenology (cf. McAtee, 1947; van der Pijl, 1969; Della-Bianca, 1979; Snodderly, 1979; Stiles, 1980; Burkhardt, 1982; Herrera, 1982b, 1984, 1987; Stiles and White, 1982; Janson, 1983; McDonnell et al., 1984; Gautier-Hion et al., 1985; Gorchov, 1985, 1990; Izhaki and Safriel, 1985; Johnson et al., 1985; Rathcke and Lacey, 1985; Van Roosmalen, 1985; Wheelwright, 1985; Platt and Hermann, 1986; DeBussche et al., 1987; Fleming et al., 1987; Levey, 1987a, b; Jordano, 1987a; Borowicz, 1988a, b; Lee et al., 1988; Poston and Middendorf, 1988; Lambert, 1989; White, 1989; Willson et al., 1989; Willson

38

M. L. CIPOLLINI A N D E. W. STILES

and O’Dowd, 1989; Willson and Whelan, 1989). Many of these attributes are suspected of influencing the dispersal of seeds by animals, and thus a large number of hypotheses have been proposed concerning the potential selective influence of frugivores on such fruit traits (cf. Harper et af., 1970; Snow, 1970; McKey, 1973; Morton, 1973; Regal, 1977; Howe, 1979, 1984, 1985; Janzen, 1979, 1981a, b; Thompson and Wilson, 1979; Stiles, 1980, 1982; Herrera, 1982a, 1985; Wheelwright and Orians, 1982; Willson and Thompson, 1982; Sorenson, 1983; Tiffney, 1984; Herbst, 1986; Murray, 1987; Gorchov, 1988; Willson and Whelan, 1990a). Field studies concerning fruits and frugivores suggest that complex, multi-species interactions may be the rule rather than the exception (cf. Howe and Primack, 1975; Howe and Estabrook, 1977; Howe and Smallwood, 1982; Moore and Wilson, 1982; Davidar, 1983; Pratt and Stiles, 1983; Levey et al., 1984; Sorensen, 1984; Janzen, 1985; Beehler, 1986; Fleming, 1986; Herrera, 1986, 1988a, b; KeelerWolf, 1988; Murray, 1988; O’Donnell, 1989; Palmeirim et af., 1989; Willson et al., 1989; Loiselle, 1990; Willson and Whelan, 1990b; Willson et al., 1990; Loiselle and Blake, 1991; Witmer, 1991). The general consensus is that pairwise coevolution is an unlikely result of such “diffuse” fruit-frugivore interaction, and that only consistent broad-scale interactions may be expected to result in coadaptive evolutionary patterns in fruit traits (Janzen, 1980; Stiles, 1980; Wheelwright and Orians, 1982; Gould, 1988; Spencer, 1988; Thompson, 1989). Through broad-scale differences in their selection of fruits and treatment of seeds, frugivores certainly have the potential to influence fruit traits, although it has been suggested that the converse (i.e. frugivores adapting to fruit traits) may be more likely (Herrera, 1985). Yet, due to the lack of empirical data concerning these relationships, hypotheses concerning the degree (or lack) of coadaptation among plants and frugivores remain largely untested (Howe, 1984; Herrera, 1986; Jordano, 1987b; Berenbaum and Zangerl, 1988; Thompson, 1989; Witmer, 1991). We suggest that consideration of the selective influence of other players in the game (i.e. fungi), and of the influence of secondary chemistry in mediating multi-way interactions, may help to answer some of these general questions.

C. Interspecific Variation in Secondary Defense Chemistry Interspecific variation in chemical and physical characteristics suggests that fruits may vary in susceptibility to microbes, seed predators and pests, and thus in their response to selective pressure for the evolution of secondary defenses (Stiles, 1980; Herrera, 1982a). Extensive evidence from horticultural and medicinal plants indicates that ripe fruits vary considerably in secondary chemistry (cf. Nelson, 1927; Goldstein and Swain, 1963; Chirboga and Francis, 1970; Hulme, 1971; Somers, 1971; Du and Francis, 1973; Jankowski, 1973; Stohr and Herrmann, 1975; Aoki et al., 1976; Starke and

RELATIVE RISKS OF MICROBIAL ROT FOR FLESHY FRUITS

39

Herrmann, 1976; Lea, 1978; Wang et al., 1978; Baj et al., 1983; Ivanic et al., 1983; Moller, 1983; Narstedt et al., 1983; Ojewole and Adesina, 1983; Samanta et al., 1983; Hikino et al., 1984; Perera et al., 1984; Jaworski and Lee, 1987; Roddick, 1987; Edwards, 1988; Morozumi et al., 1989; Bandyopadhyay et al., 1990; Janovitz-Klapp et al., 1990a, b). Outside of detailed and taxonomically significant knowledge of the anthocyanin pigments of fleshy fruits in the family Ericaceae (Francis et al., 1966; Fuleki and Francis, 1967; Harborne, 1967; Ballinger et al., 1972, 1979; Sapers et al., 1984; Andersen, 1985, 1987; Ballington et al., 1987), general patterns of secondary chemistry for fruits of wild species remain virtually unknown (Dement and Mooney, 1974; Janzen, 1979, 1983; McKey, 1979; Wrangham and Waterman, 1983). The extent to which dispersal characteristics relate to patterns of secondary chemistry is also virtually unexplored (Herrera, 1982a). Focusing specifically upon the question of antimicrobial defense chemistry, we base this chapter upon the following general questions: (i) To what extent does secondary chemistry influence the antimicrobial activity and persistence time of ripe fruit? (ii) How is ripe fruit choice by avian frugivores affected by factors relating to microbial degradation, including: (a) microbial modification of the pulp substrate, (b) accumulation of microbial metabolites, and (c) presence of constitutive and induced antimicrobial defenses? (iii) What are the evolutionary consequences for plants using different modes of antimicrobial defense (e.g. physical defenses, secondary chemicals, escape through time)? (iv) Can variation in antimicrobial chemistry be related to differences in other fruit characteristics, particularly those associated with temperate seasonality and fruit phenology? In order to evaluate these questions, we first present background information and indirect evidence that can be used to estimate the intrinsic risks of fruit rot for particular plant species, and thus the degree of selective pressure for antimicrobial defense. Using this background information, we then generate several predictions concerning variation in selection pressure for antifungal defense, with particular reference to fruits and fruit-rot fungi of eastern North America.

111. FRUIT ROT AND EFFECTS ON DISPERSAL

A. Influence of Fruit Rot on Dispersal 1. Variation in Fruit Quality and Persistence Fruit persistence is necessarily estimated from counts of fruits present in discrete quality categories such as “green”, “ripe”, “rotted” and “dropped”

40

M. L . CIPOLLINI A N D E. W . STILES

(cf. Gorchov, 1990). In the eyes of the actual consumer, however, these discrete categories do not exist. True fruit “quality” varies throughout the ripening period in a continuous fashion, and independently with respect to individual fruiting plants and their frugivores. Variation in quality can occur through actual chemical and physiological changes in the fruit, or by extrinsic changes occurring within the plant-frugivore community that alter general patterns of resource availability. Fruit ripening involves changes in many metabolic pathways, usually resulting in an increase in volume and water content, and changes in pigmentation, texture, firmness, cell wall composition, nutrient chemistry, and secondary chemistry (Tukey and Young, 1939; Goldstein and Swain, 1963; Crane, 1964; Boland et al., 1968; Hulme, 1971; Makus and Ballinger, 1973; Stohr and Herrmann, 1975; Starke and Herrmann, 1976; Samanta et al., 1983; Gross and Sams, 1984; Brady, 1987; Gross, 1987; Blanke and Lenz, 1989). These changes are brought about by changes in various hormone levels, especially increases of ethylene (Biale, 1975; Bruinisma et al., 1975; Pratt, 1975; Rhodes and Reid, 1975; Yang, 1975; Young et al., 1975; Lieberman, 1979; Shimokawa, 1983; Yang and Hoffman, 1984; Brady, 1987), and decreases in indole acetic acid (Frenkel, 1972, 1975; Cohen and Bandurski, 1982), gibberellin (Hedden et al., 1978), and cytokinin (Crane, 1964; Letham and Palni, 1983). Ripening may be abrupt or gradual in its initiation, and may vary in synchrony among plants and among fruits on a plant (Stiles, 1980; Janzen, 1983; Gorchov, 1990). Abscission of ripe fruit is also highly variable among and within plant species (Gough and Litke, 1980; Stephenson, 1981; Janzen, 1983; Sutherland, 1986). The intrinsic changes in fruit pulp due to natural maturation do not necessarily represent tightlylinked phase changes; many physiological traits have been shown to vary independently during the ripening process (Goldstein and Swain, 1963; Ballinger and Kushman, 1970; Moore et al., 1972; Makus and Ballinger, 1973; Markakis et al., 1963; Stohr and Herrmann, 1975; Starke and Herrmann, 1976; Willson and Thompson, 1982; Gross and Sams, 1984; Eck, 1988, 1990; Poston and Middendorf, 1988). Moreover, variation in frugivore community composition, frugivore experience and hunger level, presence of predators, spatial display pattern, and presence of other available fruiting plants may each independently affect the perception of fruit quality by an individual frugivore (Howe and Primack, 1975; Howe and Estabrook, 1977; Howe, 1979; Howe and Vande Kerchove, 1979; Herrera, 1982b, 1985, 1988a,b; Moore and Willson, 1982; Real et al., 1982; Stapanian, 1982; Wheelwright and Orians, 1982; Pratt and Stiles, 1983; Levey et al., 1984; Izhaki and Safriel, 1985; Murray, 1987; Levey, 1988a, b; White, 1989). Variation in relative quality toward dispersers and damaging agents may thus be a key factor influencing selection on pulp secondary chemistry.

RELATIVE RISKS OF MICROBIAL ROT FOR FLESHY FRUITS

41

2. Definition of “Fruit Rot” Fruit rot may be defined as the chemical and physical alteration of fruit pulp tissue due to infection by microbes (Janzen, 1977). Because the high acidity of fruit tissue tends to inhibit bacterial growth, and because dispersal and tissue penetration capabilities of bacteria may be limited, filamentous fungi are the most important agents of fruit rot (Stevens, 1913; Stevens and Hall, 1926; Ainsworth, 1971; Dennis, 1983; Nel, 1985; Rossman et al., 1987; Agrios, 1988; Farr et al., 1989). Infection by various insect-dispersed yeasts, particularly Saccharomyces spp., is also very common (Begon, 1982; Dennis, 1983; Nel, 1985; Starmer et al., 1990). While recognizing the potential importance of other organisms as fruit degraders, we restrict our remaining discussion to filamentous fungi and yeasts, which we refer to as “fungi”. Fungal fruit rot results in the alteration of pulp nutrient patterns (Hawkins, 1915; Schiffman-Nadel, 1975; Cooke, 1979; Pucheu-Plante and Mercier, 1982; Dennis, 1983; Pitt and Hocking, 1985; Snowdon, 1990; Starmer et al., 1990), physical breakdown of tissues (White and Fabian, 1953; Cooke, 1979; Rujkenberg et al., 1980; Cooper and Wood, 1975; Ceponis and Stretch, 1983; Barmore and Ngyen, 1985; Pitt and Hocking, 1985), accumulation of mycotoxins (Bilai, 1963; Rodricks, 1976; Pitt and Hocking, 1985; Marasas and Nelson, 1987; Hsieh, 1989), alteration of ripening rates (Cohen and Schiffman-Nadel, 1971 ; Zauberman and Schiffman-Nadel, 1973), and changes in taste, odor and color (Dennis, 1983; Nel, 1985; Pitt and Hocking, 1985; Leistner et al., 1989; Newsome, 1990). In general, therefore, fruit rot should be detrimental to seed dispersal.

3. Relationship to Seed Dispersal It has been suggested that saprophytic fungi attacking senescent tissues should exert little selection pressure upon the plants they infect, e.g. Drosophila-dispersed yeasts infecting cactus tissues (Starmer and Fogelman, 1986; Starmer et al., 1990). We suggest that the case of ripe fruit rot is an exception to this general assertion, because variation among individual plants in their ability to prevent or delay ripe fruit rot should have decided fitness consequences, due to variation in seed dispersal. Thus, selective pressure for plants to prevent fruit rot should relate directly to variation in the probability of seed dispersal (Janzen, 1977). Fitness benefits due to successful dispersal may include: (a) reduction in local seedling density or intraspecific competition (Howe and Smallwood, 1982; Augspurger, 1984; Howe and Schupp, 1985), (b) removal away from the competitively dominant maternal plant (Janzen, 1970; Howe, 1979), (c) increased colonization of spatially distant or ephemeral habitats (Smith, 1975; McDonnell and Stiles, 1983; Denslow, 1987; Hoppes, 1988; Levey, 1988b; Murray, 1988; Schupp et al., 1989), (d) escape from density- and distance-dependent predators and

42

M. L. CIPOLLINI AND E. W. STILES

diseases (Janzen, 1970; Connell, 1971; Augspurger, 1984; but see Hubbell, 1979), (e) dispersal to microenvironmental “safe-sites’’ (Harper et al., 1970; Davidar, 1983; Piper, 1986; Schneider and Sharitz, 1988), and (f) removal of inhibitory effects of surrounding pulp or fulfillment of the need for seed scarification (Lieberman and Lieberman, 1986; Robinson, 1986; Janzen, 1977; Barnea er al., 1991). Alterations of fruit tissue via fungal degradation can negatively affect dispersal in two important ways: (a) by decreasing fruit retention time by hastening ripening and abscission, and (b) by direct and indirect negative effects upon foraging by frugivores. Fungi may hasten fruit ripening and abscission via the production or induction of plant hormones critically associated with senescence and ripening. For instance, the autocatalytic release of ethylene may be induced by fungal infection (Cohen and Schiffman-Nadel, 1971; Zauberman and Schiffman-Nadel, 1973; Fleuriet and Macheix, 1975; Schiffman-Nadel, 1975; Yang and Hoffman, 1984). Fruit drop from the plant does not necessarily preclude dispersal, because potential dispersal agents may forage for fallen fruits on the ground (Howe and Smallwood, 1982), and seeds commonly germinate beneath parent plants (Hubble, 1979). However, dropped fruits may be subjected to increased fungal rot and seed predation (Janzen, 1970). Fruit rot can also directly and indirectly discourage foraging by dispersal agents (Janzen, 1977; Herrera, 1982a). Direct effects include the alteration of the physical structure, odor, appearance, and palatability of the fruit, such that frugwores are discouraged specifically from feeding upon the infected fruit (Borowicz, 1988b; Buchholz and Levey, 1990). Indirect effects may include a decrease in the dispersal of non-infected fruit brought about by associative feeding aversions (Chapman and Blaney, 1979; Wicklow, 1988), or by alterations in overall fruit display pattern (Murray, 1988). Although fruit-rot fungi may alter pulp chemistry and physical structure by any or all of these mechanisms, such changes cannot be considered in absolute terms, because interactions among fruit characteristics and various extrinsic factors may make rotted (or otherwise “poor” quality) fruit acceptable at certain times or under certain conditions. This may be especially true during times of resource depletion (cf. Foster, 1977), when animals may be less discriminating while feeding. Yeast rots that produce acids and ethanol may even delay fruit abscission (M. Cipollini, personal observation of Vaccinium macrocarpon), and it is possible that physical and chemical changes produced by these and other fungi may actually enhance the apparent attractiveness of fruit under some circumstances (Janzen, 1977; Pirozynski and Hawksworth, 1988). Some fungal rots (e.g. “noble” rot produced by Botrytis cinerea on Sauterne grapes) may actually increase levels

RELATIVE RISKS OF MICROBIAL ROT FOR FLESHY FRUITS

43

of simple sugars within the pulp (Pucheu-Plante and Mercier, 1982). Pirozynski and Hawksworth (1988) go so far as to suggest that fungal intermediaries may be necessary to enhance fruit quality for dispersers, and thus act to promote dispersal. It is clear, therefore, that study of the effects of fungal rot on fruit “quality” must take into account not only measures of fruit persistence under field conditions, but must also evaluate potential effects of specific fungi upon dispersal. One method of accomplishing this is to estimate removal rates by foragers in the field for naturally rotted and unrotted fruits. Aside from logistical problems, this method suffers from the difficulty of simultaneously determining physical and chemical fruit parameters, determining fruit-rot status, measuring dispersal rates, and controlling for extrinsic factors that may account for a large degree of the variation that occurs in dispersal from year to year. Additionally, the fungal agents responsible for fruit rot usually remain unidentified (cf. Borowicz, 1988b; Buchholz and Levey, 1990). Feeding trials with captive frugivores provide an alternative method of evaluating the degree of deterrence produced by particular fruitrot agents, while experimentally controlling many potentially interacting extrinsic factors. Field experimentation (artificial inoculations, etc.) may also provide pertinent empirical data.

B. Factors that Affect Risk of Microbial Rot 1. Pulp Nutrient Chemistry Much evidence exists for interspecific variation in the primary nutrient content of ripe fruit pulp, including water, simple sugars, polysaccharides, protein, lipids, and minerals (Boland et al., 1968; Snow, 1970; Hulme, 1971; Landers et al., 1979; Stiles, 1980; Herrera, 1982b, 1984, 1987; Stiles and White, 1982; Sorensen, 1984; Gautier-Hion et al., 1985; Izhaki and Safriel, 1985; Johnson et al., 1985; Wheelwright and Janson, 1985; Herbst, 1986; DeBussche et al., 1987; Eck, 1988, 1990; White, 1989; Peters and Hammond, 1990; Cipollini, 1991; Gagiullo and Stiles, 1991). Evidence is also accumulating which shows potentially important levels of intraspecific variation in nutrient content, as well as important qualitative differences in pulp constituents (Galleta et al., 1971; Vander Kloet and Austin-Smith, 1986; Herrera, 1988a, b; Keeler-Wolf, 1988; Poston and Middendorf, 1988; Gargiullo and Stiles, 1991; E. Stiles and M. Cipollini, unpublished). Growth of fungi has been shown to respond to variation in the chemical nutrient make-up of media and host plants (Trelease and Trelease, 1929; Bilai, 1963; Muys et al., 1966; Ballinger and Kushman, 1970; Ballinger et al., 1978; Parkinson, 1981; Vanderplank, 1984; Dhingra and Sinclair, 1985; Pitt and Hocking, 1985; Cihlar and Hoberg, 1987; Kerwin, 1987; Verhoeff et al., 1988; Lacey, 1989; Cipollini, 1991). It

44

M. L. CIPOLLINI AND E. W. STILES

follows that variation in fruit-pulp nutrient chemistry should be related to variation in the intrinsic potential for fungal growth. Although it is generally expected that higher nutrient levels should enhance microbial growth rates, non-linear responses may be common. For instance, increases in sugar content of host tissues may increase susceptibility to certain fungi, but only up to the point at which osmotic effects of high sugar content begin to inhibit hyphal growth (Trelease and Trelease, 1929; Janzen, 1983; Vanderplank, 1984; Cipollini, 1991). Due to substrate specialization, fungi may vary interspecifically in their utilization of particular nutrient components, and thus in their response to quantitative variation in these components. For example, mutualistic interactions among yeasts have apparently resulted in complementary metabolic capabilities that may facilitate nutrient degradation within fruits (Starmer and Fogelman, 1986). It should be noted that the allocation of plant constituents has been conventionally reported as per cent dry mass, but from the perspective of fungal growth potential, per cent wet mass may be a more appropriate measure. A succession of microbes generally accompanies the saprophytic decomposition of senescent plant tissue (e.g. rapid colonizers that reduce simple sugars, followed by slower reducers of lipids, cellulose, and finally lignins; Cooke, 1979). Growth studies suggest that, despite their heterotrophic nature, fungi are generally less limited in growth by quantitative variation in organic substrates, and more limited by variation in the mineral content of the medium (Ballinger and Kushman, 1970; Cooke, 1979; Cihlar and Hoberg, 1987; Kerwin, 1987). This is apparently especially true for available nitrogen. Fruit-pulp tissue is generally low in nitrogen and other minerals, despite considerable variation in organic constituents (Ballinger and Kushman, 1966; Boland et al., 1968; Hulme, 1971; Landers et al., 1979; Stiles, 1980; Herrera, 1982b, 1984, 1987; Stiles and White, 1982; Izhaki and Safriel, 1985; Johnson et al., 1985; Herbst, 1986; DeBussche et al., 1987; Poston and Middendorf, 1988; White, 1989). The rate of hyphal growth may be very important in influencing the rate at which the physical structure and nutrient patterns of the pulp are altered (White and Fabian, 1953; Cooper and Wood, 1975; Verhoeff et al., 1988; Lacey, 1989), and the overall ability of the fruit pulp to sustain fungal growth (total available nutrients) may influence the degree of mycotoxin production (Bilai, 1963; Rodricks, 1976; Pitt and Hocking, 1985; Marasas and Nelson, 1987; Wicklow, 1988; Lacey, 1989; Scott, 1989). Thus the effect of fungal growth on fruit chemistry may be considered in both absolute and relative terms, that is, similar absolute changes in nutrient level may differentially affect the relative quality of fruit for dispersers. For instance, fruits with a low initial level of pulp sugar may be more rapidly reduced to a nonacceptable sugar threshold than fruits of high initial sugar content. A higher rate of fungal growth could offset such an effect by reducing sugar content

RELATIVE RISKS OF MICROBIAL ROT FOR FLESHY FRUITS

45

more quickly in high-sugar fruits, but this is unlikely to ameliorate completely differences between initially high- and low-sugar fruits (Janzen, 1977; Herrera, 1982a). High-sugar fruits should thus always be more profitable for dispersers than low-sugar fruits, and they should therefore be preferred when availabilities are equal (cf. Lundberg and Astrom, 1990). This is where variation in toxin production by fungi may be critically important (Janzen, 1977). Increased toxin production by fungi on nutrient-rich substrates may reduce pulp quality to a greater degree than may be apparent from simple examination of fungal growth rates (Rodricks, 1976; Marasas and Nelson, 1987; Lacey, 1989; Leistner et al., 1989). Fungal species may vary in the quality of toxins produced on various substrates (Marasas and Nelson, 1987; Mills, 1989; Tanaka and Ueno, 1989; Yabe et af., 1989), in the effect of quantitative nutrient variation on toxin production (Cooke, 1979), and in the effects of their toxins upon different frugivore species (Janzen, 1977; Marasas and Nelson, 1987; Wicklow, 1988; Kiessling, 1989). Additionally, synergistic toxin interactions may occur due to coinfection by toxigenic fungi (Janzen, 1977; Wicklow, 1988). In short, the influence of primary nutrient variation on the rates and effects of fungal decomposition may be very dependent upon the particular plant-frugivore-fungus species combination under consideration, and experimental tests should employ media that reflect the basic nutrient and chemical background of the plant tissue of concern.

2. Ambient Environmental Conditions Extrinsic environmental factors that vary among plant species should result in variation in the risk of fungal rot. Extrinsic factors that may affect colonization, spore germination, mycelial growth, and toxin production include temperature, light regime (including UV irradiation), humidity, rainfall, and disease vectors (McKeen, 1958; Bilai, 1963; Cappellini et al., 1972, 1982, 1983; Rodricks, 1976; Ballinger et af., 1978; Cooke, 1979; Hartung et a[., 1981; Ceponis and Stretch, 1983; Jarvis and Traquair, 1984; Pitt and Hocking, 1985; Marasas and Nelson, 1987; Agrios, 1988; Biggs and Northover, 1988; Lacey, 1989; Snowdon, 1990; Starmer et al., 1990). Environmental conditions during the time of colonization may be just as important in influencing fruit rot as conditions during colony growth and sporulation (Varney and Stretch, 1966; Hartung et al., 1981; Cappellini et al., 1983; Milholland and Daykin, 1983; Daykin, 1984; Agrios, 1988; Dashwood and Fox, 1988; Arauz and Sutton, 1990; Daykin and Milholland, 1990; Yang et al., 1990). Also, some environmental effects, such as bruising, insect damage, abscission scarring, sun-scald and freeze-rupturing, may predispose fruits to rot at later times (Graham et al., 1967; Cappellini and Ceponis, 1977; Dennis, 1983; Jarvis and Traquair, 1984; Howard er al., 1985; Nel, 1985; Schwarz and Boone, 1985; Starmer et al., 1990).

46

M. L. CIPOLLINI A N D E. W. STILES

Interactions among various abiotic and biotic factors may be important influences on fungal growth and toxin production (Lacey, 1989; Scott, 1989). For instance, climatic variation may affect the abundance and activities of insect vectors of various fruit diseases (Batra, 1983; Dennis, 1983; Batra and Batra, 1985; Nel, 1985). Climatic variables may themselves interact, resulting in non-linear effects on fungal growth. For instance, high temperatures may be commonly associated with low levels of humidity and high levels of potentially fungicidal UV radiation. Generally speaking, warm ( f 25” C), humid or moist conditions are optimum for spore germination and mycelial growth of fruit-rot fungi (Cappellini et al., 1972; Ballinger et al., 1978; Cooke, 1979; Hartung et al., 1981; Johnson and Booth, 1983; Jarvis and Traquair, 1984; Agrios, 1988; Biggs and Northover, 1988; Lacey, 1989; Scott, 1989; Arauz and Sutton, 1990; Snowdon, 1990), but exceptions to this general pattern may occur (Scott, 1989). Experimental work should attempt to account for potential environmental variation, or such variation may form the focus of testable predictions.

3. Identity and Quantity of Spore Inoculum Initiation of fruit rot necessarily begins with inoculation of fruit tissue by microbes. Seasonal, habitat-associated, and environmental effects may all contribute to variation in the abundance and species composition of fungal inoculum (Dye and Vernon, 1952; Pady, 1957; Williams et al., 1957; Leben et al., 1968; Cooke, 1979; Wong and Kwan, 1980; Lacey, 1981; Wicklow, 1981; Shivas and Brown, 1984; Pandey, 1990; Starmer et al., 1990; Yang et al., 1990). Ripe fruit rot is most commonly associated with “opportunistic”, or “generalist” fungal species (Varney and Stretch, 1966; Dennis, 1983; Nel, 1985; Rossman et al., 1987; Agrios, 1988; Farr et al., 1989; Snowdon, 1990). Fungal species noted for causing ripe fruit rot belong to: Alternaria (Cappellini et al., 1972), Aspergillus (Raper and Fennell, 1965; Christensen and Tuthill, 1986), Botrytis (Cappellini et al., 1972; Rujkenberg et al., 1980; Pucheu-Plante and Mercier, 1982; Dashwood and Fox, 1988; Malathrakis, 1989), Cladosporium (Dennis, 1983; Nel, 1985), Colletotrichum (Gloeosporium) (Stanghellini and Aragaki, 1966; Cappellini et al., 1972; Hartung et al., 1981; Daykin, 1984; Yang et al., 1990), Fusarium (Zauberman and Schiffman-Nadel, 1973; Nelson et al., 1983), Ceotrichum (Cooke, 1979), Monilinia (Hawkins, 19 15; Biggs and Northover, 1988), Penicillium, Phoma (Malloch, 198l), Phomopsis (Wilcox, 1939, 1940; Cappellini et al., 1982; Milholland and Daykin, 1983), Rhizopus (Nel, 1985), Saccharomyces and other yeasts (Cappellini et al., 1972; Starmer and Fogelman, 1986), and various Mycelia sterilia (Barnett and Hunter, 1972). Latent (endophytic) fungal infections may be a very important and widespread mechanism of fruit-rot initiation (McKeen, 1958; Stanghellini and Aragaki, 1966; Graham et al., 1967;

RELATIVE RISKS OF MICROBIAL ROT FOR FLESHY FRUITS

47

Schiffman-Nadel, 1975; Rujkenberg et al., 1980; Hartung et al., 1981; Pucheu-Plante and Mercier, 1982; Daykin, 1984; Agrios, 1988; Biggs and Northover, 1988; Carroll, 1988; Dashwood and Fox, 1988). Latent infections result from fungal spore germination and penetration in the flower bud, flower, or green fruit stages, but d o not cause symptoms of rot until fruits mature (Agrios, 1988; Snowdon, 1990). For latent fungi especially, environmental conditions at the time of spore germination may be as important as conditions at the time fruit rot commences (Snowdon, 1990). Highly host-specific fungi may be important pathogens associated with the loss of green fruit, but apparently rarely cause ripe fruit rot (Nel, 1985; Agrios, 1988; Farr et al., 1989). For example, the Monilinia spp. that cause Mummy-berry disease of Vaccinium and Gaylussacia spp. are the only well known host-specific blueberry fruit pathogens (Varney and Stretch, 1966; Batra, 1983; Batra and Batra, 1985). All of these Monilinia fungi cause loss of green fruit prior to ripening (Batra, 1983). Although primarily generalists, saprophytic fungal species associated with ripe fruit rot may vary considerably in the mechanisms by which they infect fruit tissue, in the breadth of their plant species and tissue affinities, and in their ability to produce changes in fruit pulp (cf. Starmer et al., 1990). Fungi may likewise vary considerably in the manner and extent to which they affect fruit dispersal. Experimental work should focus on fruit-associated organisms that both d o and do not typically cause rot, because it is equally important to elucidate mechanisms responsible for general defense, as it is to identify variation in resistance to known pathogens.

4 . Synchrony and Other Display Characteristics Synchronous ripening increases the physical proximity of susceptible fruits, and may thus facilitate the within-plant spread of fruit-rot agents. Examples from commercial fruits include webbing and nesting that occurs with several post-harvest fungi (Dennis, 1983; Nel, 1985), the enhancement via physical proximity of splash-dispersal infection of strawberry fruits by the anthraconose fungus Colletotrichum acutatum (Yang et al., 1990), and the increased likelihood of bunch rot (Aspergillus acleatus) within tight, synchronously ripening grape clusters (Jarvis and Traquair, 1984). As susceptible host plants and their fruits may be thought of as “islands” in space and time (Janzen, 1968, 1973; Kuris et al., 1980; but see Janzen, 1979), synchronous within- and among-plant fruit ripening should accelerate colonization rates, and thus increase the epidemiological spread from fruit to fruit and plant to plant (Wicklow, 1981; Vanderplank, 1984). As with extrinsic risk variables, the degree of synchrony or clustering may be controlled in experimental work by the selection of comparable plant species, or such variation may form the focus of testable hypotheses.

48

M. L. CIPOLLINI AND E. W. STILES

5 . Removal Rate, or Time of Exposure Other factors being equal, the risk of fruit rot should be directly proportional to the rate at which susceptible fruits are removed by frugivores (Herrera, 1982a). Plants whose fruits are normally removed quickly should be under less risk of becoming colonized. Even if the fruit is colonized, rapid removal should allow less time for fruit-rot agents to grow and thereby negatively affect fruit characteristics. Because plant, fruit, and environmental characteristics may each independently influence removal rates, as well as fruit-rot rates, interactions among these factors must be taken into account when generating predictions concerning fruit defense patterns (Pirozynski and Hawksworth, 1988).

C. Natural Selection for Fruit Defenses I . Secondary Chemicals as Defense Agents The recognition that secondary chemicals may act as defense agents revolutionized studies of plant-herbivore and plant-pathogen interactions (Rhoades, 1985). We propose that the study of fruit secondary defenses will be equally informative and stimulating to the field of dispersal ecology. Although fruit chemistry has been little studied from this perspective, patterns of immature fruit defenses are likely to be basically similar to the broad-spectrum defenses of leaves. Thus, as a background to the discussion of potential chemical patterns in ripe fruit, we present the following general description of plant defenses. ( a ) Constitutive defenses. Secondary chemicals that may defend plants against pests, herbivores, other plants, and pathogens can be placed into two major classes: constitutive and induced (Harborne and Ingham, 1978; Vanderplank, 1984; Agrios, 1988). Constitutive defenses are present in plant tissue prior to feeding damage or the invasion of pathogens (Stoessl, 1970; Mitscher, 1975; Schonbeck and Schlosser, 1976; Harborne and Ingham, 1978). Examples of structural constituents include cellulose, lignin (Vance et af., 1980), epicuticular waxes (Franich et al., 1983), and trichomes (Levin, 1973; Rathcke and Poole, 1975). A variety of toxic compounds may enhance structural constituents (Nickell, 1959; Raffauf, 1970; Stoessl, 1970; Freeland and Janzen, 1974; McKey, 1974; Robinson, 1975; Levin, 1976; Swain, 1977; Harborne and Ingham, 1978). Specific examples of biologically active constituents include numerous classes of phenolic compounds (Hulme and Edney, 1960; Cruikshank and Perrin, 1964; Mathis, 1966; Towers et al., 1966; Feeny, 1968; Levin, 1971; Janzen, 1974; Bate-Smith, 1977; Rhoades, 1977; Harborne, 1979, 1980, 1989; McClure, 1979; Swain, 1979a, b; Gartlan et af., 1980; Berenbaum, 1981; Lane and Shuster, 1981; Johnson, 1983;

RELATIVE RISKS OF MICROBIAL ROT FOR FLESHY FRUITS

49

Jaworski and Lee, 1987; Lee et al., 1987; Laks, 1989; Meyer and Karasov, 1989; Morozumi et al., 1989; Van den Berg and Labadie, 1989; Van Sumere, 1989; Williams and Harborne, 1989a, b; Bandyopadhyay et al., 1990; Henderson, 1990), organic acids (Markakis et al., 1963; Moller, 1983), alkaloids (Raffauf, 1970; McKey, 1974; Robinson, 1979; Manske and Rodrigo, 1981; Giral and Hidalgo, 1983; Ojewole and Adesina, 1983; Samanta et al., 1983; Perera et al., 1984; Roddick, 1987), terpenes (Mabry and Gill, 1979; Camazine et al., 1983; Hubbell et al., 1983; Ivanic et al., 1983; Hubbell and Howard, 1984), terpenoid and steroidal saponins (Aoki et al., 1976; Applebaum and Birk, 1979; Roddick, 1987), fatty and resin acids (Seigler, 1979; Franich et al., 1983), phytohemagglutinins (Liener, 1979), and non-protein amino acids (Janzen, 1969; Rosenthal and Bell, 1979; Blieler et al., 1988). Condensed and hydrolyzable tannins, two classes of polyphenolic compounds, are among the most important secondary constituents because of their general antifungal, antibacterial, and antiherbivore activities, and widespread occurrence among plant species (Levin, 1971, 1976; Bate-Smith, 1972, 1977; Freeland and Jazen, 1974; Swain, 1978, 1979a,b; Lane and Schuster, 198I ; Galloway, 1989; Porter, 1989; Walkinshaw, 1989). Tannins complex with a broad spectrum of biomolecules, especially proteins and polymeric carbohydrates, and thus have astringent properties (Bate-Smith, 1977; Wang et al., 1978; Hagerman and Butler, 1980; Martin and Martin, 1982; Asquith and Butler, 1986; Hagerman, 1987, 1989; Mole and Waterman, 1987a,b; Laks, 1989; Lewis and Yamamoto, 1989). Apparently due to their astringency, tannins are distasteful for many herbivores, and ingestion has been hypothesized to interfere with digestion by binding digestive enzymes, food proteins, and digestive membranes (Goldstein and Swain, 1963; Feeny, 1968, 1969, 1973; Bate-Smith, 1972, 1977; Reese, 1979; Wrangham and Waterman, 1983; Faeth, 1985; Coley, 1986; Bernays and Janzen, 1988; Rossiter et al., 1988; Bernays et al., 1989; Butler, 1989; Karowe, 1989; Meyer and Karazov, 1989; Schultz, 1989; Clausen et al., 1990; Koenig, 1991). Other constitutive chemicals are less generally deterrent for herbivores (Feeny, 1969; Bate-Smith, 1972; Chapman and Blaney, 1979; Crawley, 1984). In fact, for specialist herbivores, specific secondary constituents may be responsible for evolved feeding preferences (Benson et al., 1975; Rhoades and Cates, 1976; Janzen, 1979; Rhoades, 1979, 1985; McDonald, 1983; but see Jermy, 1984; Smiley, 1985). ( b ) Induced defenses. Induced defenses are elicited by physical damage or by microbial entry into plant tissues (Fleuriet and Macheix, 1975; Harborne and Ingham, 1978). Induced structural responses include cell wall lignification and callose formation (Stanghellini and Aragaki, 1966; Allison and Shalla, 1973; Rujkenberg et al., 1980; Vance et al., 1980), deposition of silica

50

M. L. CIPOLLINI AND E. W. STILES

(Heath, 1981), and hypersensitive death of cells (Maclean et al., 1974; Bailey et al., 1980). Bioactive agents that can be induced include bound toxins that are released upon damage or microbial entry (post-inhibitins: Harborne and Ingham, 1978), various other phenolics (Harborne, 1980; Rossiter et al., 1988; Tuomi et al., 1990; Zangerl, 1990), cyanogenic glycosides (Dement and Mooney, 1974; Conn, 1979; Narstedt et al., 1983; Briggs and Schultz, 1990), glucosinolates (Van Etten and Tookey, 1979), various enzymatic proteins (Esquerre-Tugaye et al., 1979; Ryan, 1979; Malamy et al., 1990; Metraux et al., 1990), alkaloids (Baldwin, 1988), and small molecular weight phenolics synthesized de novo following damage or invasion (phytoalexins: Bailey and Deverall, 1971; Mansfield and Hargreaves, 1974; Paxton et al., 1974; Anderson-Prouty and Albersheim, 1975; Albersheim and Valent, 1978; Anderson, 1978; Hahn and Albersheim, 1978; DeWit and Roseboom, 1980; Burdon and Marshall, 1981; DeWit and Kodde, 1981; Hahlbrock et al., 1981; Bruce and West, 1982; Yamazaki et al., 1983; Darvill and Albersheim, 1984). Induction of defense is not necessarily restricted to the damage site of infected tissues, as various transduction signals may induce systemic resistance (Lynn and Chang, 1990; Malamy et al., 1990; Metraux et al., 1990), and it had even been suggested that resistance in nearby conspecifics may be induced via airborne chemical signals (Baldwin and Schultz, 1984; Rhoades, 1985). The importance of metabolically induced defenses in the resistance of plants to leaf and stem pathogens is well established (Keen, 1975). These defenses can be overcome by host-specific pathogens, thus paving the way for coevolutionary feedback (Bailey and Deverall, 1971; Denny and Van Etten, 1983). Such host-specific pathogens are commonly associated with immature fruits (Agrios, 1988). In having photosynthetic capability, active cell division, and active anabolic metabolism, immature fruits may be physiologically similar to green leaves, and may be quite capable of responding to fungal infection via de novo induced defenses. As discussed previously, infection of ripe fruit is more commonly due to facultative or opportunistically saprophytic fungi. The catabolic changes that accompany fruit ripening may render mature fruit susceptible to generalist saprophytes, and may thus diminish the opportunity for coevolutionary feedback between host and fungus necessary for reciprocal selection leading to host-specificity (Rhoades, 1985; Roddick, 1987; Thompson, 1989). Defense induction in ripe fruits is thus likely to result primarily from the release of less specific bound toxins (post-inhibitins) following fungal invasion or tissue damage. The epidermal layers of the fruit pericarp (fruit skin) are important structural barriers to microbial invasion, but as with most physical features, epidermal structure may be enhanced by bioactive agents (Croteau and Fagerson, 1971; Croteau, 1977; Franich et al., 1983; Janzen, 1983). There are several reasons to expect that defense should be concentrated in fruit skin:

RELATIVE RISKS OF MICROBIAL ROT FOR FLESHY FRUITS

51

(i) Direct intra- and inter-cellular penetration through the epidermis is a primary source of fungal infection (White and Fabian, 1953; Stanghellini and Aragaki, 1966; Stiles and Abdalla, 1966; Graham et al., 1967; Cappellini and Ceponis, 1977; Cooke, 1979; Rujkenberg et al., 1980; Pucheu-Plante and Mercier, 1982; Ceponis and Stretch, 1983; Milholland and Daykin, 1983; Daykin, 1984; Jarvis and Traqair, 1984; Schwarz and Boone, 1985; Agrios, 1988). Only a few fruit infections are systemic, and are commonly transmitted through vascular tissues (e.g. Xanthomonas infecting plum fruit (Du Plessis, 1990)). (ii) Because fruit skin comprises a relatively small fraction of fruit mass, concentrating chemicals there is unlikely to result in a total fruit concentration that is toxic to frugivores (Janzen, 1979, 1983). (iii) Animals may avoid detrimental effects by regurgitating or defecating relatively intact fruit skins (M. Cipollini, personal observation). Indirect evidence for the importance of fruit skin as a defense stems from studies showing that host-specific fruit pathogens may depend upon insect vectors for entry into fruit tissue. For instance, Monilinia (Mummy-berry) diseases of Vaccinium spp. are transferred to flower stamens via bees acting as pollen vectors (Batra, 1983; Batra and Batra, 1985), and specific bacterial and yeast rots are introduced into fruits by foraging Drosophila and Rhagoletis fruit-flies (Howard et al., 1985; Starmer and Fogelman, 1986; Pirozynski and Hawksworth, 1988; Starmer et al., 1990). ( c ) Interaction among plant chemicals. Because plant tissues contain a mixture of potentially interacting chemicals, synergistic or counteractive interactions among various components may be common (McKey, 1979; Reese, 1979; Berenbaum and Zangerl, 1988; Cipollini, 1991). For instance, the toxicity of many small phenolics is enhanced at low pH by high concentrations of organic acids such as citric acid (Hoffman et al., 1941; Cruikshank and Perrin, 1964; Constantine et al., 1966; Graham et al., 1967; Swartz and Medrick, 1968; Stoessl, 1970; Lacey, 1989; Davidson and Juneja, 1990). Although phenolics are commonly held at medium pH within vacuoles (Robinson, 1975; McKey, 1979; Harborne, 1980), cellular disruption following physical damage or microbial entry may allow them to come in contact with the acidic cytoplasm. For example, the level of organic acid in cultivated blueberry fruit is known to be related to its resistance to fungal rot (Ballinger and Kushman, 1970; Galleta et al., 1971; Ballinger et al., 1978). This occurs in spite of the fact that equivalent in vitro variation in organic acids alone has only a small negative or even a positive effect on growth of the causative agents of these rots (Ballinger and Kushman, 1970; Galleta et al., 1971; Cipollini, 1991). This suggests that interactions of acids with small phenolics in the fruit tissue may be partially responsible for the antifungal activity.

52

M. L. CIPOLLINI A N D E. W. STILES

Fungi may counter this pH/phenolic interaction effect by utilizing organic acids as substrates during hyphal extension, and thus increasing tissue pH to levels at which phenolics are less active (Verhoeff et al., 1988). Further increases in pH may result in phenolic oxidation products (e.g. quinones) that may be more toxic to fungi (Cilliers and Singleton, 1990). Many other synergistic defense patterns have emerged in the literature: (i) The activity of the most effective phenolic within grape (Vitis vinifera) berries, pterostilbene, against the fungus Botrytis cinerea, depends in part upon its association with glycolic, tartaric and malic acids (Pont and Pezet, 1990). (ii) The antimicrobial properties of bearberry leaf extract (Arctostaphy10s spp.) and cranberries (Vaccinium spp.) have been attributed to a complex synergism among various organic acids and small phenolics (Clague and Fellers, 1934; Constantine et al., 1966; Papas et al., 1966; Graham et al., 1967; Swartz and Medrick, 1968; Frohne, 1969; Matzner, 1971; Eck, 1990). (iii) The antifungal activity of Citrus spp. fruit skin has been attributed to a synergistic interaction between nootkatene and citric acid (Morozumi et al., 1989). (iv) Complexing of plant protein with tannins in protein-rich tissues during consumption may reduce the negative effects of tannins on digestion by herbivores (Asquith and Butler, 1986; Bernays et al., 1989). (v) Storage lipids act in an unknown fashion to enhance the effect of condensed tannins in reducing digestive efficiencies for the acorn woodpecker, Melanerpes formicivorus (Koenig, 1991). (vi) Coingestion of saponins and tannins by rodents has been shown to reduce the negative effects of these compounds on growth, and rodents may balance intake of each compound so as to ameliorate toxic effects (Freeland et al., 1985). (vii) Fusaric acid, a toxin produced by Fusarium spp., acts synergistically to increase the antifungal activity of plant phytoalexins (Wicklow, 1988). (viii) Increased pH levels may reduce growth of fungi attacking Lycoperiscon esculentum fruits, due to a positive effect on the antifungal activity of the steroid glycoalkaloids solanine, tomatine, and demissine (Roddick, 1987). (ix) The overall inhibitory effects of the mixture of furanocoumarins present in the herb Pastinaca sativa toward insect herbivores is greater than the sum of the activities of the individual compounds (Berenbaum et al., 1991).

RELATIVE RISKS OF MICROBIAL ROT FOR FLESHY FRUITS

53

Coupled with our generally poor knowledge of the compartmentalization of various chemical constituents within plant cells and tissues, as well as our limited understanding of mechanisms of toxicity, these potential interactions make it very difficult to attribute in vivo antifungal defense, and/or feeding patterns of herbivores and frugivores to individual isolated chemicals (Janzen, 1979; Berenbaum and Zangerl, 1988).

2. Biotic Defenses Much interest has been generated recently concerning the potential role of phylloplane and endophytic microbes as antiherbivore and antidisease agents in plants (Blakeman and Fokema, 1982; Cooke and Baker, 1983; Windels and Lindow, 1985; Carroll, 1988; Scott, 1989; Wilson et al., 1991). Endophytic fungi may be very important as antiherbivore defense agents in grasses via ergot alkaloid production (Clay, 1988a, b; Buckley and Halisky, 1990). Additionally: (i) Phylloplane micro-organisms have been shown to be antagonistic toward plant pathogens via both competitive and antibiotic mechanisms (Newhook, 1957; Upadhyay, 1981; Anagnostakis, 1982; Mercier and Reeleder, 1985; Wilson and Lawrence, 1985; Boland and Hunter, 1988; Janisiewicz, 1988; Stretch, 1989). (ii) Mycoparasites, notably Trichoderma species, have likewise been shown to be antagonistic towards potential plant pathogens (Agrios, 1988). (iii) Growth of mycotoxin-producing fungi has been controlled by inoculations of grain products with non-toxic, but more competitive, fungal strains (Scott, 1989). (iv) Non-pathogenic fungi have been shown to induce defense responses that protect plants against pathogenic organisms (Bailey and Deverall, 1971; Keen, 1975; Bailey et al., 1980; Pirozynski and Hawksworth, 1988; Stretch, 1989). (v) Horizontal gene flow may occur between fungi and plants (Pirozynski, 1988). For instance, genes putatively transferred from Myrothecium fungi to the tropical shrub Baccharis are responsible for the plant’s ability to produce mycotoxins in the seed coat (Jarvis et al., 1989). Such interactions among plants, phylloplane, endophytic and potentially pathogenic micro-organisms are presently poorly understood, and virtually all information has been derived from applied horticultural studies (Pirozynski and Hawksworth, 1988; Pandey, 1990). The distinction between

54

M. L. CIPOLLINI A N D E. W. STILES

pathogens and mutualists may become blurred in nature, as intermediary and facultative interactions may be common (Carroll, 1988). Nevertheless, the potential exists for positive interactions among plants and microbial associates with respect to defense against herbivores and pathogens (Pirozynski and Hawksworth, 1988; Wicklow, 1988). The commonness of superficial yeast-like, bacterial, and latent (endophytic) infections of fleshy fruit suggests possible roles for these micro-organisms in the inhibition of predators and pathogens that may entirely destroy flowers or immature fruits (Newhook, 1957; Janisiewicz, 1988; Stretch, 1989; Cipollini, 1991; Wilson er al., 1991). If so, plants may tolerate infection by micro-organisms that provide fitness benefits through their negative effects upon more damaging biotic agents (Carroll, 1988).

3. Selection for Antimicrobial Defense of Ripe Fruit As discussed at the outset, selection pressure by frugivores on fruiting plants is likely to be due to broad-scale disperser selective pressure, rather than by species-specific coevolution. Although fruit-frugivore interactions appear to be complex in nature, numerous laboratory and field studies have demonstrated that vertebrates can be very selective when foraging. Animals commonly make foraging choices that reflect slight differences in nutrient chemistry, palatability, size, design, color, and presentation pattern (cf. Duncan, 1960; Turcek, 1963; Berthold, 1976; Sorensen, 1983, 1984; Levey et al., 1984; Blem and Shelor, 1986; Herbst, 1986; Bairlein, 1987; Levey, 1987a, b, c; Borowicz, 1988a, b; Chai, 1988; Levey and Karasov, 1989; Martinez del Rio et al., 1989; Roper and Cook, 1989; Whelan, 1989; Willson and Whelan, 1989; Willson et al., 1990; Buchholz and Levey, 1991; Cipollini, 1991; E. Stiles, unpublished). Despite these observations, we are still uncertain about the extent to which even broadly divergent selective pressures due to frugivore foraging preferences and patterns effect variation in chemical and physical characteristics of fleshy fruits. Like plant-animal interactions, plant-animal-fungus interactions in seed dispersal systems may be complex (Pirozynski and Hawksworth, 1988). Fruits that remain undispersed eventually succumb to microbial attack; like most defenses of plant organs, fruit defenses are often overcome by microbes in nature (Janzen, 1977; Stiles, 1980; Herrera, 1982a). Assuming that foraging animals make choices among available fruits, differential seed dispersal (fitness) among individual plants, resulting from differences in their susceptibility to fruit rot, should result in selection for protection from microbial degradation. Generally speaking, plants should be under selection pressure to avoid fruit rot in order to increase retention time, or to otherwise enhance dispersal by volent or arboreal dispersal agents.

RELATIVE RISKS OF MICROBIAL ROT FOR FLESHY FRUITS

55

4. Retention of Structural and Chemical Defenses Selection pressure for the avoidance of fruit rot can result in diverse adaptive defense mechanisms. According to Herrera (1 982a), there are four basic mechanisms for the defense of ripe fruit: (i) To ripen fruits when pest pressure is lowest. (ii) To reduce the exposure time of ripe fruits to damaging agents. (iii) To reduce the nutritive quality of tissue for pests and pathogens by providing unbalanced, or poor quality fruits. (iv) To retain some degree of structural or chemical defense. Because many traits of fleshy fruits, unlike those of most other plant parts, are thought to have evolved specifically for the attraction of vertebrates (Bate-Smith, 1972; Swain, 1978; but see Pirozynski, 1988), retention of toxic compounds during fruit ripening has been generally considered to be a compromise between optimal defense and optimal attraction (Dement and Mooney, 1974; Herrera, 1982a; but see Janzen, 1975). Plants may vary in their ability to recover defense chemicals from senescent leaf and stem tissues, and microbial growth may be affected by the residual chemical make-up long after abscission (Wong and Kwan, 1980; Bernays et al., 1989; Harper, 1989; Blair et al., 1990). Reduction or alteration of secondary defense constituents during fruit ripening may be a similar non-adaptive result of senescence. Alternatively, patterns of secondary chemistry in ripe fruits may be influenced by selective pressures relating to dispersal. If dispersal-related selective pressures are an important influence on defense traits of ripe fruit, then interspecific variation in defense should be related to broad differences in dispersal patterns, or “syndromes” (sensu van der Pijl, 1969). Janzen (1979, for instance, suggested that secondary chemicals within fruit tissue may act to “filter” out non-effective dispersers or seed predators from more functional dispersal agents. In the sense of this discussion, fruit-rot agents may be thought of as seed predators (Janzen, 1977; Janzen, 1979). Secondary chemistry of plant tissues may also vary with respect to extrinsic environmental conditions via proximate physiological mechanisms (Ballinger and Kushman, 1963; Devlin et al., 1969; Rossiter, 1969; Lawanson et al., 1971; Kushman and Ballinger, 1975; Camm and Towers, 1977; Margna, 1977; Chew and Rodman, 1979; McClure, 1979; Jones, 1984; Mancinelli and Rabino, 1984; Jensen, 1985). However, such variation may be a response to evolutionary pressures (Rhoades, 1979; Coley et al., 1985; Coley, 1986). For instance, Vanderplank (1984) suggests that the susceptibility of cultivated plants to pathogenic disease is primarily a result of artificial selection for high reproductive output, which results in tradeoffs with resistance factors. According to Vanderplank, the “low-sugar’’

56

M. L. CJPOLLJNI A N D E. W. STJLES

disease syndrome in cultivated species is caused by artificially high levels of reproduction that tax carbohydrate supplies, and thus make reproductive plants more susceptible to pathogens. When plants are further stressed due to poor growth conditions or damage, defenses against pathogens may be compromised. Vanderplank also suggests that under natural selective regimes plants are much more in balance with their disease agents, and thus rampant disease is rare. These observations indirectly support the hypothesis that defenses are costly, and that they should be expressed in direct relation to fitness benefits relative to those costs (sensu Rhoades, 1979).

IV. GENERAL HYPOTHESES AND PREDICTIONS

A. General Deterrent Nature of Fruit Rot Although anecdotal evidence suggests that animals may sometimes be attracted to rotting fruit (Janzen, 1977; Pirozynski and Hawksworth, 1988), the nutrient alteration and secondary metabolites produced by fruit-rot fungi should generally result in fitness costs to consumers (Janzen, 1977). When rotting ripe fruit, fungi should generally exhibit negative effects upon potential consumers (Borowicz, 198813; Buchholz and Levey, 1990; Cipollini, 1991). Nevertheless, fungi that cause rot may vary considerably in the degree of deterrence produced, and in the degree of defense allocated against them by plants. Although all fungi should demonstrate direct negative effects at the time fruit rot occurs, indirect positive effects upon plant fitness may result from fungi acting as biotic defenses against pathogens during flowering and green fruit stages, and against more deterrent fruit-rot fungi in the ripe fruit stage.

B. Microbe-specific Defenses Retention of chemical defense in ripe fruits has been considered a compromise of simultaneous selection for the maintenance of palatable and nontoxic fruit pulp for frugivores, and defense against pests and pathogens (Janzen, 1977; Herrera, 1982a). It is well known that immature fruits accumulate various classes of defensive compounds, many of which are catabolized, translocated, or detoxified by complexation during ripening (Goldstein and Swain, 1974; Dement and Mooney, 1974; Janzen, 1977, 1983; McKey, 1979; Eltayeb and Roddick, 1984; Roddick, 1987). Because secondary compounds are often toxic and deter vertebrates, retention of chemicals with these properties is considered to be maladaptive. It follows that chemicals with

RELATIVE RISKS OF MICROBIAL ROT FOR FLESHY FRUITS

57

antifrugivore effects should be at a low level in ripe fruit (McKey, 1979), with an accompanying increase in nutrients and attractants (e.g. non-bioactive colors, odors and flavors (van der Pijl, 1969; Batesmith, 1972; Wilson and Thompson, 1982; Newsome, 1990; Sinki and Schlegel, 1990). But as we have discussed, microbial pathogens likewise have the potential of reducing the palatability and increasing the toxicity of fruits to frugivores. Retention of features that have little or no effect upon dispersal agents (microbe-specific defense) should result from dispersal-driven selection for antimicrobial defense (Janzen, 1975, 1983). Simple examples of microbe-specific defenses include the concentration of defenses in fruit skin (for reasons addressed previously), osmotically high levels of organic metabolites, and waxy ‘blooms’ that promote water-shedding or otherwise inhibit fungal spore germination and penetration (Burkhardt, 1982; Janzen, 1983; Willson and Whelan, 1989). The evolution of microbe-specific toxins is also within the range of possible results (Janzen, 1983). Palatability is not a unique feature of a specific chemical or chemical group, but most likely a manifestation of selection in animals to recognize resources that provide large fitness benefits, with little or no toxic effects (Bate-Smith, 1972; Janzen, 1979; McKey, 1979). Thus, plant species under selection pressure to provide nutrient-rich fruits as a means of attracting frugivores, should likewise be under selection pressure to defend pulp with secondary chemicals having little or no negative effects to dispersers. Microbe-specific chemical defense should thus be likely in plants that require high palatability for effective dispersal, yet are at high risk of fungal attack due to intrinsic or environmental factors. Conversely, plant species that depend upon mimicry of high-quality species, or that are otherwise under low selection pressure to provide rich resources to obtain dispersal, may be under less selection for palatability in defense chemistry (Rhoades, 1979; Lundberg and Astrom, 1990).

C. Interspecific Variation in Defense Effectiveness Selective pressure for antifungal defense in ripe fruit can result in at least two outcomes: (a) quantitative increases during ripening in the retention of secondary compounds already present in immature fruit, or (b) qualitative chemical changes that arise de n o w during fruit maturation. Assuming that a response to selection in either direction may increase overall defense effectiveness, we propose two alternate models for predicting the magnitude of the result.

1 . Relative-risk Model This model assumes that variation in chemical and physical characteristics of vertebrate-dispersed plants primarily influences the relative risk of microbial

58

M. L. CIPOLLINI AND E. W. STILES

rot (potential rotting rate), and thus the degree of selection pressure for antimicrobial defense. Under this model, selection for antimicrobial defense should be highest for plants whose fruits are at high risk of rotting due to nutrient content, water content, season of ripening (environmental conditions), or ripening synchrony.

2. Removal-rate Model This model assumes that variation in chemical and physical characteristics primarily influences preferences and removal rates by frugivores, and thus the degree of selection for antimicrobial defense. Under this model, selection for antimicrobial defense should be high for fruits that are less preferred due to low nutrient quality, or are for other reasons removed slowly by frugivores. Being more “apparent” to fungal rot agents (sensu Feeny, 1973), such fruits are expected to exhibit a higher level of defense. While these models are not necessarily mutually exclusive, they provide an appropriate initial framework for making predictions concerning the relative influence of selective factors upon secondary defense of ripe fruits.

V. PREDICTIONS FOR TEMPERATE SEED DISPERSAL SYSTEMS

A. Temperate Fruiting Classes Fleshy-fruited plant species of eastern North America have fruiting patterns that are influenced by many factors, including temporal fluctuations in frugivore type and availability, and temperature, humidity, and moisture (Stiles, 1980; Stiles and White, 1982; Willson and Thompson, 1982; Rathcke and Lacey, 1985; White, 1989). While some plants have fruits available for dispersal by resident frugivores (summer and mid-winter), the majority of species ripen fruit during fall migration, when highly mobile birds are in need of high-energy food sources (Thompson and Willson, 1979; Stiles, 1980). Stiles (1980) and Stiles and White (1982) suggested that vertebratedispersed plants of the northeastern United States differ enough in ripening patterns, fruit design, and fruit nutrient quality to fall into four broad fruiting classes: (a) summer small-seeded ( S S ) , (b) summer large-seeded (SL), (c) fall high-quality (FQ), and (d) fall low-quality (FL). The characteristics and the significance of this temporal classification were extensively reexamined by White (1989). Table 1 provides a summary of the physical and chemical characteristics of these classes as determined in this reanalysis. Both Stiles (1980) and White (1989) included the majority of temperate species in the nutrient-poor FL class. Fruits of FL plants were reported to be slowly or sporadically removed by fall migrant, winter resident, or spring migrant

Table 1 Mean chemical and physical characteristics of fleshy fruits of temperate northeastern North America (adapted from Stiles, 1980; Stiles and White, 1982; White, 1989). Chemical data are mean estimates for water-soluble carbohydrates (CHO), petroleum ether-soluble lipids (lipid), and protein from micro-Kjeldahl analyses of total nitrogen (protein) presented as per cent of dry pulp mass Fruiting class Summer species: Small-seeded (SS) Large-seeded (SL) Fall high-quality (FQ) Fall low-quality species (FL): Herbaceous Deciduous Evergreen Waxy Sumac

Fruit mass (mg)

Pulp water

CHO

Lipid

(%)

(%I

(%I 0.4 29.7

590 515 248

3 89 81

86 81 65

62.8 49.5 20.2

375 317 321 26 18

58 36 48 15 10

87 76 62 13 14

26.5 42.7 40.2

1.1

1 .O

1.2 2.4 1.2 44.0

8.8

15.1

Protein

3.5 3.4 6.7 5.4 4.0 4.1

3.3 2.6

6 ZE

60

M . L. CIPOLLINI AND E. W . STILES

birds. The FL class includes such species as Ilex opaca, Crataegus crusgalli, Rhus fyphina and Smilax rotundifolia. A small group of other fall-fruiting species, including Cornus Jlorida, Lindera benzoin, Nyssa sylvatica, and Sassafras albidum, were placed in the FQ class. The FQ class was noted for rather synchronous ripening patterns coinciding with peak fall migration, and high-lipid pulp. The summer groups (SS and SL), notably Vaccinium spp., Gaylussacia spp., Rubus spp., and Prunus spp., are reportedly taken by summer resident birds and mammals, tend to ripen asynchronously, and contain high levels of carbohydrates and water. Within this chapter, we refer to this seasonal classification primarily to focus arguments concerning the potential influence of interspecific variation on selection for defense characteristics, and not in an attempt to evaluate the significance of the grouping per se.

B. Predictions for Temperate Species I . Deterrence ox and Defense against Fruit-rot Fungi Regardless of fruit nutrient chemistry or season of fruiting, fruit-rot fungi should always produce direct negative effects upon feeding by frugivores. Antifungal characteristics and chemical defense should thus be common in fruit tissues. However, fungi may vary considerably in their negative effects, and thus fruit defenses should be allocated in direct relationship to costs to seed dispersal resulting from particular fungal infections. Thus: (a) higher levels of defense should be allocated against agents causing loss of immature fruit, relative to those infecting ripe fruit, (b) antifungal defenses should fall for ripening fruit, as removal by frugivores and successful seed dispersal becomes probable, (c) within ripe fruit, defense should be directed primarily against fungi that are toxic and deterrent, with lower levels of defense directed towards agents with only slight deterrent effects, and (d) fungal species having indirect positive effects because of interactions with other micro-organisms or pests should be tolerated more than fungal species with strictly negative effects. In a study of temperate Ericaceous species bearing avian-dispersed fleshy fruits, we have accumulated data that generally support the prediction of variation among fungal species. Plant species examined in this study included three summer (SS) species: Vaccinium corymbosum (VC), V. vacillans (W), and Gaylussacia frondosa (GF), and three fall (FL) species: Vaccinium macrocarpon (VM), Caultheria procumbens (GP), and Arctostaphylos uva-ursi (AU). Based upon feeding trials in which surface-sterilized fruits were inoculated with a suite of fruit-rot fungi (Table 2), fruit rot was generally deterrent to avian frugivores (Fig. I ) . However, consumption of rotted fruits varied significantly among the fungal species used to inoculate the fruits (Fig. 2), with putatively toxic fungi being more deterrent to frugivores than

Table 2 Fungi used in fruit rot and antifungal tests. For identification of plant species, refer to text. “Field plate” refers to aerial spore collection plates placed in the field during the ripening season ~

Species name

Plant source

Symbol

Groupa

Alternaria tenuis Nees. Aspergillus niger Tiegh. Botrytis cinerea Pers.: Fr. Colletotrichum gloeosporioides De Vries Cladosporium cladosporioides Penz. & Sacc. Fusarium sporotrichioides Sherb. Geotrichum candidum Link. Penicillium spp. No. 65 Penicillium rubrum 0. Stoll Pestalotiopsis maculans Nag Raj Phoma vaccinii Dearn & House Rhizopus stolonifer Vuill. Saccharomyces cerevisiae Sacc. Phomopsis spp.

Rotted VC fruit Rotted VC fruit Rotted VC fruit Rotted VC fruit Rotted AU fruit Field plate Rotted G F fruit Rotted G F fruit Field plate Rotted VM fruit Rotted VM fruit Rotted G F fruit Rotted VC fruit Rotted VC fruit

ALT ASP BOT COL CLD FUS GEO PNU PNR PES PHM RHP SAC UNK

TOX TOX NON NON NON TOX NON TOX TOX NON NON NON NON TOX

Putative toxicity status (TOX= “toxic”, NON= “non-toxic”) based upon literature reports for each genus (Rodricks, 1976; Marasas and Nelson, 1987; Hsieh, 1989; Mills, 1989; Tanaka and Ueno, 1989; Yabe et al., 1989).

a

62

M. L. CIPOLLINI A N D E. W. STILES

SUMMER- SPECIES: AUG.-SEPT. 1986

0

5

ROlTEDFRUlT INTACT FRUIT

4

3

A

2

F

1

a -

2

0

W

+ a t 3 a

5

LL LL

4

z

GF

FALL SPECIES: N0V.-DEC. 1986

W

0 0

vc

3

ROTTEDFRUIT

1

INTACT FRUIT

a

2 1

0 VM

AU

PLANT SPECIES Fig. 1. Effect of artificially induced fruit rot on consumption of summer (SS) and fall (FL) Ericaceous fruits by captive birds in 1986. Data are mean numbers of fruits consumed in pairwise (five rotted: five intact) 15-min feeding trials with summer fruit species using catbirds and veeries, and fall fruit species using robins, across nine fungal species (n=6-16 replicates of each fruit: fungus combination). The effect of fruit state (rotted vs. intact) was significant in full factorial ANOVA for both summer (F= 1354.15, P < 0.0001) and fall (F=425.83, P < 0.0001) fruits. Identical letters denote means that did not differ significantly ( P > 0.05) based upon Bonferroni Ttests. For key to plant species, refer to text.

z W

2 w P

w

t

0 U

FUNGAL SPECIES Fig. 2. Proportion of rotted fruit eaten by captive birds during feeding trials in 1986 (refer to Fig. I ) , showing differences among fungal species. Results are means for each fungus across all birds. The effect of fungal species was significant in full factorial ANOVA for both summer (F= 179-56, PO.O5) based upon Bonferroni T-tests. For key to fungal species, see Table 2.

M. L. CIPOLLINI AND E. W. STILES

* -r

“TOXIC” FUNGI

* T

vc

GF

“NONTOXIC’ FUNGI

AU

VM

ss

FL

PLANT SPECIES Fig. 3. The effect of fungal toxicity status on consumption of rotted Ericaceous fruits by captive birds in 1986 (refer to Fig. 1). Data are mean numbers of rotted fruits eaten for fruit-rot fungi classified according to toxicity status (refer to Table 2) across all birds. The effect of fungal toxicity status was significant in nested ANOVA (i.e. toxicity status, with fungal species nested within toxicity status) for summer species ( F = 4.99, P= 0.048) and for fall species ( F = 5.19, P=0.0307). Asterisks denote significant ( P 6 0.05) differences between means for “toxic” and “non-toxic” fungi within plant species based upon one-way ANOVA.

putatively non-toxic species (Fig. 3). And, based on fungal growth trials using artificial fruit-pulp media, the antifungal activity of secondary extracts of ripe fruit was greater against the toxic fungi, relative to the non-toxic group (Fig. 4).

2. Microbe-specific Defense Microbe-specific defense should be most advantageous for fall fruits with unusually high lipid and/or sugar content (FQ fruits). For such high-quality fruit, increased persistence may increase the probability of dissemination-

65

RELATIVE RISKS OF MICROBIAL ROT FOR FLESHY FRUITS

loo n

8

1

0

*

80

W

-

r

“TOXIC” FUNGI “NONTOXIC’ FUNGI

*

I

5

60

0

K

*

4

(3

w

>

40

-

20

-

F

a

J

w

K 0-

vc

VV

GF

VM

ss

AU

GP

FL

PLANT SPECIES Fig. 4. The effect of fungal toxicity status on the degree of antifungal activity present in secondary extracts of ripe fruit pulp. Data are mean radial mycelial growth of “toxic” and “non-toxic” fungi on media containing secondary ethanolic extracts (test agars), relative to growth on media (control agars) that mimicked the nutrient makeup of each fruit species (n = 2 replicates for each fruit: fungus pair). The effect of fungal toxicity status was significant in nested ANOVA (i.e. toxicity status, with fungal species nested within toxicity status: F = 6.33, P
but only if fruit palatability remains high. Sugary, watery fruits disseminated by territorial residents during times of resource abundance (SS and SL fruits) are similarly predicted to exhibit highly microbe-specific defense patterns. Variation in fruit quality may be an important determinant of seed dispersal for each of the above groups, especially when competition for available dispersers is high (Foster, 1977; Howe and Estabrook, 1977; Stiles, 1980; Stapanian, 1982; Rathcke and Lacey, 1985). Conversely, species that ripen fruit in late fall and winter, and whose fruits contain low amounts of nutritive

66

M. L. CIPOLLINI AND E. W. STILES

material and/or water (FL species), are less likely to exhibit microbe-specific defense. For these species, palatability may be relatively unimportant for dispersal, and some may even rely on mimicry of higher quality species for effective dispersal (White and Stiles, 1990; Willson and Whelan, 1990a; E. Stiles, unpublished). Fruits under low selection for palatability are more likely to retain general (potentially antifrugivorous) chemicals. Evidence supporting this hypothesis includes that of White and Stiles (1990), who found that diet mixing (plant species co-occurrences) for frugivorous birds in New Jersey was highest when feeding upon low-quality (FL) species during the late fall and winter, despite a substantial decrease in the species richness of fruiting plants relative to other seasons. One explanation for this result is that frugivores must mix diets in order to avoid toxic or nutritional effects that could result from constant feeding upon one or a few species (Izhaki and Safriel, 1989, 1990; Mack, 1990; Sedinger, 1990), although an alternative explanation is that birds must mix diets to balance nutrient intake. Levey and Karasov (1989) similarly suggested that observed switches in preference patterns of captive birds fed FL fruits may have resulted from avoidance responses to toxins within initially preferred species. And, based upon artificial fruit-pulp feeding trials using our suite of temperate Ericaceous plants, secondary pulp chemistry of the three lowquality (FL) species was considerably more deterrent to avian frugivores than that of the three higher quality (SS) species (Fig. 5).

3. Interspecijk Variation in Pulp Antifungal Activity Two alternative predictions are forwarded, following directly from the two models of selection for defense described previously:

( a ) Relative-risk model. Fruits with high risk of microbial degradation due to season of ripening or nutrient-rich fruit pulp (FQ, SS and SL species) should employ more effective antifungal defenses than fruits with a low risk of microbial degradation (FL species).

( b ) Removal-rate model. Plants that bear fruits with physical and chemical characteristics or ripening patterns that result in high removal rates should devote fewer resources to antifungal defense. Fruits with high-quality pulp (FQ, SS and SL species) may be more profitable to birds, or are for other reasons taken more quickly upon ripening, and should be under less selection for antifungal defense than lower quality fruits (FL species). Note that the removal-rate model leads to exact opposite predictions for the temperate fruit classes in comparison to the previous (relative-risk) model. In our study of temperate Ericaceous plants, we have found evidence supporting the removal-rate model. While not apparently differing from SS species in intrinsic risk factors (relative-risk), fruits of FL species were eaten at relatively lower rates in feeding trials with captive birds, and were more

67

RELATIVE RISKS OF MICROBIAL ROT FOR FLESHY FRUITS

a

a

a

41

'1

0

TESTAGARS CONTROLAGARS

b

C

vc

GF

VM

ss

AU FL

PLANT SPECIES Fig. 5. The relative palatability of artificial fruit-pulp media containing secondary methanolic extracts of ripe Ericaceous fruits (test agars) with respect to that of control media that mimicked the nutrient makeup of each fruit species (control agars). Results are mean feeding rates for 15-min pairwise trials using captive catbirds, wood thrushes and robins (n = 12 replicate trials with each plant species/agar type combination), expressed across all birds combined. The effect of agar type (test vs. control) was significant in full factorial ANOVA (F=25.12, P=OWOl), with a significant plant species x agar type interaction (F= 137.1 1, P = 0.0001). Identical letters denote means that did not differ significantly (P> 0.05) based upon Bonferroni T-tests.

resistent to fungal rot in the field (Cipollini, 1991). Fruits of FL species not only contained secondary chemicals that were more deterrent to frugivores (as discussed previously), but based upon fungal growth trials using artificial fruit-pulp media, FL fruits also contained a much higher degree of antifungal activity (Fig. 4).

VI. CONCLUSIONS Aside from the recently described coadaptive interactions involving Clavicipitaceous endophytic fungi, grasses, and their herbivores (Clay, 1988a, b), the Moniliniu (Mummy-berry) system involving Ericaceous plants (Batra,

68

M. L. CIPOLLINI AND E. W. STILES

1983; Batra and Batra, 1985), and plant-Drosophilu-yeast complexes (Begon, 1982; Starmer and Fogelman, 1986; Starmer et ul., 1990), few data exist concerning patterns and dynamics of plant-animal-fungus evolutionary interactions in the wild (Pirozynski and Hawksworth, 1988; Barbosa et al., 1991). Evolutionary data necessary for the determination of coadaptive patterns are lacking for even the well-described and ecologically important mycorrhizal fungi and their plant associates (Pirozynski and Malloch, 1988). The simple predictions forwarded in this chapter provide an initial platform for addressing, from an evolutionary perspective, the area of plant-frugivore-fungus interactions in seed-dispersal systems. These predictions are based largely upon assumptions of how plants and microbes should react to various selective pressures, and are therefore subject to re-evaluation if certain assumptions are incorrect, or if certain factors are overriding in importance. It is therefore necessary to address more specifically these assumptions, and to weigh various factors, before making specific predictions for particular plant-frugivore-fungus interactions.

ACKNOWLEDGEMENTS We would like to thank C. Augspurger, A. Brash, L. Carris, S. Cipollini, E. DeVito, C. Frenkel, M. Gargiullo, S. Handel, C. Herrera, D. Levey, P. Morin, J. Pickering, A. Stretch, D. White, M. Willson, and an anonymous reviewer for constructive criticism and comments concerning the ideas and hypotheses presented in this chapter. D. Levey and D. White kindly provided relevant pre-prints. This work was supported in part by Hutcheson Memorial Forest Fellowships and Steinetz-Leathem-Stauber Awards, Rutgers Department of Biological Sciences, to MLC, and a Bureau of Biological Research Grant, Rutgers Department of Biological Sciences, to EWS. Final work on this chapter was made possible through facilities provided by the Smithsonian Environmental Research Center, and a Smithsonian postdoctoral fellowship to MLC.

REFERENCES Agrios, G. N. (1988). Plant Pathology. Academic Press, San Diego. Ainsworth, G. C. (1971). Ainsworth and Bisby’s Dictionary of the Fungi, 7th Edn. Commonwealth Mycological Institute, Kew, Surrey. Albersheim, P. and Valent, B. S. (1978). Host-pathogen interactions in plants: plants when exposed to oligosaccharides of fungal origin, defend themselves by accumulating antibiotics. J . Cell Biol. 78, 627-643. Allison, A. V. and Shalla, T. A. (1973). The ultrastructure of local lesions induced by potato virus x: a sequence of cytological events in the course of infection. Phytopathology 64, 784-793.

RELATIVE RlSKS OF MICROBIAL ROT FOR FLESHY FRUITS

69

Anagnostakis, S. L. (1982). Biological control of chestnut blight. Science 215, 4 6 6 471. Andersen, 0. M . (1985). Chromatographic separation of anthocyanins in Cowberry (Lingonberry) Vaccinium vitis-idaea L. J. Food. Sci. 50, 123c1232. Andersen, 0.M. (1987). Anthocyanins in fruits of Vaccinium japonicum. Phytochemistry 26, 122g1221. Anderson, A. J. (1978). Isolation from three species of Colletotrichium of glucancontaining polysaccharides that elicit phytoalexin production in bean. Phytopatho b g y 68, 189-1 94. Anderson-Prouty, A. J. and Albersheim, P. (1975). Host-pathogen interactions VIII. Isolation of a pathogen-synthesized fraction rich in glucan that elicits a defense reaction in the host. Plant Physiol. 56, 286291. Aoki, T., Tanio, Y. and Suga, T. (1976). Triterpenoid saponins from Fatsia japonica. Phytochemistry 15, 781-784. Applebaum, S. W. and Birk, Y . (1979). Saponins. In: Herbivores- Their Interaction with Secondary Plant Metabolites (Ed. by G. A. Rosenthal and D. H. Janzen), pp. 53S566. Academic Press, New York. Arauz, L. F. and Sutton, T. B. (1990). Effect of interrupted wetness periods on spore germination and apple infection by Botryosphaeria obtusa. Phytopathology 80, 1218-1 233. Asquith, T. N. and Butler, L. G. (1986). Interactions of condensed tannins with selected proteins. Phytochemistry 25, 1591-1 593. Augspurger, C. K. (1 984). Seedling survival of tropical tree species: interactions of dispersal distance, light-gaps, and pathogens. Ecology 65, 1705-1 7 12. Bailey, J. A. and Deverall, B. J. (1971). Formation and activity of phaseolin in the interaction between bean hypocotyls (Phaseolus vulgaris) and Colletotrichum lindemuthianum. Physiol. Plant Pathol. 1, 4 3 5 4 9 . Bailey, J. A., Rowell, P. M. and Arnold, G. M. (1980). The temporal relationship between host cell death, phytoalexin accumulation and fungal inhibition during hypersensitive reactions of Phaseolus vulgaris to Colletotrichum lindemuthianum. Physiol. Plant Pathol. 17, 329-339. Bairlein, F. (1987). Nutritional requirements for maintenance of body weight and fat deposition in the long-distance migratory garden warbler, Sylvia borin (Boddaert). Comp. Biochem. Physiol. 86A, 337-347. Baj, A.,Bombardelli, E., Gabetta, B. and Martinelli, E. M. (1983). Qualitative and quantitative evaluation of Vaccinium myrtillus anthocyanins by high-resolution gas chromatography and high-performance liquid chromatography. J. Chromatogr. 279,365-372. Baldwin, I. T. (1988). Short-term damage-induced increases in tobacco alkaloids protect plants. Oecologia (Berl.) 75, 367-370. Baldwin, I. T. and Schultz, J. C. (1984). Rapid induced changes in tree leafchemistry and evidence of communication between plants. Science 221,277-279. Ballinger, W. E. and Kushman, L. J. (1963). Influence of crop load and nitrogen applications upon yield and fruit qualities of Wolcott blueberries. J . Amer. SOC. hort. Sci. 82,264-276. Ballinger, W. E. and Kushman, L. J. (1966). Factors influencing the mineral-element content of leaves and fruit of Wolcott blueberries. J . Amer. SOC.hort. Sci.88, 325330. Ballinger, W. E. and Kushman, L. J. (1970). Relationship of stage of ripeness to composition and keeping quality of highbush blueberries. J. Amer. hort. SOC.95, 239-242.

70

M. L. ClPOLLINl AND E. W. STILES

Ballinger, W. E., Maness, E. P., Galleta, G. J. and Kushman, L. J. (1972). Anthocyanins of ripe fruit of a “Pink-fruited’’ hybrid of highbush blueberries, Vaccinium corymbosum L. J . Amer. SOC.hort. Sci. 91, 381-384. Ballinger, W. E., Maness, E. P. and McClure, W. F. (1978). Relationship of stage of ripeness and holding temperature to decay development of blueberries. J. Amer. hort. SOC.103, 13G134. Ballinger, W. E., Galleta, G . J. and Maness, E. P. (1979). Anthocyanins of fruits of Vaccinium, sub-genera Cyanococcus and Polycodium. J . Amer. SOC.hort. Sci. 104, 554-557. Ballington, J. R., Ballinger, W. E. and Maness, E. P. (1987). Interspecific differences in the percentage of anthocyanins, aglycones and aglycone-sugars in fruit of seven species of blueberries. J. Amer. SOC.hort. Sci. 11, 859-864. Ballington, J. R., Ballinger, W. E. and Maness, E. P. (1988a). Anthocyanin, aglycone and aglycone-sugar content of fruits of temperate North American species of Gaylussacia (Ericaceae). Can. J . Plant Sci. 68, 247-253. Ballington, J. R., Ballinger, W. E., Maness, E. P. and Luby, J. J. (1988b). Anthocyanin, aglycone and aglycone-sugar content of the fruits of five species of Vaccinium Section Myrtillus. Can. J. Plant Sci. 68, 241-246. Ballington, J. R., Kirkman, W. B., Ballinger, W. E. and Maness, E. P. (1988~). Anthocyanins, aglycone, and aglycone-sugar content in the fruits of temperate North American species of four Sections in Vaccinium. J. Amer. SOC.hort. Sci. 113, 746749. Bandyopadhyay, C., Narayan, V. S. and Variyar, P. S. (1990). Phenolics of green pepper berries (Piper nigrum L.). J . Agric. Food Chem. 38, 16961699. Barbosa, P., Krischnik, V. A. and Jones, C. G. (Eds) (1991). Microbial Mediation of Plant-Herbivore Interactions. John Wiley, New York. Bannore, C. R. and Ngyen, T. K. (1985). Polygalacturonase inhibition in rind of Valencia orange infected with Diplodia natalensis. Phytopathology 75, 44-9. Barnea, A., Yom-Tov, Y. and Friedman, J. (1991). Does ingestion by birds affect seed germination? Funct. Ecol. 5, 394-402. Barnett, H. L. and Hunter, B. B. (1972). Illustrated Genera of Imperfect Fungi. Burgess, Minneapolis. Bate-Smith, E. C. (1972). Attractant and repellents in higher animals. In: Phytochemical Ecology (Ed. by J. B. Harborne), pp. 45-56. Academic Press, New York. Bate-Smith, E. C. (1977). Astringent tannins of Acer species. Phytochemistry 16, 1421-1426. Batra, L. R. (1 983). Monilinia vaccinii-corymbosi (Sclerotiniaceae): its biology and comparison with related species. Mycologia 75, 131-1 52. Batra, L. R. and Batra, S. W. T. (1985). Floral mimicry induced by mummy-berry fungus exploits host’s pollinators as vectors. Science 228, 1011-1012. Beehler, B. M. (1986). Patterns of frugivory and the evolution of birds-of-paradise. Acta Congr. Intern. Ornith. XIX, 816826. Begon, M. (1982). Yeasts and Drosophila. In: The Genetics and Biology of Drosophila, Vol. 3B (Ed. by M. Ashburner, H. L. Carson, and J. N. Thompson, Jr), pp. 345384. Academic Press, New York. Benson, W. W., Brown, K. S. and Gilbert, L. E. (1975). Coevolution of plants and herbivores: Passion Flower butterflies. Evolution 29, 659-680. Berenbaum, M. R. (1981). Patterns of furanocoumarin distribution and insect herbivory in the Umbelliferae: plant chemistry and community structure. Ecology 62, 12541266. Berenbaum, M. R. and Zangerl, A. R. (1988). Stalemates in the coevolutionary arms

RELATIVE RISKS OF MICROBIAL ROT FOR FLESHY FRUITS

71

race: syntheses, synergisms, and sundry other sins. In: Chemical Mediation of Coevolution (Ed. by K. C. Spencer), pp. 113-132. Academic Press, New York. Berenbaum, M. R., Nitao, J. K. and Zangerl, A. R.(1991). Adaptive significance of furanocoumarin diversity in Pastinaca sativa (Apiaceae). J . chem. Ecol. 17, 207215. Bernays, E. A. and Janzen, D. H. (1988). Saturniid and sphingid caterpillars: two ways to eat leaves. Ecology 69, 1153-1 160. Bernays, E. A,, Driver, G. C. and Bilgener, M. (1989). Herbivores and plant tannins. Adv. ecol. Res. 19, 263-302. Berthold, P. (1976). The control and significance of animal and vegetable nutrition in omnivorous songbirds. Ardea 64, 14G153. Biale, J. B. (1975). Allocution: internal regulation of fruit maturation. In: Facteurs et Regulation de la Maturation des Fruits (Ed. by R. Ulrich), pp. 367-368. Centre National de la Recherche Scientifique, Anatole. Biggs, A. R. and Northover, J. (1988). Influence of temperature and wetness duration on infection of peach and sweet cherry fruits by Monilinia fruticola. Phytopathology 78, 1352-1356. Bilai, V. I. (1963). Antibiotic-producing Microscopic Fungi. Elsevier, Amsterdam. Blair, J. M., Parmalee, R. W. and Beare, M. H. (1990). Decay rates, nitrogen fluxes, and decomposer communities of single- and mixed-species foliar litter. Ecology 71, 1976-1985. Blakeman, J. P. and Fokema, N. J. (1982). Potential for biological control of plant diseases on the phylloplane. Ann. Rev. Phytopathol. 20, 167-192. Blanke, M. M. and Lenz, F. (1989). Fruit photosynthesis. Plant Cell Envir. 12,3146. Blem, C. R. and Shelor, M. H. (1986). Multiple regression analyses of midwinter fattening of the white-throated sparrow. Can. J. Zool. 64, 2405-241 1. Blieler, J. A., Rosenthal, G. A. and Janzen, D. H. (1988). Biochemical ecology of canavanine-eating seed predators. Ecology 69, 427433. Boland, F. E., Blomquist, V. H. and Estrin, B. (1968). Chemical composition of fruits. J. Amer. Org. Agric. Chem. 51, 1203-1209. Boland, G. J. and Hunter, J. E. (1988). Influence of Alternaria alternata and Cladosporium cladosporioides on white mold of bean caused by Sclerotinia sclerotiorum. Can. J . Plant Pathol. 10, 172-177. Borowicz, V. A. (1988a). Fruit consumption by birds in relation to fat content of pulp. Amer. Midl. Nut. 119, 121-127. Borowicz, V. A. (1988b). Do vertebrates reject decaying fruit?: an experimental test with Cornus arnomum fruit. Oikos 53, 74-78. Brady, C. J. (1987). Fruit ripening. Ann. Rev. Plant Physiol. 38, 155-178. Briggs, M. A. and Schultz, J. C. (1990). Chemical defense production in Lotus corniculatus L. 11. Trade-offs among growth, reproduction and defense. Oecologia (Berl.) 83, 32-37. Bruce, R. J. and West, C. A. (1982). Elicitation of casbene synthetase activity in castor bean. Plant Physiol. 69, 11 8 1-1 188. Bruinisma, J., Knegt, E. and Varga, A. (1975). The role of growth-regulating substances in fruit ripening. In: Facteurs et Regulation de la Maturation des Fruits (Ed. by R. Ulrich), pp. 193-200. Centre National de la Recherche Scientifique, Anatole. Buchholz, R. and Levey, D. J. (1990). The evolutionary triad of microbes, fruits, and seed dispersers: an experiment in fruit choice by cedar waxwings, Bombycilla cedrorum. Oikos 59, 20G204. Buckley, R. J. and Halisky, P. M. (1990). Feeding deterrence of the corn leaf aphid by

72

M. L. CIPOLLINI AND E. W. STILES

an Acremonium endophyte in perennial ryegrass. (Abstr.) Bull. New Jersey Acad. Sci. 35, 36. Burdon, J. J. and Marshall, D. R. (1981). Inter- and intra-specific diversity in the disease response of Glycine species to the leaf rust fungus Phakopsora pachyrhizi. J . E c o ~69, . 381-390. Burkhardt, D. (1982). Birds, berries and UV. Naturwissenschaften 69, 153-157. Butler, L. G. (1989). Effect of condensed tannin on animal nutrition. In: Chemistry and Signijicance of Condensed Tannins (Ed. by R. W. Hemingway and J. J. Karchesy), pp. 391402. Plenum Press, New York. Camazine, S. M., Resch, J. F., Eisner, T. and Meinwald, J. (1983). Mushroom chemical defense: pungent sesquiterpenoid dialdehyde antifeedant to opossum. J . chem. Ecol. 9, 1439-1447. Camm, E. L. and Towers, G. H. N. (1977). Phenylalanine ammonia lyase. Prog. Phytochem. 4, 169-188. Cappellini, R. A. and Ceponis, M. J. (1977). Vulnerability of stem-end scars of blueberry fruits to post-harvest decays. Phytopathology 67, 1 18-1 19. Cappellini, R. A., Stretch, A. W. and Maiello, J. M. (1972). Fungi associated with blueberries held at various storage times and temperatures. Phytopathology 62, 6849. Cappellini. R. A., Ceponis, M. J. and Koslow, G. (1982). Nature and extent of losses in consumer-grade samples of blueberries in greater New York. HortScience 17, 55-56. Cappellini, R. A., Ceponis, M. J. and Schulze, C. P. Jr. (1983). The influence of “sweating” on post-harvest decay of blueberries. Plant Dis. 67, 38 1-382. Carroll, G. (1988). Fungal endophytes in stems and leaves: from latent pathogen to mutualistic symbiont. Ecology 69, 2-9. Ceponis, M. J. and Stretch, A. W. (1983). Berry color, water immersion time, rot, and physiological breakdown of cold-stored cranberry fruits. HortScience 18,484485. Chai, P. (1988). Wing coloration of free-flying neotropical butterflies as a signal learned by a specialized avian predator. Biotropica 20, 20-30. Chapman, R. F. and Blaney, W. M. (1979). How animals perceive secondary compounds. In: Herbivores- Their Interaction with Secondary Plant Metabolites (Ed. by G. A. Rosenthal and D. H. Janzen), pp. 161-199. Academic Press, New York. Chew, F. S. and Rodman, J. E. (1979). Plant resources for chemical defense. In: Herbivores- Their Interaction with Secondary Plant Metabolites (Ed. by G. A. Rosenthal and D. H. Janzen), pp. 271-308. Academic Press, New York. Chirboga, C. and Francis, F. J. (1970). An anthocyanin recovery system from cranberry pomace. J. Amer. SOC.hort. Sci. 95, 233-236. Christensen, M. and Tuthill, D. E. (1986). Aspergillus: an overview. In: Advances in Penicillium and Aspergillus Systematics (Ed. by R. A. Sampson, R. A. and J. I. Pitt), pp. 195-210. Plenum Press, New York. Cihlar, R. L. and Hoberg, K. A. (1987). Use of cerulenin and butyrate in the study of Candida albicans germination. In: Ecology and Metabolism of Plant Lipids (Ed. by G. Fuller and W. D. Nes), pp. 355-365. American Chemical Society, Washington, DC. Cilliers, J. J. L. and Singleton, V. L. (1990). Autooxidative phenolic ring opening under alkaline conditions as a model for natural polyphenols in foods. J . Agric. Food Chem. 38, 1797-1798. Cipollini, M. L. (1991). Chemical Defense of Ripe Fruits: Antifungal Defense Versus Frugivore Attraction in the Family Ericaceae. PhD thesis, Rutgers The State University of New Jersey, New Brunswick.

RELATIVE RISKS OF MICROBIAL ROT FOR FLESHY FRUITS

73

Clague, J. A. and Fellers, C. R. (1934). Relation of benzoic acid content and other constituents of cranberry to keeping quality. Plant Physiol. 9, 631-636. Clausen, T. P., Provensa, F. D., Burritt, E. A., Reichardt, P. B. and Bryant, J. P. (1990). Ecological implications of condensed tannin structure: a case study. J . chem. Ecol. 16, 2381-2392. Clay, K. (1988a). Clavicipitaceous fungal endophytes of grasses: coevolution and the change from parasitism to mutualism. In: Coevolution sf Fungi with Plants and Animals (Ed. by K. A. Pirozynski and D. L. Hawksworth), pp. 79-105. Academic Press, London. Clay, K. (1988b). Fungal endophytes of grasses: a defensive mutualism between plants and fungi. Ecology 69, 10-16. Cohen, E. and Schiffman-Nadel, M . (1971). Respiration pattern of lemon fruit infected with Phytophthora citrophora. Phytopathology 62, 932-933. Cohen, J. D. and Bandurski, R. S. (1982). Chemistry and physiology of the bound auxins. Ann. Rev. Plant Physiol. 33, 403430. Coley, P. D. (1986). Costs and benefits of defense by tannins in a neotropical tree. Oecologia (Berl.) 70, 238-24 1. Coley, P. D., Bryant, J. P. and Chapin, F. S. 111. (1985). Resource availability and plant antiherbivore defense. Science 230, 895-899. Conn, E. E. (1979). Cyanide and cyanogenic glycosides. In: Herbivores- Their Interaction with Secondary Plant Metabolites (Ed. by G . A. Rosenthal and D. H. Janzen), pp. 387412. Academic Press, New York. Connell, J. H. (1971). On the role of natural enemies in preventing competitive exclusion in some marine animals and in rain forests. In: Dynamics ofPopularion (Ed. by P. J. Den Boer and G . R. Gradwell), pp. 298-312. Pudoc, Wageningen. Constantine, G . H., Catalfomo, P., Sheth, K. and Sciuchetti, L. A. (1966). Phytochemical investigation of Arctostaphylos colurnbiana Piper and Arctostaphylos patula Greene (Ericaceae). J . pharm. Sci. 55, 1378-1382. Cooke, R. J. and Baker, K. F. (1983). The Nature and Practice of’Biologica1 Control of Plant Pathogens. American Phytopathological Society, St Paul. Cooke, W. B. (1979). The Ecology qfFungi. C R C Press, Boca Raton, FL. Cooper, R. M . and Wood, R. K. S. (1975). Regulation of cell wall degrading enzymes by Verticillium albo-atrum and Fusarium o.xysporum f. sp. lycopersici. Plant Physiol. 5. 135-156. Crane, J. C. (1964). Growth substances in fruit setting and development. Ann. Physiol. Plant Pathol. 15, 303-326. Crawley, M . J. (Ed.) (1984). Herbivory- The Dynamics qfAnimal-Plant Interactions. University of California Press, Berkeley. Croteau, R. (1977). Biosynthesis of benzaldehyde, benzyl alcohol, and benzyl benzoate from benzoic acid in cranberry (Vaccinium macrocarpon). J . Food Biochem. 1, 3 17-326. Croteau, R. and Fagerson, I. S. (1971). Volatile substances of the cuticular wax of cranberry. Phytochemistry 10, 3239-3245. Cruikshank, I. A. M. and Perrin, D. R. (1964). Pathological function of phenolic compounds in plants. In: Biochemistry of Phenolic Compounds (Ed. by J. B. Harborne), pp. 51 1-544. Academic Press, New York. Darvill, A. G . and Albersheim, P. (1984). Phytoalexins and their elicitors-a defense against microbial infection in plants. Ann. Rev. Plant Physiol. 35, 243-275. Dashwood, E. P. and Fox, R. A. (1988). Infection of flowers and fruits of red raspberry by Botrytis cinerea. Plant Pathol. 37, 423430. Davidar, P. (1983). Birds and neotropical mistletoes: effects on seedling recruitment. Oecologia (Berl.) 60, 271-273.

74

M. L. CIPOLLINI A N D E. W. STILES

Davidson, P. M. and Juneja, V. K. (1990). Antimicrobial agents. In: Food Additives (Ed. by A. L. Branen, P. M. Davidson and S. Salminen), pp. 83-137. Marcel Dekker, New York. Daykin, M. E. (1984). Infection of blueberry fruit by Colletotrichum gloeosporioides. Plant Dis. 68, 948-950. Daykin, M. E. and Milholland, R. D. (1990). Histopathology of blueberry twig blight caused by Phomopsis vaccinii. Phytopathology 80, 736740. DeBussche, M., Cortez, J. and Rimbault, I. (1987). Variation in fleshy fruit composition in the Mediterranean region: the importance of ripening season, lifeform, fruit type, and geographical distribution. Oikos 49, 244-252. Della-Bianca, L. (1979). Fruit production by summer grape in the southern Appalachians. J. Wildl. Manage. 43, 579-583. Dement, W. A. and Mooney, H. A. (1974). Seasonal variation in the production of tannins and cyanogenic glucosides in the chaparral shrub, Heteromeles arbutifolia. Oecologia (Berl.) 15, 65-76. Dennis, C. (1983). Soft fruits. In: Post-harvest Pathology of Fruits and Vegetables (Ed. by C. Dennis), pp. 2 3 4 2 . Academic Press, London. Denny, T. P. and Van Etten, H. D. (1983). Characterization of an inducible, nondegradative tolerance of Nectria haematococca MPVI to phytoalexins. J. gen. Microbiol. 129, 2903-2913. Denslow, J. S. (1987). Tropical rainforest gaps and tree species diversity. Ann. Rev. ecol. Syst. 18, 431451. Devlin, R. M., Zuckerman, B. M. and Demonarville, I. E. (1969). Influence of preharvest application of malathion and indole-3-acetic acid on anthocyanin development in Vaccinium macrocarpon, var. “Early Black”. J. Amer. Soc. hort. Sci. 94, 52-55. DeWit, P. J. and Kodde, E. (1981). Induction of plyacetylenic phytoalexins in Lycopersicon esculentum after inoculation with Cladosporium fulvum. Physiol. Plant Pathol. 18, 143-148. DeWit, P. J. and Roseboom, P. H. (1980). Isolation, partial characterization, and specificity of glycoprotein elicitors from culture filtrates, mycelium and cell walls of Cladosporium fulvum. Physiol. Plant Pathol. 16, 391408. Dhingra, 0. D. and Sinclair, J. B. (1985). Basic Plant Pathology Methods. CRC Press, Boca Raton, FL. Du, C.-T. and Francis, F. J. (1973). Anthocynanins from Cornus mas. Phytochemistry 12, 2487-2489. Duncan, C. J. (1960). The sense of taste in birds. Annul. appl. Biol. 48, 409414. Dye, M. H. and Vernon, T. R. (1952). Air-borne mould spores: a New Zealand survey. New Zealand J. Sci. Technol. Sept. 1952, 118-127. Eck, P. (1988). Blueberry Science. Rutgers University Press, New Brunswick. Eck, P. (1990). The American Cranberry. Rutgers University Press, New Brunswick. Edwards, D. (1988). Berry good and nutty anticancer agent. Sci. News 133, 216. Eltayeb, E. A. and Roddick, J. G. (1984). Changes in the alkaloid content of developing fruits to tomato (Lycopersiconesculentum Mill). 1. Analysis of cultivars and mutants with differing ripening characteristics. J. exper. Bot. 35, 252-260. Esquerre-Tugaye, M.-T., Lafitte, C., Mazau, D., Toppan, A. and Touse, A. (1979). Cell surfaces in plant-microorganism interactions. 11. Evidence for the accumulation of hydroxyproline-rich glycoproteins in the cell wall of diseased plants as a defense mechanism. Plant Physiol. 64, 32Ck326. Faeth, S. H. (1985). Quantitative defense theory and patterns of feeding by oak insects. Oecologia (Berl.) 68, 34-40,

RELATIVE RISKS OF MICROBIAL ROT FOR FLESHY FRUITS

75

Farr, D. F., Bills, G. F. Chamaris, G. P. and Rossman, A. Y. (1989). Fungi on Plants and Plant Products in the United States. American Phytopathology Society, St Paul. Feeny, P. P. (1968). Effect of oak leaf tannins on larval growth of the winter moth Operophthera brumata. J. Insect Physiol. 14, 805-8 17. Feeny, P. P. (1969). Inhibitory effect of oak leaf tannins on the hydrolysis of proteins by trypsin. Phytochemistry 8, 2 1 19-2 126. Feeny, P. P. (1973). Biochemical coevolution between plants and their insect herbivores. In: Coevolution of Animals and Plants (Ed. by L. E. Gilbert and P. H. Raven), pp. 3-19. University of Texas Press, Austin. Fleming, T. H. (1986). Opportunism versus specialization: the evolution of feeding strategies in frugivorous bats. In: Frugivores and Seed Dispersal (Ed. by A. Estrada and T. H. Fleming), pp. 105-1 18. Dr W. Junk, Dordrecht. Fleming, T. H., Breitwisch, R. and Whitesides, G. H. (1987). Patterns of tropical vertebrate frugivore diversity. Ann. Rev. Ecol. Syst. 18, 91-109. Fleuriet, A. and Macheix, J. J. (1975). Relations entre la maturation acceleree de fruits blesses et leur teneur en composes phenoliques. In: Facteurs et Regulation de la Maturation des Fruits (Ed. by R. Ulrich), pp. 147-152. Centre National de la Recherche Scientifique, Anatole. Foster, M. S. (1977). Ecological and nutritional effects of food scarcity on a tropical frugivorous bird and its fruit source. Ecology 58, 73-85. Francis, F. J., Harborne, J. B. and Baker, W. G. (1966). Anthocyanins in the lowbush blueberry, Vaccinium angustifolium. J. Food. Sci. 31, 583-587. Franich, R. A., Gadgh, P. D. and Shain, L. (1983). Fungistatic effects of Pinus radiata needle epicuticular fatty and resin acids on Dothisfromapini Physiol. Plant Pathol. 23, 183-195. Freeland, W. J. and Janzen, D. H. (1974). Strategies in herbivory by mammals: the role of secondary compounds. Am. Nut. 108, 269-289. Freeland, W. J., Calcott, P. H. and Anderson, L. R. (1985). Tannins and saponins: interaction in herbivore diets. Biochem. Syst. Ecol. 13, 189-193. Frenkel, C. (1 972). Involvement of peroxidase and indole-3-acetic acid oxidase isozymes from pear, tomato, and blueberry fruit in ripening. Plant Physiol. 49, 757-763. Frenkel, C. (1975). Role of oxidative metabolism in regulation of fruit ripening. In: Facteurs et Regulation de la Maturation des Fruits (Ed. by R. Ulrich), pp. 201-210. Centre National de la Recherche Scientifique, Anatole. Frohne, D. (1969). Urinary disinfectant activity of bearberry leaf extracts. PIanta Medica 18, 1-7. Fuleki, T. and Francis, F. J. (1967). The co-occurrence of monoglucosides and monogalactosides of cyanidin and peonidin in the American Cranberry, Vaccinium macrocarpon. Phytochemistry 6 , 1705-1708. Galleta, G . J., Ballinger, W. E., Monroe, R. J. and Kushman, L. J. (1971). Relationships between fruit acidity and soluble solids levels of highbush blueberries and fruit keeping quality. J. Amer. SOC.hort. Sci.%, 758-762. Galloway, D. F. (1989). The biological significance of tannins: an overview. In: Chemistry and Signijicance of Condensed Tannins (Ed. by R. W. Hemingway and J. J. Karchesy), pp. 447456. Plenum Press, New York. Gargiullo, M. B. and Stiles, E. W. (1991). Chemical and nutritional differences between 2 bird dispersed fruits-Ilex opaca and Ilex verticillata. J. chem. Ecol. 17, 1091-1 106. Gartlan, J. S., McKey, D. B., Waterman, P. G., Mbi, C. N. and Struhsaker, T. T.

76

M. L. CIPOLLINI A N D E. W. STILES

(1980). A comparative study of the phytochemistry of two African rain forests. Biochem. Syst. Ecol. 8, 401422. Gautier-Hion, A., Duplantier, J.-M., Quris, R., Feer, F., Sourd, C., Decoux, J.-P. et al. (1985). Fruit characters as a basis of fruit choice and seed dispersal in a tropical forest vertebrate community. Oecologia (Berl.) 65, 324-337. Giral, F. and Hidalgo, C . (1983). Presence of alkaloids in Mexican plants. Int. J . Crude Drug Res. 21, 1-13. Goldstein, J. L. and Swain, T. (1963). Changes in tannins in ripening fruits. Phytochemistry 2, 37 1-383. Gorchov. D. L. (1985). Fruit ripening asynchrony is related to variable seed number in Amelanchier and Vaccinium. Amer. J . Bot. 72, 1939-1943. Gorchov, D. L. (1988). Does asynchronous fruit ripening avoid satiation of seed dispersers?: a field test. Ecology 69, 1545-1 55 1. Gorchov, D. L. (1990). Pattern, adaptation, and constraint in fruiting synchrony within vertebrate-dispersed woody plants. Oikos 58, 169-1 80. Gough, R. E. and Litke, W. (1980). An anatomical and morphological study of abscission in highbush blueberry fruit. J . Amer. hort. Soc. 105, 335-341. Could, F. (1988). Genetics of pairwise and multispecies plant- herbivore coevolution. In: Chemical Mediation of Coevolution (Ed. by K . C. Spencer), pp. 13-55. Academic Press, New York. Graham, S. 0.. Patterson, M. E. and Allen, B. (1967). Bruising as a predisposing factor in the decay of stored cranberries. Phytopathology 57, 497-501. Gross, J. (1987). Pigments in Fruits. Academic Press, London. Gross, K. C. and Sams, C. E. (1984). Changes in cell wall neutral sugar composition during fruit ripening: a species survey. Phytochemistry 23, 2457-2461. Hagerman, A. E. (1987). Radial diffusion method for determining tannins in plant extracts. J . chem. Ecol. 13, 437449. Hagerman, A. E. (1989). Chemistry of tannin-protein complexation. In: Chemistry and Significance of Condensed Tannins (Ed. by R. W. Hemingway and J. J. Karchesy). pp. 323-334. Plenum Press, New York. Hagerman, A. E. and Butler, L. G . (1980). Determination of protein in tanninprotein precipitates. J . Agric. Food Chem. 28, 947-952. Hahn, M. G. and Albersheim, P. (1978). Host-pathogen interactions XIV. Isolation and partial characterization of an elicitor from yeast extract. Plant Physiol. 62, 107-1 1 1 . Hahlbrock, K., Lamb, C. J., Purwin, C., Ebel, J., Foutz, E. and Schafer, E. (1981). Rapid response of suspension-cultured parsley cells to the elicitor from Phytophthora megaspermae var. sojae. Plant Physiol. 67, 768-773. Harborne. J. B. (1967). Comparative Biochemistry of the Flavanoids. Academic Press, London. Harborne, J. B. (1979). Flavonoid pigments. In: Herbivores- Their Interaction with Secondary Plant Metabolites (Ed. by G . A. Rosenthal and D. H. Janzen), pp. 619656. Academic Press, New York. Harborne, J. B. (1980). Plant phenolics. In: Encyclopedia of Plant Physiology (Ed. by A. Pirson and M. Zimmermann), Vol. 8. Secondary Plant Products (Ed. by E. A. Bell and B. V. Charlwood). Springer-Verlag, Berlin. Harborne, J. B. (1989). General procedures and measurements of total phenolics. In: Methods in Plant Biochemistry (Ed. by P. M. Dey and J. B. Harborne), Vol. 1 . Plant Phenolics (Ed. by J. B. Harborne), pp. 1-28. Academic Press, London. Harborne. J. B. and Ingham, J. L. (1978). Biochemical aspects of the coevolution of higher plants with their fungal parasites. In: Biochemical Aspects of Plant and

RELATIVE RISKS OF MICROBIAL ROT FOR FLESHY FRUITS

77

Animal Coevolution (Ed. by J. B. Harborne), pp. 343406. Academic Press, London. Harper, J. L. (1989). The value of a leaf. Oecologia (Berl.) 80, 53-58. Harper, J. L., Lovell, P. H. and Moore, K . G. (1970). The shapes and sizes of seeds. Ann. Rev. Ecol. Syst. 1, 327-356. Hartung, J. S., Burton, C. L. and Ramsdell, D. C. (1981). Epidemiological studies of blueberry anthracnose disease caused by Colletotrichum gloeosporioides. Phytopathology 71, 449-453. Hawkins, L. A. (1915). Some effects of the brown-rot fungus upon the composition of the peach. Amer. J. Bot. 2, 71-81. Heath, M. C. (1981). Insoluble silicon in necrotic cowpea cells following infection with an incompatible isolate of the cowpea rust fungus. Physiol. Plant Pathol. 19, 273-276. Hedden, P., MacMillan, J. and Phinney, B. D. (1978). The metabolism of the gibberellins. Ann. Rev. Plant Physiol. 29, 149-192. Henderson, C. B. (1990). The influence of seed apparency, nutrient content and chemical defenses on dietary preference in Dipodomys ordii. Oecologia (Berl.) 82, 333-341. Herbst, L. H. (1986). The role of nitrogen from fruit pulp in the nutrition of the frugivorous bat Carollia perspicillata. Biotropica 18, 3 9 4 4 . Herrera, C. M. (1982a). Defense of ripe fruit from pests: its significance in relation to plantkdisperser interactions. Am. Nut. 120, 218-241. Herrera, C. M. (1982b). Seasonal variation in the quality of fruits and diffuse competition between plants and avian dispersers. Ecology 63, 773-785. Herrera, C. M. (1984). A study of avian frugivores, bird-dispersed plants, and their interaction in Mediterranean scrublands. Ecol. Monogr. 54, 1-23. Herrera, C. M. (1985). Determinants of plant-animal coevolution: the case of mutualistic dispersal of seeds by vertebrates. Oikos 44, 132-141. Herrera, C. M. (1986). Vertebrate-dispersed plants: why they don’t behave the way they should. In: Frugivores and Seed Dispersal (Ed. by A. Estrada and T. H. Fleming), pp. 5-18. Dr W. Junk, Dordrecht. Herrera, C. M. (1987). Vertebrate-dispersed plants of the Iberian peninsula: a study of fruit characteristics. Ecol. Monogr. 57, 305-33 1. Herrera, C. M. (l988a). The fruiting ecology of Osyris quadripartita: individual variation and evolutionary potential. Ecology 69, 233-249. Herrera, C. M. (1988b). Habitat-shaping, host plant use by a hemiparasitic shrub, and the importance of gut fellows. Oikos 51, 383-386. Hikino, H., Kiso, Y . , Taguchi, H. and Ikeya, Y . (1984). Antihepatotoxic actions of lignoids from Schizandra chinensis fruits. Planta Medica 50, 21 3-218. Hoffman, C., Schweitzer, T. R. and Dalby, G. (1941). Fungistatic properties of antiseptic and related compounds. Effect of pH. Ind. Eng. Chem. 33, 749-751. Hoppes, W. G. (1988). Seedfall pattern of several species of bird-dispersed plants in an Illinois woodland. Ecology 69, 32&329. Howard, D. J., Bush, G. L. and Breznak, J. A. (1985). The evolutionary significance of bacteria associated with Rhagoletis. Evolution 39, 405417. Howe, H. F. (1979). Fear and frugivory. Am. Nut. 114, 925-931. Howe, H. F. (1984). Constraints on the evolution of mutualisms. Am. Nut. 123, 7 6 4 777. Howe, H. F. (1985). Gomphothere fruits: a critique. Am. Nut. 125, 853-865. Howe, H. F. and Estabrook, G. F. (1977). On intraspecific competition for avian dispersers in tropical trees. Am. Nut. 111 , 8 17-83 1.

78

M. L. CIPOLLINI AND E. W. STILES

Howe, H. F. and Primack, R. B. (1975). Differential seed dispersal by birds of the tree Caseria nitida. Biotropica 7, 278-283. Howe, H. F. and Schupp, E. W. (1985). Early consequences of seed dispersal for a neotropical tree (Virola suranimensis). Ecology 66, 781-791. Howe, H. F. and Smallwood, J. (1982). Ecology and seed dispersal. Ann. Rev. Ecol. Syst. 13, 201-228. Howe, H. F. and Vande Kerckhove, G. A. (1979). Fecundity and seed dispersal of a tropical tree. Ecology 60,18G189. Hsieh, D. P. H. (1989). Potential human health hazards of mycotoxins. In: Mycotoxins and Phycotoxins ’88(Ed. by S. Natori, K. Hashimoto and Y. Ueno), pp. 69-80. Elsevier, Amsterdam. Hubbell, S. P. (1979). Tree dispersion, abundance, and diversity in a tropical dry forest. Science 203, 1299-1309. Hubbell, S. P. and Howard, J. J. (1984). Chemical leaf repellency to an attine ant: seasonal distribution among potential host plants. Ecology 65, 1067-1076. Hubbell, S. P., Wiemer, D. F. and Adejare, A. (1983). An antifungal terpenoid defends a neotropical tree (Hymenaea) against attack by fungus-growing ants (Atta). Oecologia (Berl.) 60, 321-327. Hulme, A. C. (Ed.) (1971). The Biochemistry of Fruits and Their Products, Vol. 11. Academic Press, London. Hulme, A. C. and Edney, K. L. (1960). Phenolic substances in the peel of Cox’s orange pippin apples with reference to infection by G . perennans. In: Phenolics in Plants in Health and Disease (Ed. by J. B. Pridham), pp. 87-94. Pergamon Press, Oxford. Ivanic, R., Savin, K. and Robinson, F. V. (1983). Essential oil from Pimpinella serbica fruits. PIanta Medica 48, 6&61. Izhaki, I. and Safriel, U. N. (1985). Why do fleshy-fruit plants of the mediterranean scrub intercept fall- but not spring-passage of seed dispersing migratory birds? Oecologia (Berl.) 67, 4 W 3 . Izhaki, I. and Safriel, U. N. (1989). Why are their so few exclusively frugivorous birds? Experiments on fruit digestibility. Oikos 54, 23-32. Izhaki, I. and Safriel, U. N. (1990). Weight losses due to exclusive fruit dietinterpretation and evolutionary implications: a reply to Mack and Sedinger. Oikos 57, 14G142. Janisiewicz, W. J. (1988). Biocontrol of postharvest diseases of apples with antagonistic mixtures. Phytopathology 78, 194-198. Jankowski, K. (1973). Alkaloids of cranberries. Experientia 29, 1334-1335. Janovitz-Klapp, A. H., Richard, F. C., Goupy P. M. and Nicolas, J. J. (1990a). Inhibition studies on apple polyphenol oxidase. J. Agric. Food Chem. 38,926931. Janovitz-Klapp, A. H., Richard, F. C., Goupy P. M. and Nicolas, J. J. (1990b). Kinetic studies on apple polyphenol oxidase. J . Agric. Food Chem. 38, 1437-1441. Janson, C. H. (1983). Adaptation of fruit morphology to dispersal agents in a neotropical forest. Science 219, 187-189. Janzen, D. H. (1968). Host plants as islands in evolutionary and contemporary time. Am. Nut. 102, 592-595. Janzen, D. H. (1969). Seed-eaters versus seed size, number, toxicity and dispersal. Evolution 23, 1-27. Janzen, D. H. (1970). Herbivores and the number of tree species in tropical forests. Am. Nat. 104, 501-528. Janzen, D. H. (1973). Host plants as islands 11. Competition in evolutionary and contemporary time. Am. Nut. 107, 786-790.

RELATIVE RISKS OF MICROBIAL ROT FOR FLESHY FRUITS

79

Janzen, D. H. (1974). Tropical blackwater rivers, animals, and mast fruiting by the Dipterocarpaceae. Biotropica 6, 69-1 03. Janzen, D. H. (1975). Ecology of Plants in the Tropics. Arnold, London. Janzen, D. H. (1977). Why fruits rot, seeds mold, and meat spoils. Am. Nut. 111,691713. Janzen, D. H. (1979). New horizons in the biology of plant defenses. In: HerbivoresTheir Interaction with Secondary Plant Metabolites (Ed. by G. A. Rosenlhal and D. H. Janzen), pp. 331-350. Academic Press, New York. Janzen, D. H. (1980). When is it coevolution? Evolution 34, 61 1-612. Janzen, D. H. (1981a). Enterolobium cyclocarpum seed passage rate and survival in horses, Costa Rican Pleistocene seed dispersal agents. Ecology 62, 593401. Janzen, D. H. (1981b). Guanacaste tree seed-swallowing by Costa Rican range horses. Ecology 62, 587-592. Janzen, D. H. (1983). Physiological ecology of fruits and their seeds. In: Physiological Plant Ecology III (Ed. by 0. Lange, P. S. Nobel, C. B. Osmond and H. Ziegler), pp. 625-655. Springer-Verlag, Berlin. Janzen, D. H. (1985). On ecological fitting. Oikos 45, 308-310. Jarvis, B. B., Bean, G. A., Midiwo, J. D., Aboul-Nasr, M. B., Kuti, J. and Mokhtari, N. (1989). The mystery of trichothecenes in higher plants. In: Mycotoxins and Phycotoxins '88 (Ed. by S. Natori, K. Hashimoto and Y. Ueno), pp. 197-204. Elsevier, Amsterdam. Jarvis, W. R. and Traquair, J. A. (1984). Bunch rot of grapes caused by Aspergillus acleatus. Plant Dis. 68, 718-719. Jaworski, A. W. and Lee, C. Y. (1987). Fractionation and HPLC determination of grape phenolics. J . Agric. Food Chem. 35, 257-259. Jensen, R. A. (1985). The shikimate/arogenate pathway: link between carbohydrate metabolism and secondary metabolism. Physiol. Plant. 66, 164-168. Jenny, T. (1984). Evolution of insect/host plant relationships. Am. Nut. 124,609-630. Johnson, A. and Booth, C . (Eds). (1983). Plant Pathologist's Pocketbook. Commonwealth Mycological Institute, Kew, Surrey. Johnson, N. D. (1983). Flavonoid aglycones from Eriodictyon californicum resin and their implications for herbivory and UV screening. Biochem. Syst. Ecol. 11, 21 1215. Johnson, R. A,, Willson, M. F., Thompson, J. N. and Bertin, R. I. (1985). Nutritional values of wild fruits and consumption by migrant frugivorous birds. Ecology 66, 8 19-827. Jones, D. H. (1984). Phenylalanine ammonia lyase: regulation of its induction and its role in plant development. Phytochemistry 23, 1349-1 359. Jordano, P. (1987a). Avian fruit removal: effects of fruit variation, crop size, and insect damage. Ecology 68, 1711-1723. Jordano, P. (1987b). Patterns of mutualistic interactions in pollination and seed dispersal: connectance, dependence asymmetries, and coevolution. Am. Nut. 129, 651477. Karowe, D. N. (1989). Differential effect of tannin acid on two tree-feeding Lepidoptera: implications for theories of plant anti-herbivore chemistry. Oecologia (Berl.) 80, 506512. Keeler-Wolf, T. (1988). Fruit and consumer differences in three species of trees shared by Trinidad and Tobago. Biotropica 20, 3 8 4 8 . Keen, N. T. (1975). Specific elicitors of plant phytoalexin production: determinants of race specificity in pathogens? Science 187, 74-75. Kerwin, J. L. (1987). Fatty acids and fungal development: structure activity relation-

80

M. L. CIPOLLINI AND E. W. STILES

ships. In: Ecology and Metabolism of Plant Lipids (Ed. by G. Fuller and W. D. Nes), pp. 329-342. American Chemical Society, Washington, DC. Kiessling, K.-H. (1989). A first line defence in animals against mycotoxins. In: Mycotoxins and Phycotoxins '88 (Ed. by S. Natori, K. Hashimoto and Y . Ueno), pp. 119-126. Elsevier, Amsterdam. Koenig, W. D. (1991). The effects of tannins and lipids on digestion of acorns by acorn woodpeckers. Auk 108, 79-88. Kuris, A. M., Blaustein, A. R. and Alio, J. J. (1980). Hosts as islands. Am. Nut. 116, 57Ck586. Kushman, L. J. and Ballinger, W. E. (1975). Effect of season, location, cultivar, and fruit size upon quality of light-sorted blueberries. J . Arner. SOC.Hort. Sci. 100, 564-569. Lacey, J. (1981). The aerobiology of conidial fungi. In: Biology of Conidial Fungi, Vol. I (Ed. by G. T. Cole and B. Kendrick), pp. 3 7 3 4 1 4 . Academic Press, London. Lacey, J. (1989). Prevention of mold growth and mycotoxin production through control of environmental factors. In: Mycotoxins and Phycotoxins '88 (Ed. by S. Natori, K. Hashimoto and Y . Ueno), pp. 161-168. Elsevier, Amsterdam. Laks, P. E. (1989). An overview of condensed tannin structure. In: Chemistry and SigniJicance of Condensed Tannins (Ed. by R. W. Hemingway and J. J. Karchesy), pp. 131-138. Plenum Press, New York. Lambert, F. (1989). Fig-eating by birds in a Malaysian lowland rain forest. J . trop. Ecol. 5, 4 0 1 4 1 2 . Landers, J. L., Hamilton, R. J., Johnson, A. S. and Marchinton, R. L. (1979). Foods and habitats of black bears in southeastern North Carolina. J. Wildl. Manage. 43, 143-1 53. Lane, H. C. and Schuster, M . F. (1981). Condensed tannins of cotton leaves. Phytochemistry 20, 425-427. Lawanson, A. O., Akindele, B. B., Fasalojo, P. B. and Akpe, B. L. (1971). Timecourse of anthocyanin formation during deficiencies of nitrogen, phosphorus and potassium in seedlings of Zea mays Linn. var. E. S. Z . P'anzenphysiol. Bd. 66, 25 1-253. Lea, A. G. H. (1978). The phenolics of ciders: oligomeric and polymeric procyanidins. J . Sci. Food Agric. 29, 4 7 1 4 7 7 . Leben, C., Daft, G. C. and Schmitthenner, C. (1968). Bacterial blight of soybeans: population levels of Pseudomonas glycinea in relation to symptom development. Phytopathology 58, 1143-1 146. Lee, D. W., Brammeier, S. and Smith, A. P. (1987). The selective advantages of anthocyanins in developing leaves of Mango and Cacao. Biotropica 19, 4 M 9 . Lee, W. G., Wilson, J. B. and Johnson, P. N. (1988). Fruit colour in relation to the ecology and habit of Coprosoma (Rubiaceae) species in New Zealand. Oikos 53, 325-33 I . Leistner, L., Geisen, R. and Fink-Gremmels, J. (1989). Mold-fermented foods of Europe: hazards and developments. In: Mycotoxins and Phycotoxins '88 (Ed. by S. Natori, K. Hashimoto and Y. Ueno), pp. 145154. Elsevier, Amsterdam. Letham, D. S. and Palni, L. M. S. (1983). The biosynthesis and metabolism of cytokinins. Ann. Rev. Plant Physiol. 34, 163-197. Levey, D. J. (1987a). Facultative ripening in Hamelia patens (Rubiaceae): effects of fruit removal and rotting. Oecologia (Berl.) 74, 203-208. Levey, D. J. (1987b). Seed size and fruit-handling techniques of avian frugivores. Am. Nut. 129, 4 7 1 4 8 5 .

RELATIVE RISKS OF MICROBIAL ROT FOR FLESHY FRUITS

81

Levey, D. J. (1987~).Sugar-tasting ability and fruit selection in tropical fruit-eating birds. Auk 104, 173-179. Levey, D. J. (1988a). Spatial and temporal variation in Costa Rican fruit and fruiteating bird abundance. Ecol. Monogr. 58, 251-269. Levy, D. J. (1988b). Tropical wet forest treefall gaps and distributions of understory birds and plants. Ecology 69, 10761089. Levey, D. J. and Karasov, W. H. (1989). Digestive responses of temperate birds switched to fruit or insect diets. Auk 106, 675-686. Levey, D. J., Moermond, T. C. and Denslow, J. S. (1984). Fruit choice in neotropical birds: the effect of distance between fruits on preference patterns. Ecology 65,844850. Levin, D. A. (1971). Plant phenolics: an ecological perspective. Am. Nut. 105, 157181. Levin, D. A. (1973). The role of trichomes in plant defense. Q. Rev. Biol. 48, 3-15. Levin, D. A. (1976). The chemical defenses of plants to pathogens and herbivores. Ann. Rev. Ecol. Syst. 7, 121-159. Lewis, N. G. and Yamamoto, E. (1989). Tannins- their place in plant metabolism. In: Chemistry and SigniJicance of Condensed Tannins (Ed. by R. W. Hemingway and J. J. Karchesy), pp. 23-26. Plenum Press, New York. Lieberman, M. (1979). Biosynthesis and action of ethylene. Ann. Rev. Plant Physiol. 30, 533-591. Lieberman, M . and Lieberman, D. (1986). An experimental study of seed ingestion and germination in a plant-animal assemblage in Ghana. J . trop. Ecol. 2, 113126. Liener, I. E. (1979). Phytohemagglutinins. In: Herbivores- Their Interaction with Secondary Plant Metabolites (Ed. by G. A. Rosenthal and D. H. Janzen), pp. 567598. Academic Press, New York. Loiselle, B. A. (1990). Seeds in droppings of tropical fruit-eating birds: importance of considering seed composition. Oecologia ( B e d . ) 82, 494-500. Loiselle, B. A. and Blake, J. G. (1991). Temporal variation in birds and fruits along an elevational gradient in Costa Rica. Ecology 72, 18&193. Lundberg, P. and Astrom, M. (1990). Low nutritive quality as a defense against optimally foraging herbivores. Am. Nut. 135, 547-562. Lynn, D. G. and Chang, M. (1990). Phenolic signals in cohabitation: implications for plant development. Ann. Rev. Plant Physiol. Plant Mol. Biol. 41, 497-526. Mabry, T. J. and Gill, J. E. (1979). Sesquiterpene lactones and other terpenes. In: Herbivores- Their Interaction with Secondary Plant Metabolites (Ed. by G. A. Rosenthal and D. H. Janzen), pp. 502-538. Academic Press, New York. Mack, A. (1990). Is frugivory limitkd by secondary compounds in fruits? Oikos 57, 135-1 38. Maclean, D. J., Sargent, J. A., Tommerup, I. C. and Ingram, D. S. (1974). Hypersensitivity as the primary event in resistance to fungal parasites. Nature 249, 186-187. Makus, D. J. and Ballinger, W. E. (1973). Characterization of anthocyanins during ripening of fruit of Vaccinium corymbosum L. cv. Wolcott. J . Amer. Soc. hort. Sci. 98, 99-101. Malamy, J., Carr, J. P., Klessig, D. F. and Raskin, I. (1990). Salicylic acid: a likely endogenous signal in the resistance response of tobacco to viral infection. Science 250, 1002- 1004. Malathrakis, N. E. (1989). Resistance of Botrytis cinerea to dichlofluanid in greenhouse vegetables. Plant Dis. 73, 138-141.

82

M. L. CIPOLLINI A N D E. W. STILES

Malloch, D. ( 1 98 I). Moulds: Their Isolation, Cultivation and Identijication. University of Toronto Press, Toronto. Mancinelli, A. L. and Rabino, I. (1984). Photoregulation of anthocyanin synthesis X. Dependence on photosynthesis of high irradiance response anthocyanin synthesis in Brassica oleraceae leaf disks and Spirodela polyrrhiza. Plant Cell. Physiol. 25, 1153-1 156. Mansfield, J. W. and Hargreaves, J. A. (1974). Phytoalexin production by live cells in broad bean leaves infected with Botrytis cinerea. Nature 252, 316. Manske, R. H. F. and Rodrigo R. G. A. (Eds) (1981). The Alkaloids. Academic Press, New York. Marasas, W. F. 0. and Nelson, P. E. (1987). Mycotoxicology: Introduction to the Mycology, Plant Pathology, Chemistry, Toxicology, and Pathology of Naturafly Occurring Mycotoxicoses in Animals and Man. Pennsylvania State University Press, University Park. Margna, U. (1977). Control at the level of substrate supply-An alternative in the regulation of phenylpropanoid accumulation in plant cells. Phytochemistry 16, 4 19426. Markakis, P., Jarczyk, A. and Krishna, S. P. (1963). Nonvolatile acids of blueberries. J. Agric. Food Chem. 11, 8-1 1. Martin, J. S. and Martin, M. M. (1982). Tannin assays in ecological studies: lack of correlation between phenolics, proanthocyanidins and protein-precipitating constituents in mature foliage of six oak species. Oecologia (Berl.) 54, 205-21 I . Martinez del Rio, C., Karosov, W. H. and Levey, D. J. (1989). Physiological basis and ecological consequences of sugar preferences in Cedar Waxwings. Auk 106, 64-71. Mathis, C. (1 966). Comparative biochemistry of the hydroxyquinones. In: Comparative Phytochemistry (Ed. by T. Swain), pp. 245-270. Academic Press, London. Matzner, V. F. (1971). Quality and components of cultured American cranberry ( V . macrocarpon), European cranberry ( V . oxycoccus), and wild cranberry (V. vitisidaea). Indust. 0bst.-Gemueseverwertung 56, 27-32. McAtee, W. L. (1947). Distribution of seeds by birds. Amer. Midf. Nut. 38, 214-223. McClure, J. W. (1979). Physiology of phenolic compounds in plants. Rec. Adv. Phytochem. 12, 525-555. McDonald, J. F. (1983). The molecular basis of adaptation: a critical review of relevant ideas and observations. Ann. Rev. Ecol. Syst. 14, 77-102. McDonnell, M. J. and Stiles, E. W. (1983). The structural complexity of old field vegetation and recruitment of bird-dispersed plant species. Oecologia (Berl.) 56, 109-1 16. McDonnell, M. J., Stiles, E. W., Cheplick, G . P. and Armesto, J. J. (1984). Birddispersal of Phytolacca americuna L. and influence of fruit removal on subsequent fruit development. Amer. J . Bot. 71, 895-901. McKeen, W. F. (1958). Blueberry canker in British Columbia. Phytopathology 48, 277-280. McKey, D. (1973). The ecology of coevolved seed dispersal systems. In: Coevolution of Animals and Plants (Ed. by L. E. Gilbert and P. H. Raven), pp. 159-191. University of Texas Press, Austin. McKey, D. (1974). Adaptive patterns in alkaloid physiology. Am. Nut. 108,305320. McKey, D. F. (1979). The distribution of secondary compounds within plants. In: Herbivores- Their Interaction with Secondary Plant Metabolites (Ed. by G. A. Rosenthal and D. H. Janzen), pp. 56134. Academic Press, New York. Mercier, J. and Reeleder, R. D. (1985). Use of phylloplane fungi to control Sclerotinia sclerotiorum on lettuce. Phytoprotection 66, 176.

RELATIVE RISKS OF MICROBIAL ROT FOR FLESHY FRUITS

83

Metraux, J . P., Signer. H.. Ryals, J.. Ward, E.. Wyss-Benz, M.. Gaudin, J. et al. (1990). Increase in salicylic acid at the onset of systemic acquired resistance in cucumber. Science 250, 1004-1006. Meyer, M. W. and Karasov, W. H. (1989). Antiherbivore chemistry of Larrea tridentuta: effects on woodrat (Neotomu lepidu) feeding and nutrition. Ecology 70, 953-961, Milholland, R . D. and Daykin, M. E. (1983). Blueberry fruit rot caused by Phomopsis vuccinii. Plant Dis. 67, 325-326. Mills. J. T. (1989). Ecology of toxigenic fungi associated with grains in Manitoba, Canada. In: M.vcoto.vins and Phycoto.uins '88 (Ed. by S. Natori, K. Hashimoto and Y. Ueno), pp. 13-20, Elsevier, Amsterdam. Mitscher, L. A. (1975). Antimicrobial agents from higher plants. Rec. Adv. Phytochem. 9. 243-282. Mole, S. and Waterman, P. G . (1987a). A critical analysis of techniques for measuring tannins in ecological studies. 11. Techniques for biochemically defining tannins. Oecologiu (Berl.) 72, 148-1 56. Mole, S. and Waterman, P. G. (1987b). A critical analysis of techniques for measuring tannins in ecological studies. I. Techniques for chemically defining tannins. Oecologia ( B e r l . ) 72, 137-147. Moller, B. (1983). Quinic acid esters of hydroxycinnamic acids in stone and pome fruit. Phytochemistry 22, 4 7 7 4 8 I . Moore, J. N., Reynolds, B. D. and Brown, G . R . (1972). Effects of seed number, size, and development on fruit size of cultivated blueberries. HortScience 7 , 268-269. Moore, L. A. and Willson. M. F. (1982). The effect of microhabitat, spatial distribution, and display size on dispersal of Lindera benzoin by avian frugivores. Can. J . Bot. 60, 557-560. Morozumi, S., Wauke, T., Kudoh, Y. and Hitokotu. H . (1989). Antifungal effects of commercial foods and spices, and their components. In: M.vcoto.vins and Phycoto.uins '88 (Ed. by S. Natori, K. Hashimoto and Y . Ueno), pp. 155-160. Elsevier, Amsterdam. Morton, E. S. (1973). On the evolutionary advantages and disadvantages of fruiteating in tropical birds. A m . Nut. 107, 8-22. Murray, K . G . (1987). Selection for optimal fruit-crop size in bird-dispersed plants. Am. Nut. 129, 18-3 I . Murray, K . G . (1988). Avian dispersal of three Neotropical gap-dependent plants. Ecol. Monogr. 58, 271-298. Muys, G. T., Van Gils, H . W. and De Vogel, P. (1966). The determination and enumeration of the associated microflora of edible emulsions. 11. The microbiological investigation of margarine. Luh. Pruct. 15, 975-984. Narstedt, A,, Wray, V., Grotjahn, L., Fikenscher, L. and Hegnauer, R . (1983). New acylated cyanogenic diglycosides from fruits of Anthemis cuirica. Plunta Medica 49, 143-148. Nel, P. J. (1985). Deciduous Fruits and Vines- Pests and Diseases and Their Control. David Philip, Cape Town. Nelson, E. K . (1927). The non-volatile acids of the pear, quince, apple, loganberry, blueberry, cranberry, lemon and pomegranate. J . Atner. Chem. Soc. 49, 130& 1302. Nelson, P. E., Tousson, T. A. and Marasas, W. F. 0. (1983). Fusarium species-An Illustrated Munual,for Ident$cution. Pennsylvania State University Press, University Park. Newhook, F. J . (1957). The relationship of saprophytic antagonism to control Botrytis cinereu Pers. on tomatoes. Ncn* Ziwlund J . Sci. Technol. 38, 473481. ~

84

M. L. CIPOLLINI AND E. W. STILES

Newsome, R. L. (1990). Natural and synthetic coloring agents. In: Food Additives (Ed. by A. L. Branen, P. M. Davidson and S. Salminen), pp. 327-345. Marcel Dekker, New York. Nickell, L. G. (1959). Antimicrobial activity of vascular plants. Econ. Bot. 13, 281318. O’Donnell, S. (1989). A comparison of fruit removal by bats and birds from Piper hispidum, a tropical second growth shrub. Brenesia 31, 25-32. Ojewole, J. A. 0. and Adesina, S. K. (1983). Mechanism of the hypotensive effect of scopoletin isolated from the fruit of Tetrapleura tetraptera. Planta Medica 49,4& 50. Pady, S. M. (1957). Quantitative studies of fungus spores in the air. Mycologia 49. 339-3 5 3. Palmeirim, J. M., Gorchov, D. L. and Stoleson, S. (1989). Trophic structure of a neotropical frugivore community: Is there competition between birds and bats? Oecologia (Berl.) 79, 40341 1. Pandey, R. R. ( 1 990). Succession of microfungi on leaves of Psidium guajava L. Bull. Torrey But. Club 117, 153-162. Papas, P. N., Brush, C. A. and Ceresia, G. C. (1966). Cranberry juice for treating urinary infections. Southwestern Med. 47, 17. Parkinson, D. (1981). Ecology of soil fungi. In: Biology of Conidial Fungi, Vol. 1 (Ed. by G. T. Cole and B. Kendrick), pp. 277-292. Academic Press, London. Paxton, J., Goodchilde, D. J. and Cruikshank, 1. A. M. (1974). Phaseolin production by live bean endocarp. Physiol. Plant Pathol. 4, 167-1 7 1. Perera, P., Sandberg, F., Van Beek, T. A. and Verporte, R. (1984). Tertiary indole alkaloids from fruits of Tabernaemontana dichotoma. Planta Medica 50, 251-253. Peters, C . M. and Hammond, E. J. (1990). Fruits from the flooded forests of Peruvian Amazonia: yield estimates for natural populations of three promising species. Adv. econ. Bot. 8, 159-176. van der PijI, L. ( I 969). Principles of Dispersal in Higher Plants. Springer-Verlag, Berlin. Piper, J. K. (1986). Germination and growth of bird-dispersed plants: effects of seed size and light on seedling vigor and biomass allocation. Amer. J . Bot. 73,959-965. Pirozynski, K. A. (1988). Coevolution by horizontal gene transfer: a speculation on the role of fungi. In: Coevolution of Fungi with Plants and Animals (Ed. by K. A. Pirozynski and D. L. Hawksworth), pp. 247-268. Academic Press, London. Pirozynski, K . A. and Hawksworth, D. L. (1988). Coevolution of fungi with plants and animals: introduction and overview. In: Coevolution of Fungi with Plants and Animals (Ed. by K. A. Pirozynski and D. L. Hawksworth), pp. 1-29. Academic Press, London. Pirozynski, K. A. and Malloch, D. W. (1988). Seeds, spores and stomachs: coevolution in seed dispersal mutualisms. In: Coevolution of Fungi with Plants and Animals (Ed. by K. A. Pirozynski and D. L. Hawksworth), pp. 227-246. Academic Press, London. Pitt, J. I. and Hocking, A. D. (1985). Fungi and Food Spoilage. Academic Press, Sydney. Platt, W. J . and Hermann, S. M . (1986). Relationships between dispersal syndrome and characteristics of populations of trees in a mixed-species forest. In: Frugivores and Seed Dispersal (Ed. by A. Estrada and T. H. Fleming), pp. 309-322. Dr W. Junk, Dordrecht. Du Plessis, H. J. (1990). Systemic invasion of plum seed and fruit by Xanthomonas campestris pv. pruni through stalks. J . Phytopathol. 130, 3 7 4 5 .

RELATIVE RISKS OF MICROBIAL ROT FOR FLESHY FRUITS

85

Pont, V. and Pezet, R. (1990). Relation between the chemical structure and the biological activity of hydroxystilbenes against Botrytis cinerea. J . Phytopathol. 130, 1-8. Porter, L. J. (1989). Tannins. In: Methods in Plant Biochemistry (Ed. by P. M. Dey and J. B. Harborne), Vol. I , Plant Phenolics (Ed. by J. B. Harborne), pp. 389420. Academic Press, London. Poston, M. E. and Middendorf, G. A. 111. (1988). Maturation characteristics of Rubuspennsylvanicus fruit: Are black and red the same? Oecologia (Berl.) 77, 6972. Pratt, H. K . (1975). The role of ethylene in fruit ripening. In: Facteurs et Regulation de la Maturation des Fruits (Ed. by R. Ulrich), pp. 153-160. Centre National de la Recherche Scientifique, Anatole. Pratt, T. K. and Stiles, E. W. (1983). How long fruit-eating birds stay in the plants where they feed: implications for seed dispersal. Am. Nut. 122, 797-805. Pucheu-Plante, B. and Mercier, M. (1982). Etude ultrastructural de l’interrelation hote-parasite entre le raisin et le champignon Botrytis cinerea: exemple de la pourriture nobel en Sauternaise. Can. J . Bot. 61, 1785-1797. Raffauf, R. F. (1970). A Handbook of Alkaloids and Alkaloid-containing Plants. Wiley-Interscience, New York. Raper, K. B. and Fennell, D. I. (1965). The Genus Aspergillus. Williams and Wilkins, Baltimore. Rathcke, B. J. and Lacey, E. P. (1985). Phenological patterns of terrestrial plants. Ann. Rev. ecol. Syst. 16, 179-214. Rathcke, B. J. and Poole, D. 0. (1975). Coevolutionary race continues: butterfly larval adaptations to plant trichomes. Science 187, 175. Real, L., Ott, J. and Silverfine, E. (1982). On the tradeoff between the mean and the variance in foraging: effect of spatial distribution and color preference. Ecology 63, 1617- 1 623. Reese, J. C. (1979). Interactions of allelochemicals with nutrients in herbivore food. In: Herbivores- Their Interaction with Secondary Plant Metabolites (Ed. by G. A. Rosenthal and D. H. Janzen), pp. 309-330. Academic Press, New York. Regal, P. J. (1977). Ecology and evolution of flowering plant dominance. Science 196, 622-628. Rhoades, D. F. (1977). Integrated antiherbivore, antidesiccant and ultraviolet screening properties of Creosotebush resin. Biochem. syst. Ecol. 5, 28 1-290. Rhoades, D. F. (1979). Evolution of plant chemical defense against herbivores. In: Herbivores- Their Interaction with Secondary Plant Metabolites (Ed. by G. A. Rosenthal and D. H . Janzen), pp. 4 5 5 . Academic Press, New York. Rhoades, D. F. (1985). Offensive4efensive interactions between herbivores and plants: their relevance in herbivore population dynamics and ecological theory. Am. Nut. 125, 205-236. Rhoades, D. F. and Cates, R. G. (1976). Toward a general theory of plant antiherbivore chemistry. Rec. Adv. Phytochem. 10, 168-21 3. Rhodes, M. J. C. and Reid, M. S. (1975). The production of ethylene and its relationship with the onset of the respiration climacteric in the apple. In: Facteurs et Regulation de la Maturation des Fruits (Ed. by R. Ulrich), pp. 189-192. Centre National de la Recherche Scientifique, Anatole. Ridley, H. N. (1930). The Dispersal of Plants Throughout the World. L. Reeve, Ashford, Kent. Robinson, T. (1975). The Organic Constituents of Higher Plants. Cordus, North Amherst.

86

M. L. CIPOLLINI AND E. W. STILES

Robinson, T. (1979). The evolutionary ecology of alkaloids. In: Herbivores- Their Interaction with Secondary Plant Metabolites (Ed. by G . A. Rosenthal and D. H. Janzen), pp. 4 1 3 4 8 . Academic Press, New York. Robinson, W. A. (1986). Effect of fruit ingestion on Amelanchier seed germination. Bull. Torrey Bot. Club 113, 131-134. Roddick, J. G. (1987). Antifungal activity of plant steroids. In: Ecology and Metabolism of Plant Lipids (Ed. by G. Fuller and W. D. Nes), pp. 286303. American Chemical Society, Washington, DC. Rodricks. J. V. (1976). Mycotoxins and Other Fungal Related Food Problems. American Chemical Society, Washington, DC. Rosenthal, G. A. and Bell, E. A. (1979). Naturally occurring, toxic nonprotein amino acids. In: Herbivores- Their Interaction with Secondary Plant Metabolites (Ed. by G. A. Rosenthal and D. H. Janzen), pp. 351-386. Academic Press, New York. Rossiter, M. C., Schultz, J. C. and Baldwin, I. T. (1988). Relationships among defoliation, red oak phenolics, and gypsy moth growth and reproduction. Ecology 69, 267-277. Rossiter, R. C. (1969). Physiological and ecological studies on the oestrogenic isoflavones in subterranean clover ( T . subterraneum L.). Aust. J . agric. Res. 20, 1043-1 05 1. Rossman, A. Y., Palm, M . E. and Spielman, L. J. (1987). A Literature Guide for the Identijication of Plant Pathogenic Fungi. American Phytopathological Society Press, St Paul. Rujkenberg, F. H., De Leeuw, G . T. and Verhoeff, K. (1980). Light and electron microscopy studies on the infection of tomato fruits by Botrytis cinerea. Can. J . Bot. 58, 13941404. Ryan, C. A. (1979). Proteinase inhibitors. In: Herbivores- Their Interaction with Secondary Plant Metabolites (Ed. by G . A. Rosenthal and D. H. Janzen), pp. 599618. Academic Press, New York. Samanta, S. K., Chose, S. K., Sen, S. and Bhattacharyya, N. K. (1983). Associations between berry characters and solasodine content of Solanum khasianum. Planta Medica 48, 94-96. Sapers, G. M., Burgher, A. M . , Phillips, J. G., Jones, S. B. and Stone, E. G . (1984). Color and composition of highbush blueberry cultivars. J . Amer. Soc. hort. Sci. 109, 105-1 11.

Schiffman-Nadel, M . (1975). Relation between fungal attack and post harvest fruit maturation. In: Facteurs et Regulation de la Maturation des Fruits (Ed. by R. Ulrich) pp. 139-146. Centre National de la Recherche Scientifique, Anatole. Schneider, R. L. and Sharitz, R. R. (1988). Hydrochory and regeneration in a bald cypress-water tupelo swamp forest. Ecology 69, 1055-1063. Schonbeck, F. and Schlosser, E. (1976). Preformed substances as potential protectants. In: Physiological Plant Pathology (Ed. by R. Heitefuss and P. H. Williams), pp. 653-678. Springer-Verlag, Berlin. Schultz, J. C. (1989). Tannin-insect interactions. In: Chemistry and Significance of Condensed Tannins (Ed. by R. W. Hemingway and J. J. Karchesy), pp. 417434. Plenum Press, New York. Schupp, E. W., Howe, H. F., Augspurger, C. K. and Levey, D. J. (1989). Arrival and survival in tropical treefall gaps. Ecology 70, 562-564. Schwarz, M . R. and Boone, D. M . (1985). Effect of wounding on incidence of black rot of cranberry in Wisconsin. Plant Dis. 69, 225-227. Scott, P. M. (1989). Control of mycotoxins-an overview. In: Mycotoxins and Phycotoxins '88 (Ed. by S . Natori, K. Hashimoto and Y. Ueno), pp. 127-134. Elsevier, Amsterdam.

RELATIVE RISKS OF MICROBIAL ROT FOR FLESHY FRUITS

87

Sedinger, J. S. (1990). Are secondary compounds responsible for negative apparent metabolizability of fruits by passerine birds? A comment on lzhaki and Safriel. 0 i k 0 ~57, 138-140. Seigler. D. S. (1979). Toxic seed lipids. In: Herbivores- Their Interaction with Secondury Plant Metabolites (Ed. by G . A. Rosenthal and D. H. Janzen), pp. 449470. Academic Press, New York. Shimokawa, K. (1 983). An ethylene-forming enzyme in Citrus unsliiu fruits. Phytochemistry 22, 1903-1908. Shivas, R. G. and Brown, J. F. (1984). Identification and enumeration of yeasts on Banksiu collina and Callistemon viminalis leaves. Truns. Br. Mycol. Soc. 83, 687689. Sinki, G . S. and Schlegel, W. A. F. (1990). Flavoring agents. In: Food Additives (Ed. by A. L. Branen, P. M. Davidson and S. Salminen), pp. 195-258. Marcel Dekker, New York. Smiley, J. T. (1985). Are chemical barriers necessary for evolution of butterfly-plant associations? Oecologia (Berl.) 65, 58@583. Smith, A. J. (1975). Invasion and ecesis of bird-disseminated woody plants in a temperate forest sere. Ecology 56, 19-34, Snodderly, D. M. (1979). Visual discriminations encountered in food foraging by a neotropical primate: implications for the evolution of color vision. In: The Behavioral Significance of Color (Ed. by E. H. Burtt, Jr), pp. 237-280. Garland, New York. Snow, D. W. (1970). Evolutionary aspects of fruit-eating by birds. Ibis 113, 194-202. Snowdon, A. L. (1990). Post-Harvest Diseases and Disorders of Fruits and Vegetables, Vol. 1. General Introduction and Fruits. C R C Press, Boca Raton. Somers, T. C. (1971). The polymeric nature of wine pigments. Phyfochemistry 10, 2 175-2 186. Sorensen, A. E. (1983). Taste aversion and frugivore preference. Oecologia (Berl.) 56, 117-120. Sorensen, A. E. (1984). Nutrition, energy and passage time: experiments with fruit preference in European Blackbirds (Turdus merula). J . Anim. E d . 53, 545-557. Spencer, K. C. (1988). Introduction: chemistry and coevolution. In: Chernicul Mediation ($Coevolution (Ed. by K . C. Spencer), pp. 1-1 1. Academic Press, New York. Stanghellini, M. E. and Aragaki, M. (1966). Relation of periderm formation and callose deposition to anthracnose resistance in papaya fruit. Phytopathology 56, 444450. Stapanian, M. (1982). A model for fruiting display: seed dispersal by birds for mulberry trees. Ecolrigy 63, 1432-1443. Starke, H. and Herrmann, K . (1976). The phenolics of fruits. VIII. Changes in flavanol concentrations during fruit development. Z . Lebensm. Unters.-Forsch 161, 131-135. Starmer, W. T. and Fogelman. J. C. (1986). Coadaptation of Drosi)p/iila and yeasts in their natural habitat. J . chem. Ecol. 12, 1037-1055. Starmer, W. T., Lachance, M.-A,. Phaff, H. J. and Heed, W. B. (1990). The biogeography of yeasts associated with decaying cactus tissue in North America. the Caribbean, and northern Venezuela. Evol. Biol. 24. 253-296. Stephenson, A. G. (1981). Flower and fruit abortion: proximate causes and ultimate functions. Ann. Rev. Ecol. Syst. 12, 253-279. Stevens, F. L. (1913). The Fungi Which Cause Plant Disease. MacMillan, New York. Stevens, F. L. and Hall, J. (1926). Diseuses of’ Economic Plants. MacMillan, New York.

88

M. L. CIPOLLINI AND E. W. STILES

Stiles, E. W. (1980). Patterns of fruit presentation and seed dispersal in birddisseminated woody plants in the eastern deciduous forest. Am. Nut. 116,67&687. Stiles, E. W. (1982). Fruit flags: two hypotheses. Am. Nut. 120, 500-509. Stiles, E. W. and White, D. (1982). Additional information on temperate birddisseminated fruits: response to Herrerd’s comments. Am. N u t . 120, 823-827. Stiles, W. C. and Abdalla, D. A. (1966). Harvesting, processing, and storage. In: Blueberry Culture (Ed. by P. Eck and N. F. Childers), pp. 28&301. Rutgers University Press, New Brunswick. Stoessl, A. (1970). Antifungal compounds produced by higher plants. Rec. Adv. Phytochem. I , 143-180. Stohr, H. and Herrmann, K. (1975). The phenolics of fruits. VI. The phenolics of currents, gooseberries, and blueberries. Changes in phenolic acids and catechins during development of black currents. Z. Lehensm. Linters.-Forsch 160, 210-215. Stretch, A. W. (1989). Biological control of blueberry and cranberry fruit rots ( Vaccinium corymbosum L. and Vaccinium macrocarpon Ait.). Acta Horticultura 241, 301-306. Sutherland, S. (1986). Patterns of fruit-set: What controls fruit-flower ratios in plants? Evolution 40, 117-128. Swain, T. (1977). Secondary compounds as protective agents. Ann. Rev. Plant Physiol. 28, 479-50 I . Swain, T. (1978). Plant-animal coevolution: a synoptic view of the Paleozoic and Mesozoic. In: Biochemical Aspects of’ Plunt and Aniniul Coevolution (Ed. by J. B. Harborne), pp. 3-20. Academic Press, London. Swain, T. (1979a). Phenolics in the environment. Rec. Adv. Phytochem. 12, 617-640. Swain, T. (1979b). Tannins and lignins. In: Herbivores- Their Interaction with Secondary Plant Metabolites (Ed. by G . A. Rosenthal and D. H. Janzen), pp. 657682. Academic Press, New York. Swartz, J . H. and Medrick, T. F. (1968). Antifungal properties of cranberry juice. J . appl. Microhiol. 16, 1 5 2 4 1527. Tanaka, T. and Ueno, Y. (1989). Worldwide occurrence of Fusurium mycotoxins nivalenol, deoxynivalenol, and zearalenone. In: Mycotoxins and Phycoto-uins ’88 (Ed. by S. Natori, K . Hashimoto and Y. Ueno), pp. 51-56. Elsevier, Amsterdam. Thompson, J . N. (1989). Concepts o f coevolution. Trends Ecol. Evol. 4, 179-181. Thompson, J. N. and Willson, M. F. (1979). Evolution of temperate fruit/bird interactions: phenological strategies. Evolution 33, 973-982. Tiffney, B. H. (1984). Seed size, dispersal syndromes, and the rise of the angiosperms: evidence and hypotheses. Annul. Missouri Bot. Garden 71, 551-576. Towers, G. H . N., Tso, A. and Maass, W. S. G . (1966). Phenolic acids and phenolic glycosides of Gaultheria species. Phytochemistry 5, 677-68 1. Trelease. S. F. and Trelease, H. M. (1929). Susceptability of wheat to mildew as influenced by carbohydrate supply. Bull. Torrey Bot. Club 56, 65-92. Tukey, H. B. and Young, J . D. (1939). Histological study of the developing fruit of the sour cherry. Bot. Gaz. 100, 723-749. Tuomi, J . , Niemela, P. and Siren, S. (1990). The Panglossian paradigm and delayed inducible accumulation of foliar phenolics in mountain birch. Oikos 59, 399410. Upadhyay, R. Y. (1981). Sporostasis between phylloplane microfungi and a foliar pathogen. E.uperientia 37, 833-835. Van den Berg, A. J . J. and Labadie, R. P. (1989). Quinones. In: Methods in Plant Biochemistry (Ed. by P. M. Dey and J. B. Harborne), Vol. 1. Plant Phenolics (Ed. by J . B. Harborne), pp. 451492. Academic Press, London. Van Etten, C. H. and Tookey, H. L. (1979). Chemistry and biological effects of

RELATIVE RISKS OF MICROBIAL ROT FOR FLESHY FRUITS

89

glucosinolates. In: Herbivores- Their Interaction with Secondary Plant Metabolites (Ed. by G. A. Rosenthal and D. H. Janzen), pp. 471-501. Academic Press, New York. Van Roosmalen, M. G. M. (1985). Fruits ofthe Guianan Flora. Institute of Systematic Botany, Utrecht. Van Sumere, C. F. (1989). Phenols and phenolic acids. In: Methods in Plant Biochemistry (Ed. by P. M. Dey and J. B. Harborne), Vol. 1. Plant Phenolics (Ed. by J. B. Harborne) pp. 29-76. Academic Press, London. Vance, C. P., Kirk, T. K. and Sherwood, R. T. (1980). Lignification as a mechanism of disease resistance. Ann. Rev. Phytopathol. 18, 259-288. Vander Kloet, S. P. and Austin-Smith, P. J. (1986). Energetics, patterns of ripening, and timing of seed dispersal in Vaccinium Section Cyanococcus. Amer. Midl. Nut. 115, 386-396. Vanderplank, J. E. (1984). Disease Resistance in Plants, 2nd Edn. Academic Press, Orlando. Varney, E. H. and Stretch, A. W. (1966). Diseases and their control. In: Blueberry Culture (Ed. by P. Eck and N. F. Childers), pp. 236-279. Rutgers University Press, New Brunswick. Verhoeff, K., Leeman, M., Van Peer, R., Posthuma, L., Schot, N. and Van Eijk, G. W. (1988). Changes in pH and the production of organic acids during colonization of tomato petioles by Botrytis cinerea. J . Phytopathol. 122, 327-336. Walkinshaw, C. H. (1989). Are tannins resistance factors against rust fungi? In: Chemistry and Signijicance of Condensed Tannins (Ed. by R. W. Hemingway and J. J. Karchesy), pp. 435446. Plenum Press, New York. Wang, P. L., Du, C. T. and Francis, F. J. (1978). Isolation and characterization of polyphenolic compounds in cranberries. J . Food. Sci. 43, 1402-1404. Wheelwright, N. T. (1985). Fruit size, gape width, and the diets of fruit-eating birds. Ecology 66, 808-8 18. Wheelwright, N. T. and Janson, C. H. (1985). Colors of fruit displays of birddispersed plants in two tropical forests. Am. Nut. 126, 777-799. Wheelwright, N. T. and Orians, G. H. (1982). Seed dispersal by animals: contrasts with pollen dispersal, problems of terminology, and constraints on coevolution. Am. Nut. 119,402413. White, D. W. (1989). North American Bird-dispersed Fruits: Ecological and Adaptive Signijicance of Nutritional and Structural Traits. PhD thesis, Rutgers The State University of New Jersey, New Brunswick. White, D. W. and Stiles, E. W. (1990). Co-occurrences of foods in stomachs and feces of fruit-eating birds. Condor 92, 291-293. White, L. S. and Fabian, F. W. (1953). The pectolytic activity of molds isolated from black raspberries. Appl. Microbiol. 1, 243-247. Wicklow, D. T. (1981). Biogeography and conidial fungi. In: Biology of Conidial Fungi, Vol. 1 (Ed. by G. T. Cole and B. Kendrick), pp. 417-445. Academic Press, London. Wicklow, D. T. (1988). Metabolites in the coevolution of fungal defence systems. In: Coevolution of Fungi with Plants and Animals (Ed. by K. A. Pirozynski and D. L. Hawksworth), pp. 173-201. Academic Press, London. Wilcox, M. S. (1939). Phomopsis twig blight of blueberry. Phytopathology 29, 136142. Wilcox, R. B. (1940). Cranberry fruit rots in New Jersey. New Jersey Agric. exp. Stat. Circ. No. 403. Williams, A. J.. Wallace. R. H. and Clark, D. S. (1957). Changes in the yeast

90

M. L. CIPOLLINI AND E. W. STILES

population of Quebec apple fruits during ripening. Can. J . Microbiol. 2, 645-648. Williams, C. A. and Harborne, J. B. (1989a). Biflavonoids. In: Methods in Plant Biochemistry (Ed. by P. M. Dey and J. B. Harborne), Vol. 1. Plant Phenolics (Ed. by J . B. Harborne), pp. 357-388. Academic Press, London. Williams, C. A. and Harborne, J. B. (1989b). Isoflavonoids. In: Methods in Plant Biochrmistry (Ed. by P. M. Dey and J. B. Harborne), Vol. 1. Plant Phenolics (Ed. by J. B. Harborne). pp. 421450. Academic Press, London. Willson. M. F. and O’Dowd, D. J. (1989). Fruit color polymorphism in a birddispersed shrub (Rhugodia parabolica) in Australia. Evol. E d . 3, 40-50. Willson, M. F. and Thompson, J. N. (1982). Phenology and ecology of color in birddispersed fruits, or why some fruits are red when they are “green”. Can. J . Bot. 60, 70 1-71 3. Willson, M. F. and Whelan, C. J . (1989). Ultraviolet reflectance of fruits of vertebrate-dispersed plants. Oikos 55, 341-348. Willson, M. F. and Whelan, C. J. (1990a). The evolution of fruit color in fleshyfruited plants. Am. Nut. 136, 790-809. Willson, M. F. and Whelan, C . J. (1990b). Variation in postdispersal survival of vertebrate-dispersed seeds: effects of density, habitat, location, season, and species. Oikos 57, 191-198. Willson, M. F., Irvine, A. K. and Walsh, N. G . (1989). Vertebrate dispersal syndromes in some Australian and New Zealand plant communities, with geographic comparisons. Biotropica 21, 133-147. Willson, M. F., Graff, D. A. and Whelan, C . J. (1990). Color preferences of frugivorous birds in relation to the colors of fleshy fruits. Condor 92, 545-555. Wilson, C. L. and Lawrence, P. P. (1985). Potential for biological control of postharvest diseases. Plant Dis. 69, 375-379. Wilson, C. L., Wisniewski, M. E., Biles, C. L., McIaughlin, R., Chalutz, E. and Droby, S. (1991). Biological control of post-harvest diseases of fruits and vegetables alternatives to synthetic fungicides. Crop Protect. 10, 172-177. Windels, C. E. and Lindow, S. E. (1985). Biological Control in the Phylloplane. American Phytopathological Society, St Paul. Witmer, M. C. (1991). The dodo and the tambalacoque tree: an obligate mutualism reconsidered. Oikos 61, 133-137. Wong, M. H. and Kwan, S. H. (1980). Fungal species isolated from the senescent leaves of Nayraudiic arundinacea growing on areas disturbed by iron-ore mining activities. Environ. Res. 22, 322-330. Wrangham, R. W. and Waterman, P. G. (1983). Condensed tannins in fruits eaten by chimpanzees. Biorropica 15, 21 7-222. Yabe, K., Ando, Y., Terakado, N., Hashimoto, J., Nakajima. H. and Hamasaki, T. (1989). Aflatoxin biosynthesis by mutants of Asprq$llzrs species. In: Mycotoxins and Phycotoxins ’88(Ed. by S. Natori, K. Hashimoto and Y. Ueno), pp. 21-28. Elsevier, Amsterdam. Yang, S. F. (1975). Ethylene biosynthesis in fruit tissues. In: Fucteurs ef Regulation dc. la Maturation cles Fruits (Ed. by R. Ulrich), pp. 171-176. Centre National de la Recherche Scientifique, Anatole. Yang, S. F. and Hoffman, N. E. (1984). Ethylene biosynthesis and its regulation in higher plants. Ann. Rev. Plant Physiol. 35, 155-189. Yang, X., Wilson, L. L., Madden. L. V. and Willis. M. A. (1990). Rain splash dispersal of Colletotrichutn acutatum from infected strawberry fruit. Phj!topathology 80, 59&595.

RELATIVE RISKS OF MICROBIAL ROT FOR FLESHY FRUITS

91

Yamazaki, N., Fry, S. C . , Darvill, A. G . and Albersheim, P. (1983). Host-pathogen interactions XXIV. Fragments isolated from suspcnsion-cultured sycamore cell I into proteins. Plani walls inhibit the ability of cells to incorporate ['Tleucine Phj,siol. 72, 864-869. Young, R. E., Salminen, S. and Sornsrivichai, P. (1975). Enzyme regulation associated with ripening in banana fruit. In: Facteurs ei Regulation de lr Mnturuiion des Fruits (Ed. by R. Ulrich). pp. 271-280. Centre National de la Recherche Scientifique, Anatole. Zangerl, A. R. (1990). Furanocoumarin induction in wild parsnip: evidence for an induced defense against herbivores. Ecology 71, 19261932. Zauberman, G. and Schiffman-Nadel, M . (1973). Changes in the ripening process of avocado fruit infected by Fustrriuni s o l t r i i . Pliytopathology 64, 188-190.

This Page Intentionally Left Blank

Oxygen Availability as an Ecological Limit to Plant Distribution R . M . M . CRAWFORD

I . Introduction . . . . . . . . . . . . . . . . . . . I1 . Plant Organs Liable to Oxygen Deprivation . . . . . . . . A . The HypoxicSeed . . . . . . . . . . . . . . . . B . Underground Organs . . . . . . . . . . . . . . . . . . C . Above-ground Organs with Limited Access to Oxygen 111. Plant Structure and Oxygen Supply . . . . . . . . . . . A . Distribution and Function ofAerenchyma . . . . . . . B . Mass Movement of Air in Aquatic Species . . . . . . . IV . Symbiosis and Oxygen Supply . . . . . . . . . . . . . A . Root Nodules . . . . . . . . . . . . . . . . . B . Nitrogen Fixation in the Rhizosphere of Aquatic Plants . . . C . Mycorrhizas . . . . . . . . . . . . . . . . . . V . Consequences of Oxygen Deprivation for Survival and Metabolism . A . Sensing Oxygen Deficiency in Plant Tissues . . . . . . . B . Cellular Effects of Oxygen Deprivation . . . . . . . . . C . Metabolic Adaptations to Anoxia . . . . . . . . . . D . Causes and Prevention of Post-anoxic Injury . . . . . . . E . Mineral Nutrition and Flooding Tolerance . . . . . . . VI . Oxygen and Plant Competition . . . . . . . . . . . . . VII . Consequences of Climatic Change for the Vegetation of Oxygen-deficient Habitats . . . . . . . . . . . . . . . VIII . Conclusions . . . . . . . . . . . . . . . . . . . Acknowledgements . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . .

93 97 97 110

118 120 120 123 126 126 129 129 133 134 141 144 150

152 158 161 171 171 171

I . INTRODUCTION The oxygen concentration of the Earth's atmosphere during the past 500 million years has increased unevenly rising to a peak at about twice its present concentration in the Upper Carboniferous (Berner and Canfield. 1989). By contrast since the Early Carboniferous. atmospheric levels of ADVANCES IN ECOLOGICAL RESEARCH VOL . 23

Copsright $: I Y Y Z Acudemic Press Limired All righrs of reproducrion in uny form rrsrrwd

ISBN s 1 2 4 1 3 9 2 3 - 5

93

94

R . M . M . CRAWFORD

carbon dioxide have declined from 0.3% to only one tenth of their former levels (Budyko et a/., 1987) with the result that it is now a limiting factor for plant growth. The seemingly plentiful present-day supply of atmospheric oxygen, as compared with the current trace amounts of carbon dioxide in the atmosphere, has led to oxygen being ignored as a discriminating factor in plant ecology. Even biochemically, oxygen limitations to plant survival have received little attention. James (1953), in his classic work Plunt Respirution, dismissed anaerobiosis as “not a way of life for higher plants”. The capacity for cytochrome oxidase to maintain maximal reaction velocity, even at low concentrations of oxygen (down to 1 kPa-the Pasteur Point-at the site of metabolic consumption), strengthened the belief that, apart from extreme conditions in anaerobic soils and very large organs, oxygen was rarely a limitation to plant survival. The well-developed aerenchyma tissue found in many wetland and aquatic species has reinforced the view that plants that live in poorly aerated soils avoid rather than tolerate anoxia (Armstrong, 1979; Kawase, 1981). Although the ventilation systems of wetland plants are widespread and undoubtedly effective, they d o not guarantee that underground or submerged plant organs always have an adequate supply of oxygen. Measurement and modelling of the ability of plants to supply air to underground organs has shown that oxygen supply will always be a limitation on depth of root penetration (Armstrong, 1979). However roots are not the only plant organs liable to suffer from inadequate supplies of oxygen. Depending on soil conditions and the season of the year, periodic oxygen deficits occur in many plant organs both above and below ground (Brix, 1989; Waters et a/., 1989; Weisner and Graneli, 1989). As this chapter will attempt to show. despite the relative abundance of oxygen in the Earth’s atmosphere and the facilitation of gaseous diffusion pathways, together with the efficiency of cytochrome oxidase, there are many situations where oxygen supply limits the growth and survival of higher plants. This limitation does not act equally on all species, consequently variation in oxygen availability is an important ecological determinant in a wide variety of habitats. There is also a case for considering oxygen as a resource in terms of competition. In common with other resources, oxygen is consumed and, depending o n the habitat and the size of the tissue concerned, it is not always replenished at a rate which matches consumption. The relatively high demand of the soil microbiota for oxygen ensures that there is universal competition for oxygen in poorly aerated soils. Consequently, oxygen availability can be a limiting factor and potentially lethal at various times in the life cycle of many species. The relative capacities of different plant species to survive during periods of shortage will therefore play a crucial role in plant

OXYGEN AND PLANT DISTRIBUTION

95

competition. When oxygen is viewed in this way as an ecological determinant, it forces a reconsideration of some established views on the mechanisms of competition, especially when competition is viewed as “the simultaneous demand by two species for the same resource which is in limiting supply” (Lincoln et al., 1982). In the penultimate section of this chapter a case will be made for removing this time constriction in defining competition. The effect of oxygen availability also provides, in some cases, a further example of the Montgomery effect (Montgomery, 1912) where ecological advantage is conferred by low growth rates in areas of low environmental potential. All eukaryotes, with the exception of some protozoa and yeasts, require oxygen for mitosis. All species and most organs therefore have a need for oxygen. However, not all species or all plant tissues or organs require a constant supply of oxygen. This phenological differentiation with regard to oxygen needs, often irrespective of growth rate and total annual oxygen demand, will allow some species to gain an advantage over others which are unable to suffer any interruption in oxygen supply. Three aspects of plant form give rise to tissues which risk suffering from oxygen deficits, namely the seed, underground organs (roots, rhizomes and tubers) and any other organs, aerial or subterranean which, by virtue of their size or external protecting tissue, impede the ready diffusion of oxygen from the atmosphere. Geological evidence indicates that there have been considerable fluctuations in the oxygen content of the atmosphere during the period that aerobic metabolism has been dominant on Earth. The extent of the fluctuations and the numbers of peaks and troughs varies depending on the method of calculation. Budyko et al. (1987) suggest three main peaks and troughs with oscillations between 4.45kPa and 50.1 kPa. However, a more recent conservative estimate by Berner and Canfield (1989) suggests only two peaks, one at about twice the present concentration of atmospheric oxygen in the late Devonian-Carboniferous followed by a plunge in the Permian and a lesser peak in the Cretaceous-Early Tertiary (Fig. 1 -for review see Raven, 1991). Given these extensive fluctuations it has been argued that the evolution and spread of the angiosperms would have been impossible before the Cretaceous due to limitations in oxygen supply. Taktajan (1980) considers that the angiosperms emerged at the beginning of the Cretaceous and that this coincided with recovery from the Permian oxygen plunge. Higher levels of atmospheric oxygen would have been needed for the development of flowers and the germination of seeds of flowering plants, as both processes require considerable amounts of oxygen to be available to compact tissues. However, the lower oxygen concentrations of oxygen in the atmosphere in the pre-Cretaceous period does not seem to have been any obstacle to the diversification of gymnospermous seed plants.

96

R . M. M. CRAWFORD

Time (my)

Fig. 1. A “best estimate” of atmospheric oxygen fluctuations versus time including a crude (shaded) error zone from a study of Berner and Canfield (1989).

A possible interaction between plant form and oxygen availability could have been responsible for the evolution of the monocots. The evolution of the monocots may have been an adaptation to aquatic habitats where roots were anchored in oxygen-deficient muds and consequently liable to periods of oxygen deprivation. The simplified structure of the monocots has long been regarded as an ancestral response of certain groups of land plants readapting to either an amphibious or aquatic habitat (Henslow, 1893; Arber, 1920).The simplification of leaf form, the reduction in secondary thickening, and the reliance on an adventitious root system, which are marked features in the monocots, suggest their suitability to aquatic or wetland habitats. Many authors have commented on the preponderance of monocot species in aquatic habitats. Only 4% of the dicotyledonous families can be regarded as aquatic compared with 33% in the monocots (Arber, 1920). Given the uncertain divisions between families, this figure can be regarded only as an approximation. However, as there are approximately 4.5 times as many dicot as monocot species in the world flora, it is all the more remarkable that the monocots hold such dominance over wetland and aquatic habitats (Fig. 2). In addition, the pre-eminent position of monocots in aquatic fringe habitats is

OXYGEN AND PLANT DISTRIBUTION

97

exemplified by the dominance of the sedges, rushes and reeds in the wetlands and swamps of the world. The ecological hazards and opportunities of oxygen-deprived areas are evident in the sharp community boundaries that are to be observed in vegetation in wetland habitats. It is sufficient for an area to be flooded only two or three times in a decade for the imprint of the high water mark to be clearly defined in the distribution of plant species. Flood-prone wetland habitats exhibit some of the best-defined boundaries to be found between plant communities (Fig. 3). Although the reduced availability of oxygen in flooded soils can be shown to be the predominant causal effect in producing these boundaries and creating distinctive communities, it does not follow that all species that live in such areas are tolerant of oxygen deprivation. Some species survive in these habitats because they are anoxia-tolerant, but others owe their success to adaptations which allow them to avoid an oxygen deficit. Nevertheless, it is the restriction of the oxygen supply that has imposed the environmental limitation that has evoked these varying adaptive responses. From the Arctic to the tropics, the pattern of plant distribution frequently bears the imprint of oxygen as a limiting factor, either directly by acting on plant metabolism or indirectly by altering the environment, and it is this phenomenon which forms the subject of this chapter.

11. PLANT ORGANS LIABLE TO OXYGEN DEPRIVATION A. The Hypoxic Seed The impermeability of seed coats to oxygen causes most seeds to undergo a period of natural anaerobiosis at germination even although the seed itself has not been subjected to any impediment in oxygen supply through burial or inundation. After imbibition there is a gradual acceleration of metabolism and the resulting demand for oxygen exceeds the rate of replacement until the testa is ruptured by the swelling of the cotyledons or the protrusion of the radicle. This natural anaerobiosis is most readily observed in large seeds with rapid germination. Thus in pea (Pisum sativum), french bean (Phaseolus vulguris) and chickpea (Cicer arietinum) there are marked accumulations of ethanol in the early stages of germination which decline only after rupture of the testa. In some species, for example P . vulgaris (Sherwin and Simon, 1969) and Fagopyrum esculentum (Effer and Ranson, 1967), the natural period of ethanolic fermentation is preceded by a period of lactic fermentation, while in chickpea (Aldasoro and Nicolas, 1980) simultaneous accumulations of ethanol, lactic and malic acids have been reported. The endurance of this

Fig. 2. The highly flood-tolerant Brazilian palm Mauriria fle.~uo.sa(“Buriti”) with its zonation following the course of a flood-prone stream. This species is yet a further example of the capacity of many monocotyledonous species to survive in wetland habitats.

OXYGEN AND PLANT DISTRIBUTION

99

Fig. 3. An aerial view of the dune and slack system at Tentsmuir National Nature Reserve (Fife, Scotland). The boundaries between the flood-prone slacks and the drier dunes are very marked. Mapping of these zonations in a number of locations has shown that they represent the maximum level of flooding and that their position has not changed by more than 20cm in the last 25 years. (Photo J. K . S. St Joseph, Crown Copyright Reserved.)

100

R . M. M. CRAWFORD

period of natural anaerobiosis by seeds is strictly limited and when it is prolonged by excessive burial, flooding or compaction then emergence is greatly reduced (Crawford, 1977).

I . Hypoxic and Anoxic Germination An early simple test used to distinguish seeds that require unrestricted access to oxygen for successful germination from those that can achieve the process (at least as far as radicle protrusion) with reduced oxygen supplies is to observe which species are capable of germination under water (Moringa, 1926). Thus celery, rice, carrot and lettuce are all capable of rupturing the testa and extruding the radicle under water, while peas, broad beans and maize are not. A more severe test is to determine which species are capable of germination in total absence of oxygen. The number of species that can achieve germination as far as radicle or coleoptile protrusion under total anoxia is more limited but still remarkable. The list includes not only rice (Taylor, 1942; Vartapetian et al., 1978; Alpi and Beevers, 1983) but also wild rice (Zizania ayuatica; Campiranon and Koukkari, 1977). In addition, various species of the genus Echinochloa, a graminaceous weed of rice fields, can also germinate under total anoxia at varying rates. This ability of Echinochlou to match the anoxia tolerance of rice in germination makes the genus a troublesome weed of paddy fields (Rumpho and Kennedy, 1981; Kennedy ef ul., 1983; Kennedy el al., 1987). The ability to germinate under anaerobic conditions is not restricted to aquatic species. In the myxospermous seeds (seeds with a coat that produces mucilage when wet) of the Brazilian dry forest tree Chorisiu speciosa, germination is capable of proceeding as far as radicle protrusion in total absence of oxygen (Joly, 1982). The advantages of mucilage in reducing desiccation injury and possibly also fungal attack have to be combined with an ability to extend the radicle anaerobically as the mucilage restricts oxygen access to the interior of the seed. If on protrusion the radicle still finds no oxygen available, further development ceases. A similar example of a dryland species being able to germinate under total anoxia is found in the S.E. African leguminous plant, the Kaffir broom Erythrina cufra (Small et al., 1989). The ecological advantage for this species probably stems from the ability of the seedling to break open the hard testa which, while intact, impedes the entry of oxygen. As with germinating seeds of C. sprciosa, after the rupture of the testa there is no further development in E. cuffra unless oxygen becomes available. Other species which are recorded as being able to germinate under very low concentration of oxygen include Cucumis sativus and Celosia argentea as well as the Indian lotus (Nelumho nucifera). This latter species is reported as being able to germinate with the help of the oxygen trapped within the seed as well as being capable of germination after 200 years of burial in anaerobic mud (Ohga, 1926).

OXYGEN AND PLANT DISTRIBUTION

101

The first stages of tissue expansion in the germination process are produced entirely by imbibition and cell expansion. Defining germination as radicle or coleoptile emergence, allows germination to be considered as being possible under total anoxia for some species, whereas all that has taken place is a physical expansion of preformed cells through water uptake. This limited definition, which in species such as radish amounts to little more than a rupture of the seed coat, is a questionable use of the term germination as there is no evidence of new growth or cell division. The initial stage of anaerobic germination of rice with full extension of the coleoptile under anoxia appears more vital, but even in this species the extent of development under total anoxia is also strictly limited and does not proceed beyond extension of the coleoptile (Vartapetian et ul., 1978; Crawford, 1982a). As pointed out by Kordan (1972) the normal under-water germination and development of rice is prevented if exchange of air with the water is impeded. The only stages of germination that even submersion-tolerant rice can achieve without oxygen are coleoptile protrusion and the formation of adventitious root primordia. There is no visible growth of adventitious roots in rice under anoxia (Kordan, 1976).

2. Energy Metabolism and Oxygen A vailability .for Germination A more revealing approach to understanding interspecific variation to oxygen tolerance is to examine the effects of anoxia quantitatively in terms of metabolic rate of the processes involved in the early stages of germination, namely mobilization of food reserves and respiration. Most seeds accumulate glycolytic end-products, notably ethanol during the first 24-48 h of germination. Access to oxygen is usually not sufficient for the needs of the accelerating metabolism of the germinating embryo. Consequently there is a reduction in ATP supply and this can be clearly quantified in terms of energy charge (EC). Energy charge can be defined as the ratio: EC =

(ATP) + O.S(ADP) (ATP) + (ADP) + (AMP)

The energy-charge ratio has proved a very sensitive indicator of any limitation to the proper functioning of aerobic metabolism (Pradet and Raymond, 1983) and has been used to determine the onset of anaerobic respiration in a variety of plant tissues including seeds, roots and rhizomes. The differences in oxygen requirement for germination in lipid- and starchcontaining seeds have been clearly demonstrated by comparative studies of EC values. From an energy standpoint it might be expected that lipids, with their greater energy supply per unit weight, would provide the ideal and therefore universal energy reserve for seeds. This is far from the case. Classification of seeds in terms of their food reserves shows that among those seeds with larger size and rapid germination, carbohydrates are probably the

102

R. M. M. CRAWFORD

i

\\\

\ \

\

‘ \

Starch seeds

\

*.

20

10 00

24

\\

;<, I

I I

48

72

120

96

Hours of anoxia 0-0 Wheat

U -U Turnip

0 -0 Rice

Cabbage

A -A Chickpea

- Flax

A

-A

Maize

v -v Lettuce

Fig. 4. Post-anoxic survival of seedlings subjected to various length of anaerobic incubation immediately after radicle emergence. Note the greater tolerance of anoxia in the starch seeds. (Crawford. unpublished.)

commonest energy reserve. Lipid respiration consumes more oxygen than carbohydrates and as seeds with hard, oxygen-impervious seed coats tend to suffer at least a partial oxygen deficit during the initial stage of germination, the metabolism of lipids will be subjected to greater hindrance from lack of oxygen than if the energy reserves were carbohydrates. Starch seeds invariably demonstrate a greater tolerance of anoxia (Fig. 4) and, as might be expected, the drop in energy charge, where lipid is the energy reserve in germinating seeds (e.g. turnip, radish, lettuce and cabbage), is particularly marked and can be observed as soon as the partial pressure of oxygen falls below the relatively high figure of 10 kPa. In carbohydrate-containing species the decline in EC values is less pronounced and only becomes noticeable at partial pressures of oxygen less than I kPa (A1 Ani et uf., 1982, 1985). Figure 5 shows the effect of progressive reduction in the partial pressure of oxygen on the ability to germinate of a range of species (Al-Ani et ul., 1985). Emergence of the coleoptile as in rice, which can take place in total anoxia, is not in itself a definition of germination in this particular test as radicle emergence in rice will not take place under total anoxia (see above, Kordan, 1976). The oxygen concentration values are plotted on a log scale to emphasize the low concentrations that are necessary to prevent germination in the starch-containing seeds. By contrast, in lipid-containing seeds germination is seriously retarded when the oxygen partial pressure falls below

OXYGEN AND PLANT DISTRIBUTION

103

Fig. 5. Relationship between oxygen concentration and germination rate in seeds with varying food reserves. The non-starch seeds have either lipid or protein reserves: Non-sturch seeds: I , radish; 2, turnip; 3, cabbage; 4,sunflower; 5, lettuce; 6, flax; 7, soya. Starch seeds: 8, rice; 9, wheat; 10, sorghum; 1 I , maize; 12, pea. Note that the reduction in oxygen concentration required to inhibit germination is 1-2 orders of magnitude lower in the starch seeds than in the non-starch seeds. Germination is defined as radicle emergence therefore the emergence of the coleoptile under anoxia in rice does not signify germination. (Data adapted with permission from A1 Ani et ml., 1985.)

104

R. M. M. CRAWFORD

10 kPa and ceases at 2 kPa, while carbohydrate seeds, although retarded in germination rate at partial pressures below 10 kPa, can still eventually achieve radicle protrusion at extremely low partial pressures, with rice, maize, wheat, sorghum and pea all able to germinate in the range 0.20.9 kPa. Experiments of this type clearly show the distinction between a viable metabolic activity leading to successful germination and processes that are limited to expansion of pre-existing cells through water uptake alone as in coleoptile extension. Although in some cultivars of lipid-containing seeds (lettuce, radish) there can be an appearance of germination, the radicle protrusion appears merely to be an expansion of preformed cells and there is no evidence of continued growth in the absence of oxygen.

3. Aquutic Seeds For over a century various botanists have commented on the peculiarities of seeds of hydrophytes in relation to germination. Crocker ( 1907) reviewed earlier experiments and noted that for many aquatic species scarification is necessary for germination which suggests that gaseous impermeability and possible lack of oxygen is a restricting factor in their germination. The role of oxygen as an ecological determinant for the germination of aquatic plants, however, is more clearly evident if species are grouped into those that will germinate with limited access to oxygen and those that will delay germination until after flooding has subsided. In some cases a period of flooding can enhance subsequent post-flooding germination as in Punicum larum. a grass native to Central and South America (Cole, 1977), a property that has helped this species to colonize grass-herb swamps in West Africa where it has been introduced. Seeds of the tropical and subtropical wetland tree Nyssa sy/vu/ica var. hiflora d o not germinate while flooded but d o so readily when flooding subsides. The seeds will germinate in water, but germination is enhanced by aeration (DeBell and Naylor, 1972). Many aquatic species are dependent on periodic lowering of water levels for reproduction by seed. With present-day regulation of river flow many of these seasonal or occasional reductions of river levels no longer take place to the detriment of bottomland forest regeneration. Most bottomland tree species require a period in which the water table drops for germination and seedling establishment. In the bottomland forests of North America it was sufficient in the past for this to take place once every 20-30 years. However, the absence of even this amount of occasional draw-down now seriously threatens regeneration of these forests in several areas such as the Mississippi basin and the swamp forests of Louisiana. Even the recruitment of such flood-tolerant trees as the bald cypress ( Taroilium distirhum, Fig. 6) and water tupelo N ~ ~ s s.Yjdvu/icu u is prevented when there is no relief from constant flooding (Conner e / (11.. 1981 ).

OXYGEN AND PLANT DISTRIBUTION

105

Fig. 6 . Pneumatophore of the flood-tolerant swamp cypress Ta.uodium distichurn. These pneumatophores develop best in areas with fluctuating water tables. Constant inundation due to artificial over-regulation of water levels is now threatening the regeneration of this species and other bottomland forest trees.

I06

R. M . M . CRAWFORD

In the large Amazon floodplain forests ( v a r x a and igapo, Figs. 7-9) resistance of the trees to flooding increases with size and age in common with most woody species. During the dry season seedlings of many species start to colonize low-lying areas but are subsequently eliminated during the following flood period. Observations of seed germination and seedling growth in Parkia auriculata (a common leguminous tree of the igapo forests of the Rio Negro) show that the dormancy of the seeds is due to an impermeable seed coat. Provided this is not scarified, the seeds can survive 6 months’ inundation and still have 85% germination on re-exposure to air. However, if the seeds are scarified and then submerged, death follows in a few days. Although this seedling does not grow while submerged, once established, it will remain alive throughout a 7-month inundation (Coutinho and Struffaldi, 1971). Seedlings of Cecropia species which are found in the Amazon flood plain forests are more sensitive to prolonged flooding and when completely inundated show stem-tip necrosis which increases the longer the seedlings are submerged. Successful establishment of seedlings in low-lying areas is therefore possible only during prolonged low-water periods or when there is a run of several consecutive dry years. This sensitivity to flooding, even in a swamp forest, demonstrates that in the Amazon floodplain forests and the bottomland forests of the southern States of the USA, the factors controlling species distribution can only be understood when flood events are considered over decades or even centuries (Junk, 1989). The oxygen dependence of germination in some aquatic species is frequently combined with an ability to detect the impedence of aeration through other associated environmental variables. Thus, although Phragmites australis is a wide-ranging species, able to survive considerable depths of flooding, seed germination is inhibited while the seeds are submerged, even by as little as 5cm of water (Spence, 1964). Many aquatic species such as wild rice, Zizania aquatica (Simpson, 1966) are stimulated to germinate by alternating temperatures which are indicative of shallower and usually better-oxygenated water. A double confirmation of shallow flooding can be sensed by seeds through the presence of light together with fluctuating temperatures. Species known to require this double stimulus for germination include Bidens tripartita, Lycopus europaeus, Lythrum salicaria (Frankland et al., 1987); Cyperus odoratus, Gratiola viscidula, Penthorum sedoides and Scirpus lineatus (Baskin at al., 1989). Some hydrophytic species, however, do germinate better if there is a reduction in oxygen availability. This requirement is particularly characteristic of the rooted hydrophytes. Seeds of these species may be dispersed by floating but germination is delayed until the seed sinks. Thus seeds of rvpha latifolia, although able to germinate o n a moist surface exposed to air, show an improvement when covered by water or exposed to reduced oxygen concentrations (Sifton, 1959). A more extreme condition is found in the

Fig. 7. The understorey of the flood tolerant “igapo” forest of the Rio Negro. These trees are normally flooded annually to a depth of 6 m for up to 6 months. In this particular forest the bare trunks are usually inundated for this period with only the crowns of the trees emerging above the flood water. How the roots survive the prolonged period of inundation in presumably anaerobic soils at a temperature of 27°C is unknown.

Fig. 8. Surface roots of a tree from the normally flood tolerant Rio Negro riverside forest. Although these surface roots, which have been exposed since the death of the tree after a period of prolonged flooding, may have access to oxygen in the moving river water the problem remains of how the deeper anchoring roots survive.

OXYGEN AND PLANT DISTRIBUTION

109

Fig. 9. The giant water lily Victoria amuzonica a frequent species of the nutrient-rich pools found in the flood-plain “varzea” forests of the Amazon basin.

obligately submerged cosmopolitan dicot of brackish waters, Najas marina, where germination takes place only in complete absence of oxygen (Van Vierssen, 1982). When seeds germinate under water they generally become lighter than water and tend to float to the surface once aerenchymatous tissues develop, unless they have specialized organs which either anchor them

110

R . M. M. CRAWFORD

or weigh them down (for review see Cook, 1987). Species that float to the surface after submerged germination can be found in Typha and Nymphoides spp. This contrasts with species that require anchoring, where some means of maintaining contact with the submerged soil is necessary to prevent the germinating seed floating to the surface. One method of remaining submerged is by the production of numerous ephemeral hairs which anchor the seedlings before the roots and leaves develop. The anchoring hairs then disintegrate once the seedling is rooted. Such prop hairs can be found in Limnocharis, Vallisneria and Stratiotes. Alternatively, species can be weighed down so as to remain on the bottom as in Myriophyllum, Sparzanium and Trapa where heavy endocarps hold the seedlings down until they are rooted (Cook, 1987). In view of the physiological problems of germination in oxygen-poor environments, it is not surprising that propagation from seeds is not the most frequent dispersal method in aquatic plants. The seed is often interpreted as a means of propagation especially adapted to terrestrial habitats where a drought-avoidance mechanism is essential for reproduction. Even the need for sexual recombination to facilitate adaptation to habitat alterations is minimized in the more constant conditions of the aquatic habitat. The need for drought avoidance is absent from the aquatic habitat and reproduction of many aquatic species relies instead on vegetative propagules which can act as both perennating organs (hibernaculae) and dispersal units. Such vegetative propagules in the form of specialized dwarf shoots, known as turions, are found in Hydrocharis, Myriophyllum and Utricularia. Morphologically different, but serving the same purpose, are the small tubers produced late in the season on the end of stolons in Sagittaria and Potamogeton. These organs, laden with food reserves and often protected by mucilage and a thick cuticle, lie dormant throughout the winter. Over-wintering dwarf tubers of Potamogeton$liformis when brought into warmer conditions can extend new shoots even in total absence of oxygen (Barclay and Crawford, 1982).

B. Underground Organs 1. Variation in Anoxia-tolerance in Perennating Organs Roots and underground stems (rhizomes and tubers) because of their size, variable porosity and epidermal protection, together with their burial to varying depths, are the plant organs most likely to be subjected to periods of oxygen deficiency. However, in any physiological study of adaptation to an environmental stress, it is important to examine not only the organ that is most likely to be subjected to the adverse conditions, but also to carry out the investigation at that season of the year when the stress is most likely to occur. Roots have been the natural choice of experimental material for many

OXYGEN AND PLANT DISTRIBUTION

111

researchers investigating the effects of flooding on higher plants. However, the choice of roots for study in relation to oxygen deprivation, although seemingly logical at first sight does not take into account the manner in which plants adapt to anoxia, particularly with regard to monocots where perennation depends on the survival of rhizomes and basal buds and is not dependent on maintaining an intact rooting system (Braendle and Crawford, 1987). Roots of most species, particularly when excised, have a very limited life when deprived of oxygen. In cotton and soybean roots, death from anoxia can take place in as little as 3 and 5 h respectively with a proportion of the roots being killed within 30 min (Huck, 1970). Survival in the absence of oxygen depends, not on an anoxia-tolerant root system per se, but on the possession of organs capable of tolerating anoxia and able to regenerate new root systems once an air supply is restored. Such organs are most frequently rhizomes and tubers, both organs capable of storing large quantities of carbohydrate. The only type of roots to fall within this category are the swollen tap roots of both herbaceous and woody plants. When the anoxia tolerance of such perennating organs, and in particular rhizomes, is compared with fibrous roots, their superior ability to withstand oxygen deprivation is very striking. There are no examples of excised roots (either monocot or dicot) of herbaceous plants that can survive more than a few days of anoxia, while detached rhizomes of some wetland species can be kept under total anoxia for 2-3 months. Thus the physiological tolerance of roots to anoxia shows only small differences between species which can be measured in hours or days, while the tolerance variation between the perennating organs (rhizomes and tubers) can range from a few days to 3 months (Barclay and Crawford, 1982). These variations in anoxia tolerance can also be related to differing ecological distributions of the species (Braendle and Crawford, 1987). Not all wetland species possessing rhizomes are, however, tolerant of anoxia. Rhizomes of Carex rostrata, Juncus efusus, J. conglomeratus and Glyceria maxima can be killed in as little as 4 days of total anoxia (Barclay and Crawford, 1982). The length of time that the rhizomes can survive under anoxia appears to be a function of the level of carbohydrate supply and respiratory activity. Thus in early spring when carbohydrate supplies are high, before shoot elongation has taken place, the rhizomes of Glyceria maxima are capable o f withstanding up to 3 weeks of anoxia and are still capable of shoot expansion and normal growth on return to air. The same treatment in June, after shoot elongation and when carbohydrate reserves are minimal, results in death in less than a week (Section VA and Steinmann and Braendle, 1984a; Braendle and Crawford, 1987). Similarly, cutting reeds (Phragmites australis) in June will reduce subsequent regrowth of shoots, while cutting in August when carbohydrate reserves are recharged has no adverse effects (Weisner and Graneli, 1989).

112

R. M. M. CRAWFORD

2. Re-emergence from Anaerobic Habitats For herbaceous species, survival in wetland and aquatic habitats poses the additional problem of re-emergence in spring. Frequently the perennating organ lies buried in anaerobic mud which is in turn submerged under water. Many species maintain a contact with the atmosphere through a snorkel from the dead remains of the previous year’s shoots. Thus old stems of the bulrush Schoenoplectus lacustris (Fig. 10) tend to fracture above the highwater mark and serve as an air channel (snorkel) to the submerged and buried rhizomes in winter (Steinmann and Braendle, 1984a). The dead culms of Phragmites australis also serve for the exchange of carbon dioxide and methane, the flux rates in winter being about 10% of those measured in summer. The winter rates for carbon dioxide varied between 30 and 130 ml m-2 of substratum per day depending on the type substratum in the lake and the density of the reeds. In populations rooted in organic sediments the flux rates for O,, CO, and CH, were higher than for plants rooted in sandy sediments (Brix, 1989). In winter the low metabolic rate of the underground organs of Phragmites australis does not exceed the capacity of the dead shoots to transport oxygen. However, in summer when the soil temperature reaches 15°C some regions of the roots can suffer an overall daily oxygen deficit and have to compensate with anaerobic respiration at night when the photosynthetically generated oxygen supply is not available (Gries et al., 1990) and the humidity-induced convection is less. This is a condition very similar to that observed in rice (see Waters et al., 1989). In spring, however, with rising soil temperatures and the absence of a fully developed aerial shoot, many aquatic plants appear to rely on energy from anaerobic respiration for re-emergence, irrespective of whether or not they have a snorkel-aided ventilation system still intact from last year’s shoots. Figure 11 shows the annual cycle in alcohol dehydrogenase (ADH) activity and ethanol content of four rhizomatous wetland species growing under natural conditions (Haldemann and Braendle, 1986). The high levels of ADH activity and detectable levels of ethanol in winter and spring are indicative of the use of anaerobic respiration in winter and spring. The high levels of ADH in late spring in Phragmites australis match the relatively late shoot emergence that takes place in this species. Although access to air hastens re-emergence in spring, a number of wetland and rhizomatous species can achieve shoot extension in total absence of oxygen. This capacity for shoot extension under total anoxia can be observed in Schoenoplectus lacustris as well as in Typha latijolia, T. angustifolia, Scirpus maritimus (Fig. 12) and S. americanus (Braendle and Crawford, 1987). It is not clear how rhizomes that are buried in anaerobic mud, under a considerable depth of water, sense the appropriate time to begin shoot extension. The temperature fluctuations and light levels under

Fig. 10. Beds of bulrush (Schoenoplectus lucustris) colonizing a fresh water loch in Orkney. The rhizomes of this species are highly tolerant of anoxia and can not only survive 2-3 months total deprivation of oxygen but also produce new shoots under total anoxia.

1 I4

R . M. M . CRAWFORD 1

1

1

1

1

1

1

1

1

1

1

G.maxima

A. calamus

-

1

2 0 00

T

h

I

.-C 0,

c

2

a a E

1000

I

.E

0

E

-0

P. australis

E

C

I

500

n

a

1

25G

0

J F M A M J J A S O N D

J F M A M J J A S O N D

Fig. 11. Annual cycle of alcohol dehydrogenase (ADH) activity and ethanol content of rhizomes in their natural habitats of Acorus calamus, Glyceria muxima, Phrugmites australis and Typha lutijofia. (Reproduced with permission from Haldemann and Braendle, 1986.)

mud covered by up to 1 m of water are minimal. At a depth of 2 m, daily fluctuations in temperature d o not exceed 0.5"C (Frankland et al., 1987). Ethylene, which is often implicated in shoot extension responses, is unlikely to play a role in anoxic shoot extension as oxygen is required for its synthesis from its precursor 1-aminocyclopropane- 1-carboxylic acid (ACC; Young and Hoffman, 1984).

3. Root Apex Hypoxia There have been many demonstrations that, even with unrestricted access to air, the growing meristems of both monocot and dicot roots have to rely partially on anaerobic respiration. The naturally hypoxic zone of the root is usually the distal 5-10 mm in species such as pea and maize, as evidenced by

Fig. 12. Rhizomes of Scirpus maritimus which have grown new shoots during a 2-week anaerobic incubation. Rhizomes of this species can survive at least 10 weeks under total anoxia and still produce viable shoots when transferred to air.

1 I6

R . M. M. CRAWFORD

ethanol synthesis in root tips and respiratory quotient values greater than unity. Whether the anaerobic activity was due to a lack of oxygen or merely to young newly-formed tissues not having developed a full complement of the enzymatic apparatus necessary for normal aerobic metabolism was for a time a subject of controversy (Ruhland and Ramshorn, 1938; Berry and Norris, 1949; Betz, 1958). Feeding radioactive glucose to different portions of maize roots either under normal atmospheric oxygen concentrations or with additional oxygen and then carefully distilling off the resulting ethanol and measuring its specific radioactivity showed that it was lack of aeration and not a metabolic deficiency that causes root meristem hypoxia (Crawford, 1976; Fig. 13). The generality of root-tip hypoxia has recently been extended to the roots of aquatic plants. In a detailed series of studies on the diurnal changes in radial oxygen loss and ethanol metabolism in roots of submerged and nonsubmerged rice seedlings (Waters et al., 1989) it was found that when rice plants were submerged, the onset of darkness caused a rapid fall in radial oxygen loss from the roots, a cessation of root extension and a rapid accumulation of ethanol. When light was restored, oxygen loss resumed within 3 min followed by a decline in ethanol concentration in the root medium. The authors attributed these diurnal cycles in submerged plants to the effect of the high boundary-layer resistance to gas exchange between leaf and water which hinders oxygen escape during the day and likewise the reverse flow at night. Still more revealing was the observation in non-flooded rice plants of a similar, although less marked, diurnal pattern indicating that even with their shoots in air, flood-tolerant rice roots can suffer an oxygen deficit during the night. The authors (Waters et al., 1989) expressed some surprise at this generation of ethanol at night in non-flooded rice plants, especially in view of the well-developed aerenchyma that rice possesses and the reports of aqueous oxygen concentrations as high as 0.16 mol m-3 at the surface of the root extension zone (Armstrong and Webb, 1985). However, in view of the earlier experiments cited above (Betz, 1958; Crawford, 1976) it is not unexpected that the dense meristematic root tissues situated at the distal end of a passive diffusive pathway are liable to remain hypoxic. The proximity of an aerenchymatous oxygen source is not in itself sufficient to ensure the passage of oxygen into a denser tissue if a readier means of escape is available. Diffusion will always be greater along the path of least resistance and therefore the high root-surface values for oxygen noted by Armstrong and Webb (1985) just before the extreme distal root zones could coexist with a state of hypoxia in the denser root meristem. The zone of oxygen diffusion represents the readiest point of exit of oxygen and may disguise the fact that there is an adjacent hypoxic root-tip. Given the difficulties in providing an adequate nocturnal oxygen supply, even to the aerenchyma-endowed roots of rice, it is not surprising that wheat

5 8000 0

0

0

0 7000 L Q)

a

E

a

-a 6000 V

600

-

c

I

z

2

O 2 uptake

5000

0 )

e

0,

500

0

I

r

0 '

0

00

2

/

2

E 'c

-3-

400

4000

-3w 1

0

c

0 3

-

C

c

-0

300

3000

W

2000

0

g X

\

C

0

W

0

200

0

c

V

3

t5

+5

Root seq m e n t s

I00

mm

\

Fig. 13. Manifestation of natural hypoxia that exists in the roots of maize seedlings when grown in air. Plotted against the root segments position is the specific activity of the combusted ethanol from the root segments that were incubated with 0.05 M 3I4Cpyruvate for 4 h under air, together with the oxygen consumption of the same segments.

118

R . M. M. CRAWFORD

roots also exhibit a similar shortage of oxygen (Thomson et al., 1989a, b). In these studies the effects of hypoxia were investigated in relation to K + / N a + selectivity. Energy is required to maintain the electrochemical gradient which controls the extrusion of N a + and permits the uptake of K' and therefore abnormal imbalances of these ions will give an indirect indication of an energy deficiency such as would be diagnosed more precisely by low energycharge values. Root apices of wheat seedlings were found to have low K + / N a + selectivity, even on exposure to high oxygen concentrations (0.1 15 mol m-3. Furthermore, it was concluded that the stele, like the root apex, is also oxygen deficient as there is no resumption of K + transport to the shoot for the first 2 h after re-aeration (Thomson et al., 1989b). Clear metabolic evidence for stelar anoxia has recently been reported for maize roots grown in air, based on the selective accumulation of ethanol in the stele as compared with the cortex (Thomson and Greenway, 1991). Thus oxygen diffusion into the stele, like that to the apex, is inadequate to fulfil the demands made by normal aerobic respiration, even in seedlings (Fiscus and Kramer, 1970).

C. Above-ground Organs with Limited Access to Oxygen I. Fruits Any bulky tissue with high rates of respiration is liable to suffer periods of oxygen deprivation. The particular case of germinating seeds has already been discussed (Section IIA). Fruits such as apples, pears and bananas can also produce ethanol as a result of high respiration rates and low surface to volume ratios (Bufler and Bangerth, 1982). Reduced tissue ventilation causes fruits to experience carbon dioxide concentrations above their level of tolerance and can result in accumulations of acetaldehyde and ethanol without the imposition of hypoxic conditions (Zagory and Kader, 1988). Limited access to oxygen can interfere with metabolic processes other than the main respiratory pathways. Reduced oxygen supplies limit oxidases such as ascorbic acid oxidase, polyphenol oxidase and glycolic acid oxidase where affinities for oxygen are five to six times lower than that of cytochrome oxidase (Solomos, 1982). The whole process of natural senescence is dependent on oxygen. Oxygen concentrations below 8 % can decrease ethylene production and sensitivity in both fresh fruit and vegetables, possibly due to the need for oxygen for the conversion of the ethylene precursor 1aminocyclopropane- 1-carbox ylic acid.

2. Wood Woody tissues such as tree stems and branches have long been suspected of having to endure spontaneous oxygen shortages described as asphyxia by Devaux (1899). Ethanol can frequently be detected in the trunks and roots of

OXYGEN AND PLANT DISTRIBUTION

119

forest trees (Crawford and Baines, 1977). Whether the ethanol found in the stems originates from roots and is found in the stem as a result of dispersion, or whether it is generated in the stem as a result of impeded aeration, has been the subject of a number of studies. A correlation between root ethanol content and that found in the stem has been observed in forest-grown trees of Pinus sylvestris and Picea abies, where peak ethanol values were recorded in spring and autumn (Crawford and Baines, 1977). Mass spectrometry of gases extracted from woody stems of Acacia mearnsii showed that the most actively metabolizing tissues existed in a gaseous environment that was almost exclusively carbon dioxide (Carrodus and Triffett, 1975). Such high concentrations of carbon dioxide, as well as implying a hypoxic condition, will also mean that the tissues are surviving in CO, concentrations that are known to inhibit enzymes such as succinic dehydrogenase (Miller and Hsu, 1965). Carrodus and Triffett were under the impression that ethanol did not accumulate in the metabolic tissues of woody organs. However, more recent studies (Kimmerer and Stringer, 1988) have clearly demonstrated that the cambium of several tree species is naturally hypoxic as shown by the presence of significant concentrations of ethanol, acetaldehyde and ADH activity. Contrary to expectations, these authors did not find any positive relationship between bark thickness and ethanol concentrations. Consequently, it was concluded that it was the rate of consumption of oxygen by the cambium rather than the relative impedence of oxygen diffusion by the cork layer which gave rise to the state of hypoxia. Easily detectable activities of pyruvate decarboxylase, which is often taken as a sign of at least hypoxic conditions, were also observed in the bark of Populus deltoides. Defoliation of this species also led to almost a 10-fold increase in ethanol concentration in the xylem sap. Kimmerer and Stringer (1988) also pointed out that although at first sight reliance on anaerobic respiration may appear energetically wasteful, this is only so if the end-product of glycolysis, ethanol, is lost to the atmosphere. If the ethanol is remetabolized by tissues with better aeration, such as the leaves, then the energy loss is minimized. ADH is present at high activity in the leaves of Populus delroides (Kimmerer, 1987) and may serve to remetabolize ethanol from the transpiration stream. In extensive experiments, Kimmerer and Kozlowski (1982) were unable to detect ethanol evaporating from leaves of unstressed trees, suggesting that this mechanism is at least sufficient to scavenge the ethanol from the transpiration stream of trees growing under optimal conditions. However, given the relatively low concentrations of ethanol in the leaf tissue and the heavy balance in favour of ethanol remaining in the liquid phase from the gas/liquid partition coefficient, the negligible rate of exit of ethanol from leaves in the vapour phases is perhaps not surprising. The role of lenticels in aerating woody stems has recently been examined both in their role in providing aeration for flooded trees and for the dispersal

120

R. M. M. CRAWFORD

of the volatile products of glycolysis. It is a common observation in many flood-tolerant trees that there is a hypertrophic development of the lenticels after a period of flooding (Hook and Scholtens, 1978). It is usually assumed that this hypertrophy serves to aid aeration to the submerged parts of the tree. However, sealing the bark of Acer rubrum and Betula nigra seedlings with lanolin and “Parafilm” does not affect either the growth or survival of these trees when flooded (Tripepi and Mitchell, 1984). Quantitative assessment of the role of lenticels in venting ethanol vapour from flooded roots of Pinus contorta (Fig. 14) has shown that although ethanol can be readily detected (for review see Gill, 1970), nevertheless the amount that exits from the lenticels is less than 0.2% of that produced by the roots. The greater proportion of that generated in the flooded roots of P . contorta moves upwards via the transpiration stream to the needles and not through the lenticels (Crawford and Finegan, 1989). For an alternative explanation of the possible functions of hypertrophy and lenticels see the discussion of “Knudsen diffusion” or “thermo-transpiration” in Section IIIB.

111. PLANT STRUCTURE AND OXYGEN SUPPLY

A. Distribution and Function of Aerenchyma Cell division in eukaryotes requires oxygen. Consequently, any plant which is able to extend roots or rhizomes into permanently waterlogged and anaerobic soils must be able to supply these organs with an adequate supply of oxygen, at least during periods of growth involving cell division. Welldeveloped aerenchyma, which is such a conspicuous feature of many wetland and aquatic species, provides large and continuous air spaces which facilitate the downward diffusion of oxygen from shoot to root (Armstrong, 1979). When flooded, many species rapidly increase their root porosity through aerenchyma development. A recent review (Jackson, 1990) shows that there is much evidence to support the view that plant hormones mediate this response by virtue of their speed of activity and susceptibility to rapid change in concentration following the imposition of aeration stress. There are many striking correlations between the development of aerenchyma and the ability of plants to inhabit wetland sites (Kawase, 1981; Justin and Armstrong, 1987). Typical of such studies is the comparison of six species of Rumex taken along an elevation gradient in a river ecosystem in the Netherlands (Laan et al., 1989a, b) where it was shown that low-elevation species produced new roots more rapidly on flooding than the high elevation species. The low-elevation species also had roots with a root porosity greater than 10% of root volume while the species from upper elevations had porosities of

Fig. 14. Sampling gas that has exited from the stem of a young tree of Pinus contorta for analysis of ethanol content by gas liquid chromatography. Note the collar fixed around the stem of the tree to trap gases exiting from the lenticels.

122

R . M . M . CRAWFORD

typically 6 7 % . Smirnoff and Crawford (1983) in a study of a number of species found that a value of more than 10% for root porosity distinguished flood-tolerant from flood-intolerant species. The other function of the aerenchyma-facilitated oxygen pathway to the roots that has been related to wetland survival is the ability to provide an oxidizing atmosphere at the root surface for the oxidation and immobilization of potentially toxic concentrations of ferrous iron and reduced manganese (Armstrong, 1979; Gries et al., 1990). As with all adaptations that are not universal, it is necessary to consider the disadvantages that result from their existence. A possible drawback to the presence of aerenchyma is a reduced capacity for nutrient uptake (KonEalova, 1990). Roots adapted to flooding with aerenchyma developments are usually thick and with little branching and with consequently a reduced surface/volume ratio. An adverse consequence of this strategy may be a diminution of the total surface available for nutrient uptake. This will be further compounded when the roots are suberized as an adaptation against premature oxygen leakage. A solution to this problem may be reflected in the prevalence of dimorphic roots systems as found in graminoids growing in shallow water, e.g. Carex gracilis (Koncalova and Pazourek, 1987). These plants form soil roots which are usually thick and poorly branched and aquatic roots with numerous finely branched hairs. (Note: a metabolic dimorphism in relation to ADH activity is also seen in barley roots (Section VCI; Fagerstedt and Crawford, 1986). However, the ease with which oxidized iron cores can be observed along roots in reducing soils, together with the undisputed need for oxygen for root growth and the obvious existence of an aeration pathway through the aerenchyma in many wetland species, may have satisfied scientific curiosity too easily in relation to explaining satisfactorily all the functions and the remarkable size of some aerenchymatous tissues. Two aspects of the generally accepted role of aerenchyma and diffusive internal movement of gas within flood-tolerant plants have, however, been questioned, namely the excessive size of the air spaces in many aerenchymatous species and the dependence on diffusion alone for the internal movement of gas in plant tissues. The first aspect relates to the size of the intercellular spaces in many species which has prompted the question that the internal air space may be bigger than is necessary just to facilitate the downward diffusion of air (Williams and Barber, 1961). This suggestion has been examined quantitatively and refuted by Armstrong (1972) on the basis that rooting depth and hence exploitation of soil volume is dependent on root porosity (see also Justin and Armstrong, 1987). Although this quantitative evidence is convincing, the occurrence of very large air spaces along the entire length of the rhizomes of some wetland species, e.g. Potentillu pulustris, as well as in the distal regions of roots immediately within 2cm of the apex (Smirnoff and

OXYGEN AND PLANT DISTRIBUTION

123

Crawford, 1983) does raise the question of whether in these particular locations their size is necessary to facilitate oxygen diffusion in its final arrival at sites of metabolic activity. In order to explain the evolutionary forces which could have brought about such an extensive development of aerenchyma in the most distal root zones it is necessary to postulate a functional significance for the size at these particular locations. The suggestion that aerenchyma may serve as an oxygen reservoir is not tenable. Even those species with the best developed aerenchyma d o not contain enough oxygen to sustain their oxygen requirements for more than 120 min (Crawford, 1982a). An alternative explanation, however, could be that large air spaces will provide an immediate and efficient means of diluting toxic volatile compounds produced in roots under conditions of poor aeration (Crawford, 1989). The removal of volatile products of anaerobiosis has been shown to increase the length of time that many plants species can be kept under anoxia (Crawford and Zochowski, 1984; Crawford et al., 1987). There is no reason for supposing that the function of aerenchyma in the upward movement of the volatile respiration products (carbon dioxide and acetaldehyde) from roots and rhizomes is in any way less important for survival than the movement of oxygen downwards. Although ethanol can be detected by smell (e.g. the popular name “brandy bottle” for the yellow water lily Nuphar lutea), it is unlikely to be removed in very great quantities by gas movement since the partition coefficient for ethanol favours the liquid phase (see above, Crawford and Finnegan, 1989). Acetaldehyde and carbon dioxide will be more readily removed in this fashion and their depletion in moving anaerobic gas streams may account for the increased survival of germinating seeds when kept in circulating as opposed to static anaerobic atmospheres (see Section VC2).

B. Mass Movement of Air in Aquatic Species The movement of the internal atmosphere of water lilies has also been the basis of the second aspect of internal aeration in aquatic plants recently to have received renewed attention and to have questioned the dogma that submerged plant organs rely solely on diffusion as the sole means of moving oxygen to their rhizomes and roots. For the past 80 years there has been little questioning of the passive nature of oxygen movement by diffusion. The view that the mass movement of the interchange of gases in plants was inconceivable (Brown and Escombe, 1900) remained unchallenged until recently, despite reports by early French workers (Raffineau-Delile, 1841; Mergrt, 1874; see Dacey, 1987) that the movement of gas within certain aquatic species was not entirely diffusive but was aided by internal pressurization made possible by small pore-diameter barriers to diffusion. The terms used in describing the kinetic properties of gases in relation to their passage through

124

R . M. M. CRAWFORD

small pores have become somewhat confused in the biological literature with various authors using different terms with seeming synonymity. To clarify this discussion the following definitions have been kindly provided by Dr W. Armstrong (see also Armstrong at al., 1991). Knudsen diffusion describes gas movement through tubes (or porous media) where pressure is so low that collisions between molecules are so infrequent compared with collision with the walls of the tube, that they can be disregarded. As with normal diffusion, net movements of a species are determined by concentration gradients. At atmospheric pressure the Knudsen regime is realized only when tube/pore diameters are very small (molecular mean free path length, < 0.1 pm) and Knudsen diffusion coefficients are numerically smaller than those of normal diffusion in which moleculemolecule collisions dominate (Leuning, 1983). The terms thermo-osmosis and thermal transpiration often seem to have been used synonymously. In context, however, thermal transpiration seems to have historical precedence and to have been more rigorously defined (Armstrong et al., 1991). If two chambers filled with the same gas are separated by a porous partition in which pore diameters are less (usually much less) than the mean free path length of the molecules and if the chambers are maintained at different temperatures, there will be a net flow of gas from the colder to the warmer chamber. This flow, by Knudsen diffusion, is known as thermal frunspirution and will result in the raising of pressure in the warmer chamber. I f the two chambers are also connected by a large diameter pipe to allow a return flow, the process of thermal transpiration will continue indefinitely (Armstrong, personal communication). On the basis of Knudsen diffusion and thermal transpiration, the entirely passive movement of gas through plant tissues has now been challenged for a number of aquatic species (Dacey, 1980; Dacey and Klug, 1982) as well as for alder trees (Grosse and Schroder, 1984). The fact that water lily leaves can maintain a pressure differential means that the pore diameters surrounding the internal leaf air cavities must be very small. In Nuphar lutea, Schroder er al. ( 1 986) estimate the pores to lie in the region between the palisade tissues and the spongy parenchyma and to have diameters ranging from 0.7 to 1.2 pm. Such pores are indeed small and less than the residual aperture of even closed stomata (about 5.6 x 2.4 pm). In these conditions the difference in temperature inside and outside the leaf results in a thermal transpiration pressure which will persist as long as a temperature gradient exists. Evaporation of water inside the leaf also tends to increase the total internal gas pressure (hygrometric pressure). Recent research (Armstrong and Armstrong, 1990) suggests there are good grounds for believing that a humidityinduced diffusion of oxygen and nitrogen is of much greater significance for causing pressure flows than is thermal transpiration.

OXYGEN AND PLANT DISTRIBUTION

125

Integration of flow rates from data with tracer studies using ethane in the yellow water lily (Nuphar lutea) has shown that the total internal gas pressure can be sufficient to produce a flow through a single petiole of 22 litres of air per day (4.6 litres of oxygen) with maximum rates of 60 ml min-' (Dacey, 1980). In terms of linear flow this can be 8.5 mm s-I in water lily petioles and 1.7 mm s- in Phragmites australis rhizomes (Armstrong and Armstrong, 1988, 1990). The efficiency of the solar pumping system will depend on the various resistances encountered in petioles, stalks and rhizomes. However the process seems functional even for long-distance transport. In Phragmites australis rhizomes 1 m long with a pith cavity radius of 7.5mm and diaphragms at 100-mm intervals, a total pressure deficit of 100 Pa is sufficient to drive a convective gas flow of 100cm3 min-' through the rhizome which theoretically would make it possible to support fully the aeration of rhizomes up to 1 m in length. However this maximal figure could be significantly reduced depending on culm resistance (Armstrong et al., 1988). Solar energy can thus be considered as a means of forced ventilation for submerged rhizomes and roots. In addition to supplying oxygen to submerged and buried organs this pressurized flow will also aid the removal of volatile products of anaerobiosis, COz, CH, and also acetaldehyde which may result as an oxidation product of anaerobically generated ethanol. In alder trees (Alnus glutinosa), a species capable of surviving long periods of flooding, the downward movement of gas from stem to roots has similarly been shown to be actively driven by a thermo-osmotic pressurization within the air-space system of the stems. The brownish pigments of the bark warm up in sunlight and can create a 3.6"C gradient between the stem and the surrounding air (Grosse and Schroder, 1985). This temperature gradient establishes a pressure gradient between the inner tissues of the stem and the surrounding air when the inner tissues are separated from the atmosphere by a porous layer with pore diameters <0.1 pm (Grosse and Schroder, 1985). The assumption from these observations is that the internal atmosphere in the alder tree is then forced downwards under pressure to the roots where presumably a more porous connection between the root and the outside atmosphere allows its ready escape. The analogy with the through-flow system of rhizomatous plants, however, is not exact and raises the problem of whether the system for trees will be as effective as that for rhizomatous wetland species. There may be an experimental artefact in the tree experiments as the roots were allowed to emerge into a gas-filled head space where broken lateral roots may have allowed greater than normal egress of gas. However, putting these reservations on one side, it is remarkable that the capacity of the alder trees to provide their roots with an adequate supply of oxygen is increased eight-fold as a consequence of hypertrophy of the lenticels during a prolonged period of flooding. Grosse and Schroder therefore assume that the phellogen is the region of tissue where warming

126

R . M. M. CRAWFORD

drives the gas downwards to the roots. Comparison of the capacity for oxygen diffusion from shoot to root between a number of European tree species shows that gas diffusion is relatively slow in alder but increases to rates comparable with the other tree species tested once the stems are exposed to light. In 12-month-old leafless alder trees after an equilibration time of 20 min in light, the air influx to the roots was estimated to reach 5.82 p1 air min-I. When the trees had leaves, air influxes of 14.7 pI air min- I (2.7 pl oxygen) were transported by this mechanism (Grosse and Schroder, 1985). This capacity for thermal transpiration was found not only in alder but also in Aesculus hippocastanurn which also had a low dark diffusion rate. Mass flow of gas as an aid to ventilation of submerged and underground organs has also been suggested as a possible mechanism of air movement into pneumatophores in the black mangrove Avicennia nitida (Scholander et al., 1955). Mangroves are subject to regular tidal inundation and each time the pneumatophores are submerged, the internal atmosphere is isolated from outside air, oxygen is consumed and the carbon dioxide produced dissolves in the surrounding water. When the tide recedes, a partial vacuum due to carbon dioxide depletion then draws in a fresh supply of air. Further studies have shown that the oxygen concentration in the root system of mangroves is rarely less than 50% of that found in air (Chapman, 1976). In a detailed quantitative study, Curran et al. (1986) calculate that, contrary to the suggestions of Dacey (1980), diffusion alone is sufficient to supply the needs of roots of Avicennia marina and there is no need to invoke any explanation based on an active system of gas transport as in water lilies. Curran et al. ( 1986) in their pneumatophore study, however, did observe respiratory quotients of less than one, with the amount of carbon dioxide emitted being less than would be expected. This phenomenon of gas movement driven by C0,-solubilization has been observed in other submerged root systems (Brix, 1988; KonEalova et al., 1988). Possible metabolic consequences of this observation are discussed below (Section V).

IV._SYMBIOSISAND OXYGEN SUPPLY A. Root Nodules The nitrogen-fixing bacteria and actinomycetes have a particularly delicate relationship with regard to aeration. The accomplishment of nitrogen-fixing reactions is dependent on an adequate supply of ATP from aerobic respiration. However, oxygen concentrations must not be too high in those regions of the cell where nitrogen fixation takes place as it will inhibit the reducing nitrogenase activity. In the legumes this dilemma is resolved by

OXYGEN AND PLANT DISTRIBUTION

127

limiting oxygen concentration by the diffusive resistance of the root nodule while augmenting the supply of oxygen at low concentrations with the help of leghaemoglobin and cytochrome P450.Leghaemoglobin, due to its property of combining with oxygen, can maintain a steady flux of bound oxygen which greatly exceeds that of free dissolved oxygen. Thus, although the oxygen concentration at the bacteroid surfaces is low due to diffusive resistance, the leghaemoglobin-facilitated supply mechanism ensures that ATP generation is not impaired. It has now become evident that leghaemoglobin is considerably more widespread in its occurrence than was previously realized. Both the nodules of leguminous and non-leguminous species (actinorhizal plants) possess leghaemoglobin (Appleby et al., 1983) and it has recently been suggested that genes for haemoglobins are of widespread occurrence in plants and may act as oxygen-sensing mechanisms synthesizing haemoglobin possibly as a signal molecule indicating an oxygen deficit and the need to shift plant metabolism from an oxidative to a fermentative pathway of energy generation (Appleby et al., 1988; Section V). Nitrogen-fixing nodules appear to use a combination of haemoglobin and variable resistance to oxygen diffusion to achieve the necessary degree of anoxia for nitrogen fixation without exposing the entire metabolism of the nodule to the limitations of an anaerobic existence. Thus the relatively high levels of haemoglobin (for actinorhizal plants) in Casuarina and Myrica nodules, where infected cells have typically low PO, levels, suggest a need for oxygen transport similar to that in legume nodules (see review in Sprent and Sprent, 1990). Nevertheless, despite these adaptations the problem of aeration of a compact underground organ such as a root nodule in a soil with fluctuating water and air content is delicately balanced and environmental stresses such as drought or flooding can reduce the oxygen supply and lower the capacity to fix nitrogen. Root nodules can be classified into two types, determinate and indeterminate. In the former, growth is produced by division of infected cells and the resulting nodule is more or less spherical, with a minimal surface to volume ratio and therefore more likely to suffer problems in diffusive oxygen supply. To this determinate type belong soybean nodules, as well as those of Phaseolus and cowpea ( Vigna unguiculata). Indeterminate nodules have a longer-lived meristem, are less spherical and persistent infection threads are required to spread the bacteria to newly formed cells. Indeterminate nodules are found in Pisum sativum and Trifolium spp. as well as in actinorhizal species. Indeterminate nodules apparently lack lenticels but may have gaps allowing gas exchange to take place over the surface of the nodule (Sprent, 1979). Of the two types of nodule, it is the spherical, determinate nodules that appear more sensitive to even minor perturbations in aeration with adverse effects on their capacity to fix nitrogen.

128

R. M. M. C R A W F O R D

On soybean nodules (determinate type) lenticels on the surface of the nodule connect via intercellular air spaces with the centre of the nodule, though there may be a possible diffusion barrier in the region of the cortex (Bergersen and Goodchild, 1973; Pankhurst and Sprent, 1975). This arrangement appears to be a common feature for most determinate nodules. In a study of nitrogen fixation in soybean root nodules (Sprent and Gallacher, 1976) it was found that both drought and flooding lowered C,H,-reducing activity and enhanced CO, and ethanol evolution. The induced anaerobiosis could be relieved by supplying oxygen both to droughted and flooded nodules. It appears that the drought-induced shrinkage of the root nodules induced a state of at least partial anaerobiosis. In addition to being sensitive to drought, many legumes are intolerant of waterlogging. Comparisons of waterlogged nitrogen-fixing plants with waterlogged nitrate-fed plants show that the nitrogen-fixing plants suffer more, and in the least-tolerant species, such as Vigna unguicufata, a large proportion of nodules can be lost (Sprent, 1979). The sensitivity of nitrogen fixation to flooding has surprised some workers, given the ability of many plants to transport oxygen from the shoots to the roots by diffusion. However it has to be remembered that this diffusion pathway, just as with the root tip (Section IIB3), operates best through the path of least resistance; hence nodules with their relatively compact tissues are unlikely to receive an adequate supply of oxygen in flooded soils. In the British flora two wetland species are notable as being actinorhizal (non-leguminous) nitrogen fixers, namely Myrica gale and Alnus glutinosa, with related actinorhizal species in other parts of the world sharing this property. Excavation of the nodules of these species frequently shows that the root emerges from the nodule with a reversed geotropism, growing upwards to reach the soil surface. Bond (1952) suggested that this property helped Myrica gale to colonize bogs. Physiological confirmation was subsequently obtained by Tjepkema (1978) who showed that the upwardly directed nodule roots wer? important for oxygen uptake in water culture experiments at low oxygen tensions. Comparison of the oxygen demands of isolated cultures of Frankia, the nitrogen-fixing organism of Myrica, Afnus and other actinorhizal plants, shows surprisingly that nitrogen fixation is maximal at or about atmospheric oxygen levels. This contrasts with Rhizobium where nitrogenase activity is expressed only at extremely low oxygen concentrations (for review see Warwick et af., 1990). This difference is associated with the different means used to exclude oxygen from the nitrogenase site in the two species. Rhizobium relies on the structure of the nodule impeding the entry of oxygen to the central nitrogen-fixing area, while in Frankia the production of vesicles by the actinomycete provides its own intrinsic oxygen protection mechanism.

OXYGEN AND PLANT DISTRIBUTION

129

B. Nitrogen Fixation in the Rhizosphere of Aquatic Plants The problem that even terrestrial nodule-bearing plants face in providing a suitable oxygen supply for the energetic needs of nitrogen fixation, would suggest that this process is unlikely to be found in aquatic plants. However, nitrogen fixation can be detected in aquatic species. Studies of Potamogeton jiliformis growing in Loch Leven in Scotland have revealed considerable acetylene-reducing activity in the roots (Sylvester-Bradley, 1976). This was attributed to organisms very similar to Spirillum lipoferum which found the micro-aerophilic conditions resulting from oxygen transported to the rhizosphere suitable for growth. Marine angiosperms have also been found to support significant amounts of heterotrophic nitrogen fixation. In a study of three tropical species Diplanthera wrightii, SyringodiumJiliforme and Thalassia testudinium and one temperate species, Zostera marina, significant amounts of fixation were found only in the tropical species. In this case the capacity of the plants to support nitrogen fixation was attributed, not to the existence of a micro-aerophilic site, but on the contrary, to well-developed anaerobic conditions. Thalassia testudinium rhizomes are located at considerable depths (at least 60cm) in anaerobic mud. Anaerobic metabolism can result in the release of organic material suitable for the growth of anaerobic nitrogen-fixing rhizosphere bacteria (Patriquin and Knowles, 1972). The salt marshes of Georgia with their lush stands of Spartina altern@ora are noted for their high productivity and this may also be due to the anaerobic substrate causing an organic exudate to leak from the roots and rhizomes of the Spartina and thus providing the rhizosphere conditions suitable for nitrogen fixers (Hanson, 1977). Despite this lack of direct experimental evidence for symbiotic fixation of nitrogen in Zostera, the lush growth of this species in some habitats could conceivably be due to nitrogen fixation.

C. Mycorrhizas 1 . Tree mycorrhizas The practical importance of mycorrhizas in facilitating plant nutrition, particularly of trees in boreal forests, has prompted a number of investigations as to the effects of environmental variables on their viability and efficacy (Brundrett and Kendrick, 1988). In practical forestry it is commonly noted that even when trees are pre-inoculated with mycorrhizal fungi, if they are planted out into wet peaty soils this association can disappear and only after the trees have become established do the typical mycorrhizal root systems again become evident. Investigations into the adverse effects of flooding on mycorrhizal associations have been complicated by the need to determine experimentally whether it is a simple lack of aeration that causes the association to break down or whether the host tree, suffering itself from

130

R . M. M . CRAWFORD

flooding, fails to provide the fungus with an adequate supply of carbohydrate. In an attempt to solve this problem, Read and Armstrong (1972) compared the growth of the mycorrhizal fungus Boletus variegatus on living coniferous roots and on hollow silicone rubber “roots” in anaerobic media. In the living roots it was possible to show in older suberized roots that fungal growth was restricted to areas of the main axis where laterals appeared and where oxygen leakage could be detected polarographically. To answer the question of whether it was just the oxygen leakage that was essential or whether it was also better access to carbohydrate at these leakage points that encouraged the growth, the fungus was cultured in a carbohydrate medium and oxygen delivered through an artificial silicone rubber “root”. This use of an artificial “root” showed that access to oxygen was clearly a prerequisite for fungal growth. The above experiment, although showing that oxygen and carbohydrate are necessary for mycorrhizal growth, demonstrates yet again the basic conditions required by all eukaryotes for growth involving cell division, namely a physiological requirement for oxygen (Section I). Ecologically however, it is important to recognize that the relative tolerance of species or associations of species to survive a stress and grow again when conditions become more favourable is not addressed by this type of experiment. To answer the question of why young trees in wet soils are often nonmycorrhizal, it is necessary to know if the fungus is so badly injured by a lack of oxygen that it is not available for growth during periods of inadequate aeration. Figure 15 shows two mycorrhizal fungal species that have been grown in culture and kept under anoxia for 12 months. During this period photographic records of the cultures taken at monthly intervals indicate their ability to grow (Crawford, unpublished). Although this experiment does not provide any direct information on the effects of anoxia on a mycorrhizal root, it has shown that a number of mycorrhizal species in addition to those shown in Fig. 15, e.g. Heheloma mesophaeum, Paxillus involutus. Laccaria proxima, and Laccaria laccata can all endure periods of total anoxia that extend between 6 months and a year. Recent investigations on the tolerance of Sitka spruce roots to waterlogging (Coutts and Nicoll, 1990) showed that although the hyphae of the ectorrhizal mycelium and roots were killed by early flooding in October, when the roots were metabolically active, later flooding was less damaging. With regard to the anoxia tolerance of the mycorrhizal fungi in the field it was noted in this study that hyphal strands survived flooding and new hyphae grew from them after flooding subsided (Fig. 16).

2. Vesicular-arbuscular mycorrhizas Vesicular-arbuscular (VA) mycorrhizas are widespread in terrestrial plants. However, in common with other mycorrhizal associations they have in the

OXYGEN AND PLANT DISTRIBUTION

131

Fig. 15. Two species of mycorrhizal fungi which under pure culture have survived and grown during 12 months of incubation under total anoxia (a) Paxillus involutus (b) Heheloma crustuliniforme.

past been considered as unsuited to aquatic and wetland habitats. Khan (1974) observed no mycorrhizal associations in 16 hydrophytes and noted that terrestrial plants became non-mycorrhizal when flooded. Other reports (Harley, 1969; Read et al., 1976) have also concluded that aquatic plants were not infected by VA mycorrhizal fungi or else the infections are low and present only during dry periods. Similarly in an investigation of the influence of aeration on the efficiency of VA mycorrhizas it was shown that at low oxygen tensions mycorrhizal plants of Eupatorium odoratum infected with Glomus macrocarpus lost their efficacy in facilitating nutrient uptake (Saif, 1981). The first report of VA mycorrhizas in shallow-water aquatic plants was by Sondergaard and Laegaard (1977), who found an Endogone sp. forming VA mycorrhizas in Litorella unijora, Lobelia dortmanna, Callitriche

Fig. 16. The effects of flooding on the mycorrhizal association between Thelophara terrestris and the root system of Picea sitchensis. (a) The extramatrical hyphal system has grown down from healthy unflooded roots grown in an acrylic tube of peat. The horizontal lines mark previous positions of the mycelium. (b) The ectomycorrhizal mycelium and roots have been killed as a result of flooding in early October. Note however that the hyphal strands have survived and from these new hyphae grow after flooding has subsided. (Reproduced with permission from Coutts and Nicoll, 1990.)

OXYGEN AND PLANT DISTRIBUTION

133

hamulata, Eleocharis palustris and Phragmites australis but not in MJv-iophj+ lum alternflorum and Isoetes lacustris in samples taken from four oligotrophic lakes in Denmark. More recently a study of 22 submerged water-plant species in New Zealand found VA mycorrhizal associations with records down to depths of 6 m in one lake and at 2-3 m depths in two other lakes (Clayton and Bagyaraj, 1984). All these depths were lower than the minimum water levels ever recorded for these lakes and thus demonstrate that a period of emergence is not necessary for VA mycorrhizal development. Nevertheless, the degree of infection showed a general decline with increasing water depth with the highest frequency of infection occurring in shallow-water plants, particularly those either without root hairs or with only sparse root hair development. This matches the observation with terrestrial plants that as root hair cover increases the percentage infection decreases (Chilvers and Daft, 1981). The suggestion that the mycorrhizal association is a feature of water-plants from oligotrophic lakes (Sondergaard and Laergaard, 1977) does not appear to be a valid generalization. In the New Zealand study no relationship was found between the incidence of infection and the trophic status of the lake. Furthermore, VA mycorrhizas have been reported with infection levels as high as 96% in water hyacinth plants growing in nutrientenriched drains (Venkataramanan, 1982). Whether or not VA mycorrhizal associations have a survival value to aquatic plants will depend ultimately on their competitive ability in relation to resource exploitation. Persistent growth depression can result from mycorrhizal infections if nutrients are not limiting host growth (Smith, 1980). Consequently the ability of water-plants to exploit the deeper water zones could be adversely affected by mycorrhizal infections reducing carbohydrate supply in a habitat where light rather than nutrients is the limiting factor. The role of VA mycorrhizas in aquatic plants is therefore most likely to depend on the growth form of the plant rather than the environmental conditions of the lake. Thus in plants with an extensive absorptive shoot surface available for nutrient uptake such as in Myriophyllum (a non-infected plant in the Danish study-see above) there would be little advantage in extending the absorptive capacity of roots with a VA mycorrhizal infection.

V. CONSEQUENCES OF OXYGEN DEPRIVATION FOR SURVIVAL AND METABOLISM The opening sections of this chapter presented evidence for the extent of variation in tolerance to anoxia found in plant organs. Some of the greatest variations in anoxia-tolerance are encountered in rhizomes where survival can vary between 4 days and 3 months. Large differences can also be found in germinating seedlings, provided the experimental temperatures are no higher

134

R . M. M. CRAWFORD

than those likely to be experienced in the field. If seedlings that have just germinated to the stage of producing their radicles are placed under total anoxia at temperatures resembling those found in spring in northern Europe (5-10”C) their tolerance of anoxia can range from a few days to several weeks (Barclay and Crawford, 1982; Fagerstedt and Crawford, 1987). As these differences in anoxia tolerance in seedlings and rhizomes have been obtained by incubating the plants in closed anaerobic incubators with no external access to air, they must reflect variations at a cellular level in anaerobic viability. The last few years have seen an increase in interest in this aspect of plant variability, partly from the losses sustained in agriculture, horticulture and forestry from winter flooding and poor seed emergence in wet springs, and partly from an increased ecological interest in the physiology of wetland vegetation. Irrigation can also have dangers for crop plants. The cotton crop in Soviet Central Asia suffers from root anoxia before the water demand of the growing shoots is fully met. There is also a tendency in plant breeding and tree selection to concentrate on yield to man with resilience to environmental stresses such as flooding being ignored. Some of the newer varieties of rice have proved to be more susceptible to flood-damage from monsoon flooding than the lower yielding older varieties (Crawford, 1989). In forestry the current tendency to select for “plus” trees, based on maximum growth of the trunk, and with complete disregard of the root and its ability to over-winter in flood-prone soils, is likely to aggravate the problem of wind-throw, which already causes serious losses to forestry in northern Britain. For all these various reasons there is a need, both in wetland conservation and in the selection of new varieties of both crop species and forest trees, to know what aspects of cell metabolism are changed when roots and perennating organs are deprived of oxygen. Oxygen deprivation can act both directly and indirectly on plant tissues. Direct effects are those where restriction of oxygen supply impinges on the metabolic function of the plant tissues. Indirect effects are those caused by environmental alterations due to a lack of oxygen which exposes plants to secondary effects of anoxia such as increased concentrations of ferrous or manganous ions. In the following sections of this chapter the term “anoxic injury” will wherever possible be restricted to the direct effects of oxygen deprivation on tissues. Where the effect is indirect or secondary, as with exposure to reducing conditions in the soil, this will be described as flooding tolerance.

A. Sensing Oxygen Deficiency in Plant Tissues The concept of a critical oxygen pressure (COP) above which oxygen supplies are “normal” or adequate (Armstrong and Gaynard, 1976) can be misleading unless the tissues concerned and the environmental history are very

OXYGEN AND PLANT DISTRIBUTION

135

precisely defined (Bertani and Brambilla, 1982). Adjustments, both metabolic and morphological, can alter the rate of supply and consumption of oxygen so that concentrations that are adequate at one season of the year may be totally inadequate at another. Some tissues such as apical meristems, due to their high metabolic rate, coupled with lack of porosity for ready access to oxygen, may have to endure a perpetual state of hypoxia (see Section IIB3). In maize roots tips at 25°C the COP,,, value is 30 kPa (Saglio et al., 1984) which, being greater than that of air, indicates that these roots must always be suffering a metabolic deficit in relation to oxygen. Such a condition may be regarded as inadequate in terms of satisfying the potential oxygen demand of these tissues but as it is a constant condition in most root meristems (Berry and Norris, 1949; Griffin, 1968) it cannot be considered as abnormal. Drew (1990) draws attention to the fact that in roots the oxygen status of the tissues is probably heterogeneous with a core of anoxic cells lacking molecular oxygen surrounded by a cylinder of hypoxic and then fully aerobic cells. The low K , values for cytochrome oxidase (cytochrome a3 K,,, 0.11.0mmol m-3, Bonner, 1973) with respect to oxygen would appear to hinder any plant or animal tissue from sensing directly any reduction in oxygen supply before it has been depleted to near the Pasteur point (1 % 0,). Numerous studies have been carried out on plant and animal tissues to search for mechanisms which might sense low oxygen concentrations and initiate an adaptive response either metabolically or morphologically. Changes in form (lenticel proliferation, adventitious root formation, aerenchyma development) are obvious indicators that plants d o adapt to low oxygen concentrations and have been extensively reviewed (Gill, 1970; Hook and Scholtens, 1978; Drew, 1990). If a plant can respond morphologically to changing aeration it should follow that metabolic adaptation is also possible. This present discussion therefore concentrates on the ability of plants to alter their metabolism so that their survival under low oxygen conditions is enhanced. Metabolic sensing of low oxygen concentrations can be considered under three headings: (a) Compensating mechanisms which alter respiration rate in relation to oxygen supply. (b) Use of signal molecules which respond to changes in oxygen concentration (e.g. haemoglobin). (c) Responses to environmental signals other than oxygen which are associated with low oxygen concentrations (e.g. carbon dioxide). ( a ) Metabolic oxygen sensing. Low oxygen concentrations per se can be seen to be sensed directly by free-living protozoans such as Tetrahymena pyriformis in which the motile cells, when placed in an oxygen gradient, swim

136

R . M . M . CRAWFORD

towards the region of higher oxygen concentration. In this case the respiration rate of the organisms has a direct dependence on the cytosolic [ATPI/ [ADP][P,I and intramitochondrial [NAD']/[NADH] ratios. Both these ratios have been studied extensively in animal tissues particularly in relation to their effect on respiratory rate (Wilson et a / . , 1979). Although the respiratory rate was essentially independent of oxygen concentration down to 10 PM with an apparent K,,, of less than 2 PM, progressive cytochrome c reduction occurred with decreasing oxygen tension at all concentrations below 200 PM. Thus, in this protozoan, as in other examples of animal cells, mitochondria1 oxidative phosphorylation is oxygen dependent throughout the physiological range of air-saturated water and below (Wilson et al., 1977, 1979). Similar situations are reported for other micro-organisms where cytochrome content is very variable and changes with dissolved oxygen content (Harrison, 1976). This aspect of plant response to low oxygen supplies has not yet been investigated in relation to flooding tolerance, but would appear nevertheless to offer scope for a degree of capacity adaptation that would have clear survival potential in low oxygen environments. ( h ) Signal molecules. Haemoglobin until recently had been recorded only in the nitrogen-fixing nodules of plants (Section IV). Low concentrations of haemoglobin have, however, now been found not only in the roots of plants that are capable of nodulation, but also in the roots of species that are not known to nodulate, and it has been suggested that it may be a component of the genome of all plants (Appleby et a / . , 1988). The low concentrations of the molecule present in roots rule out any function similar to that served by haemoglobin in rhizobium or actinorhizal root nodules of facilitating oxygen supply at low concentrations. An alternative function according to Appleby and co-workers might be for haemoglobin to serve as a signal molecule indicating an oxygen deficit and the need to shift plant metabolism from an oxidative to a fermentative pathway. In common with many other postulated signal molecules (e.g. phytochrome) the manner of transfer of the signal to the site of action remains elusive. Despite the lack of a causal explanation, the possibility of signal molecules for oxygen nevertheless remains an attractive hypothesis and raises the question of whether any other molecules could serve this purpose? ( c ) Environmental signals associated with oxygen deficiency. For both plant and animal tissues any depletion of oxygen supply is usually associated with an increase in carbon dioxide concentration. In plants, most experiments on carbon dioxide effects have been concerned with photosynthetic efficiency and experiments on other possible regulatory functions have been confined largely to fruit ripening and seed germination. However, in human and animal physiology the danger of narcosis from high carbon dioxide concentrations has prompted considerable research. In animal tissues increases in

OXYGEN AND PLANT DISTRIBUTION

137

carbon dioxide concentration stimulate adaptive respiratory responses such as hyperventilation which can be interpreted as an environmental signal which alerts the organism to the possibility of an impending oxygen deficiency (Milic-Emil and Tyler, 1963). The existence of such a phenomenon in animals raises the possibility that plants may also be able to operate such an adaptive sensing system. It is therefore of interest to note that some aquatic species d o respond to high carbon dioxide concentrations by reducing their metabolic rates. In studies of the respiration of whole mangrove seedlings (Avicennia, Bruguiera, Rhizophoru) in oxygen concentrations varying from 0 to 21 %, gas exchange was found to be markedly reduced by the presence of external carbon dioxide (Brown et al., 1969). By contrast exposure of germinating seedlings of the anoxia-intolerant chickpea (Cicer arietinum) to anaerobic environments with high carbon dioxide concentrations ( I 5%) increases glycolytic activity and reduces viability (Fig. 17). In the rhizomes of the dryland species Tussilago furfura and Elymus repens there is a similar response with glycolytic rate increasing and post-anoxic survival being reduced (Figs 18 and 19). Wetland species such as Iris pseudacorus and Scirpus americanus by contrast are relatively unaffected in glycolytic rate or anoxia survival by high carbon dioxide concentrations (Crawford, unpublished). Most of the research on the effect of high carbon dioxide concentrations on plant organs has been concerned with determining the optimum conditions for storage of various fruits and vegetables (Kader, 1986). Due to their size and the use of low oxygen concentrations, coupled with high carbon dioxide, fruits and vegetables stored under controlled atmospheres experience a certain degree of hypoxia. In a study of the effect of 10% carbon dioxide concentrations on pears (Kerbel et al., 1988), it was found that the reduction in respiratory activity that this treatment induced was associated with a reduction in the activity of ATP : phosphofructokinase (PFK) and PPi : phosphofructokinase (PFP) which suggests that high carbon dioxide concentrations may affect the control points in the glycolytic pathway mediated by these enzymes. Regulation of glycolysis by PFK is influenced by intracellular pH (Turner and Turner, 1980) and theoretical calculations indicate that carbon dioxide concentrations in excess of 5% will generally lower intracellular pH (Bown, 1985; Siriphanich and Kader, 1986). These studies together with those on the effect of carbon dioxide on anoxia-tolerant and anoxia-intolerant plants discussed above suggest that, depending on the pH and buffering capacity of the cell cytoplasm, plants may respond differently to accumulating carbon dioxide concentrations in anaerobic atmospheres. As suggested by Roberts et al. (1984a, b) the regulation of cytoplasmic pH is an important factor for survival under hypoxia and high accumulations of carbon dioxide may be a factor in accelerating the damage to intolerant species under anoxia.

Fig. 17. Adverse effects of adding 15% carbon dioxide to the anaerobic atmosphere of germinating chickpea seedlings. Carbon dioxide appears to trigger an acceleration of glycolysis with increased ethanol accumulation by the seedlings (see text, also Figs 18 and 19).

OXYGEN AND PLANT DISTRIBUTION

139

Per cent survival of Tussilago farfara

Fig. 18. Reduction in post-anoxic survival of Tussilugo,furfururhizomes as a result of addition of 15% carbon dioxide to the anaerobic incubation atmosphere. (Crawford, unpublished.)

Pyruvate decarboxylase (PDC) also shows large increases in activity when maize roots (Wignarajah and Greenway, 1976) and rice coleoptiles (Morrell et al., 1990) are transferred from air to hypoxic conditions. This transfer will of itself lead to a reduction in cytoplasmic pH (anaerobic acidosis-Roberts et al. (1984b). PDC has optimal activity under acid conditions (pH 6.0) and it has been estimated from incubation experiments of purified extracts that during incubation at pH 7.4 there would be a long-term decrease in activity to at most 7% of the activity in anoxic cells due to changes in substrate levels and cytoplasmic pH alteration (Morrell et al., 1990). This enzyme is therefore

140

R . M. M. CRAWFORD

Per cent survival of Elymus repens

Days of Anoxia Anoxia+CO,

~

Anoxia

Fig. 19. Reduction in post-anoxic survival in E/.vrnus repens rhizomes as a result of addition of 15% carbon dioxide to the anaerobic incubation atmosphere. (Crawford, unpublished.)

another possible point where carbon dioxide concentration in the anaerobic atmosphere may exert a controlling influence on metabolic rate. In dryland species periods of hypoxia and reliance on anaerobic respiration are likely to be transitory. Under such conditions it is therefore no long-term disadvantage to accelerate anaerobic respiration and draw on carbohydrate reserves to make good a temporary short-fall in metabolic energy through anaerobiosis. This dichotomy in behaviour could account for the difference in wetland and dryland rhizomes in their response to carbon dioxide in the anaerobic atmosphere. Similarly in germinating seeds, those that normally germinate rapidly in non-flooded soils will probably achieve

OXYGEN AND PLANT DISTRIBUTION

141

the greatest survival by counteracting temporary periods of anaerobiosis with continued growth, especially if it facilitates the early rupturing of the testa and greater accessibility of oxygen to the germinating seed. If this interpretation is correct then carbon dioxide concentrations in hypoxic or anoxic atmospheres could serve as an environmental signal that allows plant tissues to anticipate the effects of oxygen depletion, something which cannot be sensed directly due to the low K , values for cytochrome oxidase.

B. Cellular Effects of Oxygen Deprivation I . Immediate Eflects of Oxygen Deprivation Irrespective of whether or not tissues can survive more than a few days under anoxia the immediate effect of oxygen deprivation is to cause a rapid reduction in energy charge (Section IIA2). In the anoxia-intolerant seeds of lettuce, the energy charge falls to values as low as 0 . 2 4 . 3 in a matter of minutes and there is no recovery from these low values (Hourmant and Pradet, 1981). In species that can endure prolonged periods of anoxia, as in the bulrush (Schoenoplecrus lucustris), the energy charge falls rapidly during the first 24 h of anoxia and then stabilizes at about 0.5 in over-wintering rhizomes. In summer when the rhizomes are metabolically more active, recovery takes place during continued anaerobic incubation and after 2 days the energy charge values rise to values between 0.7 and 0.8 (Monk and Braendle, 1982; Steinmann and Braendle, 1984b; Braendle, 1990). Processes dependent on ATP which are inhibited as a result of oxygen starvation include the H' transport system from cytoplasm to vacuole (Bennett and Spanswick, 1984), fatty acid desaturation, sterol biosynthesis and cation selectivity (Thomson et al., 1989b). The inhibition of H' transport from the cytoplasm to the vacuole leads to a reduction in cytoplasmic pH (glycosidic or anaerobic acidosis). In a non-tolerant species such as Zea mays, tissue acidification can take place within 2 min of the onset of hypoxia leading to root death in 12-24 h (Roberts et al., 1984a,b, 1989). One of the most revealing recent studies in relation to anoxia tolerance reported recently has been an examination of the ability of young maize plants to acclimatize to oxygen deficiency by a hypoxic pretreatment before being subjected to total anoxia (Saglio et ul., 1988). Most experimental studies on oxygen deprivation are performed by abruptly transferring plants or detached organs to a totally anoxic environment so that the transition from aerobic to anaerobic metabolism is rapid. Such a sudden withdrawal of oxygen may not allow the expression of an alternative metabolic pathway, as total anoxia is likely to suppress more synthetic reactions than a partial oxygen deficit. Under hypoxic conditions some acclimatization and alteration of metabolic mechanisms and enzyme induction may take place. Estimates of the average lifetime of ATP in aerobic maize roots tips are as

142

R . M. M. CRAWFORD

low as 8 s (Roberts et a/., 1985),consequently under total anoxia there will be a rapid decline in nucleotide triphosphates in a few minutes. Under natural conditions such changes would be expected to take place over a few hours or days allowing cells the possibility of acclimation. When young intact maize plants were treated with a 2 4 kPa partial oxygen pressure (hypoxia) for 18 h before their root tips were excised, there was an improved energy metabolism during the subsequent period of anoxia with higher ATP/ADP ratios. These metabolic changes were accompanied by an increase in anoxia tolerance from 7 h for non-acclimatized tissues to 22 h in the roots that had received the hypoxia pretreatment (Saglio et al., 1988). A subsequent study using intact maize root tips showed that bubbling the culture solution of maize seedlings with 4% oxygen for 18 h increased the subsequent tolerance of anoxia to 96 h compared with a limit of only 24 h with seedlings that had been bubbled with 40% oxygen (Johnson et al., 1989). A similar case of metabolic acclimation has been found in winter cereals. In winter wheat it has been shown that seedling death through ice encasement is due to anaerobic injury (Andrews and Pomeroy, 1983). The extent of this injury can also be reduced by hypoxic acclimatization. Exposure of cold-hardened winter wheat (Triticum aestivum) to low-temperature flooding provides a sub-lethal hypoxic treatment which increases subsequent tolerance to anoxia resulting from ice encasement (Andrews and Pomeroy, 1989).

2. Long-term Eflects of Oxygen Deprivation As discussed above, it is the plant species which inhabit the most extreme anaerobic habitats that show the greatest ability to survive prolonged periods of oxygen deprivation (Fig. 20). Compared with aerobic respiration, anaerobic metabolism is costly in terms of carbohydrate reserves. Not surprisingly, therefore, species that can survive extended periods of anoxia and live in the most anaerobic environments typically have large carbohydrate reserves. These reserves show considerable fluctuations throughout the year with a typical maximum in autumn and a minimum in early summer after shoot extension (Haldemann and Braendle, 1986). In both Typha latifolia and T . angustifofia the highest dry-weight values for the rhizomes are found in late autumn (Fiala, 1978). Similarly tracer experiments with I4CO, on Spartina alterngora show that the rhizome is the major carbohydrate sink in the growing season (Lyttle and Hull, 1980). Over winter, a sharp fall in the carbohydrate content has been observed in the rhizomes of Typha latifolia from 45 to 27% of their total dry weight (Kausch et al., 1981). In Spartina alterngora the high winter levels of carbohydrate reserves are most severely depleted in spring in zones of rapid development (Gallagher, 1983; Gallagher et al., 1984). A similar pattern is observed in Schoenoplectus lacustris (Steinmann and Braendle, I984b). Such marked fluctuations in carbohydrate levels indicate the high metabolic cost to the plant of resuming growth in

Fig. 20. Rhizomes of Iris pseudacorus prepared for prolonged anaerobic incubation inside an anaerobe jar containing a gas mixture of hydrogen, nitrogen and carbon dioxide in the presence of a catalyst which ensures the reduction by hydrogen of any oxygen that may enter the system. Under conditions like these Iris pseudacorus rhizomes can be kept alive for more than 6 weeks. Anaerobic life can be further extended if the jars are placed in a refrigerated incubator at 10°C.

144

R . M . M. CRAWFORD

spring from rhizomes that are deeply submerged in an anaerobic habitat. In the surface-living rhizomes of A . calamus it is noticeable that the fluctuations in carbohydrate are not so pronounced and never sink in early summer to as low a level as that found in those of the deeper-buried species (Haldemann and Braendle, 1986). Surface-living rhizomes apparently make lower demands on their reserves, presumably by not having to extend shoots as far before reaching light. There also appears to be a difference between anoxia-tolerant and anoxiaintolerant species in their ability to conserve carbohydrate supplies when experimentally deprived of oxygen. As carbohydrate consumption by plant tissues is highly temperature dependent, it is striking that after 4 days of experimental anoxia (22°C) Schoenoplectus lacustris showed only a 2% reduction in total non-structural carbohydrate reserves. By contrast under the same conditions and same length of time Glyceria maxima (which is intolerant of prolonged periods of anaerobiosis) showed a 46% reduction in total non-structural carbohydrate (Barclay and Crawford, 1983). The soluble sugars also showed a similar pattern of behaviour with a marked reduction over 4 days in Glyceria maxima and little change in Schoenoplectus lacustris. Thus, even although Glyceria maxima rhizomes may normally possess considerable carbohydrate reserves, they are likely to suffer rapid depletion when exposed to low-oxygen stress. Such behaviour is maladaptive in habitats where the oxygen supplies may be limited. There appears therefore to be an ecological relationship between carbohydrate conservation and the degree of anoxia that is found in the preferred habitats of these species. Ecologically, it is interesting that musk rat colonies in the Mississippi delta are found most frequently in the ar.eas dominated by Scirpus americanus, a species closely related to Schoenoplectus lacustris and sharing its high tolerance of anoxia. Similarly, in the marshes of southern Spain, overwintering geese seek out the rhizomes of Schoenoplectus lacustris. In both cases the foraging animal chooses the more anoxia-tolerant species which would be expected to have greater carbohydrate reserves than the lesstolerant species.

C. Metabolic Adaptations to Anoxia I . Glycolytic Rate and Alcohol Dehydrogenase ( A D H ) Induct ion A frequent phenomenon associated with both anoxia and hypoxia is the induction of ADH activity. The conditions which give rise to ADH induction have been a subject of much study and debate as to whether the increased ADH activity is an inductive response which aids survival or merely signals a metabolic injury due to insuficient oxygen. Alternatively, some authors consider that the survival value of increases in ADH activity are not related

OXYGEN AND PLANT DISTRIBUTION

I45

to energy supply but may be involved in the regulation of cellular pH (Cerana et al., 1989; Roberts et al., 1989). Flooding of intact plants and subsequent estimation of ADH activity in root extracts have frequently revealed marked inductions of ADH activity (Crawford, 1967; Smith and a p Rees, 1979; Jenkin and ap Rees, 1983; Harberd and Edwards, 1983). Further detailed studies have shown this to be a true induction with specific effects on certain isoenzymes of ADH (Freeling and Schwartz, 1973; Ferl et a/., 1979, 1980). A painstaking study of the actual levels of ADH in roots growing naturally at different soil depths (Smith et al., 1986) found that ADH activity varied very considerably among species and at varying depths in dry and waterlogged soil. In Filipendula ulmaria, a species with limited aerenchyma, ADH activity at depth was found to be 14 times greater than at the surface, while with Curex riparia, which has well-developed aerenchyma, there was little change with depth. These authors however were unable to find any correlation between ADH activity and ethanol production, either in detached roots supplied with sucrose or roots receiving carbohydrate from photosynthetically active shoots, and therefore concluded that the physiological role of high activities of ADH in the roots of these species was unclear. However, since the studies on the role of hypoxic acclimation of Saglio et a/. ( I 988) and Johnson et al. (1989) there does appear to be a logical connection between the hypoxia-promoted induction of ADH activity and the subsequent ability to survive under anoxia. Roots that are kept in a highly oxygenated environment have low ADH activities, which remain low if they are transferred immediately to anoxia (Johnson et al., 1989). However a hypoxic acclimation period produces a 20-fold increase in ADH activity and is accompanied by increased ATP concentrations and improved anoxic viability. Genetic studies have shown that under aerobic conditions ADH activity is associated with the Adh, and Adh, genes. Under anoxia this activity is strengthened and other ADH isozymes also appear (Hanson and Brown, 1984). The isozyme content changes (Mayne and Lea, 1984) and the metabolic characteristics of ADH can also alter as a result of anoxia. In mature barley roots anaerobiosis causes a decrease in the K,,, of ADH for acetaldehyde (Fagerstedt and Crawford, 1986). A model for the possible reactions taking place during short-term oxygen deficiency in root tissues has been suggested by Hanson et al. (1984) which proposes that as lactate dehydrogenase (LDH) activity rises during early periods of oxygen deficit, lactate glycolysis will begin to compete with ethanol glycolysis for pyruvate and NADH (Davies, 1980). Thus the higher affinities for NADH and acetaldehyde in the induced ADH isozymes, could increase the ability of ethanol glycolysis to compete with lactate glycolysis, which would otherwise lead to acidification of the cytoplasm which in anoxia-intolerant plants would bring

146

R. M. M. CRAWFORD

about rapid cell death (Roberts et al., 1984a, 1985). Acetaldehyde has been shown to induce ADH activity in root tissues (Crawford and McManmon, 1968). In maize mutants that lack ADH, cytoplasm acidification continues and finally leads to cell death (Harberd and Edwards, 1983; Roberts et af., 1985). More recent studies (Roberts et al., 1989) have shown that maize is able to survive extreme anoxia, independent of ADH activity, except for genetic lines in which the activity was very low (less than 20 milliunits mg-' protein). In this later study the authors conclude that possibly the induction is a reaction to the stressper se or to a sensed shortage of proteins and that it is possible that the signals and mechanism responsible for the selective synthesis and translation of Adh mRNAs, observed during hypoxia, operate independently of the amount of ADH in cells without reference to whether or not ADH activity is limiting cell function. However, the increasing weight of evidence, at least in tissues which remain metabolically active under anoxia, points to a role for changes in ethanol, lactate and malate relative to one another to be the mechanisms whereby the fine-tuning of the cytosolic pH is achieved (Davies, 1986; Cerana et al., 1989). The existence of a connection between ADH activity and adaptation for survival in oxygen-poor habitats is strengthened by the existence of balanced polymorphisms in root ADH in many species. Studies of flooding tolerance in maize in relation to the distribution of the two alleles AdhlFand Adhls in relation to flooding tolerance (Marshall et al., 1973) showed that flooding reduced growth in both genotypes but the reduction was greater in the AdhlF homzygotes, which produced the catalytically more active enzyme. Repetition of these experiments with Bromus mollis (Brown et al., 1974) and Trifolium pratense (Francis et al., 1974) suggested that the maintenance of polymorphism in isozymes of ADH confers improved fitness in populations that come from sites with variable soil moisture. In wild sunflower populations the frequency of alleles for certain ADH isozymes was found to differ, depending on whether they had experienced a run of wet or dry years (Torres and Diedenhofen, 1979). The allele that was in lower frequency on the dry site, and in greater frequency in the wet sites, declined in the population in the wet site after a number of years of reduced flooding. Such a balanced polymorphism would not exist if different forms of the isozymes did not confer distinct advantages in wet and dry conditions. Further study is still necessary, but the induction of ADH by hypoxic conditions appears to be a general phenomenon. The isozymes that are most readily induced have been those that are catalytically more active and their induction has been observed to take place most readily in species that are characteristic of nonflooded sites. Thus in agreement with the observations of Crawford (1967) and Smith et al. (1986) the genetic studies discussed above also show the general tendency that it is species, or varieties of species, that would not expect to be flooded which show the greatest inductions of ADH activity (20-

OXYGEN AND PLANT DISTRIBUTION

147

fold). Tolerant species such as those investigated by Haldemann and Braendle, ( 1 986) show less extreme levels of change. The extent of change in ADH levels as a result of anaerobic conditions also depends on the type of roots. In barley the greatest levels of ADH induction have been observed in the adventitious roots while the effect on the seminal roots was less marked (Fagerstedt and Crawford, 1986).

2. End-products of Glycolysis A possible source of injury to plants deprived of oxygen arises when prolonged periods of anaerobiosis causes excessive accumulations of glycolytic end-products such as ethanol and carbon dioxide. Whether or not such accumulations are harmful to the plant tissue in question depends partly on the rate of production and partly on the rate at which the products can be removed either by diffusion or through transport mechanisms such as the transpiration stream or the mass movement of gases (Section 111). For this reason any discussion of the susceptibility of plant tissues to the potential hazards of accumulating glycolytic end-products needs to be related to the physiology of the whole plant. The high degree of porosity of the rhizomes of aquatic species such as Iris pseudacorus which aids aeration also facilitates the dispersal of ethanol. Thus when a comparison was made of ethanol accumulation in a number of wetland species with a dryland species such as Iris germanica, it was found that after 4 days of anaerobic incubation the wetland species all reached a plateau of tissue ethanol concentration, where production equalled dispersion, while the dryland species continued to accumulate ethanol through the 16 days of the experiment (Monk et al., 1984). Doubt has often been expressed as to whether or not ethanol is toxic to plant tissues in the concentrations that are commonly found after periods of anaerobic incubation (Jackson et al., 1982). If anoxia-sensitive organs such as rhizomes or germinating seedlings are kept under anoxic environments where the anaerobic atmosphere is kept moving, so that ethanol and carbon dioxide d o not accumulate, then the anaerobic life of the tissues is commonly prolonged. This has been observed for several species of germinating seeds (Crawford et al., 1987). Factorial experiments in which static and moving environments are combined with the addition or removal of carbon dioxide showed that the greatest mortality is found when static environments cause the tissues to be exposed to high carbon dioxide concentrations without ethanol removal (Crawford et al., 1987). The presence of carbon dioxide induced an increase in glycolytic activity and it was concluded that the deleterious interaction between carbon dioxide and ethanol was due to the increase in ethanol accumulated during the anoxia treatment. The moment when ethanol becomes dangerous to the tissues appears to be on re-admission to air, as this causes a surge in acetaldehyde concentrations

148

R . M. M. C R A W O R D

within the tissues as the anaerobically accumulated ethanol is oxidized to acetaldehyde which is considerably more phytotoxic (Studer and Braendle, 1988). Studies which follow the course of ethanol accumulation and its subsequent dangers to plant tissues during the post-anoxic phase, help to explain the earlier controversy as to whether or not ethanol is toxic to plant tissues. Many aerobically respiring tissues are either able to disperse ethanol into the atmosphere (e.g. barley; Cossins and Turner, 1962) or else deplete its concentration metabolically (Cossins, 1978). Anaerobically treated organs, especially when kept under non-circulating environments, are in an entirely different situation. Aerobic metabolism has been arrested during the anaerobic incubation and when suddenly exposed to oxygen the tissues not only suffer a surge in acetaldehyde concentration from ethanol oxidation, but are also subject to a number of other post-anoxic dangers for which their period of anaerobiosis has left them ill prepared (Section VD). The question also arises whether plant tissues make use of any other endproducts of glycolysis apart from ethanol. When intact plants are flooded a number of organic acids have been reported as increasing in concentration. These have included malic acid (Crawford and Tyler, 1969; Linhart and Baker, 1973; McKee et al., 1989), shikimic acid (Crawford, 1982b) and oxalic acid (Ernst and Lugtenborg, 1980) and succinic acid (Menegus et al., 1989). Prolonged lactate production is also found in barley (Hoffman et al., 1986) as well as in Ranunculus sceleratus and Senecio aquaticus (Smith and ap Rees, 1979). Amino acids, principally alanine and r-amino butyric acid, have also been found from NMR spectra to accumulate in rice seedlings under restricted oxygen supply (Menegus et al., 1989; see also Section VD1). In some cases the. concentrations observed are not sufficiently high for them to contribute significantly to the absorption of accumulating oxygen debts although they may have survival value in other ways, as for instance regulating cellular pH (p. 146). One of the dangers of prolonged anoxia as mentioned above is anaerobic acidosis. It has therefore been suggested that the regulation of cytoplasmic pH is an important factor in determining root survival. Ethanolic rather than lactate fermentation will prevent acidosis as will the accumulation of alanine and r-amino-butyric acid (GABA).

3. EfSects of Root Anoxia on Whole Plant Physiology Unravelling the effects of anoxia by itself on plant tissues without confusion with other effects of flooding presents serious experimental problems. Flooding with nutrient solution, particularly if it is rich in iron, can in itself change the organic acid content of the roots. Experiments with detached roots under anoxia are, however, equally likely to be misleading due to the short life of the roots and their probable inability to adapt to environmental change when excised or under total anoxia (Johnson et al., 1989). Ernst

OXYGEN AND PLANT DISTRIBUTION

149

(1 990), in discussing the problems in interpreting the changes in organic acids that can be observed on flooding, suggests that progress will only be made when instead of comparing species such as peas and Glyceria maxima, the experimentation focuses on differences between genotypes, as in the studies on Nyssa sylvatica by Keeley (1979). In this study of a swamp and an upland population of the swamp tupelo tree, the swamp population showed a threefold higher level of malate than the upland population after a year’s flooding. Ernst (1982) interpreted this increase to a genetic difference in the swamp plant’s metabolism in relation to the increased availability of metals and suggests that malic acid fluctuations are more likely to be involved in balancing ionic disequilibrium than in transporting an oxygen debt. Nevertheless proton consumption, when it takes place to a degree sufficient to have an adaptive effect in regulating ionic disequilibrium, will also automatically stabilize the NAD/NADH ratio and thus also service the oxygen debt and facilitate the transfer of this debt from roots to shoots by the translocation of organic and amino acids. In the temperate seagrass Zostera marina, a clear use of amino acids in accumulating the oxygen debt has been found operating on a diurnal rhythm (Pregnall et al., 1984). During the night anoxic conditions prevail with little production of ethanol. There is instead an accumulation of alanine and r-amino butyric acid which can account for 70% of the total amino acid pool. Upon resumption of shoot photosynthesis and oxygen transport to the roots, the accumulated alanine and r-amino butyric acid decline rapidly. When studies are carried out on intact plants such as flooded trees, then the effect of flooding can be observed over a prolonged period during which a sequence of changes can be observed both in the metabolic activity of the roots and the compounds being transferred from root to shoot in the xylem sap. When examined in terms of the intact plant, then, anaerobic respiration is not as energetically wasteful as it might appear at first sight, as the carbon skeletons are not lost but remetabolized in the aerial portions of the plants (see Kimmerer and Stringer (1988); Section IIC2). Trees, with their spatial separation between an extensive rooting system and their aerial parts, appear particularly disposed to the transfer of metabolic hydrogen from root to shoot in the transpiration stream as an alternative to the movement of gaseous oxygen from shoot to root. The annual metabolic cycle of trees also facilitates the use of hydrogen-transporting compounds (ethanol, organic acids, amides) for the upward movement of the oxygen debt. The storage of starch in the xylem parenchyma of the roots in late summer and early autumn and its breakdown in spring, often while the soil is poorly aerated, causes a period of anaerobiosis in the root in which rapid use is made of the accumulated starch. The sap of flooded trees is often enriched in ethanol (Crawford and Baines, 1977; Crawford, 1982b) and as with the reported cases of cambial anaerobiosis in tree trunks (Kimmerer and Stringer, 1988),

150

R. M. M. CRAWFORD

the ethanol generated under hypoxia serves to transport the oxygen debt to another plant of the tree where it is remetabolized. Polyamine synthesis also appears to be triggered by anaerobic conditions in a number of species. In rice seedlings exposed to oxygen-deficit stress there is a large increase in free putrescine. This is accompanied by an increase in arginine decarboxylase and labelling experiments have shown that arginine is the precursor of the anaerobically accumulated putrescine (Reggiani et al., 1989a,b). The production of free and bound putrescine observed by these authors after 24 h of anaerobiosis was 0.8 and I .O pmol g - ' fresh wt in the coleoptile and 1.0 and 0.3 pmol g - ' fresh wt for the roots. These concentrations matched the increase of organic acids in the tissues and it was suggested that this would serve to reduce the risk of anaerobic acidosis to the germinating anaerobic seedling. Rice seedlings also show significant differences in polyamine accumulation in roots and coleoptiles in terms of their ability to form conjugates. The bound soluble forms of polyamines in plants are mostly found as amides of hydrocinnamic acid (HCAAs: hydrocinnamic acid amides). It is interesting to note in the study of Reggiani et al. (1989a) that putrescine conjugates are mainly found in the coleoptile and not in the root since it is the former that is capable of growth during anoxia. Under anoxia, rice coleoptile elongation does not respond to IAA applications, however exogenous application of putrescine will stimulate elongation and the anoxic titre on putrescine content is correlated with coleoptile elongation (Fig. 21; Reggiani et a f . , 1989b). These authors suggest two possible roles for putrescine on anaerobic growth. The first is as controller of anaerobic extension growth and the second as a homeostatic buffering mechanism to stabilize intracellular pH. As the concentrations of polyamines are undeniably low, it remains at present uncertain in most tissues as to whether polyamines have a general adaptive function in assisting the anaerobic survival of plant tissues.

D. Causes and Prevention of Post-anoxic Injury Re-exposure to oxygen after a period of deprivation can cause serious injury both to plant and animal tissues. In animals this type of damage is usually referred to as post-ischaemic injury and is a potential source of tissue damage after heart attacks and organ transplants. In plant tissues, re-exposure of underground perennating organs and roots to air after a period of inundation can cause post-anoxic injury. Thus in Iris germanica rhizomes there is little sign of injury when they are first taken from an anaerobic incubator after 15 days, oxygen deprivation. However, after 6 h exposure to air there is extensive peroxidative damage and the condition of the rhizomes rapidly deteriorates (Hunter et al., 1983). Post-anoxic injury can arise from a number of causes. After periods of anoxia, re-admission of oxygen to tissues can

OXYGEN AND PLANT DISTRIBUTION

151

Fig. 21. Regression of growth in length under anoxia in rice coleoptiles against putrescine content. (Reproduced with permission from Reggiani et af., 1989b.)

generate superoxide radicals (Halliwell, 1984; Monk et a / . , 1989). Other potential sources of post-anoxic injury come from the rapid oxidation of anaerobically accumulated metabolites. Thus the return of oxygen to anaerobic tissues can cause a surge in acetaldehyde and ethylene production from their respective anaerobically accumulated precursors, namely ethanol and ACC (Studer and Braendle, 1987). Plants are equipped with a variety of mechanisms to counteract the hazards of the post-anoxic environment. Comparative studies on the effects of anoxia on perennating organs (rhizomes, stolons and tubers) have shown that the anoxia tolerant Iris pseudacorus differs from I. germanica, which is prone to post-anoxic injury in having high levels of superoxide dismutase (SOD) activity (Monk et al., 1987). Apart from enzymatic protection against superoxide radicals, rhizomes and tubers are frequently rich in natural antioxidants, e.g. reduced ascorbic acid, and glutathione and a-tocopherol. Experimentally it has been demonstrated that transferring anaerobically incubated seedlings of chickpea (Cicer arietinum) to a 1 O m ~solution of ascorbic acid (pH 5.6) before re-exposing them to air enhanced growth and recovery in the post-anoxic phase (Fig. 22; Crawford and WollenweberRatzer, 1992). Regenerating enzymes for the commoner antioxidants (dehydroascorbate reductase, glutathione reductase, as well as the directly protective enzymes SOD, catalase and peroxidase) are present in most

152

R . M. M. CRAWFORD

aerobic tissues (Larson, 1988). The functional significance of these substrates and enzymes has not as yet been systematically investigated in relation to post-anoxic injury and flooding tolerance in wetland vegetation. Fluctuating water tables have been observed to provide a more serious hazard than constant flooding to some species, particularly trees (Hook and Denslow, 1987). Alternation between the anoxic and the aerobic condition with the consequent hazards of post-anoxic injury are therefore likely to play a role in many situations. The increased use of irrigation may also expose plants to post-anoxic injury as short-term periods of anoxia when irrigated are followed by subsequent aeration. Post-anoxic injury is also likely to be of importance for the post-harvest physiology and storage of plant products. Many crops, e.g. apples and pears, are stored under conditions of low oxygen and high carbon dioxide and on removal from storage can suffer postoxidative damage while en route to market. This is at present frequently prevented by the use of antioxidant drenches. A better knowledge of postanoxic injury could help to dispense with this potential source of food contamination by selection for species with adequate natural protection against post-anoxic injury. In medicine, the levels of natural antioxidants are known to influence the recovery of tissues after oxygen deprivation but in plant sciences this possibility has just begun to be studied (Monk et al., 1989). It is clearly necessary to investigate the impact of antioxidants on tissue recovery capacity from periods of anaerobiosis.

E. Mineral Nutrition and Flooding Tolerance Oxygen starvation in flooded soils exposes plants to the possible uptake of a range of soluble reduced ions, many of which are potentially toxic to plants. Thus manganous, ferrous, nitrite and sulphide ions are all potential soil toxins under reduced soil conditions. In addition the reduction of organic matter can give rise to short-chain organic acids which are phytotoxic in their undissociated form (Lynch, 1977; Drew and Lynch, 1980). As a result of reducing conditions, the yse of NO,- as a microbial electron acceptor, and the inhibition of the nitrifying process, nitrogen is almost solely available in wetland habitats as ammonium. Such conditions are inimical to the growth of unadapted plants and represent yet another situation where plant distribution is influenced by the lack of oxygen.

1. Nitrogen Nutrition in Anaerobic Environments Plants unadapted to flooded habitats typically show signs of nitrogen deficiency on flooding. Measurements of total nutrient content in barley shoots show a rapid decline in net nitrogen uptake shortly after the onset of flooding (for review see Drew, 1990). The lack of distribution of nitrogen to shoots on flooding is paralleled by the decline in P and K in barley (Drew and

Fig. 22. Ascorbic acid acting as an antidote to anoxic injury. The chickpea seedlings on the right were transferred to a 10 mM solution of ascorbic acid (pH 5.6) before emerging from anaerobic treatments of 48 and 72 h respectively. The antioxidant properties of ascorbic acid appear to have minimized post-anoxic injury and aided recovery particularly in the less damaging condition of the seedlings that had only 48 h anoxia. (Crawford and Wollenweber. 1992.)

154

R . M. M. CRAWFORD

Sisworo, 1979), suggesting that the principal cause may be the inhibition of ion transport in roots at low oxygen concentrations. If plants are preloaded with nitrogen before flooding, growth retardation as a result of flooding is delayed (Trought and Drew, 1981). Nitrate has been reported as improving plant growth after flooding but speculation that this may be due to “nitrate respiration” (dissimilatory nitrate reduction), which is well known in anaerobic bacteria, has not been satisfactorily confirmed in higher plants. It appears more likely that the beneficial effects of nitrate to flooded plants may come from alleviation of the rate of fall of soil redox potential by supplying the soil microflora with an alternative terminal electron acceptor, or else by reducing cytoplasmic acidosis (Roberts et al., 1985). From NMR studies of shifts in cytoplasmic pH it was found that the inclusion of 2 5 m calcium ~ nitrate in the medium could extend the anoxic survival of maize root tips from 24 to 32 h. This nitrate might compete with lactic dehydrogenase for NADH thus slowing lactic acid synthesis and delaying an early decrease in cytoplasmic pH. However, if this is a significant pathway, clearer evidence of “nitrate respiration” would be expected than has yet come to light. Although nitrogen, once absorbed by plants, is reduced to ammonium which is then the immediate precursor of organic nitrogen, most species of higher plants (excluding those from wetlands) do not grow well if ammonium is supplied as the sole or predominant source of nitrogen. This has been observed both in pot-grown plants and in many species cultivated in water culture (Kirkby and Mengel, 1967; Cox and Reisenauer, 1973; Allen et al., 1988). Ammonium at high concentrations can be toxic and its adverse effects when supplied as the sole nitrogen source for non-adapted plants may be due to an inability of such species to detoxify ammonium when taken up rapidly as well as to the extrusion of protons which can cause a fall in both cellular and extra-cellular pH values. One mole of NH,’ has been calculated as producing approximately 1.2 mol H’ (Raven and Smith, 1976). Ammonium, when it is taken up by roots, has to be immediately detoxified by incorporation into amino acids and proteins and for this ample supplies of carbohydrate are necessary. When plant carbohydrate content has been diminished by periods of darkness, ammonium uptake is drastically reduced. However low energy supply during the period of nitrogen uptake affects nitrate absorption more than ammonium (Mengel and Viro, 1978). Plants adapted to acidic conditions face similar situations as ammonium also predominates as the nitrogen source in soils of low pH. Thus Deschampsia$exuosa will grow as well on ammonium as a nitrogen source but Rumex acetosa and Nardus stricta grow poorly if at all on NH,+ only (Gigon and Rorison, 1972). Among woody plants, A h u s glutinosa grows well on ammonium (Troelstra et al., 1985), as do Iris pseudacorus and Glyceria maxima (Fig. 23; Allen, personal observation). In rice, bubbling water cultures with nitrogen reduced nitrate uptake by about 75% but ammonium uptake was reduced only

OXYGEN AND PLANT DISTRIBUTION

155

slightly (Sasakawa and Yamamoto, 1978). Paradoxically, ammonium uptake was not affected by bubbling with oxygen while nitrate uptake was reduced. There are few studies which provide any cellular explanation for the differential tolerance to ammonium in wetland as opposed to dryland species other than rate of uptake as mentioned above. Wetland habitats, however, do oscillate between the period of flooding when nitrification is drastically reduced and the summer period when the water table may drop and rapid nitrification can take place. The extent of this oscillation can have marked effects on the plant communities. In a study of the Carex species in flood-line alder associations (Janiesch, 1986) it was found that when the water table remained high then C . pseudocyperus, C . elata and C . acutformis were predominant and relied almost exclusively on ammonium as their nitrogen source and showed little induction of nitrate reductase. Species that lived in the areas with oscillating water tables, i.e. C . remota and C . elongata, were capable of using nitrate and had high nitrate reductase activity. Whether or not plants rely on ammonium nitrogen or utilize nitrate and whether they reduce nitrate in the shoot or the root can be reflected in the ratio of carboxylate to organic nitrogen. When nitrate is the predominant nitrogen form and is reduced mainly in the shoots then the ratio of carboxylate to organic nitrogen should be about 1. Ratios of less than 1 indicate either participation of root reduction with transport of carboxylate from shoots to roots or else ammonium nitrogen nutrition (Wollenweber and Kinzel, 1988). This technique, when applied to a survey of species from five different sites in Austria, gave the lowest carboxylate/organic nitrogen ratios in plants from bogs with permanently high water tables. In purely aquatic species it is possible to investigate the effect of changing carbohydrate supply and oxygen levels on nitrogen nutrition without the complication of oscillating water tables and in this connection there have been a number of illuminating studies on ammonium utilization by eelgrass (Zostera marina) in relation to the depth of the eelgrass bed. Most of the nitrogen taken up for eelgrass growth is absorbed by the roots as ammonium (Short and McRoy, 1984). Ammonium assimilation is mainly brought about by the dual action of glutamine synthetase (GS) and glutamate 2-oxoglutarate transaminase (GOGAT). Aerobic respiration in eelgrass is dependent on oxygen from photosynthesis (Smith et al., 1984) which results in eelgrass experiencing a diurnal shift between aerobiosis and hypoxia or anoxia. Free pool sizes of glutamate and glutamine increase in aerobically treated eelgrass roots and decrease rapidly under anaerobic conditions (Pregnall et al., 1984). When plants are growing at greater depths they would be expected to experience a shorter photosynthetic period and consequently have a more prolonged exposure to anaerobic or hypoxic conditions. The consistently higher levels of alanine in roots of plants growing in deep water suggest that anoxia is more extensive and aerobic recovery less complete in this habitat

Fig. 23. Iris pseuducorus showing superior growth in water culture when fed with ammonium rather than nitrate nitrogen.

OXYGEN AND PLANT DISTRIBUTION

I57

(Pregnall et a/., 1984). In a study of plants taken from different depths (Pregnall et ul., 1987) it was found that plants from the deep extremity of an eelgrass meadow possess in vitro GS activities that are three-to six-times greater than the shallow-growing plants. The dramatic increase in GS activity with plants growing at greater depths and the daily increase in GS activity during dark treatments suggests that Zostera murinu responds to increased anaerobiosis by a compensation mechanism which enhances the rate of ammonium assimilation during the brief period when photosynthetic activity is sufficient to provide the necessary aerobic conditions for nitrogen uptake and assimilation. Although ammonium is the main source of nitrogen in flooded soils a number of plants from wetland habitats show surprisingly high levels of nitrate reductase. Apparently, the oxidation of the rhizosphere is sufficient to support substantial populations of nitrifying bacteria in Glyceriu maxima (see review-Laanbroek, 1990) and high levels of nitrate reductase have been found in fen plants with aerenchymatous roots (Blacquiere, 1986). In addition halophytes from the lower levels of saltmarshes have been observed to have greater nitrate reductase activity than plants from the upper, less frequently flooded levels (Stewart et a/., 1973)

2. DetoxiJcation of Harmful Ions in Anaerobic Soils The first line of defence against toxic soluble ions such as Fe2+,Mn’+ and FeS for most wetland plants is to render them insoluble at the root surface by oxidation with air that diffuses to the root surface from photosynthesizing shoots. During periods of bad weather when photosynthesis is reduced such plants can become prone to mineral poisoning. Such a condition causes a bronzing disease (Akagare) in rice. The differences in the “oxidation power” of the rice roots can be seen in the amount of ferric hydroxide precipitated on the roots when grown in flooded soil (Chen et al., 1980). Two elements which are potentially toxic to unadapted plants in flooded soils are ferrous iron and manganous ions. Tolerance of manganese levels in leaf tissue in flood-tolerant rice is almost 10 times greater than the level tolerated by non-flood-tolerant barley of 200 mg kg-’ dry mass (Vlamis and Williams, 1964). This is fortunate, for wetland vegetation antagonism between ions for uptake by roots causes the absorption of manganous ions to be hindered by the presence of ferrous iron. Manganous ions are most likely to cause toxicity in soils that are deficient in iron and phosphate. There is an extensive literature on manganese toxicity but most of the data relates to non-flooded agricultural soils where manganese toxicity can often be manifested through its action in causing deficiencies of other nutrients (Marschner, 1986). For wetland plants resistance to Fe” toxicity is based primarily on avoidance mechanisms, low uptake and subsequently low rates of transport

158

R . M. M. CRAWFORD

of absorbed iron from roots to the shoot. When exposed to high iron concentrations, plants in waterlogged soil can show a 10-fold increase of iron transfer to older leaves using senescing organs as a detoxification mechanism (Ernst, 1990). The damage caused to flooded plants when they lack the ability to exclude iron has been linked to the generation of the superoxide radical and the subsequent generation of hydrogen peroxide (Hendry and Brocklebank, 1985). This may be a general phenomenon for other free transition metals as all have the capacity to reduce dioxygen leading to the production of the superoxide radical.

+

+

Fe2+ 0, + Fe3+ 0;Superoxide dismutase (SOD) can then carry out the reaction:

0;- + 0;- + 2 H + -+ H 2 0 2 In a study of the iron-sensitive species Epilobium hirsutum the plants were unable to restrict the absorption of ferrous iron which triggered an induction of SOD. The roots of this species however contained little or no catalase with almost all catalase activity being confined to the healthy shoot. Consequently, in roots the accumulating hydrogen peroxide if not catabolized will react further with ferrous iron to produce the highly reactive hydroxy radical in the Fenton reaction: Fe2+ + H,O,

+ Fe3+OH'

+ O H-

The hydroxy radical, which will cause lipid peroxidation and subsequent membrane damage will eventually lead to tissue death which resembles the symptoms found in the Epilobium hirsutum plants (Hendry and Brocklebank, 1985). The deposition of ferrous and manganous oxides may be, depending on environmental circumstances, beneficial as opposed to the more general expectation of their deposition inhibiting root function. In reduced soils the plant nutrients K + , Ca2+,Mg2+and trace elements can be absorbed onto the deposits of ferric and manganic oxides. However, toxic cations such as Cu2+, Zn2+and Mn" may also be absorbed (Laanbroek, 1990).

VI. OXYGEN AND PLANT COMPETITION The ability to survive periods of oxygen deprivation confers certain ecological advantages as well as entailing a number of costs. The most obvious advantage is access to sites that are denied to anoxia-intolerant species. This advantage has evolutionary parallels in animals, where diving reptiles, mammals and birds have access to food resources that are unavailable to species confined to terrestrial habitats. As well as an access to resources, the

OXYGEN AND PLANT DISTRIBUTION

159

anaerobic environment provides a refuge from predation and disturbance both to plants and animals. The monotypic stands of Phragmites australis and Spartina anglica in anaerobic muds are but two examples of species which have access to habitats where they suffer little interspecific competition or grazing pressure. The marine angiosperms (Zostera spp.) are more extreme examples of higher plants with a high tolerance of anoxia and access to sites free from competition. Increased knowledge in recent years of the extent of physiological variation in wetland plants makes it possible to re-examine earlier studies on the competitive interaction of wetland species with new insight. Wetland habitats have long attracted ecologists interested in competition as they provide an opportunity to examine the boundaries between communities in relatively undisturbed habitats. In mature wetland sites, communities can be found that are clearly delineated by one or two dominant species. When mapped, these boundaries or zonations often appear to be related to small changes in water-table levels and duration of flooding. Buttery and Lambert (1965) studied such a case in relation to competition between Phragmites australis and Glyceria maxima in a fenland site where these two species have succeeded one another, during the usual time-course of succession. Where Glyceria showed maximum growth, Phragmites was depressed due to the rapid production of an extremely dense Glyceria sward in spring before the emergence of the Phragmites shoots. Away from open water, towards the back of the fen, both species showed a decline in growth and here P . australis became dominant. An examination of the nutritional status of the two species found no evidence to explain the reduction on growth rate and the change from dominance of Glyceria over Phragmites to that of Phragmites over Glyceria (Buttery et al., 1965). These authors therefore concluded that without further physiological data there was no satisfactory explanation for this change in competitive ability. We now know that these two species are markedly different in their tolerance of anoxia, with Phragmites australis being able to survive over a month under anoxia and Glyceria maxima frequently being killed in 4-7 days (Barclay and Crawford, 1982). Thus a fresh assessment can now be made of the role of physiological specialization in determining relative competitive ability in a manner which was not possible at the time of the original investigation. The high and prolonged winter levels of ADH and accompanying ethanol accumulations observed in a year-round observation of natural Phragmites sites (Haldemann and Braendle, 1986; see also Section IIB2) indicate that this species is reliant on anaerobic respiration even into late spring. In stagnant conditions this would be likely to be extended. From the laboratory experiments on relative anoxia tolerance this extension would be expected to be less detrimental to Phragmites than to Glyceria thus giving the former species an ecological advantage at the back of the fen.

160

R. M. M. CRAWFORD

An improved understanding of the extent of physiological adaptation in determining the relative success of Phragmites versus Glyceria also makes it possible to examine the current controversy on resource utilization and its implications for competition. Oxygen should also be considered as a resource as it is consumed by all eukaryotic species and can frequently be in limited supply. Although most eukaryote species are dependent on oxygen for their ultimate survival (oxygen is required for mitosis) this does not mean that oxygen is required by all species for all organs all the time. The ability to do without a resource and not to suffer any diminution in competitive ability due to its temporary absence could confer a considerable competitive advantage. In classical discussions on competition the definition commonly used refers to “the simultaneous demand by two or more species for an essential resource that is actually or potentially in limited supply” (Lincoln et al., 1982). If, however, one species can postpone its demand for a limiting resource without suffering any loss in viability, then it will have an advantage over other species that are not able to defer their need for the resource. A temporary respite in demand for an essential resource does not mean that the resource itself is not-limiting. It is merely a phenological adaptation which uses a short-term stress-tolerance to avoid a resource crisis. Oxygen can be regarded as a resource where species differ as to whether it has to be in constant supply or whether they can suffer interruptions in its supply. This phenological aspect of competition has been described as deprivation indiference (Crawford et al., 1989). Some of the species which have been discussed above in relation to anoxia tolerance belong to highly productive habitats, namely the swamps and saltmarshes of flooded river basins and deltas. In these habitats the ability to “do without” oxygen for a while does not diminish the ability to survive in these productive sites. Competition between the wetland species discussed above does not fit easily into either the classifications of Grime (1979) of “competitors, stresstoleraters or ruderals”, or the “resource-ratio’’ model of Tilman (1985). It does not appear that the anoxia-tolerant plants are stress-toleraters in the sense that is implied in Grime’s terminology. To Grime, toleraters are able to endure a reduction in biomass acquisition without reducing their chances of survival. In the anoxia-tolerant species discussed above, such as Phragmites australis and Typha latifolia, the endurance of a period of anoxia would allow the species to occupy sites that are inaccessible to their competitors. There is no evidence to suggest that the growth rates in established stands are in any way diminished as a result of the periods of anoxia. The habitat differentiation between the anoxia-tolerant and intolerant species also fails to fit the Tilman resource-ratio hypothesis as the oxygen deprivation stress is only temporary. Once the shoots have emerged into air, there are no grounds for suggesting that intolerant species such as Glyceria maxima have any different resource demands than Phragmites australis or

OXYGEN AND PLANT DISTRIBUTION

161

Typhu lutifolia. In this matter the situation could be considered in terms of pre-emptive competition (Grubb, 1977; Werner, 1979) where all the plants involved have similar demands for resources. However, the zonation of the wetland rhizomatous species does not fit the GrubbWerner hypothesis as this proposes that distribution is determined not by competition but by the pre-emptive occupation of vacant sites. The Grubb-Werner hypothesis fails therefore to make any recognition of the varying tolerances of wetland plants to different degrees of hypoxia and inundation. The zonation can however be explained in this (and other cases, e.g. Typhu lutifoliu and Schoenoplectus lucustris) on the basis of varying degrees of deprivation indiflerence or temporary resource denial. The ability to survive without oxygen, although beneficial in making accessible sites that would otherwise not be available, also has disadvantages. Anaerobic respiration is very inefficient in energy yield per unit of carbohydrate consumed. Plants that grow in such sites and use anaerobic respiration require a greater supply of carbohydrate to re-establish their shoots in spring (see Section VB2). This, if coupled with a potential delay in shoot emergence from rhizomes in more anaerobic sites, can create a competitive disadvantage. Thus, in areas where anoxia-tolerance is not necessary, the deep rhizome habit will place the plants at a disadvantage in occupying space in early spring. As a compensation mechanism, however, the aeration system of aerenchyma, which is frequently combined with hollow stems, does deliver to the wetland and aquatic plant an enriched source of carbon dioxide as compared with normal atmospheric concentrations. Areas where the adaptive significance of anoxia-tolerance versus the rapid resumption of growth in spring is most likely to play a crucial role in competition are in community boundaries (Kampfionen). Here fluctuations in environmental stress from year to year will vary the relative exposure of the vegetation to anoxia, and therefore influence the relative success of anoxia-adapted and non-adapted species. Examining these wetland competition examples from a physiological viewpoint provides a striking example of the potential ecological advantages of being able to dispense for a period with a resource that is in constant demand by non-adapted competitors. The ecological advantages of being able to “do without” or deprivation indifference is clearly a factor that has to be considered in plant competition.

VII. CONSEQUENCES OF CLIMATIC CHANGE FOR THE VEGETATION OF OXYGEN-DEFICIENT HABITATS The relatively small reductions in oxygen content of the atmosphere that are likely to arise, even as a result of doubling of the carbon dioxide levels,

162

R . M. M. CRAWFORD

should not preclude an examination of the effects of climatic change on oxygen availability to terrestrial plant communities. The palynological record, particularly for northern Europe, provides ample evidence of climate-induced changes on the distribution of plant communities (Huntley and Birks, 1983). The post-Pleistocene migrations of well-defined vegetation types, characterized by their dominant tree species are usually discussed in relation to changes in temperature and potential for dispersal. These changes are for the most part, extremely slow, with migration rates of as little as 50500 km per millennium for non-pioneer species such as oak, ash and lime (Huntly and Birks, 1983). However, in the fossil pollen and other plant remains in raised bogs there is extensive evidence of great sensitivity of the vegetation cover to changes in climate, particularly in relation to moisture (Barber, 1981). The plant cover of peat bogs is greatly influenced by the extent of depth development by the acrotelm (Ingram, 1978). The acrotelm is the layer of peat which becomes aerated during the growing season and in which there is active root development as distinct from the area beneath, the catatelm, which remains permanently unaerated and anaerobic. When there is active evapotranspiration the acrotelm increases in depth which results in the bog surface becoming drier, with the consequent spread of woody shrubs, e.g. Calluna vulgaris and Empetrum nigrum, as well as birch and pine (Fig. 24). By contrast the wetter periods see increased growth in Sphagnum and the spread of species such as Scheuchzeria palustris in English raised bogs (Godwin, 1975). As these communities of drier periods advance and retreat across bog surfaces they leave clearly marked horizons in the peat profile called recurrence surfaces. One of the most marked changes in north European bog profiles is the change that can be seen particularly in North Germany between fresh unhumified pale-coloured peat (white peat) and a lower, older, denser and highly humified “black peat”. The boundary between these two types, as seen in North Germany, is extremely sharp and was termed the “boundary horizon” (GrenThorizont) by C.A. Weber, the German pioneer of bog stratigraphy (see Godwin, 1975). This Grenzhorizont marks the division in North Germany between the warm dry conditions of the Sub-boreal peat (490G2400 BP; Mangerurd et al., 1974) and the older peat layed down in the wetter Atlantic period. During the Sub-boreal, peat formation in North Germany was either missing or very thin, until a change to extensive waterlogging with the formation of aquatic peat brought about a revitalization of bog growth. In other areas in Britain and Scandinavia it is difficult to relate any one peat boundary to the original Grenzhorizont of Weber and a number of horizons are evident to which the term “recurrence surfaces” is given. The oceanic climates of the north-western parts of Europe appear to have been very sensitive to frequent changes in climate with numerous

Fig. 24. One of Britain’s smallest raised bogs a1 rcat Inn, Fife (Scotland). After a number of dry summers the bog surface becomes invaded with birch saplings encroaching from the perimeter woodland necessitating their removal in order to preserve the characteristic wet lawn flora of the bog surface. (Photograph J. K . S. St Joseph.)

164

R . M. M. CRAWFORD

recurrence surfaces, reflecting environmental change either as a result of increased human activity or climatic change. In a series of studies at Bolton Fell Moss in North Cumbria, Barber (1981) re-examined the long-held belief that raised bog surfaces grew through a process of cyclical regeneration with hummocks and hollows alternately replacing one another as the bog surface passed from a wet species association through active peat growth to a dry one. After much painstaking research he declared this hypothesis of cyclical growth to be false and instead demonstrated that even small-scale features of bog stratigraphy were controlled by climate. T o replace the earlier view Barber (1977) proposed a phasic theory of bog growth. This theory states that: raised bog growth is controlled above all by climate, even down to the relative areas of hummock and pool and that the phase shifts in peat growth are a result of climatic shifts . . . Threshold factors may cause the operation of the theory to differ from region to region and to a lesser extent, from bog to bog, but all the factors of hydrology and drainage, life cycles of plants, pool size etc., are all subordinate to climate . . . The theory also unifies the formerly distinct phenomena of Recurrence Surfaces and Regeneration Complexes, the latter being viewed as a natural consequence of the former, so that by the process of pool-infill and hummock spread the bog tends to a drier state, until the next phase-shift (Barber, 1981).

Figure 25 shows Barber’s summary of available knowledge concerning precipitation and vegetation changes to construct a generalized wetness curve for Bolton Moss from AD 90 together with phase shifts in bog surface vegetation. This figure emphasizes the argument (Moore, 1988) that in northern oceanic climates such as occur in northern England and much of Ireland, Scotland, the Northern Isles (Orkney and Shetland) as well as Iceland and parts of western Norway, small alterations in relative soil wetness, irrespective of how caused, can induce marked vegetation changes. When these changes in bog vegetation are considered in relation to some of the currently proposed models of climatic change, a number of important consequences for the future ecology of oceanic landscapes can be seen. Model results suggest that increasing carbon dioxide may increase the global hydrologic cycle (Wigley et a f . , 1990). Estimates of global mean precipitation due to a doubling of CO, vary between + 3 and + 1 1 % (for review see Wigley et al., 1990). Given the general consensus (see Wigley et af., 1990) that there is a general tendency for rainbelts to move north, then it is to be expected that areas such as northern and western Britain, Iceland and parts of Scandinavia may receive considerably more rainfall than at present. One of the more common predictions of the various models of “greenhouse-gas’’ induced climatic change is for up to a 20% increase in precipitation in the wetter parts of northern Europe. For terrestrial plant communities exposed to the oceanic climates of the North Atlantic there is therefore likely to be a deterioration in

DRY

880 b

PHASE - S h l F T S

TO WET L A W N

-

TO DEFINITE POOL-

22

76

XJ

1 x7

Fig. 25. A generalized wetness curve for Bolton Moss, Cumbria (England) together with the phase shifts in bog surface vegetation from wet lawn to pool and hummock topography. (Reproduced with permission from Barber, 1981.)

166

R . M. M . CRAWFORD

the levels of soil aeration and acrotelm depth will be reduced, unless the increased precipitation is accompanied by an equivalent increase in evapotranspiration. The meteorological future in this respect is still uncertain but biologically, there are a number of reasons which make an increase in evapotranspiration unlikely and may cause the reverse effect, namely a decrease in evapotranspiration. First, increased carbon dioxide levels are anticipated to bring about an increase in water-use efficiency through reduced stomata1 opening and therefore reduce transpiration (Eamus and Jarvis, 1989; Bolin et al., 1990). Secondly, as a consequence of decreased acrotelm depths, tree regeneration will be hindered and the rooting depth of trees on peaty soils will be reduced making them susceptible to windthrow. An increasingly serious problem in many Scottish forests is the occurrence of extensive endemic windthrow damage on semi-mature plantations (Low, 1987). Reduced tree cover will itself reduce transpiration and thus accelerate the rise in soil water tables. There is also the possibility that increased storm frequencies as a result of climatic change will increase the risk of premature windthrow. There is no statistical proof as yet of increased wind speeds in Britain as windspeed, like rainfall, is a highly variable phenomenon that frustrates the detection of long-term trends. However, wave height in the North Atlantic (which can be taken as an integrator of wind activity) has increased over the past 25years (Carter and Draper, 1988). There is, therefore, a possibility not just of greater windiness, but also of an increase in catastrophic as opposed to endemic windthrow. Given the extensive areas of Sitka spruce that have been planted in wet peaty soils in recent years this could result in very serious losses to forestry in northern Britain by the end of the millennium. The effects of tree removal on bog growth in the past is amply illustrated in the growth of peat that took place in northern Scotland, the Northern Isles (Orkney and Shetland) and in Ireland and Norway with the arrival of Neolithic man. It is still an astonishment to many to learn that despite the brown peaty landscape that now covers over 50% of Shetland (Robertson and Jowsey, 1968), peat did not begin to spread to its present extent until after the arrival of man. Archaeological inventories of burnt mounds, Neolithic/Bronze Age houses and field systems in western Ireland (Herity, 1971; Mitchell, 1976), western Scotland (Whittington, 1983; Newell, 1988) and Shetland (Whittington, 1978; Whittle, 1979) give ample evidence of agricultural activity on a landscape that was largely free of peat, and which only later was obliterated by bog growth. An archaeological guidebook to two prehistoric farms in Mayo (Caulfield, undated) describes a Bronze Age farmer’s work in altering a wall of his Neolithic predecessor, with the Bronze Age wall rising higher and higher on the bog as it encroached on the farm. The spread of peat in western Ireland is suggested as taking place after 4000 BP (Caulfield, 1978). In Shetland a radiocarbon profile of the peat

OXYGEN A N D PLANT DISTRIBUTION

167

points to about 3671 f 65 BP for peat initiation at one typical site. Figure 26 shows a prehistoric field boundary on the island of Papa Stour in Shetland, stripped of the later surrounding peat due to local needs for fuel, soil and animal bedding. In the magnificent expanse of open landscape that is now covered by the Scottish peatlands of Caithness and Sutherland a similar environmental history has taken place (Fig. 27). This area, which is considered so unique that it should be conserved against peat removal (Ratcliffe and Oswald, 1988), arose largely from degradation of a richer and more varied cover of birch forest (Fig. 28). Whether or not the prime cause of peat invasion onto Neolithic-Bronze Age landscapes was due to man or climatic change or both, it is strikingly clear that in oceanic climates there is a delicate balance between precipitation and evapotranspiration which is very sensitive to any alteration (Green, 1964). Any extension of waterlogging as a result of climatic change, should it take place in the next 50 years, will affect upland landscapes at a vulnerable period in their evolutionary history. The depopulation of many upland areas does not necessarily provide an opportunity for natural ecosystems to reestablish themselves. Reduction in human settlement produces reduced management with the result that, in Scotland, over-grazing is increasing from uncontrolled deer populations and too many subsidy-maintained sheep on unlimed and unmanured pastures. Since the 1960s the red deer population in Scotland has increased from 198000 to over 300000 with most of the increase taking place in the Grampian and Central Highland Region (Countryside Commission for Scotland, 1990-data from Red Deer Commission). Into this scene of landscape dereliction, increased precipitation with high acidity levels will bring about yet a further decline in the remaining natural fertility of many upland sites. Consequently, acid grassland deteriorating to peat-accumulating communities will expand and the necessary soil aeration will not be available that is required for the natural regeneration of native woodlands. The possibility of a greater frequency of mild winters is another feature of climatic change that could have deleterious effects for some oxygen-deficient habitats. Mild winters aggravate the damage done by flooding, causing greater die-back in tree root systems (Coutts and Nicoll, 1990) as well as proving deleterious to winter cereals. The Roman agrarian writer Cat0 (23& 149 BC) was well aware of these dangers when he pointed out in his treatise De Agri Cultura (160 BC) that water, although it may be allowed to stand on corn fields during the cold part of winter, must be removed before the spring (White, 1970). Increased oxygen deficiencies from higher precipitation in upland oceanic areas are likely to reduce man’s capacity to win a livelihood from a landscape that has suffered too long from management practices that

Fig. 26. Prehistoric walls on the island of Papa Stour, Shetland, which have been exposed by removal of the peat that grew since the walls were built. Evidence from sites such as this and archaeological investigations of Neolithic/Bronze Age settlements in Scotland and Ireland strongly suggest that peat formation only became extensive throughout much of Scotland and Ireland as a result of human settlement and tree removal, illustrating the sensitivity of oceanic habitats to environmental alteration.

~~

Fig. 27. Bog being stripped of peat in Caithness, Scotland. These extensive peat deposits have developed mainly since the arrival of man and represent a degradation of what was formerly a more productive and ecologically more varied habitat with extensive areas of birch forest.

170

R. M. M. CRAWFORD

Fig. 28. Remains of birch lying below Neolithic peat in a Caithness bog.

have never properly understood the limitations (and potentials) of oceanic climates for plant growth. It can only be hoped that in the future, with our better understanding of hydrology, climate and plant-soil relationships, that landscape dereliction through peat growth will not be a continuing cause of reduction in fertility and ecological diversity.

OXYGEN A N D PLANT DISTRIBUTION

171

VIII. CONCLUSIONS By comparing the distribution and physiology of plants from a range of habitats this chapter has attempted to show that during the life cycle of most species of higher plants there are critical periods when oxygen is a resource that is frequently limiting for germination, growth and survival. The range of adaptations employed by plants, either to secure additional supplies of oxygen, or to be able to dispense with it for a period, or else counteract the harmful effects of its absence, are both diverse and widely employed by many species. The ecological advantages and disadvantages of the various adaptations have been pointed out particularly in relation to anoxia tolerance. Physiological adaptation, just like morphological adaptation, allows survival through selecting certain survival options which, although advantageous under one set of conditions, can penalize the adapted individual when such adaptations are not necessary. The delicate balance between the relative advantages and disadvantages of adaptations to oxygen shortages leaves a genetic imprint that is found not just between different species but between cultivars and ecotypes of the same species. In much of the current thinking on competition, resource acquisition is a predominant theme. This chapter has attempted to show that tolerance of adversity is just as important a path to survival as resource acquisition and, despite the relative abundance of oxygen in the atmosphere, it too is a resource that frequently limits the survival and competitive ability of many species. Finally, the future prospects for oxygen limitations for plants rooted in upland and oceanic areas are discussed in relation to possible changes in the hydrologic cycle as a consequence of the greenhouse effect.

ACKNOWLEDGEMENTS I am much indebted to Professor J.A. Raven, FRS, Professor R. Braendle, Dr W. Armstrong and Dr S. C. Maberly for reading the manuscript of this chapter and for making many helpful suggestions and to Professor J. Sprent, Dr M.P. Coutts, Dr M.B. Jackson and Dr G. W. Whittington for bringing much useful information to my notice.

REFERENCES Al-Ani, A., Leblanc, J.M., Raymond, P. and Pradet, A. (1982). Effet de la pression partielle d’oxygene sur la vitesse de germination des semences a reserves lipidiques et amylacees: role du metabolisme fermentaire. C. R. Acad Sci. Paris 295,271-274.

172

R . M. M. CRAWFORD

Al-Ani, A., Bruzau, F., Raymond, P., Saint-Ges, V., Leblanc, J.M. and Pradet, A. (1985). Germination, respiration and adenylate energy charge of seeds at various oxygen partial pressures. Plant Physiol. 79, 885-890. Aldasoro, J. and Nicolas, G. (1980). Fermentative products and dark CO, fixation during germination of seeds of Cicer arietinum. Phytochem. 19, 3-5. Allen, S., Raven, J.A. and Sprent, J.I. (1988). The role of long-distance transport in intracellular pH regulation in Phuseolus vulgaris grown with ammonium or nitrate as nitrogen source, or nodulated. J . exp. Bot. 39, 513-528. Alpi, A. and Beevers, H. (1983). Effects of O2 concentration on rice seedlings. Plant Physiol. 71, 3&34. Andrews, C.J. and Pomeroy, M.K. (1983). The influence of flooding pretreatment on metabolic changes in winter cereal seedlings during ice-encasement. Can. J . Bot. 61, 142-147. Andrews, C.J. and Pomeroy, M.K. (1989). Metabolic acclimation to hypoxia in winter cereals. PIant Physiol. 91, 1063-1068. Appleby, C.A., Tjepkema, J.D. and Trinick, M.J. (1983). Haemoglobin in a nonleguminous plant, Parasponia: possible genetic origin and function in nitrogen fixation. Science 220, 951-953. Appleby, C.A., Bogusz, D., Dennis, E.S. and Peacock, W.J. (1988). A role for haemoglobin in all plant roots? Plant Cell Environ. 11, 359-367. Arber, A. (1920). Water Plants. Cambridge University Press, Cambridge. Armstrong, J. and Armstrong, W. (1988). Phragmites australi.s-A preliminary study of soil-oxidizing sites and internal gas transport pathways. New Phytol. 108. 373382. Armstrong, J., Armstrong. W. and Beckett. P.M. (1988). Phragmites australis: a critical appraisal of the ventilating pressure concept and an analysis of resistance to pressurized gas flow and gaseous diffusion in horizontal rhizomes. New Ph.vtol. 110, 383-389. Armstrong, W. (1972). A re-examination of the functional significance of aerenchyma. Physiol. Plant. 27, 173-177. Armstrong, W. (1979). Aeration in higher plants. Adv. bot. Res. 7, 225-332. Armstrong, W. and Armstrong, J. (1990). Light enhanced convective throughflow increases oxygenation in rhizomes and rhizosphere of Phragmites austrulis (Cav.) Trin. ex Steud. New Phytol. 114, 121-128. Armstrong, W. and Gaynard, T. (1976). The critical oxygen pressures for respiration in intact plants. Physiol. Plant. 37, 20CL206. Armstrong, W. and Webb, T. (1985). A critical oxygen pressure for root extension in rice. J . cup. Bot. 36, 1573-1,582. Armstrong, W., Armstrong, J., Beckett, P.M. and Justin, S.H.F.W. (1991). Convective gas-flows in wetland plant aeration. In: Plant L(/e Under Oxygen Stress; Ecology, Physiology and Biochemistry (Ed. by M.B. Jackson, H . Lambers and B.B. Vartapetian) pp. 267-282. Academic Publishing, The Hague. Barber, K.E. (1977). Recent peat stratigraphy and climatic change. X . INQUA Congress: Ecology. Ahstracts, p. 28. Barber, K.E. (1981). Peat Stratigraphy and Climatic CIiunge. A.A. Balkema, Rotterdam. Barclay, A.M. and Crawford, R.M.M. (1982). Plant growth and survival under strict anaerobiosis. J . exp. Bot. 33, 541-549. Barclay, A.M. and Crawford, R.M.M. (1983). The effect of anaerobiosis on carbohydrate levels in storage tissues of wetland plants. Ann. Bot. 51, 255-259. Baskin, J.M., Baskin, C.C. and Spooner, D.M. (1989). Role of temperature, light and

OXYGEN AND PLANT DISTRIBUTION

I73

date seeds were exhumed from soil on germination of four wetland perennials. Aquatic Bot. 35, 387-394. Bennett, A.B. and Spanswick. R.M. (1984). H+-ATPase activity from storage tissue of Beta vulgaris. Plant Physiol. 74, 545-548. Bergersen, F.J. and Goodchild, D.J. (1973) Aeration pathways in soybean root nodules. Aust. J . biol. Sci. 26, 729-740. Berner, R.A. and Canfield, D.E. (1989). A new model for atmospheric oxygen over phanerozoic time. Amer. J . Sci. 289, 333-361. Berry, L.J. and Norris, W.E. (1949). Studies of onion root respiration. I . Velocity of oxygen consumption in different segments of root at different temperatures as a function of partial pressure of oxygen. Biochim. et Biophvs. Acta 3, 593-606. Bertani, A. and Brambilla, I. (1982). Effect of decreasing oxygen concentration on wheat roots: growth and induction of anaerobic metabolism. Z . l'flanzenphysiol. 108, 283-288. Betz, A. (19%). Athanolumsatz in mersitematischen Wurzelspitzen von Pisum sativum. Flora 146, 532-545. Blacquiire, T. (1986). Nitrate reduction in the leaves and numbers of nitrifiers in the rhizosphere of Plantago lanceolatri growing in two contrasting sites. Planr Soil 79, 305-3 16. Bolin, W., Jaer, J. and DOOs, B.R. (1990). The greenhouse effect climatic change and ecosystems. In: Scope 29: The Greenhouse Eject Climatic Change and Ecosystems (Ed. by W. Bolin. B.R. DOOs, J. Jager and R.A. Warwick), pp. 1-32. John Wiley, Chichester. Bond, G . (1952). Some features of root growth in nodulated plants of Myrica gale. Ann. Bot. 16, 467475. Bonner, W.D. ( I 973). Mitichondria and plant respiration. In: Ph.vrochemistrjj (Ed. by L.P. Miller), pp. 221-261. Academic Press, New York. Bown, A.W. (1985). COz and intercellular pH. Plant Cell Environ. 8, 459465. Braendle, R. (1990). Flooding resistance of rhizomatous amphibious plants. In: Plant Life Under O.xygen Deprivation (Ed. by M.B. Jackson, D.D. Davies and H. Lambers), pp. 3 5 4 6 . Academic Publishing, The Hague. Braendle, R. and Crawford. R.M.M. (1987). Rhizome anoxia tolerance and habitat specialization in wetland plants. In: Plant Life in Aquaric and Amphibious Habitats (Ed. by R.M.M. Crawford), pp. 397410. British Ecological Society Special Publication, Blackwell, Oxford. Brix, H . (1988). Light-dependent variations in the composition of the internal atmosphere of Phragmites australis (Cav.) Trin. ex Steudel. Aquatic Bot. 30, 319329. Brix, H. (1989) Gas exchange through dead culms of reed Phragmites australis (Cav.) Trin ex. Steudel. Aquatic Bot. 35, 81-98. Brown, H.T. and Escombe, F. (1900). Static diffusion of gases and liquids in relation to the assimilation of carbon and translocation in plants. Phil. Trans R . Soc. (Lotzd.)B. 193, 223-291. Brown, A.D.H., Marshall, D.R. and Albrecht, L. (1974). The maintenance of alcohol dehydrogenase polymorphism in Bromus mollis L. Aust. J . Sci. 27, 545-559. Brown, J.M.A.. Outred, H.A. and Hill, C.F. (1969). Respiratory metabolism in mangrove seedlings. Plant Physiol. 44, 287. Brundrett, M.C. and Kendrick, B. (1988). The mycorrhizal status, root anatomy, and phenology of plants in a sugar maple forest. Can. J. Bot. 66, 1153-1 173. Budyko, M.I., Ronov, A.B. and Yanshin. A.L. (1987). History of the Earth's Atmosphere.. Springer Verlag, Berlin.

174

R . M . M. CRAWFORD

Bufler, G. and Bangerth, F. (1982). Pyruvate decarboxylase in “golden delicious” apples: kinetics and relation to acetoin and ethanol production in different storage conditions. Sci. Hort. 16, 137-146. Buttery, B.R. and Lambert, J.M. (1965). Competition between Glyceria maxima and Phragmites communis in the region of Surlingham Broad. I. The competition mechanism. J . Ecol. 53, 163-181. Buttery, B.R., Williams, J.M. and Lambert, J.M. (1965). Competition between Glyceria maxima and Phragmites communis in the region of Surlingham Broad. 11. The fen gradient. J . Ecol. 53, 183-195. Campiranon, S. and Koukkari, W.L. (1977). Germination of wild rice Zizania aquatica seeds and the activity of alcohol dehydrogenase in young seedlings. Physiol. Plant. 41,293-297. Carrodus, B.B. and Triffett, A.C.K. (1975). Analysis of composition of respiratory gases in woody stems by mass spectrometry. New Phytol. 74,243-246. Carter, D.J.T. and Draper, I. (1988). Has the north-east Atlantic become rougher? Nature 337,494. Caulfield S. (undated). CPde and Beldberg Beg. A Guide to Two Prehistoric Farms in North Mayo. Private Publication of Local Guidebook, Ireland. Caulfield, S. (1978). Neolithic fields: the Irish evidence. In: Early Land Allotment in the British Isles: a Survey cf Recent Work (Ed. by H.C. Bowen and P.J. Fowler), pp. 137-143. Brit. Archaeo. Rep., Oxford, 48. Cerana, R., Patti, V. and Lado, P. (1989). Changes in ethanol, lactate and malate content in Acer pseudoplatanus cells: effects of fusicoccin and 0, availability. Physiol. Plant. 75, 183-187. Chapman, V.J. (1976). Mangrove Vegetation. Cramer, Vaduz. Chen, C.C., Dixon, J.B. and Turner, F.T. (1980). Iron coating on rice roots: mineralogy and quantity influencing factors. Soil Sci. Soc. Amer. J . 44,635-639. Chilvers, M.T. and Daft, M.F.J. (1981). Mycorrhizas of the Liliiflorae, 2: Mycorrhiza formation and incidence of root hairs in field grown Narcissus L. Tulipa L. and Crocus cultivars. New Phytol. 89, 247-261. Clayton, J.S. and Bagyaraj, D.J. (1984). Vesicular-arbuscular mycorrhizas in submerged aquatic plants of New Zealand. Aquatic Bot. 19, 251-262. Cole, N.H.A. (1977). Effect of light, temperature and flooding on seed germination of the neotropical Panicum laxum Sw. Biotropica 9, 191-194. Conner, W.H., Gosselink, J.G. and Parrondo, R.T. (1981). Comparison of the vegetation of three Louisiana swamp sites with different flooding regimes. Amer. J . Bot. 68, 32&331. Cook, C.D.K. (1987). Disperqion in aquatic and amphibious vascular plants. In: Plant Life in Aquatic and Amphibious Habitats (Ed. by R.M.M. Crawford), pp. 179-190. British Ecological Society Special Symposium, Blackwell, Oxford. Cossins, E.A. (1978). Ethanol metabolism in plants. In: Plant Life in Anaerobic Environments (Ed. by D.D. Hook and R.M.M. Crawford), pp. 169-202. Ann Arbor Science, Michigan. Cossins, E.A. and Turner, E.R. (1962). Losses of alcohol and alcohol dehydrogenase activity in germinating seeds. Ann. Bot. 26, 591-597. Countryside Commission for Scotland (1990). The Mountain Areas of Scotland. Conservation and Management. Report by the Countryside Commission for Scotland, Perth. Coutinho, L.M. and Struffaldi, Y. (1971). Observacoes sobre a germinacao das sementes e o crescimento das plantulas se uma leguminosa da mata amazonica de igapo (Parkia auriculata Spruce Mss). Phyton 28, 149-1 59.

OXYGEN AND PLANT DISTRIBUTION

175

Coutts, M.P. and Nicoll, B.C. (1990). Growth and survival of shoots, roots and mycorrhizal mycelium in clonal Sitka spruce during the first growing season after planting. Can. J . For. Res. 20, 861-868. Cox, W.J. and Reisenauer, H.M. (1973). Growth and ion uptake by wheat supplied with nitrogen as nitrate, o r ammonium, or both. Plant Soil 38, 363-380. Crawford, R.M.M. (1967). Alcohol dehydrogenase activity in relation to flooding tolerance in roots. J . Exp. Bot. 18, 458464. Crawford, R.M.M. (1976). Tolerance and anoxia and the regulation of glycolysis in tree roots. In: Tree Physiology and Yield Improvement (Ed. by M.G.R. Cannell and F.T. Last), pp. 387401. Academic Press, London. Crawford, R.M.M. (1977). Tolerance of anoxia and ethanol metabolism in germinating seeds. New Phytol. 79, 51 1-517. Crawford, R.M.M. (1982a). The anaerobic retreat as a survival strategy for aerobic plants and animals. Trans Bot. Soc. Edin. 44,57-63. Crawford, R.M.M. (1982b). Physiological responses to flooding. Encyclop. Plant Physiol. 12B, 453477. Crawford, R.M.M. (1989). Studies in Plant Survival. Blackwell Scientific, Oxford. Crawford, R.M.M. and Baines, M.A. (1977). Tolerance of anoxia and the metabolism of ethanol in tree roots. New Phytol. 79,519-526. Crawford, R.M.M. and Finegan, D.M. (1989). Removal of ethanol from lodgepole pine roots. Tree Physiol. 5, 53-61. Crawford, R.M.M. and McManmon, M. (1968). Inductive responses of alcohol and malic dehydrogenases in relation to flooding tolerance in roots. J . exp. Bot. 19, 43541. Crawford, R.M.M. and Tyler, P.D. (1969). Organic acid metabolism in relation to flooding tolerance in roots. J . Ecol. 57, 235-244. Crawford, R.M.M. and Woollenweber-Ratzer, B. (1992). Influence of L-ascorbic acid on post-anoxic growth and survival of chickpea seedlings (cicer arietinum L.). J. exp. Bot. 43. Crawford, R.M.M. and Zochowski, Z.M. (1984). Tolerance of anoxia and ethanol toxicity in chickpea seedlings. J . exp. Bot. 35, 1472-1480. Crawford, R.M.M., Monk, L.S. and Zochowski, Z.M. (1987). Enhancement of anoxia tolerance by removal of volatile products of anaerobiosis. In: Plant Life in Aquatic and Amphibious Habitats (Ed. by R.M.M. Crawford) pp. 375-384. British Ecological Society Special Symposium, Blackwell, Oxford. Crawford, R.M.M., Studer, C. and Studer, K. (1989). Deprivation indifference as a survival strategy in competition: advantages and disadvantages of anoxia tolerance in wetland vegetation. Flora 182, 189-201. Crocker, W. (1907). Germination of seeds of water plants. Bot. Gaz. 44,375-380. Curran, M., Cole, M. and Allaway, W.G. (1986). Root aeration and respiration in young mangrove plants (Avicennia marina (Forsk.) Vierh.). J . exp. Bot. 37, 12251233. Dacey, J.W.H. (1980). Internal winds in water lilies: an adaptation for life in anaerobic sediments. Science 210, 1017-1019. Dacey, J.W.H. (1987). Knudsen-transitional flow and gas pressurization in leaves of Nelumbo. Plant Physiol. 85, 199-203. Dacey, J.W.H. and Klug, M.J. (1982). Ventilation by floating leaves in Nuphar. Am. J . Bot. 69,999-1003. Davies, D.D. (1980). Anaerobic metabolism and the production of organic acids. In: The Biochemistry of Plants, Vol. 2 (Ed. by D.D. Davies), pp. 581-61 I . Academic Press, New York.

176

R . M. M. CRAWFORD

Davies, D.D. (1986). The fine control of cytosolic pH. Physiol. Plant. 67, 702-706. DeBell, D.S. and Naylor, A.W. (1972). Some factors affecting germination of swamp tupelo seeds. Ecology 53, 626-634. Devaux, H. (1899). Asphyxie spontanee et production d’alcool dans les tissus profonds des tiges ligneuses poussant dans les conditions naturelles. C.R . hebd. Acad. Sci. Paris 128, 1246. Drew, M.C. (1988). Effets of flooding and oxygen deficiency on plant mineral nutrition. In: Advances in Plant Nutrition, V o l 3 (Ed. by B. Tinker and A. Lauchli), pp. 115-159. Praeger, New York. Drew, M.C. (1990). Sensing soil oxygen. Plant Cell Environ. 13, 681-693. Drew, M.C. and Lynch, J.M. (1980). Soil anaerobiosis, microorganisms and root function. Ann. Rev. Phytopathol. 18, 37-66. Drew, M.C. and Sisworo, E.J. (1979). The development of waterlogging damage in young barley plants in relation to plant nutrient status and changes in soil properties. New Phytol. 82, 301-314. Eamus, D. and Jarvis, P.G. (1989). The direct effects of increase in the global atmospheric CO, concentration on natural and commercial temperate trees and forests. Adv. Ecol. Res. 19, 1-55. Effer, W.R. and Ranson, S.L. (1967). Respiratory metabolism in buckwheat seedlings. Plant Physiol. 42, 1042-1052. Ernst, W.H.O. (1982). Schwermetallpflanzen In: Pjanzeneokologie und Mineralstoflwechsel (Ed. by H. Kinzel), pp. 472-506. Ulmer Verlag, Stuttgart. Ernst, W.H.O. (1990). Ecophysiology of plants in waterlogged and flooded environments. Aquatic Bot. 38, 73-90. Ernst, W.H.O. and Lugtenborg, T.F. (1980). Vergleichende Okophysiologie von Juncus articulatus und Holcus lanatus. Flora 169, 121-1 34. Fagerstedt, K.V. and Crawford, R.M.M. (1986). Changing kinetic properties of barley alcohol dehydrogenase during hypoxic conditions. J. exp. Bot. 179, 857864. Fagerstedt, K.V. and Crawford, R.M.M. (1987). Is anoxia tolerance related to flooding tolerance? Func. Ecol. 1, 49-55. Fed, R.J., Dlouhy, S.R. and Schwartz, D. (1979). Analysis of maize alcohol dehydrogenase by native SDS two dimensional electrophoresis and autoradiography. Mol. Gen. Genet. 169, 7-12. Fed, R.J., Brennan, M.D. and Schwartz, D. (1980). In vivo translation of maize ADH: evidence for the anaerobic induction of mRNA. Biochem. Genet. 18, 681691. Fiala, K. (1978). Underground organs in Typha angustifolia and Typha latifolia, their growth propagation and production. Acta Sc. Nut. Brno. 12 (8), 1 4 3 . Fiscus, E.L. and Kramer, P.J. (1970). Radial movement of oxygen in plant roots. Plant Physiol. 45, 667-669. Francis, C.M., Devitt, A.C. and Steele, P. (1974). Influence of flooding on the alcohol dehydrogenase activity of roots of Trifolium subterraneum L. Aust. J. Plant Physiol. 1, 9-13. Frankland, B., Bartley, M.R. and Spence, D.H.N. (1987). Germination under water. In: Plant Life in Aquatic and Amphibious Habitats (Ed. by R.M.M. Crawford), pp. 397410. British Ecological Society Special Publication, Blackwell, Oxford. Freeling, M.D. and Schwartz, D. (1973). Genetic relationships between the multiple alcohol dehydrogenases of maize. Biochem. Genet. 8, 27-36. Gallagher, J.L. (1983). Seasonal patterns in recoverable underground reserves in Spartina alternipora Loisel. Am. J. Bot. 70, 212-215.

OXYGEN AND PLANT DISTRIBUTION

177

Gallagher, J.L., Wolf, P.L. and Pfeiffer, W.J. (1984). Rhizome and rootgrowth rates and cycles in protein and carbohydrate concentrations in Georgia Spartina alternijlora Loisel. plants. Am. J . Bot. 71, 165-169. Gigon, A. and Rorison, I.H. (1972). The response of some ecologically distinct plant species to nitrate- and to ammonium-nitrogen. J . Ecol. 60, 93-102. Gill, C.J. (1970). The flooding tolerance of woody species-a review. For. Abst. 31, 671488. Godwin, H. (1975). The History of the British Flora. Cambridge University Press, Cambridge. Green, F.H.W. (1964). The climate of Scotland. In: The Vegetation of Scotland (Ed. by J.H. Burnett), pp. 15-35. Oliver and Boyd, Edinburgh, London. Gries, C., Kappen, L. and Losch (1990). Mechanism of flood tolerance in reed, Phragmites australis (Cav.) Trin. ex Steudel. New Phytol. 114, 589-593. Griffin, D.M. (1968). A theoretical study relating to the concentration and diffusion of oxygen to the biology of organisms in the soil. New Phytol. 67, 561-577. Grime, J.P. (1979). Plant Strategies and Vegetation Processes. John Wiley, ChiChester. Grosse, W. and Schroder, P. (1984). Oxygen supply of roots by gas transport in aldertrees. Z . Naturforsch. 39c, 1186-1 188. Grosse, W. and Schroder, P. (1985). Aeration of the roots and chloroplast-free tissues of trees. Ber. Deutsch. Bot. Ges. 98, 31 1-318. Grubb, P.J. (1977). The maintenance of species richness in plant communities: the importance of the regeneration niche. Biol. Rev. 52, 107-145. Haldemann, C. and Braendle, R. (1986). Jahrzeitliche Unterschiede im Reservstoffgehalt und von garungsprozessen in Rhizomen von Sumpf- und Rohrichpflanzen aus dem Freiland. Flora 178, 307-313. Halliwell, B. (1984). Chloroplast Metabolism. Clarendon Press, Oxford. Hanson, R.B. (1977). Nitrogen fixation (acetylene reduction) in a salt marsh amended with sewage sludge and organic carbon and nitrogen compounds. Appl. environ. Microbiol. 33, 846852. Hanson, A.D. and Brown, A.H.D. (1984). Three alcohol dehydrogenase genes in wild and cultivated barley: characterization of the products of variant alleles. Biochem. Genet. 22,495-515. Hanson, A.D., Jacobsen, J.V. and Zwar, J.A. (1984). Regulated expression of the alcohol dehydrogenase genes in barley aleurone layers. Plant Physiol. 75,573-58 I . Harberd, N.P. and Edwards, K.J.R. (1983). Further studies on the alcohol dehydrogenases in barley: evidence for a third alcohol dehydrogenase locus and data on the effect of an alcohol dehydrogenase-1 null mutation in homozygous and in heterozygous condition. Gen. Res. in Cam. 41, 109-1 16. Harley, J.L. (1969). The Biology of Mycorrhiza, 2nd Edn. Leonard Hill, London. Harrison, D.E.F. (1 976). The Oxygen Metabolism of Microorganisms. Meadowfield Press, Shildon, England. Hendry, G.A.F. and Brocklebank, K.J. (1985). Iron-induced oxygen radical metabolism in water-logged plants. New Phytol. 101, 199-206. Henslow, G. (1893). A theoretical origin of endogens from exogens, through selfadaptation to an aquatic habit. J. Linn. SOC. Bor. 29, 485-528. Herity, M. (1971). Prehistoric fields in Ireland. Irish Univ. Review 1, 258-265. Hoffman, N.E., Bent, A.F. and Hanson, A.D. (1986). Induction of lactate dehydrogenase isozymes by oxygen deficit in barley root tissue. Plant Physiol. 82, 658-663. Hook, D.D. and Denslow, S. (1987). Metabolic responses of four families of loblolly pine to two flood regimes. In: Plant Life in Aquatic and Amphibious Habitats (Ed.

178

R . M . M . CRAWFORD

by R.M.M. Crawford), pp. 281-292. British Ecological Society Special Symposium, Blackwell, Oxford. Hook, D.D. and Scholtens, J.R. (1978). Adaptations and flood tolerance of tree species. In: Plant Life in Anaerobic Environments (Ed. by D.D. Hook and R.M.M. Crawford), pp. 119-136. Ann Arbor, Michigan. Hourmant, A. and Pradet, A. (1981). Oxidative phosphorylation in germinating lettuce seeds (Lacfuca saliva) during the first hours of imbibition. Plant Physiol. 68, 631-635. Huck, M.G. (1970). Variation in taproot elongation as influenced by composition of soil air. Agron. J . 62, 815-818. Hunter, M.I.S., Hetherington, A.M. and Crawford, R.M.M. (1983). Lipid peroxidation-a factor in anoxia intolerance in Iris species. Phytochem. 22, 1145-1 147. Huntley, B. and Birks, H.J.B. (1983). An Atlas ofPast and Present Pollen Maps.for Europe: 0-13 000 years ago. Cambridge University Press, Cambridge. Ingram, H.A.P. (1978). Soil layers in mires: functions and terminology. J . Soil Sci. 29, 224-227. Jackson, M.B. (1990). Hormones and developmental change in plants subjected to submergence or soil waterlogging. Aquatic Bot. 38, 49-72. Jackson, M.B., Hermann, B. and Goodenough, A. (1982). An examination of the importance of ethanol in causing injury to flooded plants. Plant Cell and Environ. 5, 163-172. James, W.O. (1953). Plant Respiration. Clarendon Press, Oxford. Janiesch, P. (1986). Bedeutung einer Ernahrung von Carex-Arten mit Ammonium oder Nitrat fur deren Vorkommen in Feuchtgesellschaften. Abhandlung Munster (Westf.) 48, 341-354. Jenkins, L.E.T. and a p Rees, T. (1983). Effects of anoxia and flooding on alcohol dehydrogenase in roots of Glycerin maxima and Pisum sativurn. Phytochem. 22, 2389-2393. Johnson, J., Cobb, B.G. and Drew, M.C. (1989). Hypoxic induction of anoxia tolerance in root tips of Zea mays. Plant Physiol. 91, 837-841. Jolley, K.L., Evans, L.H., Makino, M. and Mason, H.S. (1974). Oxytyrosinase. J . biol. Chem. 249, 335-345. Joly, C. (1982). Flooding Tolerance of Some Brazilian Trees. PhD thesis, University of St Andrews, Scotland. Junk. W.J. (1989). Flood tolerance and tree distributions in central Amazonian floodplains. In: Tropical Forests (Ed. by L.B. Holm-Nielsen, I.C. Nielsen and H. Balslev). pp. 47-64. Academic Press, New York. Justin, S.H.F.W. and Armstrong. W. (1987). The anatomical characteristics of roots and their reponse to flooding. New Phytol. 106, 465495. Kader, A.A. (1986). Biochemical and physiological basis for effects of controlled and modified atmospheres on fruits and vegetables. Food Technol. 40, 99-100, 102104. Kausch, A.P., Seago, J.L. and Marsh, L.C. (1981). Changes in starch distribution in the overwintering organs of Typha Iatifolia (Typhaceae). Amer. J . Bot. 68, 877880. Kawase, M. (1981). Anatomical and morphological adaptation of plants to waterlogging. Hort. Sci. 16, 3&34. Keeley, J.E. (1979). Population differentiation along a flood frequency gradient: physiological adaptation to flooding in Nyssa sylvatica. Eco. Monographs 49, 89108.

OXYGEN AND PLANT DISTRIBUTION

179

Kennedy, R.A., Rumpho, M.E. and VanderZee, D. (1983). Germination of Echinochloa crus-galli (barnyard grass) seeds under anaerobic conditions and response to metabolic inhibitors. Plant Physiol. 72, 787-794. Kennedy, R.A., Rumpho, M.E. and Fox, T.C. (1987). Germination physiology of rice and rice weeds: metabolic adaptations to anoxia. In: Plant Life in Aquatic and Amphibious Habitats (Ed. by R.M.M. Crawford), pp. 193-203. British Ecological Society Special Symposium, Blackwell, Oxford. Kerbel, E.L., Kader, A.A. and Romani, R.J. (1988). Effects of elevated CO, concentrations on glycolysis in intact “Bartlett” pear fruit. Plant Physiol. 86, 1205-1 209. Khan, A.G. (1974). The occurrence of mycorrhizae in halophytes, hydrophytes and xerophytes and of Endogone spores in adjacent soils. J. Gen. Microbiol. 81, 7-14. Kimmerer, T.W. (1 987). Alcohol dehydrogenase and pyruvate decarboxylase activity in leaves and roots of eastern cottonwood (Populus deltoides Bartr.) and soybean (Glycine max. L.). Plant Physiol. 84, 1210-1213. Kimmerer, T.W. and Kozlowski, T.T. (1982). Ethylene, ethane, acetaldehyde and ethanol production by plants under stress. Plant Physiol. 69, 840-847. Kimmerer, T.W. and Stringer, M.A. (1988). Alcohol dehydrogenase and ethanol in the stems of trees. Plant Physiol. 87, 693-697. Kirkby, E.A. and Mengel, K. (1967). Ionic balance in different tissues of the tomato plant in relation to nitrate, urea or ammonium nutrition. Plant Physiol. 42, 6 1 4 . KonEalova, H . (1990). Anatomical adaptations to waterlogging in roots of wetland graminoids: limitations and drawbacks. Aquatic Bot. 38, 127-1 34. KonEalova, H . and Pazourek, J. (1987). Root dimorphism in flooded Carex gracilis Curt. Acta Univ. Carol. Biol. 31, 43-50. KonEalova, H., Pokorny, J. and Kvet, J. (1988). Root ventilation in Carex gracilis Curt.: Diffusion or mass flow? Aquatic Bot. 30, 149-155. Kordan, H.A. (1972). Rice seedlings germinated in water with normal and impeded environmental gas exchange. J . appl. Ecol. 9, 527-533. Kordan, H.A. (1976). Adventitious root initiation and growth in relation to oxygen supply in germinating rice seedlings. New Phytol. 76, 81-86. Laan, P., Berrevoets, M.J., Lythe, S., Armstrong, W. and Blom, C.W.P.M. (1989a). Root morphology and aerenchyma formation as indicators of the flood tolerance of Rumex species. J . Ecol. 77, 693-703. Laan, P., Smolders, Blom, C.W.P.M. and Armstrong, W. (1989b). The relative roles of internal aeration, radial oxygen losses, iron exclusion and nutrient balances in flood tolerance of Rumex species. Acta Bot. Neerlandica 38, 131-145. Laanbroek, H.J. (1990). Bacterial cycling of minerals that affect plant growth in waterlogged soils: a review. Aquatic Bot. 38, 109-125. Larson, R.A. (1988). The antioxidants of higher plants. Phytochem. 27, 969-978. Leuning, R. (1983). Transport of gases into leaves. Plant Cell Environ. 6, 181-194. Lincoln, R.J., Boxshall, G.A. and Clark, P.F. (1982). A Dictionary of Ecology, Evolution and Systematics. Cambridge University Press, Cambridge. Linhart, Y.B. and Baker, G. (1973). Intra-population differentiation of physiological responses to flooding in a population of Veronica peregrina. Nature 242, 275. Low, A.J. (1987). Sitka spruce silviculture in Scottish forests. Proc. Roy. SOC.Edin. 93B, 93-106. Lynch, J.M. (1977). Phytotoxicity of acetic acid produced in the anaerobic decomposition of wheat straw. J . appl. Bacteriol. 42, 81-87.

180

R. M. M. CRAWFORD

Lyttle, R.W. and Hull, R.J. (1980). Photoassimilate distribution in Spartina alterniflora Loisel. I. Vegetative and floral development. Agron. J . 72, 933-946. Mangerurd, J., Andersen, S.T., Berglund, B.E. and Donner, J.J. (1974). Quaternary stratigraphy of Norden, a proposal for terminology and classification. Boreas 3, 109-1 28. Marschner, H. (1986). Mineral Nutrition in Higher Plants. Academic Press, London, New York. Marshall, D.R., Broue, P. and Pryor, A.J. (1973). Adaptive significance of alcohol dehydrogenase enzymes in higher plants. Nature (New Biol.) 244, 1617. Mayne, R.G. and Lea, P.J. (1984). Alcohol dehydrogenase in Hordeum vulgare: changes in levels under hypoxia. Plant Sci. Lett. 37, 73-78. McKee, K.L., Mendelssohn, I.A. and Burdick, D.M. (1989). Effect of long-term flooding on root metabolic response in five freshwater marsh plant species. Can. J . Bot. 67, 34463452. Menegus, F., Cattaruzza, L., Chersi, A. and Fronza, G. (1989). Differences in the anaerobic lactate-succinate production and in the changes of cell sap p H for plants with high and low resistance to anoxia. Plant Physiol. 90,29-32. Mengel, K. and Viro, M. (1978). The significance of plant energy status for the uptake and incorporation of NH,-nitrogen by young rice plants. Soil Sci. Plant Nutr. 24, 4 0 7 4 16. Merget, A. ( 1 874). Sur les phknomenes de thermodifusion gazeuse qui se produisent dans les feuilles et sur les mouvements circulatoires dans les feuilles et sur les mouvements circulatoires qui en resultant dans I’acte de la respiration chlorophyllienne. Comptes Rendus. Acad. Sci. Paris 77, 1468-1472. Milic-Emil, J. and Tyler, J.M. (1963). Relation between work output of respiratory muscles and end-tidal CO, tension. J . appl. Physiol. 18, 497-504. Miller, G.W. and Hsu, W.J. (1965). Effect of carbon dioxide-bicarbonate mixtures on oxidative phosphorlylation by cauliflower mitochondria. Biochem. J . 97,615-619. Mitchell, F. (1976). The Irish Landscape. Collins, London. Monk, L.S., Crawford, R.M.M. and Braendle, R. (1984). Fermentation rates and ethanol accumulation in relation to flooding tolerance in rhizomes of monocotyledonous species. J . exp. Bot. 35, 738-745. Monk, L.S., Crawford, R.M.M and Braendle, R. (1984). Fermentation rates and ethanol accumulation in relation to flooding tolerance in rhizomes of monocotyledonous species. J . exp. Bot. 35, 738-745. Monk, L.S., Fagerstedt, K.V. and Crawford, R.M.M. (1987). Superoxide dismutase as an anaerobic polypeptide. Plant Physiol. 85, 10161020. Monk, L.S., Fagerstedt, K.V. and Crawford, R.M.M. (1989). Oxygen toxicity and superoxide dismutase as an anti-oxidant in physiological stress. Physiol. Plant. 76, 45-59, Montgomery, E.G. (1912). Competition in cereals. Bull. Nebraska Agric. Stn. 26, 122. Moore, P.D. (1988). The development of moorlands and upland mires. In: Archaeology and the British Flora (Ed. by M. Jones), pp. 116122. Oxford University Committee for Archaeology, Oxbow Books, Oxford. Moringa, T. (1926). Germination of seeds under water. Amer. J . Bot. 13, 126-140. Morrell, S., Greenway, H . and Davies, D.D. (1990). Regulation of pyruvate decarboxylase in vivo and in vitro. J . exp. Bot. 41, 131-139. Newell, P.J. (1988). A buried wall in peatland by Sheshader, Isle of Lewis. Proc. SOC. Antiq. Scot. 118, 79-93.

OXYGEN AND PLANT DISTRIBUTION

181

Ohga, I. (1926). The germination of century old and recently harvested Indian Lotus fruit with special reference to the effect of oxygen supply. Amer. J . Bot. 13, 754759. Pankhurst, C.E. and Sprent, J.I. (1975). Effects of water stress on the respiratory and nitrogen-fixing activity of soybean root nodules. J . exp. Bot. 26, 287-304. Patriquin, D. and Knowles, R. (1972). Nitrogen fixation in the rhizosphere of marine angiosperms. Mar. Biol. 16, 49-58. Pradet, A. and Raymond, P. (1983). Adenine nucleotide ratios and adenylate energy charge in energy metabolism. Ann. Rev. Plant Physiol. 34, 199, 224. Pregnall, A.M., Smith, R.D., Kursar, T.A. and Alberte, R.S. (1984). Metabolic adaptation of Zostera marina (eelgrass) to diurnal periods of root anoxia. Mar. B i d . 83, 141-147. Pregnall, A.M., Smith, R.D., and Alberte, R.S. (1987). Glutamine synthetase activity and free amino acid poos of eelgrass. (Zostera marina L.) roots. J . exp. mar. Biol. Ecol. 106, 21 1-228. Raffineau-Delile, A. (1841). Evidence du mode respiratoire des feuilles de Nelumbrium. Ann Sci. Nut. II Ser. 16, 328-341. Ratcliffe, D.A. and Oswald, P.H. (Eds) (1988). The Flow Country: The Peat Lands of Caithness and Sutherland. Nature Conservancy Council, Peterborough. Raven, J.A. (1991). Plant responses to high 0, concentrations: relevance to previous high 0, episodes. In: Paleogeography, Paleoclimatology. Paleoecology (Global and Planetary Change Section). 97, 19-38. Raven, J.A. and Smith, F.A. (1976). Nitrogen assimilation and transport in vascular land plants in relation to intracellular pH regulation. New Phytol. 76, 41 5 4 3 1. Read, D.J. and Armstrong, W. (1972). A relationship between oxygen transport and the formation of the ectotrophic mycorrhizal sheath in conifer seedlings. New Phytol. 71, 49-53. Read, D.J., Koucheki and Hodgson, J. (1976). Vesicular-arbuscular mycorrhiza in natural vegetation systems. 1; The occurrence of infection. New Phytol. 77, 641653. Reggiani, R., Hochkoeppler, A. and Bertani, A. (1989a). Polyamines in rice seedlings under oxygen-deficit stress. Plant Physiol. 91, 1197-1201. Reggiani, R., Hochkoeppler, A. and Bertani, A. (1989b). Polyamines and anaerobic elongation of rice coleoptiles. Plant Cell Physiol. 30, 893-898. Roberts, J.K.M, Callis, J., Wemmer, D., Walbot, V. and Jardetzky, 0. (1984a). Mechanism of cytoplasmic pH regulation in hypoxic maize root tips and its in survival under hypoxia. Proc. Nut. Acad. Sci.81, 3379-3383. Roberts, J.K.M., Callis, J., Jardetzky, O., Walbot, V. and Freeling, M. (1984b). Cytoplasmic acidosis as a determinant of flooding intolerance in plants. Proc. Nut Acad. Sci. 81, 6029-6033. Roberts, J.K.M., Lane, A.N., Clark, R.A. and Nieman, R.H. (1985). Relationships between the rate of synthesis of ATP and the concentrations of reactants and products of ATP hydrolysis in maize root tips, determined by j'P nuclear magnetic resonance. Arch. Biochem. Biophys. 240, 7 12-722. Roberts, J.K.M., Chang, K., Webster, C., Callis, J. and Walbot, V. (1989). Dependence of ethanolic fermentation, cytoplasmic pH regulation, and viability on the activity of alcohol dehydrogenase in hypoxic maize root tips. Plant Physiol. 89, 1275-1278. Robertson, R.A. and Jowsey, P.C. (1968). Peat resources and development in the British Isles. Proc. 3rd Int. Peat ConJ Quebec. (Ed. by C. Lafleur and J. Butler).

182

R. M. M. CRAWFORD

Ruhland, W. and Ramshorn, K. (1938). Aerobe Garung in aktiven pflanzlichen Meristemen. Planta 28, 471-514. Rumpho, M.E. and Kennedy, R.A. (1981). Anaerobic metabolism in germinating seeds of Echinochloa crus-galli (barnyard grass). Plant Physiol. 68, 165-168. Saglio, P.H., Rancillac, F., Bruzaut, F. and Pradet, A. (1984). Critical oxygen for growth and respiration of excised and intact roots. Plant Physiol. 76, 151-154. Saglio, P.H., Drew, M.C. and Pradet, A. (1988). Metabolic acclimation to anoxia by low ( 2 4 kPa partial pressure) oxygen pretreatment in root tips of Zea mays. Plant Physiol. 86, 61-66. Saif, S.R. (1981). The influence of soil aeration on the efficiency of vesiculararbuscular mycorrhizae. I. Effect of soil oxygen on the growth and mineral uptake of Eupatorium odoratum L. inoculated with Glomus macrocarpus. New Phytol. 88, 649-65 9. Sasakawa, H. and Yamamoto, Y. (1978). Comparison of the uptake of nitrate and ammonium by rice seedlings. Plant Physiol. 62, 665-669. Scholander, P.F., van Dam, L. and Scholander, S.I. (1955). Gas exchange in the roots of mangroves. Amer. J. Bot. 42, 92-98. Schroder, P., Grosse, W. and Woermann, D. (1986). Localization of thermoosmotically active partitions in young leaves of Nuphar lutea. J . exp. Bot. 37, 14501461. Sherwin, T. and Simon, E. (1969). The appearance of lactic acid in Phaseolus seeds germinating under wet conditions. J . exp. Bot. 20, 776785. Short, F.T and McRoy, C.P. (1984). Nitrogen uptake by leaves and roots of the seagrass Zostera marina L. Bot. Mar. 27, 547-555. Sifton, H.B. (1959). The germination of light sensitive seeds of Typha latifolia. Can. J . Bot. 37, 719-739. Simpson, G.M. (1966). A study of germination in the seed of wild rice (Zizania aquatica). Can. J . Bot. 44, 1-9. Siriphanich, J. and Kadar, A.A. (1986). Changes in cytoplasmic and vacuolar pH in harvested lettuce tissue as influenced by COz. J. Amer. SOC.hort. Sci. 111, 73-77. Small, J.G.C., Potgeiter, G.H. and Botha, F.C. (1989). Anoxic seed germination of Erythrina caffra: ethanol fermentation and response to metabolic inhibitors. J . exp. Bot. 40,375-381. Smirnoff, N. and Crawford, R.M.M. (1983). Variation in the structure and response to flooding of root aerenchyma in some wetland plants. Ann. Bot. 51, 237-249. Smith, A.M. and ap Rees, T. (1979). Pathways of carbohydrate fermentation in the roots of marsh plants. Plqnta 146, 327-334. Smith, A.M., Hylton, C.M., Koch, L. and Woolhouse, H.W. (1986). Alcohol dehydrogenase activity in the roots of marsh plants in naturally waterlogged soils. Planta 168, 13CL-138. Smith, R.D., Dennison, W.C. and Alberte, R.S. (1984). Role of seagrass photosynthesis in root aerobic processes. Plant Physiol. 74, 1055-1058. Smith, S.E. (1980). Mycorrhizas of autotrophic higher plants. Biol. Rev. 55, 367-374. Solomos, T. (1982). Effect of low oxygen concentration on fruit respiration: Nature of respiratory diminution. In: Controlled Atmospheres f o r Storage and Transport of Perishable Agricultural Commodities (Ed. by D.G. Richardson and M. Mehehiuk), p. 161. Timber Press, Beaverton, OR. Sondergaard, M. and Laegaard, S. (1 977). Vesicular-arbuscular mycorrhiza in some aquatic plants. Nature 268, 232-233.

OXYGEN AND PLANT DISTRIBUTION

183

Spence, D.H.N. (1964). The macrophytic vegetation of freshwater lochs, swamps and associated fens. In: The Vegetation of Scotland (Ed. by J.H. Burnett), pp. 30-25. Oliver and Boyd, Edinburgh. Sprent, J.I. (1979). Nitrogen Fixation. McGraw-Hill, London. Sprent, J.I. and Gallacher, A. (1976). Anaerobiosis in soybean root nodules under water stress. Soil Biol. Biochem. 8, 3 17-320. Sprent, J.I. and Sprent, P. (1990). Nitrogen Fixing Organisms. Chapman and Hall, London. Steinmann, F. and Braendle, R. (1984a). Auswirkung von Halmverlusten auf der Kohlenhydrestoffwechsel Uberfluteter Seebinsenrhizome (Schoenoplectus lacustris L. Palla). Flora 175, 295-299. Steinmann, F. and Braendle, R. (l984b). Carbohydrate and protein metabolism in the rhizomes of the bulrush (Schoenoplectus lacustris L. Palla) in relation to natural development of the whole plants. Aquatic Bot. 19, 53-63. Stewart, G.R., Lee, J.A. and Orebamjo, T.O. (1973). Nitrogen metabolism of halophytes. 11. Nitrate availability and utilization. New Phytol. 72, 539-546. Studer, C. and Braendle, R. (1987). Ethanol, acetaldehyde, ethylene release and ACC concentration of rhizomes from marsh plants under normoxia, hypoxia and anoxia. In: Plant Life in Aquatic and Amphibious Habitats (Ed. by R.M.M. Crawford), pp. 293-301. British Ecological Society Special Symposium, Blackwell, Oxford. Studer, C. and Braendle, R. (1988). Postanoxische Effekte von Aethanol in Rhizomen von Glyceria maxima (Hartm.) Holmberg, Iris germanica (Cav.) Trin. Bot. Helvet. 98, 111-112. Sylvester-Bradley, R. (1976). Isolation of acetylene-reducing Spirilla from the roots of Potamogeton Jiliformis from Loch Leven, (Kinross). J . Microbiol. 97, 129-132. Taktajan, A.L. (1980). The order: flowering plants or Angiosperms. In: The Life of Plants 5 (I), 7-1 14. Prosvechcheniye, Moscow (in Russian). Taylor, D.L. (1942). Influence of oxygen tension on respiration, fermentation and growth in wheat and rice. Amer. J . Bot. 29, 721-738. Thomson, C.J., Atwell, B.J. and Greenway, H. (1989a). Response of wheat seedlings to low O2concentration in nutrient medium. J. exp. Bot. 218, 985-991. Thomson, C.J., Atwell, B.J. and Greenway, H. (1989b). Response of wheat seedlings to low O2concentration in nutrient medium. 11. K + / N a +selectivity of root tissues of different age. J . exp. Bot. 218, 993-999. Thomson, C.J. and Greenway, H. (1991). Metabolic evidence for stelar anoxia in maize roots exposed to low O2concentrations. Plant Physiol. 96, 12941301. Tilman, D. (1985). The resource ratio hypothesis of succession. Amer. Nut. 125,827852. Tjepkema, J.D. (1978). The role of oxygen diffusion from the shoots and nodule roots in nitrogen fixation by root nodules of Myrica gale. Can. J . Bot. 56, 1365-1371. Torres A.M. and Diedenhofen, U. (1979). Baker sunflower populations revisited. J . Hered. 70, 275-276. Tripepi, R.R. and Mitchell, C.A. (1984). Stem hypoxia and root respiration of flooded maple and birch seedlings. Physiol. Plant. 60, 567-571. Troelstra, S.R., Van Dijk, K. and Blacquiere, T. (1985). Effects of nitrogen source on proton excretion, ionic balance and growth of Alnus glutinosa (L.) Gaertner: comparison of N2fixation with single and mixed sources of NO,- and NH,'. Plant and Soil 84, 361-385.

184

R . M . M. CRAWFORD

Trought, M.C.Y. and Drew, M.C. (1981). Alleviation of injury to young wheat plants in anaerobic solution culture in relation to supply of nitrate and other inorganic nutrients. J . exp. Bot. 32, 509-522. Turner, J.F. and Turner, D.H. (1980). The regulation of glycolysis and the pentose phosphate pathway. In: The Biochemistry of’ Plants: A Comprehensive Treatise, Vol. 2. (Ed. by D.D. Davies), pp. 279-316. Academic Press, New York. Vartapetian, B.B, Andreeva, I.N. and Nurutdinov, N. (1978). Plant cells under oxygen stress. In: Plant Life in Anaerobic Environments (Ed. by D.D. Hook and R.M.M. Crawford), pp. 12-88. Ann Arbor Science, Michigan. Venkataramanan, M.N., Sing, H.D. and Baruah, J.N. (1982). Note on the occurrence of endotrophic mycorrhiza in water-hyacinth and its significance in agriculture. Indian J . Agric. Sci. 52, 39-10. Van Vierssen, W. (1982). Some notes on the germination of seeds of Najas murina L. Aquatic Bot. 12, 201-203. Vlamis, J. and Williams, E.E. (1964). Iron and manganese relations in rice and barley. Plant Soil 20, 22 1-23 1. Warwick, B.S., Harris, S.L. and Tjepkema, J.D. (1990). Oxygen regulation and hemoglobin. In: The Biology of’ Frankia and Actinorhizal Plants (Ed. by C.R. Schwintzer and J.D. Tjepkema), pp. 157-176. Academic Press, New York. Waters, I., Armstrong, W., Thompson, C.J., Setter, T.L., Adkins. S.. Gibbs, J. and Grenway, H. (1989). Diurnal changes in radial oxygen loss and ethanol mctabolism in roots of submerged and non-submerged rice seedlings. New Phytol. 113. 4 3 9 4 5 1. Weisner, S.E.B. and Graneli, W. (1989). Influence of substrate conditions on the growth of Phragmites australis after a reduction in oxygen transport to below ground parts. Aquatic Botany 35, 71-80. Werner, P.A. (1979). Competition and coexistence of similar species. In: Topics in Population Biology (Ed. by O.T. Solbrig, S. Jain, G.B. Johnson and P.H. Ravcn), pp. 287-310, Columbia University Press, New York. White, K.D. (1970). Roman Farming. Thames and Hudson, London. Whittington, G . (1978). A sub-peat dyke on Shurton Hill, Mainland, Shetland. Proc. Soc. Antiq. Scot. 109, 3CL35. Whittington, G . (1983). A palynological investigation of a second millennium BC bank-system in the Black Moss of Achnacree. J . Archaeol. Sci. 10, 283-291. Whittle, A. (1979). The Scord of Brouster. Current Archaeol. 6 , 167. Wigley, T.M.W., Jones, P.D. and Kelly, P.M. (1990). Empirical climatic studies. In: Scope 29: The Greenhoye Effect Climatic Change and Ecos.ystems (Ed. by W. Bolin, B.R. DOOs, J. Jaer and R.A. Warwick). pp. 1-32. John Wiley, Chichester. Wignarajah, K. and Greenway, H. (1976). Effect of anaerobiosis on activities of alcohol dehydrogenase and pyruvate decarboxylase in roots of Zen muys. Neri, Phytol. 77, 575-584. Williams. W.T. and Barber, D.A. (1961). The functional significance of aerenchyma in plants. Svmp. SOL..e.up. B i d . 15, 132-154. Wilson, D.F., Erecinska, M., Drown, C. and Silver, I.A. (1977). The effect of oxygen tension on cellular energetics. Amer. J . Phvsiol. 233, C I33-CI40. Wilson, D.F., Erecinska, M., Drown, C. and Silver, I.A. (1979). The oxygen dependence of cellular energy metabolism. Arch. Biochem. Biophys. 195.485493. Wollenweber, B. and Kinzel. H. (1988). Role of carboxylate in the nitrogen metabolism of plants from different natural habitats. Physiol. Plant. 72, 32 1-328.

OXYGEN A N D PLANT DISTRIBUTION

185

Young, S.F. and Hoffman, N.E. (1984). Ethylene biosynthesis and its regulation in higher plants. Ann. Rev. P l m t Plijxiol. 35, 155-1 89. Zagory, D. and Kadcr. A . M . (1988). Modified atmosphere packaging o f fresh produce. Food Teclinol. 42, 7&14. 1 6 7 7 .

This Page Intentionally Left Blank

Inherent Variation in Growth Rate Between Higher Plants: A Search for Physiological Causes and Ecological Consequences HANS LAMBERS and HENDRIK POORTER I. I1 . 111. IV . V. VI . VII .

VIII .

IX .

X.

Summary . . . . . . . . . . . . . . . . . . . . Introduction . . . . . . . . . . . . . . . . . . Growth Analyses . . . . . . . . . . . . . . . . Net Assimilation Rate and Leaf Area Ratio . . . . . . . Specific Leaf Area . . . . . . . . . . . . . . . . A . Components of SLA . . . . . . . . . . . . . . B. Plasticity in SLA . . . . . . . . . . . . . . . Biomass Allocation . . . . . . . . . . . . . . . . A . Biomass Allocation at an Optimum Nutrient Supply . . . B. Plasticity in Biomass Allocation . . . . . . . . . . Growth, Morphology and Nutrient Acquisition of Roots . . . A . Root Growth and Nutrient Acquisition at an Optimum Nutrient Supply . . . . . . . . . . . . . . . B. The Plasticity of Parameters Related to Root Growth and Nutrient Acquisition . . . . . . . . . . . . . . C. Other Root Characteristics Related to Nutrient Acquisition D . Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Chemical Composition A . Primary Compounds . . . . . . . . . . . . . . B. Secondary Compounds . . . . . . . . . . . . . C. Defence under Suboptimal Conditions . . . . . . . . D . Effects of Chemical Defence on Growth Potential . . . . . . . . . . E . The Construction Costs of Plant Material F . Conclusions . . . . . . . . . . . . . . . . . Photosynthesis . . . . . . . . . . . . . . . . . A . Species-specific Variation in the Rate of Photosynthesis . . B. Photosynthetic Nitrogen Use Efficiency . . . . . . . C . Is There a Compromise between Photosynthetic Nitrogen Use Efficiency and Water Use Efficiency? . . . . . . . . D . Photosynthesis under Suboptimal Conditions . . . . . E . Conclusions . . . . . . . . . . . . . . . . . Respiration . . . . . . . . . . . . . . . . . . A . Species-specific Variation in the Rate of Respiration . . .

ADVANCES IN ECOLOGICAL RESEARCH VOL . 23 ISBN &12413923-5

. .

188 189 191 194 196 196 198 199 199 200 200

.

200

. . . . . . .

202 205 205 206 206 201 210 211 212 214 214 214 215

. .

. . . . . .

.

. . . . . . .

. . .

219 221 222 222 222

Copyright 8 /YY2 Acudemic Press Limiled A// rinhts of reproduction in unyjorm reserved

187

I88

H. LAMBERS A N D H. POORTER

B.

Respiration a t Suboptimal Nitrogen Supply o r Q u a n t u m Flux Density . . . . . . . . . . . . . . . . . . C. Conclusions . . . . . . . . . . . . . . . . . XI. Exudation a n d Volatile Losses . . . . . . . . . . . . A . T h e Quantitative a n d Qualitative Importance of Exudation . B. T h e Quantitative a n d Qualitative Importance o f Volatile . . . . . . . . . . . . . . . . . . Losses C. Conclusions . . . . . . . . . . . . . . . . . XII. Other Differences bctween Fast- a n d Slow-growing Species . . A . Hormonal Aspects . . . . . . . . . . . . . . B. Miscellaneous Traits . . . . . . . . . . . . . . XI1 I . An Integration of Various Physiological a n d Morphological Aspects . . . . . . . . . . . . . . . . . . . . A . C a r b o n Budget . . . . . . . . . . . . . . . B . Interrelations . . . . . . . . . . . . . . . . XIV. Spccics-specific Pcrforniance under Suboptimal Conditions . . XV. T h e Ecological Consequences of Variation in Potential G r o w t h . . . . . . . . . . . Rate. , , . . . . . . A. What Ecological Advantage can be Conferred by a Plant's G row t h P o tent i al? . . . . . . . . . . . . . . €3. Selection of Traits Associated with a Low S L A . . . . . C . Sclcction for Other Traits Underlying R G R . . . . . . D. Consequences of a High G r o w t h Potential for Plant Performance in Specific Environments . . . . . . . . E. A Low G r o w t h Potcntial a n d Plant Performance in Adverse Environments. Other than Nutrient-poor Habitats . . . . F. C o n c l u s i o n s . . . . . . . . . . . . . . . . . XVI. Concluding Remarks a n d Perspectives . . . . . . . . . Acknowledgements . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . .

. .

225 226 221 227

. . . . .

228 229 229 229 23 I

.

.

23 1 23 1 233 234

.

235

. .

235 236 231

.

238

. . .

239 239 239 242 242

.

.

. .

.

.

.

I. SUMMARY When grown under optimum conditions, plant species from fertile, productive habitats tend to have inherently higher relative growth rates (RGR) than species from less favourable environments. Under these conditions, fastgrowing species produce relatively more leaf area and less root mass, which greatly contributes to their larger carbon gain per unit plant weight. They have a higher rate of photosynthesis per unit leaf dry weight and per unit leaf nitrogen, but not necessarily per unit leaf area, due to their higher leaf area per unit leaf weight. Fast-growing species also have higher respiration rates per unit organ weight, due to demands of a higher RGR and higher rate of nutrient uptake. However, expressed as a fraction of the total amount of carbon fixed per day, they use less in respiration. Fast-growing species have a greater capacity to acquire nutrients, which is likely to be a consequence, rather than the cause, of their higher RGR. There is no evidence that slow-growing species have a special ability to acquire

PHYSIOLOGY A N D ECOLOGY OF GROWTH RATE VARIATION

189

nutrients from dilute solutions, but they may have special mechanisms to release nutrients when these are sparingly soluble. We have analysed variation in morphological, physiological, chemical and allocation characteristics underlying variation in RGR, to arrive at an appraisal of its ecological significance. When grown under optimum conditions, fast-growing species contain higher concentrations of organic nitrogen and minerals. The lower specific leaf area (SLA) of slow-growing species is at least partly due to the relatively high concentration of cell-wall material and quantitative secondary compounds, which may protect against detrimental abiotic and biotic factors. As a consequence of a greater investment in protective compounds or structures, the rate of photosynthesis per unit leaf dry weight is less, but leaf longevity is increased. In short-term experiments with a limiting nutrient availability the RGR of all species is reduced, but potentially fast-growing species still grow faster than inherently slow-growing ones. Therefore, the absence of fast-growing species from infertile environments cannot be explained by their growth rate per sr. The higher leaf longevity diminishes nutrient losses and is a factor contributing to the success in nutrient-limited habitats. We postulate that natural selection for traits which are advantageous under nutrient-limited conditions has led to the low growth potential of species from infertile and some other unfavourable habitats. Other examples indicating that selection for traits which allow successful performance under adverse conditions inevitably leads to a lower potential RGR are included. We conclude that it is likely that there are trade-offs between growth potential and performance under adverse conditions, but that current ecophysiological information explaining variation in RGR is too limited to support this contention quantitatively.

11. INTRODUCTION Plants are distributed over a wide range of habitats varying from tundra to rain forests, from wetlands to deserts and from lowland to alpine regions. Coping with such contrasting, sometimes extreme, environments requires a certain degree of inherent specialization. One of the characteristics in which species of different habitats vary is their growth potential. Plants growing on nutrient-poor soils have a lower growth rate than those on fertile soils. But even when grown under optimum conditions, species which naturally occur on nutrient-poor soils still have a lower growth rate compared to plants characteristic of fertile sites (e.g. Bradshaw et al., 1964; Rorison, 1968; Christie and Moorby, 1975; Grime and Hunt, 1975; Poorter and Remkes, 1990; Fig. I). In addition, species or ecotypes which naturally occur in shaded environments (Pons, 1977; Corre, 1983a), dry habitats (Rozijn and

190

H. LAMBERS AND H. POORTER

8 L

aJ

1 3 6

E

3

z

‘ 4 Z 2

4’ Fig. 1. The relationship between the R G R of nine herbaceous C, species and the nitrogen index of the species’ habitat according to Ellenberg (1979) (high values of the N number correspond t o habitats of high nitrogen supply). R G R was determined at an optimum nutrient supply, moderate quantum flux density and fairly low vapour pressure deficit. The species described are, in order of increasing RGR: Corynephorus canescens, Festuca ovina, Pimpinella saxifraga. Phleum pratense, Anthriscus sylvestris, Poa annua, Scrophularia nodosa, Rumex crispus and Galinsoga parvijlora. (After Poorter and Remkes, 1990.)

van der Werf, 1986), alpine regions (Woodward, 1979; Atkin and Day, I990), arctic environments (Warren Wilson, 1966), saline conditions (Ball, 1988), sites which are rich in heavy metals (Wilson, 1988; Verkleij and Prast, 1989), or in other habitats adverse to plant growth all have a lower growth potential than comparable ones from favourable, fertile habitats. This close association between a species’ growth potential and the quality of its natural habitat raises two questions. First, how are the differences in growth rate between species brought about? And, second, what ecological advantage is conferred by a plant’s growth potential? These two questions are in fact closely related. A plant is a complex of organs with contrasting functions and subject to conflicting demands. A low or a high potential growth rate may either be the basis or a by-product of adaptation to a certain set of environmental conditions. Hence there may be trade-offs between adaptation to adverse conditions and growth potential. Therefore, the question on the ecological advantage of a potential growth rate cannot be answered until further ecophysiological information is available on the mechanisms explaining variation in growth potential.

PHYSIOLOGY A N D ECOLOGY OF GROWTH RATE VARIATION

191

Numerous plant characters contribute to a plant’s absolute growth rate in its natural habitat, e.g. seed size, germination time, or plant size after overwintering. In this chapter we restrict ourselves to an analysis of the different traits that contribute to a plant’s relative growth rate and discuss mechanisms which cause variation in any of these traits. We will treat the possible interdependence of various characteristics and try to quantify the importance of each of these in explaining interspecific variation in relative growth rate. Finally, we discuss the ecological implications of interspecific differences in the various traits and in the growth rate itself.

111. GROWTH ANALYSES Growth analysis is often used as a tool to obtain insight into the functioning of a plant. Different types of analyses exist, depending on what is considered a key factor for growth (cf. Lambers et al., 1989). In the most common approach, leaf area is assumed to be a key factor. The relative growth rate (RGR) (see Table 1 for a list of abbreviations), the rate of increase in plant weight per unit of plant weight already present, is then factorized into two components, the leaf area ratio and the net assimilation rate (Evans, 1972; see Table 2 for a range of published values). The leaf area ratio (LAR) is the amount of leaf area per unit total plant weight. The net assimilation rate (NAR) is defined as the rate of increase in plant weight per unit leaf area. Thus: R G R = LAR x NAR

(1)

LAR and NAR can both be divided into a further set of components. The LAR is the product of the specific leaf area (SLA), the amount of leaf area per unit leaf weight, and the leaf weight ratio (LWR), the fraction of the total plant biomass allocated to leaves. Thus: LAR = SLA x LWR

(2)

Although termed the morphological component, LAR is affected by biomass allocation, chemical composition and leaf anatomy, as will be discussed later. The NAR is the net result of dry weight gain and dry weight losses and is largely the balance of the rate of photosynthesis, expressed per unit leaf area (PS), and the rate of leaf respiration (LR), stem respiration (SR) and root respiration (RR), in this case also per unit leaf area. If these physiological processes are expressed in moles of carbon, the net balance of photosynthesis and respiration has to be divided by CC, the carbon concentration of the newly formed material, to obtain the increase in dry weight. The balance is

192

H. LAMBERS A N D H. POORTER

Table 1 Abbreviations used in this chapter and the preferred units in which they are expressed, listed in alphabetical order Abbreviation

Meaning

Preferred units

Carbon concentration Rate of exudation

mmol C g - ' mg rn-, (leaf area) day-' (mg g-l (plant wt) day-')

Leaf area ratio Rate of leaf respiration

m' kg-I pmol CO, m-, (leaf area) s - ' (nmol CO, g - ' (leaf wt) S K I ) g g-' g m-z day-' nmol (g root)-' s - ' mmol N g - ' pmol COz (mol leaf N)-' s - '

Leaf weight ratio Net assimilation rate Net nitrogen uptake rate Plant nitrogen concentration Photosynthetic nitrogen use efficiency Relative growth rate Root weight ratio Specific leaf area Rate of stem respiration Specific root length Stem weight ratio Rate of photosynthesis Rate of root respiration Rate of volatile losses

mg g - ' day-l g g-'

m' kg-' pmol COz m-2 (leaf area) s-l (nmol CO, g - ' (stem wt) S K I ) m g-' g g-' pmol CO, m-2 s-l l m o l COz m-' (leaf area) s - ' (nmol C g-' (root wt) s-I) mg rn-, (leaf area) day-') (mg g-l (plant wt) day-')

completed by subtracting losses due to volatilization (VOL) and exudation (EXU) per unit time, also expressed on a leaf area basis. Thus:

NAR = (PS, 7LR, - SR, - RR,) - EXU, - VOL,

cc

(3)

where subscript a indicates that the rates are expressed on a leaf area basis. However, leaf, stem and root respiration are not expected to be directly related to leaf area, but rather to the biomass of the different organs. Equation (3) is therefore extended to include the relations between organ biomass and leaf area:

cc

PS,- LR, .- 1 - SR,..-SWR SLA LAR

RR,.=) LAR

-EXU,-VOL,

PHYSIOLOGY A N D ECOLOGY OF GROWTH RATE VARIATION

193

Table 2 Interspecific variation in growth parameters. All values are expressed per unit dry weight. Species, grown in a controlled environment (glass house, growth room) are indicated with (C) in the specifications, references marked (F) are from plants grown in the field

Parameter Range

Mean value

RGR

74

31-151 66-3 14 120-299 1 13-365

I59 176 224

19-386

158

NAR

8-14

10

2-25

10

LAR

0.1-4.5

13-36

1.5 23

2-65

18

6-3 7 25-56

41

10-131

34

SLA

LWR

15

Specifications (C) 15 tree species (seedlings), Grime and Hunt (1975) (C) 93 perennials, Grime and Hunt (1975) (C) 22 annuals, Grime and Hunt (1975) (C) 24 herbaceous species, Poorter and Remkes ( 1990) (C,F) all species from Table 3 (C) 24 herbaceous species, Poorter and Remkes ( 1 990) (C,F) all species from Table 3 (F) 35 tropical trees, Ovington and Olson (1970) (C) 24 herbaceous species, Poorter and Remkes (1 990) (C,F) all species from Table 3 (F) 35 tropical trees, Ovington and Olson (1970) (C) 24 herbaceous species, Poorter and Remkes (1 990) (C,F) all species from Table 3

0.02434 0.43464

0.1 1 0.54

0.26-04 1

0.53

SWR

0.52486 0.07-0.27

0.70 0.17

(F) 35 tropical trees, Ovington and Olson (1970) (C) 24 herbaceous species, Poorter and Remkes ( 1 990)

RWR

0.08-0.36 0,22438

0.20 0.29

(F) 35 tropical trees, Ovington and Olson (1970) (C) 24 herbaceous species, Poorter and Remkes ( 1990)

(F) 35 tropical trees, Ovington and Olson (1970) (C) 24 herbaceous species, Poorter and Remkes ( 1990) (C,F) all species from Table 3

where subscript w indicates that rates are expressed per unit dry weight; SWR and RWR are the stem weight ratio and the root weight ratio, the fraction of biomass allocated to stem and roots, respectively. From eqn (4) it is clear that NAR is not purely a physiological component, as it is often termed, but rather a complex intermingling of a plant’s physiology, biomass allocation, chemical composition and leaf area formation. Although NAR is relatively easy to determine, it is not the most appropriate parameter to obtain a clear insight into the relation between physiology and growth. Hence, we rewrite eqns (1) and (4)into:

194 RGR =

H. LAMBERS A N D H. POORTER

(PS, x SLA X LWR - LR,,. X LWR - SR,,. x SWR - RR,,. x RWR)

cc -

EXU,,.- VOL,.

(5)

where EXU and VOL are expressed per unit total plant weight. When plants are in a steady state, i.e. when there is a fixed ratio between the increment of nutrients (e.g. nitrogen) and biomass, growth can also be considered in relation to the acquisition of such nutrients: RGR =

RWR x NIR PNC

where NIR is the net rate of nitrogen absorption, the rate of nitrogen taken up per unit root weight, and PNC the total plant nitrogen concentration. Factorizing RGR in its various components does not imply that these components are independent of each other (Hardwick, 1984). Often, an increase in one parameter affects another, either positively or negatively. In the next section we evaluate the importance of LAR and NAR in explaining variation in RGR.

IV. NET ASSIMILATION RATE AND LEAF AREA RATIO A wealth of information is available on the comparison of growth of two or three species, but few authors have investigated the relation between RGR and growth parameters for a range of species. Potter and Jones (1977) compared nine crop and weed species, Mooney et al. (1 978) investigated five Eucalyptus species and Poorter and Remkes (1 990) analysed the growth of 24 wild species common in western Europe. In all of these cases the LAR was the predominant factor explaining the inherent variation in RGR. Poorter (1989) arrived at the same conclusion after a review of 45 literature sources. An extended compilation is given in Table 3. On average, a 10% increase in RGR is associated with an 7.5% increase in LAR and a 2.4% increase in NAR. Thus, the amount of leaf area a plant realizes with a given total plant weight is an important factor determining the potential growth rate of a plant. Differences in the rate of dry weight gain per unit leaf area are of secondary importance in explaining interspecific variation in RGR. These generalizations only apply when the same types of plants are compared, e.g. C, herbs or trees. Tree species, when compared with herbs, have both a low LAR and a low NAR. C, species tend to have a higher RGR due to a higher NAR, when compared with C, species. The relatively low

PHYSIOLOGY AND ECOLOGY OF GROWTH RATE VARIATION

195

Table 3 Degree of association between the growth parameters of eqns ( I ) and (2). Means of a compilation of 78 literature references on comparative growth analyses of herbaceous C, species. For each reference a linear regression was carried out with the mean of variable A for each species as the independent variable, and the mean of variable B over the experimental period as the dependent variable. Then the change in the predicted value of variable B associated with a 10% change in variable A was calculated, starting from the mean values of A and B. A value close to 10 indicates that a 10% increase in variable A is associated with an almost equal increase in variable B, whereas a value close to zero indicates no association. Such an analysis is only fruitful provided the differences in variable A are large enough. Therefore, the degree of association was only calculated for pairs of variables in which the smallest and largest value of variable A differed at least 10% and 20 mg g - ' day-' (RGR). 10% and 2 m2 kg-' (LAR), 10% and 3 m2 kg-' (SLA) and 10% (LWR). The values for all references were then averaged. The literature references are those given in Table 2E of Poorter (1989), supplemented with the C , species of Table 2C, the sun species of Table 2D, and Tsunoda (l959), Enyi (l962), Tognoni et al. (1967), Khan and Tsunoda (1970a,b), Callaghan and Lewis (1971), Hughes and Cockshull (1971), Eze (1973), Ashenden et al. (1975), Smith and Walton (1973, Elias and Chadwick (l979), Grime (l979), Horsman et al. (1980), Cook and Evans (1983), Gray and Schlesinger (1983), Spitters and Kramer (1986), Campbell and Grime (1989), Gamier ef al. (1989), and Muller and Gamier (1990). In each of these analyses root weight determinations were carried out and all species or genotypes were grown under identical conditions. DOA

A

B

n

RGR RGR RGR RGR

NAR LAR SLA LWR

46 46 21 22

2.4 7.5 7.8 2.0

LAR LAR LAR

NAR SLA LWR

54

- 4.3

27 21

7.3 3.1

SLA SLA

NAR LWR

28 28

- 4.0

LWR

NAR

25

- 4.8

-

0.7

Sign.

* *** *** ns **t

*** * * ns

*

A , variable A ; B, variable B; n. number of references; DOA: averaged value of the degree of association between variable A and variable B; Sign.: t-test of the H, hypothesis DOA=O. ns, not significant; *, P<0.05; **, P
RGR of shade-adapted species, grown at a high quantum flux density, is caused by a low NAR, rather than a low LAR (Poorter, 1989). On average, a 10% increase in LAR is associated with a 4.3%decrease in NAR (Table 3). This is not as expected, because a high LAR decreases the respiratory burden per unit leaf area-see eqn (4). In a few cases, this discrepancy may have been caused by the lower photosynthesis and NAR,

196

H. LAMBERS A N D H. POORTER

resulting from self-shading in plants with a high LAR and a correspondingly large leaf area. However, in most cases it will be due to less well-defined interactions between the physiology, allocation, anatomy and chemical composition (Konings, 1989). Before discussing these interactions, we will first discuss inherent variation in each of these parameters (cf. eqn (5)).

V. SPECIFIC LEAF AREA As outlined in Section IV, variation in RGR is strongly correlated with that in LAR. Differences in LAR can be due to variation in LWR or in SLA. The specific leaf area is defined as the amount of leaf area per unit leaf weight. Its reciprocal, specific leaf weight or specific leaf mass, is also frequently used. Various aspects of inherent and environmentally induced variation in SLA have been reviewed by Dijkstra (1989). Large variations in SLA can be found between different types of plants and species from different habitats (Table 2) Evergreens mostly have a low SLA, whereas species with mesomorphic leaves show higher SLAs. Potter and Jones (1977), as well as Mooney et al. (1978) and Poorter and Remkes (1990; Fig. 2A), found a positive relationship between RGR and SLA. A compilation of the data available from the literature led us to the conclusion that there is a close association between the potential growth rate of a species and its SLA (Table 3). SLA can therefore be considered as the prime factor determining interspecific variation in RGR.

A. Components of SLA Which traits determine inherent variation in SLA? Starting from a simple leaf prototype, with chlorenchyma, vascular tissue and an epidermis, variation in SLA can be brought about by a change in several leaf characteristics. Firstly, a purely chemical #difference between leaves may occur, due to accumulation of, for example, starch or secondary compounds. Starch may account for up to 3&40% of total leaf dry weight (McDonald ef al., 1986; Rufty et al., 1988). Slower-growing species tend to invest relatively more in compounds which reduce the plant’s palatability, e.g. tannin and lignin (Coley, 1983, 1986; Coley et al., 1985). Accumulation of secondary compounds may be considerable. In some Australian dryland plants, resins make up 1&30% of the total leaf biomass (Dell and McComb, 1978). Lignin and other phenolic compounds form 1440% of the total leaf dry weight of Californian chaparral shrubs (Merino et al., 1984). Secondly, a lower SLA may be caused by anatomical differences, e.g. extra layers of palisade parenchyma in sun species as opposed to shade species

PHYSIOLOGY AND ECOLOGY OF GROWTH RATE VARIATION

197

50 -

40'

30'

If ,

20

1

B

P, 0.5 o.6#

v 7 h s

5

___c__

0.4-

Fig. 2. (A) The relationship between specific leaf area (SLA) and R G R and (B) the relationship between the leaf weight ratio (LWR) and R G R for the nine species described in Fig. 1. (After Poorter and Remkes, 1990.)

(Pons, 1977; Bjorkman, 1981); more support tissue, such as additional sclerenchyma, with an eight-fold difference between two Agrostis species (Pammenter ef al., 1986; see also Baruch et al., 1985); or smaller cell sizes. A reduction in cell size without altering the total leaf cell volume will drastically increase the cell-wall surface/cell volume ratio and thus decrease SLA. The

198

H. LAMBERS AND H. POORTER

size of the veinal transport system may also affect SLA. In Triticum aestivum, veins contain 10% of the leaf weight (Rawson et al., 1987). Givnish (1986) found the large veins of Podophyllum peltatum to comprise 6 2 0 % of the total leaf biomass, depending on leaf size. Thirdly, variation in SLA can be caused by a difference in investment in leaf hairs, thorns, etc. In two Espelefia species, leaf hairs comprise 4 and 20%, respectively, of the total leaf biomass (Baruch and Smith, 1979). Exceptionally, as in Encelia farinosa, leaf pubescence may account for up to 60% of total leaf dry weight (Ehleringer and Cook, 1984). However, in most plants neither pubescence nor thorns account for large differences in SLA. Thus, in general, inherent variation in SLA is not merely caused by a difference in the amount of leaf cells per unit area, but also by variation in leaf anatomy, morphology or chemical composition. These differences in anatomy or morphology will also affect the chemical composition, as each of the above-mentioned anatomical and morphological structures has a distinctive chemical composition (cf. Kimmerer and Potter, 1987). The consequences of such differences in chemical composition for plant growth are discussed in Section VIII.

B. Plasticity in SLA Plants grown under a low quantum flux density generally have a higher SLA and thinner leaves (Young and Smith, 1980), which is associated with fewer mesophyll cell layers (Pons, 1977; Bjorkman, 1981) and less non-structural carbohydrate (Waring et al., 1985). To a small extent the higher SLA is also associated with lower concentrations of phenolic compounds, including lignin (Waring et a/., 1985; Mole and Waterman, 1988). There is hardly any evidence of differences in plasticity with respect to quantum flux density between species with different growth rates (Pons, 1977; Corre, 1983a,b; Grime et al., 1989). However, the fast-growing Holcus lanatus increases SLA more at a low quantum flux density than the slow-growing Deschampsia flexuosa (Poorter, 1991). A similar difference was found in a comparison of Veronica montana, a (presumably slow-growing) woodland perennial, and V . persica, a (presumably fast-growing) annual weed. When grown under a leaf canopy as opposed to unshaded conditions, the SLA of V . persica changed considerably more than that of V. montana (Fitter and Ashmore, 1974). Plants grown at low nutrient availability either show no change (Corre, 1983~;Sage and Pearcy, 1987; van der Werf et al., 1992a) or a decrease in SLA (Sage and Pearcy, 1987; van der Werf et a/., 1992a). A decrease is at least partly due to accumulation of non-structural carbohydrates (Lambers et al., 1981a; Waring et al., 1985) or secondary compounds like lignin or other phenolics (Gershenzon, 1984; Waring et al., 1985). No differences in plasticity have been found between fast- and slow-growing species (Corre,

PHYSIOLOGY A N D ECOLOGY OF GROWTH RATE VARIATION

199

1983~;van der Werf et al., 1992a). However, data on the comparison of species grown under a range of conditions are scarce and differences in potential RGR between species sometimes small. A better-founded evaluation therefore awaits further experiments.

VI. BIOMASS ALLOCATION Biomass allocation can be defined in terms of leaf, stem and root weight ratio, the fraction of total plant biomass allocated to leaves, stems and roots, respectively. A more frequently used parameter, the shoot : root ratio or its inverse, does not acknowledge the distinct functions of leaves and stems and is avoided here. Various aspects of inherent differences in biomass partitioning between leaves and roots have been discussed by Konings (1989). A low availability of nitrogen, phosphorus and water enhances allocation to roots, whereas a low quantum flux density promotes allocation to the leaves. The mechanism behind this “functional equilibrium” (Brouwer, 1963, 1983) is still poorly understood (Lambers, 1983). Genotypic differences in biomass partitioning between leaves, stem and roots have been correlated with differences in the level of gibberellins (Zea mays, Rood et al., 1990a; Brassica r a p , Rood et a/., 1990b; Lycopersicon esculentum, Koornneef et a/., 1990; cf. Section XIIA) and abscisic acid (Zea mays, Saab et al., 1990; Lycopersicon esculentum, O.W. Nagel and H. Konings, personal communication) and with the plant’s sensitivity to endogenous gibberellin (Lycopersicon esculentum, Jupe et a/., 1988).

A. Biomass Allocation at an Optimum Nutrient Supply Some authors have found a negative correlation between LWR and RGR (Hunt et al., 1987, plants grown at low quantum flux density; Shipley and Peters, 1990, at high quantum flux density), others a positive one (Ingestad, 1981; Poorter and Remkes, 1990, at intermediate quantum flux density, Fig. 2B). Differences in growth conditions may have a decisive impact on the final result (Poorter and Lambers, 1991). Irrespective of the way the plants are grown, current information shows that LWR is less important than SLA in explaining inherent variation in RGR (Table 3). Interestingly, monocotyledonous herbaceous species invest relatively more biomass in roots and less in leaves, compared to dicotyledonous ones with the same inherent RGR (Garnier, 1991). Moreover, the general trend of increasing RGR with increasing investment in leaf weight (Poorter and Remkes, 1990) only holds for dicotyledonous species and not for grasses (Garnier, 1991).

200

H. LAMBERS AND H. POORTER

B. Plasticity in Biomass Allocation In general, plants grown at a low quantum flux density show a shift in the allocation of biomass from roots and stem to leaves. This shift is generally more pronounced in faster-growing species (Pons, 1977; Werner et a/., 1982; Grime et a/., 1989; but see Corrt, 1983b). A decrease in nutrient availability often decreases LWR and increases RWR, particularly in fast-growing species (Christie and Moorby, 1975; Tilman and Cowan, 1989; Shipley and Peters, 1990). However, as noted previously, an interaction between nutrient supply and plant size may seriously affect the observed relationship between allocation and RGR, especially at a low nutrient supply (Ingestad, 1962, cited in Corrt, 1983~). Taking into account only those references in which this artefact was most certainly avoided, a generally higher plasticity in allocation is still observed for fast-growing species (Christie and Moorby, 1975; Robinson and Rorison, 1988; Campbell and Grime, 1989; van der Werf e f al., 1992a; but see Bradshaw et a/., 1964; Crick and Grime, 1987).

VII. GROWTH, MORPHOLOGY AND NUTRIENT ACQUISITION OF ROOTS The simple growth equation: R G R = NAR x LAR, suggests that any investment in biomass other than leaf area reduces the plant’s RGR. Such an approach tends to consider the roots merely as a carbohydrate-consuming organ and does not give credit to their role in the acquisition of nutrients and water or their function in transport, storage and anchorage. In this section we will concentrate on the root’s role in the acquisition of ions. Growth can then best be approached from an alternative point of view, where R G R is defined in terms of the root weight ratio (RWR), the net rate of nitrogen acquisition (NIR) and the plant’s nitrogen concentration (PNC)-eqn (6). Equation (6) suggests that a high RGR can be achieved by a large investment in root biomass, by a high ‘rate of nitrogen uptake per unit root weight (specific ion uptake rate), or by a combination of these. However, a large investment in root weight may in fact reduce RGR because investment in leaves is reduced.

A. Root Growth and Nutrient Acquisition at an Optimum Nutrient Supply Root systems of fast- and slow-growing species differ in their architecture. Slow-growing Festuca species from nutrient-poor habitats tend to have “herringbone” morphologies, i.e. next to one main axis, the root systems have only primary laterals (Fitter et al., 1988). Results of a simulation model

PHYSIOLOGY AND ECOLOGY OF GROWTH RATE VARIATION

20 1

indicate that a herringbone morphology allows the most effective exploration and exploitation of mobile resources (Fitter, 1987). On the other hand, such a morphology is less efficient for long-distance transport, because the total transport path is longer, and hence requires a greater investment in root biomass. Fast-growing grassland species have a more random or nearly dichotomous root morphology. Slow-growing grass species from nutrient-poor habitats generally have a higher specific root length (SRL, the root length per unit root weight) and relatively more fine roots (Berendse and Elberse, 1989; Boot, 1989; Boot and Mensink, 1990). However, SRL tends to vary with age in an unpredictable manner (Fitter, 1985) and some studies show a higher SRL for fast-growing grasses (Robinson and Rorison, 1985). No correlation between SRL and RGR was found in a comparison of 24 monocotyledons and dicotyledons species, grown at an optimum nutrient supply (Poorter and Remkes, 1990). A higher SRL is likely to contribute to the acquisition of ions which diffuse slowly in the soil (Clarkson, 1985), but also of more mobile ones when plants have to compete for these. Similarly, root hairs contribute to the acquisition of relatively immobile nutrients. There is a large variation between plant species with respect to root hair density and root hair length, but variation in these root characteristics does not seem to be related to the maximum growth rate of the species or its performance in nutrient-poor environments (Robinson and Rorison, 1987; Boot, 1989). The “proteoid” roots which are found in some dicotyledonous species are discussed in Section VIIC. At an optimum nutrient supply, inherently fast-growing species have a somewhat lower RWR than slow-growing species from the same life form (Section VI). Since fast-growing species also have a higher nitrate and organic nitrogen concentration (Section VIIIA), it follows that the fastgrowing species must have higher nitrogen absorption rates (cf. eqn (6); Chapin, 1980). Nassery (1970) compared the growth and phosphate uptake of the fastgrowing Urtica dioica with that of the slow-growing DeschampsiaJtexuosa. At an optimum nutrient supply, the fast-growing species had the highest nutrient uptake rate per unit root weight. This conclusion was corroborated by Christie and Moorby (1975), Chapin and Bieleski (1982), Chapin et al. ( 1986a), Garnier et al. (1 989) and Poorter et al. ( 1 99 1 ; Fig. 3A). This is not to say that fast-growing species grow faster because their rate of nitrate or phosphate uptake is higher. Rather, the rapid uptake may be a result of their higher growth rate. At an optimum supply, the rate of ion uptake is, at least partly, determined by “demand”, which results in a strong negative feedback when the growth rate is low (Clarkson, 1986; Rodgers and Barneix, 1988). An increased demand diminishes the negative feedback, thus enhancing net uptake (Jackson et al., 1976; Doddema and Otten, 1979; Lambers et al., 1982). When pregrown at a low phosphate concentration, the rate of

202

H. LAMBERS A N D H. POORTER

phosphate uptake at a saturating concentration was 35% higher for the fastgrowing Urtica dioica than for the slow-growing Deschampsia jlexuosa. Pregrown at an optimum phosphate supply, this difference is about 300% (Nassery, 1970). Similar results have been obtained for other species, both with phosphate (Harrison and Helliwell, 1979; Clarkson and Scattergood, 1982) and a range of other ions (Lee, 1982; Glass, 1983). It is therefore very likely that a low ability to incorporate large amounts of absorbed nutrients into organic matter, rather than their low absorption capacity, controls the growth rate of plants from nutrient-poor sites at optimum nutrient supply. Is there evidence for an inherently higher affinity (lower Kn,)of the uptake system of slow-growing species from nutrient-poor sites? The phosphate uptake system of slow-growing Carex species has a high affinity for phosphate, in comparison with that of crop species (Atwell et al., 1980). Data of Muller and Garnier (1990) on growth at low nitrate concentrations, contrary to those of Freijsen and Otten (1984), suggest a lower Knl for nitrate of the uptake system in slower-growing species. However, in neither of these studies was the KmIdetermined. So far the experimental data do not support the hypothesis that the nitrate uptake system of slow-growing species from nutrient-poor habitats has a higher affinity for nitrate (van de Dijk, 1980; Bloom, 1985; Oscarson et al., 1989). However, not many systematic comparisons of kinetic parameters of nitrate uptake in contrasting species are available from the literature and experiments have been carried out under different experimental conditions, so that a final conclusion cannot yet be reached.

B. The Plasticity of Parameters Related to Root Growth and Nutrient Acquisition As discussed in Section VI, fast-growing species tend to be more plastic with respect to biomass allocation. At a growth-limiting nitrogen supply, inherently fast-growing species have a higher RWR than slow-growing ones of the same life-form. Both a low phosphate supply (Powell, 1974; Christie and Moorby, 1975) and a low nitrogen supply (Robinson and Rorison, 1985; Boot, 1989) mostly increase SRL. Slow-growing and fast-growing species respond in a similar manner (Fitter, 1985). Root hairs are of vital importance for the acquisition of ions which diffuse slowly in soil, e.g. phosphate (Clarkson, 1985), and stimulation of root hair formation at a low nutrient supply has been reported frequently (Fohse and Jungk, 1983). The inherently slow-growing species Deschampsiajlexuosa has a remarkable plasticity with respect to root hair formation, in comparison

A

8

h r

6-

3

TI

r

-

i? c n: z

4-

W

’1

,”

1

A

r

B

v)

-

z

150.

...

4

0

c c

..

/ /

/

W

.o

.

0

0

100-

/

.0 0

0

E

-I

p ,

1

Fig. 3. (A) The net influx of nitrate into roots (NIR) and (B) the rate of root respiration of fast-growing and slow-growing species, described in Fig. 1. The dashed line in (B) gives the calculated rate of root respiration assuming specific costs for maintenance, growth and ion uptake to be the same as those for a slow-growing Curex species. (After Poorter et ul., 1990, 199 I .)

204

H. LAMBERS AND H. POORTER

with fast-growing grass species (Robinson and Rorison, 1987; Boot and Mensink, 1990). Both fast-growing and slow-growing grasses respond to nutrient shortage with increased root hair length, but the tendency appears to be greatest in slow-growing ones (Boot and Mensink, 1990; Liljeroth et al., 1990). The greater plasticity in density and length of root hairs in response to nutrient supply in inherently slow-growing species is likely to contribute to their successful performance in phosphate-poor environments. Roots of crop species (Drew et al., 1973; de Jager, 1982), fast-growing herbaceous wild plants (de Jager and Posno, 1979), trees (Philipson and Coutts, 1977) and desert perennials (von Willert et al., 1992) have an amazing capacity to proliferate growth in nutrient-rich or moist patches in the root environment. Crick and Grime ( 1 987) compared morphological plasticity in the slow-growing Scirpus sylvaticus and the faster-growing Agrostis stolonifera. A . stolonifera had a greater capacity to proliferate its fine roots in nutrient-rich patches. Therefore, it can dynamically exploit a fertile root environment and successfully compete with neighbouring plants, whereas the relatively large, but unresponsive, root system of the slow-growing S. sylvaticus is more advantageous when nutrients are strongly limiting and become available in temporally unpredictable pulses. A similar conclusion is drawn from data on two cold-desert bunchgrass species (Jackson and Caldwell, 1989) differing in RGR (Eissenstat and Caldwell, 1987). However, using a somewhat different technique from the one used by Crick and Grime (1987), Grime et al. (1991) concluded that the capacity to proliferate roots locally may be similar in slow- and fast-growing species. Our current information is insufficient to draw final conclusions on variation in the root’s capacity to locally proliferate fine roots between fast-growing and slowgrowing species. Plants d o not only respond to nutrient-rich patches with an increased root production, but also with an increased capacity for nutrient uptake (Drew and Saker, 1978; Lambers et al., 1982). Jackson et al. (1990) compared the effect of a local increase in Phosphate supply on two cold-desert Agvopyron species, referred to above. Both species responded with an 80% increase in phosphate absorption capacity, and no differences between the two species were found with respect to the physiological plasticity of phosphate uptake. Apart from spatial variation in nutrient availability, there may be variation in time. Campbell and Grime (1989) found that a potentially fastgrowing species grows faster than an inherently slow-growing species when nutrient pulses are long, whereas the opposite is true under a regime of short pulses. In this case the frequency of pulses was constant, but the duration of the pulse varied, and thus the total amount of nutrients supplied. In an experiment with two Plantago major subspecies, where the frequency of pulses was increased, but total nutrient supply was constant, the faster-

PHYSIOLOGY A N D ECOLOGY OF GROWTH RATE VARIATION

205

growing subspecies achieved an increasingly higher RGR than the slowgrowing one (Poorter and Lambers, 1986). It is concluded that fast-growing species are characterized by a high degree of plasticity in root morphology, such as the adjustment of their RWR and perhaps also the local proliferation of roots in nutrient-rich patches. Such a high degree of morphological plasticity is likely to be an integral part of the mechanism of resource acquisition in productive environments (Crick and Grime, 1987). Theoretical models predict that in infertile soils such a strategy leads to net losses of mineral nutrients from the plant (Sibly and Grime, 1986). Under poor conditions, alternative mechanisms requiring a lower investment of mineral nutrients but leading to greater losses of carbon from the roots, might have greater survival value. Here rapid proliferation of fine roots might incur net nutrient losses, as the nutrient costs of investing new roots in a generally poor soil might be higher than its returns. There is as yet no evidence for a difference in plasticity of nutrient acquisition between fastgrowing and slow-growing species.

C. Other Root Characteristics Related to Nutrient Acquisition Like leaves, roots are subject to a continuous turnover. Aerts (1989) reports a turnover rate of 0 4 ~ 1 . 7 groots (g roots)-’ year-’. A high turnover rate of roots, like that of leaves, incurs a net loss of nutrients, as discussed in Section XV. We are not aware of systematic comparisons between fast-growing and slow-growing species with respect to root turnover. Specialized root structures termed “proteoid” roots, are formed on members of the Proteaceae (Lamont, 1982), many of which are slow-growing species from very nutrient-poor soils (Barrow, 1977). Proteoid roots consist of sections of dense “bottle-brush-like’’ clusters of short (5-10 mm) rootlets, covered with a dense mat of root hairs. Such structures induce the release of various nutrients from sparingly soluble sources (Section XIA). Proteoid roots, or functionally similar structures, are also known from fast-growing crop species (Gardner et al., 1982a; Hoffland et al., 1989a). Another special structure, which allows plants to grow in phosphate-poor environments, is the mycorrhiza, an association between a fungus and roots. Both fast- and slow-growing species have the capacity to form such an association, predominantly under phosphate-poor conditions. Some species are inherently non-mycorrhizal, but this characteristic is also not associated with the inherent growth rate of the species (cf. Tester et al., 1987).

D. Conclusions Variation in RGR between species is certainly associated with variation in root attributes. Slow-growing species tend to have a “herring bone” architecture, rather than the more random structure of fast-growing species, and also

206

H. LAMBERS AND H. POORTER

a higher specific root length. Fast-growing species tend to invest relatively less biomass in roots when grown at an optimum supply of nutrients, but have a greater capacity to adjust their investment in root biomass as well as associated structures than slow-growing species. The nutrient uptake rate per unit root dry weight of fast-growing species is higher than that of slowgrowing species, but this may well be an effect of rapid growth, rather than its cause.

VIII. CHEMICAL COMPOSITION Plant dry matter is composed of a number of major compounds, which can be grouped into the following seven categories: lipids, lignin, organic Ncompounds, (hemi)cellulose, non-structural sugars, organic acids and minerals. Apart from “primary” compounds, there is a wealth of “secondary” compounds, defined by the absence of a clearly defined role in the metabolic processes of the plant (Baas, 1989; Waterman and McKey, 1989). Lignin is often included in the category of “secondary” compounds and this will also be done here.

A. Primary Compounds When grown at an optimum nutrient supply, inherently fast-growing species have a higher total and organic nitrogen concentration per unit plant dry weight than slow-growing ones (Poorter et al., 1990; Fig. 4A). Their higher organic nitrogen concentration is due partly to a greater biomass investment in leaves, which tend to have a higher nitrogen concentration than other vegetative plant organs, and partly to a higher nitrogen concentration in all vegetative organs per se (Poorter et al., 1990). Soluble protein constitutes a larger fraction of the total leaf nitrogen concentration in a fast-growing Plantago major subspecies than in a slower-growing one (Dijkstra and Lambers, 1989a). Fast-growing (sub)species generally contain more minerals and organic acids per unit dry weight than slow-growing ones, when grown at an optimum nutrient supply (Chapin and Bieleski, 1982; Dijkstra and Lambers, 1989b; Poorter and Bergkotte, 1992). Some comparative studies show that phosphorus, particularly inorganic phosphate, accumulates to a greater extent in fast-growing (sub)species (Nassery, 1970; Chapin and Bieleski, 1982; Dijkstra and Lambers, 1989b). Others show the opposite (Christie and Moorby, 1975; Chapin et al., 1982). Accumulation of nitrate appears to be characteristic of fast-growing, “nitrophilous” species (Dittrich, 1931 ; Smirnoff and Stewart, 1985; Poorter and Bergkotte, 1992). Such accumulation is most pronounced at a high nitrate supply (Stulen et al., 1981) and at a

PHYSIOLOGY AND ECOLOGY OF GROWTH RATE VARIATION

207

relatively low quantum flux density, when nitrate replaces soluble carbohydrates and carboxylates as an osmotic solute (Stienstra, 1986; Veen and Kleinendorst, 1986; Blom-Zandstra and Lampe, 1985). It is sometimes claimed (cf. Chapin, 1988) that slow-growing species show more “luxury consumption” than fast-growing ones. Apart from the fact that this is not compatible with most of the comparative data cited in this section, it still remains to be demonstrated whether luxury consumption exists at all. We propose to define “luxury consumption”, as the absorption beyond a rate which leads to more growth, rather than as “vacuolar storage during the period of active growth” (Chapin, 1988). Accumulation of nitrate does not imply “luxury consumption”, since nitrate may replace organic solutes, which leads to more carbon being available for metabolic processes. For example, faster-growing genotypes of Lactuca satiua accumulate nitrate, rather than organic solutes (Blom-Zandstra et af., 1988). For phosphate, the situation may be different, but data in the literature are conflicting (e.g. Chapin and Bieleski, 1982; Chapin ef a f . , 1982). Slow-growing (sub)species contain more cell-wall components (lignin and (hemi)cellulose) than fast-growing ones, when grown at optimum nutrient supply (Dijkstra and Lambers, 1989a; Poorter and Bergkotte, 1992). Using pyrolysis-mass-spectrometry, Niemann et a/. (1 992) showed that fast-growing species contain relatively more compounds associated with the cytoplasm and slow-growing ones more of those associated with cell walls. At a relatively fixed cell size, accumulation of cell-wall components and solutes will also alter the water content (g H 2 0 gg’ DW) of the different organs. Indeed, both leaves, stems and roots of slow-growing (sub)species contain rather low amounts of water per unit dry weight (Dijkstra and Lambers, 1989b; Poorter and Bergkotte, 1992; Fig. 4C). In the Lactuca satiua genotypes mentioned above, replacement of nitrate for organic acids partly explains genotypic differences in leaf dry matter content (Reinink et al., 1987; Blom-Zandstra et a f . , 1988).

B. Secondary Compounds Plants contain a suite of “secondary” compounds, which serve a range of distinct ecological functions, including allelopathy, the deterrence of herbivores, attraction of pollinators and attraction of organisms predating on herbivores (Harborne, 1982; Chou and Kuo, 1986; Baas, 1989; Dicke and Sabelis, 1989). Much attention has been paid to the role of secondary plant compounds in reducing herbivory. In this context they are often classified as “quantitative” vs. “qualitative” defence compounds (Harborne, 1982; Waterman and McKey, 1989). Typically, quantitative secondary compounds are composed of C, H and 0 only, have low turnover rates and act as digestibility reducers when present in large amounts. Qualitative secondary

208

H. LAMBERS AND H. POORTER

9

CD

-

z

E

v

C

0 .-

P

c

6

E

C Q)

B

.r C

.Y C

m P 0

200

300

RGR (mg g-’day-’) Fig. 4. Aspects of the chemical composition of fast-growing and slow-growing species, described in Fig. 1. (A) Organic nitrogen concentration of plant dry matter. (B) Carbon concentration of plant dry matter. (C) Dry matter content (dry weight x 100/fresh weight) of plant biomass. (D) Construction costs of plant biomass, calculated as outlined in Penning de Vries et al. (1974) and Lambers and Rychter (1989). (After Poorter et al., 1990; Poorter and Bergkotte, 1992.)

209

PHYSIOLOGY AND ECOLOGY OF GROWTH RATE VARIATION

C

D

200

300

RGR (mg g-'day-') Fig. 4-continued

compounds tend to be specific toxins which occur in low concentrations and may be subject to rapid turnover. Some of these toxins contain nitrogen (cyanogenic glycosides, analogues of amino acids, alkaloids), but many do not (cardenolides, glucosinolates, saponins). Slow-growing species or genotypes accumulate more quantitative secondary plant compounds than fast-growing ones (Coley, 1986; Waterman and

210

H. LAMBERS AND H. POORTER

McKey, 1989). The concentration of lignin (Coley, 1983), condensed tannin (Waterman and McKey, 1989), volatile terpenoids (Morrow and Fox, 1980) or diplacol (Merino et al., 1984) may comprise 1&30% of a plant’s leaf dry weight. There is no clear-cut correlation between a plant’s inherent growth rate and the concentration of qualitative secondary compounds in its tissues. Perhaps the only generalization that can be made is that fast-growing species, if they have any antiherbivore chemicals at all, accumulate only qualitative compounds. There is a vast array of qualitative secondary compounds, which generally account for less than 1 % of the dry weight (Baas, 1989). However, their distribution has not been systematically investigated for a range of species grown under standard conditions. Slow-growing species may also accumulate qualitative compounds, often in a manner which complements accumulation of quantitative secondary compounds. A special case is that of young leaves. Most herbivores preferentially feed on leaves with a high protein and water content, a low leaf toughness and a low concentration of antiherbivore compounds (Coley, 1983; Kimmerer and Potter, 1987; Waterman and McKey, 1989). In young leaves of slow-growing plants, leaf toughness and the concentration of digestibility-reducing compounds is often low, whereas the nutrient and water content is relatively high compared to older leaves. Hence, young leaves appear to be attractive for herbivores, but they also accumulate toxic compounds, e.g. anthocyanins in tropical trees (Coley and Aide, 1989) or saponins in Ilex opaca (Potter and Kimmerer, 1989). Also mature leaves of Ilex opaca accumulate saponins, namely in their mesophyll cells, which are not protected by digestibility-reducing compounds like lignin, crystals and tannin. These quantitative secondary compounds accumulate instead in other cells of the same leaves (Kimmerer and Potter, 1987). One obvious risk of the accumulation of specific toxins is that herbivorous organisms coevolve and become insensitive to the defence (Harborne, 1982). In the case of Ilex opaca, mentioned above, specialist leaf miners consume only the mesophyll tissue,,apparently able to neutralize the toxic saponins in these cells. The chances for tolerance to evolve against quantitative antiherbivore compounds are considerably smaller. However, some degree of tolerance against palatability-reducing compounds has been demonstrated (Bernays et al., 1989) and some herbivores even incorporate these quantitative compounds, thus possibly gaining protection (Taper and Case, 1987).

C. Defence under Suboptimal Conditions Environmental conditions, such as nutrient supply and water stress, may restrict plant growth more than expected from their effect on photosynthesis. Under such conditions non-structural carbohydrates accumulate, leading to

PHYSIOLOGY AND ECOLOGY OF GROWTH RATE VARIATION

21 1

an excess of carbon in the plant. When plants have such an excess of carbon, accumulation of carbon-based defences is expected. Similarly, at a high availability of nutrients and a relatively low quantum flux density, accumulation of nitrogen-based secondary compounds is predicted. Confirmation of this carbon/nutrient balance theory (Bryant et al., 1983) has been found in studies where plants, grown at a low nutrient availability, show an increase in the concentration of condensed tannins, total phenols and/or phenol glycosides (Waring et al., 1985; Bryant et al., 1987; Nicolai, 1988; Margna et al., 1989; but see Denslow et al., 1987). Similarly, Johnson et al. (1 987) observed a positive correlation between N-supply and the concentration of alkaloids.

D. Effects of Chemical Defence on Growth Potential What are the costs associated with the accumulation of antiherbivore compounds? The amount of glucose needed to produce 1 g of a toxic compound may be high (cf. Baas, 1989). However, as the concentration of toxins in plant tissues is rather low, their accumulation hardly affects the cost of synthesizing plant biomass. The glucose needed to produce 1 g of digestibility-reducing compounds is generally lower than that of qualitative secondary compounds (Baas, 1989; Lambers and Rychter, 1989). In the case of tannin, approximately the same amount of glucose is required as to construct cellulose or starch. However, because these compounds may accumulate in large quantities in plant tissue, they may comprise a large part of the plant’s carbon resources. This carbon cannot be used for the construction of the photosynthetic apparatus, so that the photosynthetic return per unit weight of a well-protected leaf is less than that of a leaf which allocates less carbon to quantitative defence compounds. The costs of the accumulation of secondary compounds exceed the specific costs of synthesis. The enzyme apparatus to produce these compounds must also be maintained and there is a turnover of different compounds (e.g. monoterpenes: Burbott and Loomis, 1969; alkaloids: Waller and Nowacki, 1978; cyanogenic compounds: Adewusi, 1990). Moreover, toxic compounds must be stored in special compartments or structures in which they cannot harm the plant’s metabolism. Examples include oil glands containing essential oils in Eucalyptus species (Welch, 1920), leaf hairs containing carvone in Mentha spicata (Gershenzon et al., 1989) and the separation of cyanogenic glycosides from the enzymes releasing HCN in Phaseolus lunatus (Frehner and Conn, 1987). An alternative approach to calculate the costs of accumulating secondary compounds relates their concentration to the growth of a plant. Coley (1986) found a negative correlation between the rate of leaf production and the leaf tannin concentration of Cecropia peltata (Fig. 5). However, the negative correlation may be a reflection of correlating, but in themselves unrelated,

H. LAMBERS AND H. POORTER

Tannin (mg g-') Fig. 5. Leaf production and tannin concentration in Cecropia peltata. (After Coley, 1986.)

plant characteristics. Accumulation of secondary compounds may even be the phenotypic result of slow growth, as discussed above.

E. The Construction Costs of Plant Material At an optimum nutrient supply, fast-growing species have a lower carbon concentration in the various organs than slow-growing ones (Poorter and Bergkotte, 1992; Fig. 4B). This is a consequence of a difference in chemical composition. The different components of the plant's biomass vary in carbon concentration ranging from high in lipids to zero for minerals (cf. Table 4). Therefore, a plant with a high proportion of biomass invested in compounds with a high proportion of carbon, like lipids and protein, has to fix more carbon to construct one unit of plant weight than a plant that consists mainly of (hemi)cellulose, organic acids and minerals (cf. eqn ( 5 ) , Section 111). The difference in carbon concentration of fast- and slow-growing species is about lo%, whereas that in RGR is over 300% (Fig. 4B). Thus, variation in carbon concentration has only a rather small effect on variation in RGR. Construction of the different compounds not only requires glucose for Cskeletons, but also for the generation of ATP and NAD(P)H (Section X). Generally, compounds with a high carbon concentration are more reduced and require more glucose for their synthesis (Penning de Vries et al., 1974; Table 4). Hence, glucose costs for the synthesis of biomass can be derived

213

PHYSIOLOGY AND ECOLOGY OF GROWTH RATE VARIATION

Table 4 The carbon concentration (mmolig) of a number of primary and secondary compounds present in plant biomass, as well as the requirement of glucose (mmol g-') and oxygen (mmol g-I), and the carbon dioxide release (mmol g-I) during synthesis of these compounds from glucose and nitrate. The values for oxygen requirement and carbon dioxide release are used to calculate the expected respiratory quotient (RQ) during synthesis of these compounds. (The principles of the calculations are outlined in Penning de Vries et al. (1983) and Lambers and Rychter (1989), but different values have been used, where appropriate.) Carbon Glucose concentration costs

0,-

coy

Component

requirement

release

Volatile terpenoids Limonene

61.3

18.8

29.6

59.2

2.0

53.8 36.8

16.8 13.8

11.0

13.5

36.5 37.9

3.3 2.8

46.3 32.8

11.8

8.6

5.9 12.1

13.1 22.7

2.2 1.9

Structural carbohydrates 32.0 Hemicellulose Cellulose 30.8

7.1 6.5

3.6 2.1

3.6 2.1

1.0 1 .o

Non-structural carbohydrates Starch 30.8 Sucrose 29.3

6.5 6.1

2.1 1.5

2.1 1.5

1 .o

Organic acids Citric acid Malic acid

4.3 3.7

5.2 0.0

0.0 7.5

<1
Lipids Protein Phenolics Lignin Ellagitannin

Minerals

26.1 24.8

0.0

~

-

~

RQ

1 .o

-

from the carbon concentration of biomass, provided a correction is made for the mineral concentration (Vertregt and Penning de Vries, 1987). Alternatively, these costs can be calculated from the concentration of the various primary and secondary compounds of plant biomass (Penning de Vries et al., 1974; Lambers and Rychter, 1989). Although variation in glucose requirement exists, there is not much difference between slow-growing evergreens, slow-growing deciduous plants and faster-growing species (Chapin, 1989; Lambers and Rychter, 1989). In a comparison of a range of herbaceous species, the costs for construction of plant biomass was very similar for fastgrowing and slow-growing species (Poorter and Bergkotte, 1992; Fig. 4D). There are two reasons for this relative constancy of construction costs. Firstly, the production of protein, which is present in larger amounts in fast-growing species, requires a similar amount of glucose to that of quantitative secondary

214

H. LAMBERS A N D H. POORTER

compounds, characteristic of slow-growing species (Table 4; cf. Section VIIIA, B). And, secondly, the higher concentration of costly proteins coincides with an increased concentration of cheap compounds, such as organic acids and minerals (cf. Section VIIIA).

F. Conclusions Fast-growing species are characterized by a high organic nitrogen and mineral concentration, whereas slow-growing species accumulate relatively more quantitative secondary compounds, which play a role in reducing herbivory. The cost of constructing leaves with these contrasting chemical composition differs only marginally, but the photosynthetic return per unit weight of the leaves of fast-growing plants will be much higher.

IX. PHOTOSYNTHESIS A. Species-specific Variation in the Rate of Photosynthesis Fast-growing crop species (Evans, 1983; Makino et al., 1988) and their accompanying weeds (Sage and Pearcy, 1987) tend to have higher maximum rates of photosynthesis (expressed per unit leaf area) than evergreen trees and shrubs (Field et al., 1983; Langenheim et al., 1984). Similarly, sun species have a higher rate of light-saturated photosynthesis per unit area than slower-growing shade species, when the plants are grown at an optimum quantum flux density (e.g. Pons, 1977; Bjorkman, 1981; Seemann et al., 1987). Fast-growing tree and shrub species have higher rates of photosynthesis per unit leaf area than slower-growing ones (Mooney et al., 1978, 1983; Field et al., 1983; Oberbauer et al., 1985). Some of these differences may be phenotypic, rather than ipherent for a species, reflecting a poor nutrient or water supply in the natural habitat. Generally, fast-growing species tend to have higher rates of photosynthesis than slow-growing ones, at least when photosynthesis is expressed per unit leaf weight (Gottlieb, 1978; Dijkstra and Lambers, 1989a; Poorter et al. 1990). The difference may persist when expressed per unit leaf area (Schulze and Chapin, 1987; Evans, 1989a), as long as species of vastly different life forms are compared. In comparisons of species of similar life forms, e.g. herbaceous species (Dijkstra and Lambers, 1989a; Poorter et al., 1990), fastgrowing and slow-growing species have very similar rates of photosynthesis per unit leaf area. Hence, variation in the rate of photosynthesis per unit leaf area does not offer an explanation for differences in RGR between species of similar life form.

PHYSIOLOGY AND ECOLOGY OF GROWTH RATE VARIATION

215

Very little research has been done to elucidate differences in photosynthesis between inherently slow-growing and fast-growing species. Information providing a framework for further analysis of inherent differences in photosynthesis is discussed below and is confined in the main to C, species and not C, or CAM plants. Variation in photosynthetic capacity may reflect differences in organic nitrogen concentration. This capacity is related to a leafs nitrogen concentration, because the major part of all organic nitrogen in the mesophyll cells of a C, plant is found in the chloroplasts (Evans, 1989a). Indeed, leaves of fast-growing species have a higher nitrogen concentration (Section VIII). However, at the same nitrogen concentration in the leaf, there is still a wide variation in light-saturated rates of photosynthesis between species (Evans, 1989a).

B. Photosynthetic Nitrogen Use Efficiency The rate of photosynthesis per unit leaf nitrogen, the photosynthetic nitrogen use efficiency (PNUE), is higher for fast-growing herbaceous species than for slow-growing ones with the same life form, at least when measured at the moderate quantum flux density at which the plants were grown (Poorter et ul., 1990). Therefore, we will discuss plant traits which may explain inherent variation in PNUE. With increasing nitrogen concentration per unit leaf area, photosynthesis is saturated at an increasingly higher quantum flux density. If measured at a quantum flux density which saturates photosynthesis at a low, but not at a high nitrogen concentration in the leaf, a curvilinear photosynthesis-leafnitrogen relationship is inevitable (Evans, 1989a). Such a situation was found in a comparison of Pluntugo major subspecies (Dijkstra, 1989). The slowgrowing subspecies, with a low PNUE when determined at the relatively low quantum flux density at which plants were grown, has a higher chlorophyll concentration per unit leaf area (Dijkstra and Lambers, 1989a). At the relatively low quantum flux density, this will cause shading of the chloroplasts near the lower leaf surface. However, measured at light- and C0,saturation the PNUE was the same for both subspecies. These results agree with those on six fast- and slow-growing monocotyledonous species. When the quantum flux density during the measurements of photosynthesis was increased from that at which the plants were grown to light saturation, both the slope of the C0,-response curve and the C0,-saturated rate of photosynthesis increased significantly more for the slow-growing species than for the fast-growing ones (A. van der Werf, personal communication). Since many leaves often function at a quantum flux density well below the saturation level of photosynthesis, comparison of PNUE values determined at a low quantum flux density is valid, providing the quantum flux density is the same for all plants compared (Schulze, 1982; Karlsson, 1991).

216

H. LAMBERS A N D H. POORTER

Measuring photosynthesis at light saturation in field-grown plants, Field and Mooney (1986) found the highest PNUE values for annuals, intermediate values for drought-deciduous shrubs, and the lowest values for evergreen trees and shrubs. Though some of these differences may have been phenotypic, there is likely to be a strong inherent component as well (cf. Evans, 1989a). What could be the basis of the relatively low PNUE of slowgrowing species?

I . Partitioning of Nitrogen between Chloroplasts and other Cell Components Variation in PNUE might reflect a difference in investment of nitrogen in photosynthetic and non-photosynthetic leaf components. In C, species with a high PNUE, approximately 75% of the nitrogen in mesophyll cells is located in the chloroplasts (Evans, 1989a). It is likely that non-photosynthetic and photosynthetic cells require a similar amount of nitrogen not associated with photosynthesis. Part of this nitrogen is associated with primary cell walls, which are claimed to contain up to 20% structural proteins by weight (Jones and Robinson, 1989). These proteins (“extensin”) are rich in hydroxyproline and appear to be associated with resistance to microbial attack and some forms of abiotic stress (Esquerrt-Tugaye et al., 1979; Lamport and Catt, 198I). Apart from “extensin”, other (hydroxy)proline-rich proteins occur in plant cell walls and some of these are probably also associated with plant defence reactions (Lamport, 1980; Kleis-San Francisco and Tierny, 1990). Cell wall components such as (hemi)cellulose and lignin represent a considerably greater fraction of the leaf dry weight in slowgrowing herbaceous species than of that in fast-growing ones, whereas the reverse is true for the organic nitrogen fraction (Poorter and Bergkotte, 1992). We do not know if the amount of protein per unit weight of cell walls of fast- and slow-growing species is the same. If so, then the fraction of organic nitrogen that is tied up in cell walls is certainly much greater in slowgrowing species. This indicates that a low PNUE may be partly associated with greater investment in cell-wall components. Leaves with a very thick or multiple epidermis (Esau, 1977), crystal cells (Kimmerer and Potter, 1987), collenchyma and sclerenchyma elements (Konings et al., 1989), or cells with specific functions (e.g. water storage; Schmidt and Kaiser, 1987), must invest part of the nitrogen in their leaves in these structures. Thus, leaves which contain some of these additional elements are bound to have a lower PNUE. Similarly, accumulation of relatively large quantities of nitrogen-containing molecules as compatible solutes (e.g. proline and glycinebetaine; Wyn Jones and Gorham, 1983), storage proteins (Franceschi et al., 1983; Staswick, 1988), peptides that sequester heavy metals (Lolkema et al., 1984; Robinson and Jackson, 1986), protective compounds (e.g. polyamines;

PHYSIOLOGY AND ECOLOGY OF GROWTH RATE VARIATION

217

Galston and Sawhney, 1990; Kuehn et al., 1990), antifungal polypeptides (e.g. thionins; Reimann-Philipp et al., 1989; Ape1 et al., 1990), or toxic antiherbivore compounds (e.g. cyanogenic glycosides: Kakes, 1987; cyanolipids: Poulton 1990; alkaloids: Hartmann et al., 1989) also decreases PNUE. We conclude that a low PNUE may be a consequence of a large investment of nitrogen in cell walls, specialized cells or compounds that are not associated with photosynthesis.

2. Suboptimal Partitioning of Nitrogen within the Chloroplast When grown at optimum nitrogen supply, slow-growing herbaceous species have higher chlorophyll concentrations per unit leaf area and unit nitrogen (Poorter er al., 1990). Slow-growing herbs have double the concentration of chlorophyll found in fast-growing species (0.6 vs. 0.3 mmol m-*). This requires the extra investment of at least 15 mmol of nitrogen per square metre of leaf area, which amounts to 12% or more of a leafs total nitrogen concentration in the slow-growing herbs, but only increases the leafs absorptance by 7% (Evans, 1989b). Although extra investment in chlorophyll increases photosynthetic performance under shade conditions (Evans, 1989b), there may well be some excess in capacity of the light-harvesting machinery. But this offers only a partial explanation for the low PNUE of slow-growing species.

3. Activation of Rubisco Activation of Rubisco requires carbamylation of the enzyme (Salvucci, 1989). The degree of carbamylation depends on the quantum flux density. In some, but not all higher plant species, Rubisco is also regulated by a naturally occurring tight-binding inhibitor: 2-carboxyarabinitol 1-phosphate (Servaites, 1990). The difference in photosynthesis per unit Rubisco at high quantum flux density cannot be explained by variation in the degree of enzyme activation by either of the above mechanisms (Seemann, 1989). Intrinsic differences in the enzyme from different species, rather than regulation of its activity by activation, are the likely cause of variation in the specific activity of Rubisco.

4. Variation in Rubisco Specific Activity Variation in PNUE has been further analysed in a comparison of Alocasia macrorrhiza, a tropical understorey plant, with two crop species (Seemann et al., 1987; Seemann, 1989). Alocasia has a considerably lower photosynthetic capacity per unit leaf nitrogen or Rubisco protein. The relatively low PNUE of Alocasia is partly a consequence of a relatively low specific activity (carboxylating activity per unit enzyme) of its Rubisco. A low specific activity is not restricted to shade-tolerant or inherently slow-growing species, but has also been found in comparisons of fast-growing crop species

218

H. LAMBERS AND H. POORTER

(Seemann and Berry, 1982, cited in Evans, 1989a; Makino et al., 1988). So far we do not know the biochemical basis of variation in specific activity of Rubisco or if there is any systematic link between specific activity and ecological traits.

5. Feedback In hibit ion of Photosynthesis Comparing the rate of photosynthesis at normal and high internal partial pressure (p,) of CO,, provides insight into the extent of feedback control of photosynthesis (Sharkey et ul., 1986; Sage and Sharkey, 1987). Such control may play an important role when the products of photosynthesis, i.e. sucrose and its phosphorylated precursors, accumulate, due to for example a relatively low temperature (Sage and Sharkey, 1987) or a limited activity of the sink (Plaut et ul., 1987). Under these conditions the photosynthetic apparatus is only partly used, so that the PNUE is less than maximal. We do not know if such a situation of feedback inhibition is the norm in slowgrowing species, but if so it could offer an explanation for their relatively low PNUE.

6. Eflects of the CO, Concentration inside the Leaf Photosynthesis increases with increasing internal partial pressure of CO, inside the leaf (p,), up to a maximum. Differences in pi between species or populations have been reported (Mooney and Chu, 1983) and might offer a partial explanation for a low PNUE in leaves with a low SLA and a high nitrogen concentration per unit leaf area, as found for slow-growing species (Poorter et al., 1990). Parkhurst et al. (1988) found a significant CO, pressure gradient inside the leaf, more so for hypostomatous than for amphistomatous leaves (Parkhurst and Mott, 1990). Moreover, the CO, pressure at the site of carboxylation (p,) might be significantly lower than pi. Hence, a greater internal CO, diffusion limitation also partly explains a low PNUE. If generally valid, species with a low PNUE should show less discrimination against ' T O 2 than those with a higher PNUE. The degree of discrimination reflects the ratio of fhe CO, concentration inside the leaf and that in the atmosphere (pa) (cf. Farquhar et al., 1982). Our own data on a range of herbaceous species grown under the same conditions do not show a correlation between carbon isotope discrimination and RGR (Fig. 6B). In a comparison of species from high altitudes with related species from low altitude, the high-altitude species show less discrimination, which correlates with their lower ratio of p,/p, (Korner and Diemer, 1987; Korner et al., 1988). Leaf carbon isotope discrimination of annuals (Smedley el al., 1991) and short-lived perennials (Ehleringer and Cooper, 1988) is less than that of longer-lived (and presumably slower-growing) perennials, also when grown under the same environmental field conditions. Our plants (Fig. 6) were grown under well-watered conditions and at a

PHYSIOLOGY AND ECOLOGY OF GROWTH RATE VARIATION

219

relatively low vapour pressure deficit, which may have led to a relatively high p, (cf. Schulze et al., 1987; Brugnoli and Lauteri, 1991). We cannot fully exclude that variation in p, between species occurs when plants are supplied with a limiting amount of water or at a higher vapour-pressure deficit and that this correlates with that in PNUE. However, the variation in PNUE in comparisons of widely different species (Evans, 1989a) is certainly too great to be fully accounted for by variation in p,. Moreover, the variation in PNUE as found in a comparison of herbaceous species cannot be accounted for by variation in p, (Fig. 6A,B).

7. Conclusions To summarize the above, slow-growing species with a relatively low PNUE may invest relatively large quantities of nitrogen in components not associated with photosynthesis. They may also have a suboptimal distribution of nitrogen between elements of the photosynthetic apparatus or a Rubisco enzyme with a low catalytic capacity. A larger degree of feedback inhibition or a relatively low CO, concentration at the site of carboxylation might also play a role. There is as yet no convincing evidence for any of these possible explanations.

C. Is There a Compromise between Photosynthetic Nitrogen Use Efficiency and Water Use Efficiency? The leafs stomatal conductance tends to be regulated in such a way that a compromise is reached between gain of CO, and loss of H 2 0 (Farquhar and Sharkey, 1982). Increased conductance would lead to roughly proportionally greater transpiration, but marginally greater photosynthesis. However, different environmental conditions may require a different compromise (Cowan, 1977). Species which naturally occur in environments where the water supply is low and/or the evaporative demand is high might have a lower stomatal conductance and a lower intercellular pressure of C 0 2 (p). This tends to increase their water use efficiency of photosynthesis (WUE), but, all other parameters being equal, leads to a lower PNUE, as discussed above. Hence, a low PNUE might be a reflection of a high WUE. In a comparison of five Californian evergreen species, Field et al. (1 983) indeed found that species with a lower PNUE tended to have a higher WUE. The water-efficient species typically occurred in the driest habitats. Pavlik (1983) did not find a different WUE for two dune grasses differing in their PNUE. If there were a general trend in a wider comparison, for species with a low PNUE to have a high WUE, such species are expected to show less discrimination against "C02 (Farquhar et al., 1982; Hubick and Farquhar, 1989). However, five Californian evergreens with a low PNUE showed exactly the same discrimination as the average C, plant (Field and Mooney,

A 120

100

80

60

28' -

0

26-

24

-

0

-

0 0

22 -

C 0 0

0 0

?/'

200

300

R G R (mg g-' day-' )

PHYSIOLOGY AND ECOLOGY OF GROWTH RATE VARIATION

22 1

1986), which does not lend support to the contention that a low PNUE generally reflects a high WUE. Our own data on the PNUE and WUE of nine herbaceous species (Figs 6A,C) do not support the hypothesis either. Therefore, if such a compromise between PNUE and WUE exists, it can at best be relevant in comparisons of species with similar life form grown at a relatively high quantum flux density and with a high evaporative demand.

D. Photosynthesis under Suboptimal Conditions The photosynthetic capacity decreases with decreasing nitrogen concentration in the leaf, in a (curvi)linear fashion (Evans, 1989a). In the slow-growing Carex diandra, p, increases with decreasing steady state nitrogen supply (H. Lambers and R. Welschen, unpublished). Such an increase with decreasing leaf nitrogen has also been found for other species, including fast-growing ones (Morgan, 1986; Sage and Pearcy, 1987; Ghasghaie and Saugier, 1989; but see Sage and Pearcy, 1987). This increase in p, may contribute to curvilinearity in the photosynthesis-leaf nitrogen relationship, at least when photosynthesis is measured at an ambient CO, concentration. An alternative explanation is that the CO, gradient between the intercellular spaces and the site of carboxylation and/or the gradient in p, within the leaf increases with increasing leaf nitrogen concentration (Evans, 1989a). A relatively lower investment in chlorophyll-complexing thylakoid proteins at a low nitrogen supply, as long as this does not significantly affect the leafs absorptance, also tends to contribute to curvilinearity (Section IXB). Curvilinearity of the photosynthesis-leaf nitrogen relationship does not appear to be due to inactivation of Rubisco at a high leaf nitrogen concentration (Evans and Terashima, 1988). Despite these reasons which tend to produce curvilinearity, such a relationship is not invariably found, due to the proportionally greater investment of nitrogen in Rubisco with increasing leaf nitrogen concentration (Evans, 1989a). Curvilinearity of the photosynthetic capacity vs. leaf nitrogen relationship can be interpreted as a relatively inefficient use of nitrogen for photosynthesis at a high nitrogen concentration in the leaf and as such might be expected to occur particularly in slow-growing species adapted to nutrient-poor soils. Indeed, at high N-supply the PNUE of slow-growing species is lower than at high N-supply. A similar tendency, though not as strong, has also been

Fig. 6 (opposite). Gas exchange-related characteristics of fast-growing and slowgrowing species. described in Fig. 1. (A) The rate of photosynthesis expressed on a leaf nitrogen basis. (B) Carbon isotope discrimination. (C) The water use efficiency. Gas exchange characteristics were determined under conditions used for growing the plants. (After Poorter et al. (1990) and unpublished data of H . Poorter and G. D. Farquhar; further information, based on 24 species, is unpublished.)

222

H. LAMBERS AND H. POORTER

observed for fast-growing species, so that the PNUE of plants grown at a low N-supply is rather similar for fast- and slow-growing species (Boot er af., 1992; A. van der Werf, personal communication). Perhaps the low PNUE of slow-growing species at a high N-supply merely indicates that the rate of biomass accumulation and cell expansion of these species is saturated by the supply of N before that of N-uptake and N-assimilation. When plants are grown at a low quantum flux density, their leaves have a lower photosynthetic capacity (e.g. Pons, 1977; Young and Smith, 1981). Allocation of nitrogen to the light-harvesting machinery is increased at the expense of that to Rubisco and other enzymes of the Calvin cycle (Bjorkman, 1981). This change in pattern of nitrogen distribution over the various components of the photosynthetic apparatus leads to optimization of the use of nitrogen for photosynthesis. There is some evidence that slow-growing, shade-tolerant species, have a greater capacity to adjust their nitrogen partitioning in this manner than fast-growing ones (Evans, 1989b).

E. Conclusions From the above discussion it transpires that slow-growing species have a relatively low PNUE, when grown at optimum nitrogen supply. As yet, there is no satisfactory explanation for this difference. At the low N-supply which the slow-growing species encounter in their natural habitat, the PNUE of fast- and slow-growing species is rather similar. A further appreciation of the ecological advantage, if any, associated with a low PNUE, clearly awaits more information on the physiological, biochemical or anatomical background of the variation in PNUE and possible negative correlations of PNUE with WUE.

X. RESPIRATION Respiration provides the driving force for three major energy-requiring processes: maintenance, growth and ion uptake. Maintenance respiration is mainly associated with turnover of various cellular components and the conservation of solute gradients across membranes. Growth respiration is used to supply ATP and NADH, needed to convert glucose into the different chemical compounds. In roots, respiratory energy is also needed for the absorption of nutrients from the environment.

A. Species-specific Variation in the Rate of Respiration Fast-growing species have a higher rate of nitrate uptake (Fig. 3A) and a higher RGR. Therefore, it is not surprising that they have a higher rate of

PHYSIOLOGY AND ECOLOGY OF GROWTH RATE VARIATION

223

shoot and root respiration (Dijkstra and Lambers, 1989a; Poorter et al., 1990, 1991; van der Werf et al., 1992b; Fig. 3B). However, the rate of root respiration of fast-growing species is not as high as would be expected from their higher R G R and NIR. Using data of van der Werf et al. (1988), who determined specific costs for root growth, maintenance and ion uptake for two slow-growing Carex species, the calculated rate of root respiration of fast-growing species is approximately four times higher than that of slowgrowing ones (Poorter er al., 1991; Fig. 3B, broken line). This value is at variance with that determined experimentally, showing that the rate of root respiration of fast-growing species is only 50% higher than that of slowgrowing ones (Fig. 3B, regression line through data points). Why do the fastgrowing species respire at a rate which is so much lower than expected from their R G R and NIR?

1. Variation in Respiratory EBciency The relatively low rate of respiration of fast-growing species might be due to a relatively more efficient respiration. Apart from the cytochrome pathway, which yields three molecules of ATP per oxygen atom reduced, plants have an alternative, non-phosphorylating respiratory pathway. Engagement of this path, rather than the cytochrome path, yields only one third as much ATP (Lambers, 1985). Do fast-growing species employ this alternative pathway to a lesser extent than the slow-growing ones and therefore produce more ATP for the same amount of glucose respired and oxygen reduced? Despite a wide variation in the participation of the alternative, nonphosphorylating electron transport path in respiration, varying from 0 to 44% of total root respiration (van der Werf et al., 1989) and from 4 to 58% for leaves (Collier and Cummins, 1989; Atkin and Day, 1990), there is no evidence that this respiratory path contributes less to respiration in fastgrowing species or genotypes (Dijkstra and Lambers, 1989b; Atkin and Day, 1990; Poorter et al., 1991; van der Werf et al., 1992b). In fact, Collier and Cummins (1989) found that respiration rates of the leaves of ruderals collected in the field were higher and that their alternative path was engaged to a greater extent than that of understorey species. Next to a greater alternative path activity, the ruderals also had a greater capacity for respiration via this path. This might contribute to their functioning in fluctuating environments, in line with results on root respiration of other species (Lambers et al., 1981b). The rate of ATP production per oxygen consumed might also be higher in fast-growing species if their mitochondria operate under substrate limitation or further away from “state 4” conditions. (State 4 is the condition where the respiration is strongly restricted by the availability of ADP, as opposed to state 3, where ADP is available in saturating amounts.) If so, they d o not produce ATP whilst still reducing oxygen (cf. Whitehouse et al., 1989). In

224

H. LAMBERS A N D H. POORTER

state 3 conditions, they operate more efficiently, producing maximally three ATP per oxygen atom. If the fast-growing species would respire closer to state 3 , their ADP : 0 ratio in vivo might be higher. Straightforward methods exist to test this hypothesis (Day and Lambers, 1983), but so far no comparative data are available on fast- and slow-growing species.

2. Variation in Speclfic Costs of Energy-Requiring Processes The current data provide no indication that fast-growing species produce ATP more efficiently than slow-growing ones. Therefore, it is very likely that the relatively low respiration rate of fast-growing species is caused by lower specific respiratory costs for energy-requiring processes, such as maintenance, growth or ion acquisition, at least under optimum growth conditions. We have very little information on the biochemistry and physiology of “maintenance processes”. Protein turnover and the maintenance of solute gradients are considered major components (Penning de Vries, 1975), but quantitative information on either of these processes in roots is scarce (van der Werfet al., 1992~). Assuming the same turnover rate per unit protein, we expect the maintenance respiration of fast-growing species, which have a higher protein concentration (Section VIIIA), to be higher than that of slowgrowing species, but this is not borne out by experimental results. Other processes which require respiratory energy and which d o not contribute to growth or ion uptake will also increase the maintenance costs. For example, the relatively large turnover of carbohydrate pools in slow-growing species as compared to fast-growing ones (Farrar, 1989), might partly explain the high respiratory costs of roots in slow-growing species. However, maintenance respiration is only a small portion of total respiration, at least in young plants grown at an optimum nutrient supply (van der Werf et al., 1989; Porter et al., 1991). Thus variation in the specific costs for maintenance is unlikely to affect substantially the rate of root respiration. In the case of growth respiration, the construction costs per unit biomass may vary with the chemiFal composition of the plant. Table 4 provides information on the gas exchange which could be expected to occur with the synthesis of a range of primary and secondary compounds. The calculated oxygen uptake and carbon dioxide release include gas exchange associated with ATP production, (de)carboxylating reactions, as well as the use of oxygen in reactions catalysed by such enzymes as mixed function oxygenases. We calculated that the oxygen requirement for the synthesis of roots of fastgrowing species is higher than that for slow-growing ones (Poorter et al., 1991). Hence, differences in chemical composition cannot explain why respiration rates of slow-growing species are only marginally lower than those of species growing three times as fast and absorbing ions at over four times the rate of the slow-growing ones.

PHYSIOLOGY AND ECOLOGY OF G R O W T H RATE VARIATION

225

If the specific costs for maintenance and growth cannot explain the relatively high respiration rates of slow-growing species, then the most likely explanation for the observed differences in root respiration between slowgrowing and fast-growing species is variation in the specific cost for nutrient acquisition. Slow-growing species, which are often associated with unproductive environments (Section 11), are likely to be geared towards nutrient uptake from a very dilute solution in comparison with the fast-growing species from relatively nutrient-rich habitats. This may require more energy, either because it involves extrusion and re-uptake of compounds which release ions from complexes in the soil (Section XIA) or because the ratio of proton entry and ion absorption is higher, compared to the system of fastgrowing species (cf. Clarkson, 1986; McClure et al., 1990). For Hordeum vulgare, two transport systems for nitrate uptake have been described. One of these systems is an energy-requiring one, operating at very low external concentrations. The other one does not require metabolic energy and functions at an external nitrate concentration sufficiently high, compared to the cytoplasmic concentration, to allow passive diffusion of nitrate across the plasma membrane (Glass et al., 1990; Siddiqi et al., 1990). Perhaps fastgrowing species maintain a relatively low nitrate concentration in the cytoplasm of their roots cells, due to rapid reduction of nitrate in the cytosol, transport into their, presumably relatively large, vacuoles (cf. Section XIIA), or efficient export to the shoot, made possible by the relatively high rate of transpiration per unit root weight (Poorter et al., 1990). If so, the system which does not require metabolic energy may predominate in the roots of fast-growing species, whereas the energetically more expensive one is more important in the roots of slow-growing ones. This is not to say that fast-growing species lack a similar system. Rather, it may be constitutive in slowgrowing species and inducible under nutrient-deficient conditions in fastgrowing ones, when the electrochemical gradient across the plasma membrane does not allow passive uptake of nitrate (van der Werf et al., 1992b, cf. Section VIIA). An alternative and more attractive explanation for presumably higher costs for nitrate uptake is that the ratio between ion influx and efflux is lower in slow-growing species (cf. Pearson et al., 1981; DeaneDrummond and Glass, 1983; Oscarson et al., 1987).

B. Respiration at Suboptimal Nitrogen Supply or Quantum Flux Density With a decreasing nitrogen supply, there is a decline in the nitrogen concentration and the rate of respiration per unit mass, both in roots (Lambers et af., 1981a; Duarte et al., 1988; Granato et al., 1989; van der Werf

226

H. LAMBERS AND H. POORTER

et al., 1992b) and in leaves (Waring et al., 1985; Boot et al., 1992). The respiration rate is less, because of reduced energy requirements for biosynthetic processes, ion transport and loading of sucrose in the phloem. Upon a decrease of the nitrogen supply, both the nitrogen concentration and the rate of leaf respiration of the faster-growing Agrostis vinealis decline to a greater extent than that of the slow-growing Corynephorus canescens (Boot et al., 1992). Similar results have been obtained in a comparison of the fastgrowing Holcus lanatus and the slow-growing Deschampsia JEexuosa (C.A.D.M. van de Vijver, R.G.A. Boot and H. Poorter, unpublished). At optimum nitrogen supply, Agrostis has a higher leaf nitrogen concentration (cf. Section VIIIA), whereas it is reduced to the same level as in Corynephorus when the nitrogen supply is reduced. There is some evidence that specific costs for ion transport and/or maintenance increase at a limiting nitrogen supply, particularly in fast-growing species (A. van der Werf et al., 1992b). When plants are transferred to a low quantum flux density, the rate of root respiration declines (Kuiper and Smid, 1985; H.H. Prins, R. Hetem and H. Poorter, unpublished), as expected from the lower rate of root growth under such conditions. Upon prolonged exposure to a low quantum flux density, there is only a small, or no difference in root respiration between plants grown at high vs. low quantum flux density (Lambers and Posthumus, 1980; A. van der Werf and P. Poot, unpublished). This is attributed to the increase in LWR upon prolonged exposure to a low quantum flux density, which increases the root’s energy requirement for uptake of ions destined for the shoot. Also, the rate of leaf respiration is less for plants grown at a low quantum flux density (Pons, 1977; Waring et al., 1985). Presumably the lower rates of respiration reflect the lower energy requirement for biosynthetic and transport processes. We d o not know of any comparative data on fastgrowing and slow-growing species.

C. Conclusions At an optimum nutrient supply, the specific rate of root respiration of fastgrowing species is lower than expected from their high RGR and high NIR. A satisfactory explanation for this relatively low respiration rate cannot be provided yet, but it may well be due to a relatively low energy requirement for ion acquisition. At a nutrient supply or quantum flux density which is suboptimal for growth, the rate of respiration in both leaves and roots is less than for plants growing under optimum conditions. This is at least partly explained by reduced rates of energy-requiring processes. Upon a change in nitrogen supply, fast-growing species adjust their respiration rate to a greater extent than slow-growing ones, possibly with concomitant changes in specific costs for energy-requiring processes.

PHYSIOLOGY AND ECOLOGY OF GROWTH RATE VARIATION

227

XI. EXUDATION AND VOLATILE LOSSES Plants lose photosynthates through exudation and volatilization as well as during respiration. Exudation may occur both above- and below-ground, whereas volatilization predominantly occurs above-ground.

A. The Quantitative and Qualitative Importance of Exudation It is well established that roots exude a range of organic compounds, including sugars (McCully and Canny, 1985), organic acids (Gardner et al., 1983) and amino acids (McDougall, 1970), especially when phosphate is in short supply (Graham et al., 1981; Lipton et al., 1987). Estimates of the loss of exudates from roots vary widely, largely depending on the methods used to quantify this process and perhaps on the species under investigation (Cheshire and Mundie, 1990). Soluble exudates constitute less than 0.5% of the carbon present in the plant (Cheshire and Mundie, 1990). Their production ranges from 10 to 100 mg per gram root dry weight produced; root cap plus mucigel may provide a further 2&50 mg (Newman, 1985; Gregory and Atwell, 1991). At the very most 5 % of the photosynthates are lost through “rhizodeposition”, i.e. the loss of organic matter via both processes (Lambers, 1987), with the exception of plants with proteoid roots in which very high values are found (see below). If losses due to continual cell death are also included, losses due to rhizodeposidition (sensu lutu) may amount to 10% of all photosynthates produced (Helal and Sauerbeck, 1986; Lynch and Whipps, 1990). Exudates are of distinct importance for the acquisition of sparingly available nutrients and for interactions with symbionts. Highly efficient chelators (phytosiderophores) are excreted by roots of Gramineae and these allow the roots to absorb Fe, Zn, Mn and Cu from poorly soluble sources in calcareous soils (Romheld and Marschner, 1986; Marschner et al., 1989; Zhang er al., 1989). At least part of the excreted phytosiderophores are absorbed again by the roots as a metal-siderophore complex (Romheld and Marschner, 1990). Non-gramineous species also release chelating compounds, generally of a phenolic nature, but these are less efficient than the true phytosiderophores (Romheld, 1987). Many species, particularly those with proteoid roots, release citric acid (Gardner et al., 1983; Hoffland et al., 1989a, b) which may amount to as much as 23% of the biomass at plant harvest (Dinkelaker et al., 1989). The excretion ofcitric acid greatly enhances the root’s capacity to use insoluble phosphate (Hoffland et al., 1990). Proteoid roots also have an increased capacity for reduction of iron and manganese in the rhizosphere and, consequently, to mobilize sparingly soluble Fe or A1 phosphates (Gardner et al., 1982a, b). Excretion of citric acid may also be significant in releasing cations from humic substances (Albuzzio

228

H. LAMBERS AND H. POORTER

and Ferrari, 1989). We conclude that the release of chelating substances, such as citric acid and possibly other organic acids is important in acquiring nutrients from both calcareous substrates and acidic soils where cations are bound to a humic complex. Many slow-growing species are associated with such calcareous or acidic soils (Grime and Hunt, 1975). Hence the release of root exudates is likely to confer an advantage in such soils and should not merely be considered as a loss of carbon. Some micro-organisms exude organic compounds that precipitate heavy metals outside the cells. There is no evidence that this mechanism is important in ecotypes of higher plants that tolerate high concentrations of heavy metals in the root environment (Verkleij and Schat, 1990). Release of organic substances from roots can also play a pivotal role in symbiotic associations. For example, flavonoids, released from the roots of Leguminosae induce the nodulation genes of Rhizobium, the primary step in the nodulation process (Richardson et al., 1988; Hartwig et al., 1990). Plants capable of a mycorrhizal symbiosis release organic compounds to the rhizosphere when they contain very little phosphate in the roots, presumably due to the fact that their membranes contain less than an optimum amount of phospholipids (Ratnayaka et al., 1978). Some root exudates, predominantly of a phenolic nature, play a role in allelopathic interactions between plants. Although the exact nature of the compounds released, their biochemical effect on neighbouring plants and their ecological significance are often not known, the existence of allelopathic interactions, including those based on exuded compounds, is beyond doubt (Putnam and Tang, 1986; Kuiters, 1990). Root exudation can also affect, both negatively and positively, the rates of a number of soil biological processes, such as denitrification (Woldendorp, 1963), nitrification (Haider el al., 1987; Vitousek et al., 1989) and mineralization (Sparling et al., 1982). Different plant species affect these soil biological processes to varying degrees (Janzen and Radder, 1989; Berendse et al., 1989; van Veen el al., 1989), but the exact nature of the effects is generally not fully known. There is some evidence that losses through exudation are quantitatively more important in a slower-growing Hordeum vulgare variety than in a faster-growing one (Liljeroth et al., 1990). Losses of carbon through exudation also occur above-ground. We have very little information on their quantitative importance and inherent variation of this process (Tukey, 1970).

B. The Quantitative and Qualitative Importance of Volatile Losses Although it is probably fair to state that volatile losses are generally not of great quantitative significance, they are of fairly wide importance, e.g. in

PHYSIOLOGY AND ECOLOGY OF GROWTH RATE VARIATION

229

allelopathic interactions and herbivory (Harborne, 1982; Rhoades, 1985; Dicke and Sabelis, 1989). Emissions of isoprene in three fern species accounted for 0.02 to 2.6% of the carbon fixed during photosynthesis, increasing with photon flux density and temperature and varying between species (Tingey et al., 1987). These values are in the same range as those found for tree leaves (Sanadze, 1969; Tingey et al., 1981). They are up to 2.5 times lower than those based on field measurements for a number of tree species, which may reflect species differences in emission or effects of high quantum flux density (Flyckt et al., 1980, cited in Tingey et al., 1981). In extreme cases, such as in Ledum groenlandicum during part of the year (Prudhomme, 1983) and Populus tremuloides at high temperatures (3545°C) (Monson and Fall, 1989), up to 8% of recently fixed carbon may be lost as volatiles. Losses of specific volatiles, though quantitatively minor (less than 0.00 1 YO of the photosynthates produced daily), are responsible for attraction of predators upon attack of leaves by herbivores, e.g. spider mites, and thus reduce herbivore damage (Dicke and Sabelis, 1989).

C. Conclusions Losses through exudation sensu lato can significantly reduce a plant’s growth rate. Carbon loss through exudation and volatilization is most certainly of ecological importance in a nutrient-poor environment and in interactions of a plant species with other organisms. Although not supported by hard evidence, we hypothesize that exudation is more important in slow-growing species from nutrient-poor sites. Exudation might allow such species to acquire nutrients which are otherwise unavailable.

XII. OTHER DIFFERENCES BETWEEN FAST- AND SLOW-GROWING SPECIES Apart from the above-mentioned traits, which directly affect the growth of a plant, some other aspects of fast-growing and slow-growing species and mutants thereof have been investigated. In recent years fascinating information has become available on the role of a specific class of phytohormones in the control of a plant’s growth rate-the gibberellins.

A. Hormonal Aspects Fast-growing genotypes contain more gibberellin than slower-growing ones (Rood et al., 1983; 1990a, b, c; Dijkstra et al., 1990; H. Konings and M.

230

H. LAMBERS A N D H. POORTER

Berrevoets, personal communication). Rapid growth of hybrids (“heterosis”) has been associated with higher levels of gibberellin in both herbaceous species and trees (Rood and Pharis, 1987; Bate et al., 1988). Interestingly, in a Zea mays hybrid, both the superior growth and the higher level of gibberellin, compared to those of its parents, are restricted to favourable conditions and not displayed during growth at low temperature (Rood and Pharis, 1987). Gibberellins control leaf size (Brassica rapa, Zanewich et al., 1990; Thlaspi arvense, Metzger and Hassebrock, 1990; Lycopersicon esculentum, H. Konings and M. Berrevoets, personal communication) and it seems likely that the variation in leaf size between fast- and slow-growing species is at least partly associated with differences in the concentration of endogenous gibberellins. Mutants of Lycopersicon esculentum with reduced levels of gibberellin, have a range of characteristics similar to slow-growing species, e.g. a higher RWR, but lower LAR and SLA, relatively more dry matter per unit fresh weight, and a low rate of photosynthesis per unit leaf dry weight (H. Konings and M. Berrevoets, personal communication). Treatment with gibberellin reduces genotypic differences in RGR, indicating that this hormone plays a role in intraspecific variation in growth potential, probably via its effect on leaf area development and biomass partitioning (Dijkstra and Kuiper, 1989; Dijkstra et al., 1990; Rood et al., 1990a, b, c; Zanewich et al., 1990; H. Konings and M. Berrevoets, personal communication). The detailed mechanism of the gibberellin effects on growth is largely unknown. However, it is well documented that this phytohormone affects stem growth via both cell elongation and cell division (Mtttraux, 1987; Jupe et al., 1988; Rood et al., 1990b). Effects of gibberellin on cell enlargement could account for a number of the chemical differences between fast- and slow-growing species (cf. Section VIIIA). The larger surface-to-volume ratio of plants with smaller cells is expected to be associated with a relatively large investement in cell-wall components and hence a high dry matter percentage. Smaller cells are bound to have relatively small vacuoles, which would explain the relatively low capacity of slow-growing species to accumulate organic acids, nitrate and other minerals. The effects of gibberellin on cell division could account for differences in biomass partitioning (cf. Section VIA). A low level of gibberellins prevents rapid incorporation of photosynthates into leaf and stem biomass, so that a relatively large proportion is translocated to and incorporated into the roots. Clearly, further work is needed on the mechanism of gibberellin action on cell growth and on the level of this phytohormone in different species. This may well lead to physiological explanations for inherent variation in RGR and provide insight into evolutionary mechanisms causing such variation.

PHYSIOLOGY A N D ECOLOGY OF GROWTH RATE VARIATION

23 1

B. Miscellaneous Traits Fast-growing grass species tend to have wider leaves than slow-growing ones; leaves of dicotyledonous fast-growing species also tend to be larger than those of slow-growing ones (Christie and Moorby, 1975; Ceulemans, 1989; Korner and Pelaez Menendez-Riedl, 1989; H. Poorter, unpublished). Leaves of fast-growing poplar hybrids are larger than those of either of their parents, Populus trichocarpa and P. deltoides. Leaf cells of P . trichocarpa are larger than those of P. deltoides, whereas P. deltoides has more cells per leaf. The greater leaf size of the hybrids can be explained by inheritance of a larger cell number from P . deltoides and larger cells from P. trichocarpa (Ceulemans, 1989). In general, variation in leaf size appears to be due predominantly to fewer cells per leaf, rather than to cell size (Korner and Pelaez MenendezRiedl, 1989). The diurnal pattern of leaf growth may vary between species and hybrids thereof. Leaves of Populus trichocarpa mainly grow in the light, with little growth in the dark, those of P. deltoides grow during the dark period, with little stimulation in the light, whereas their fast-growing interspecific hybrids grow during both night and dark (Ceulemans, 1989). For a wide range of species and for families of Poa annua, a slightly negative correlation between nuclear DNA content and RGR has been reported (Grime et al., 1988). A negative correlation between RGR and seed size has been suggested (Fenner, 1978; Gross, 1984). However, no such relationship has been detected in a much larger data set (Thompson, 1987). Seeds of fast-growing species germinate more rapidly than those of slowgrowing ones (Grime er al., 1988).

XIII. AN INTEGRATION OF VARIOUS PHYSIOLOGICAL AND MORPHOLOGICAL ASPECTS In the previous sections several aspects of the physiology, morphology, allocation and biochemical composition have been discussed in relation to the potential growth rate of plant species. We now address the question of what proportion each parameter contributes to the observed differences in growth rate, using eqn (5) as a framework.

A. Carbon Budget For the nine species presented in Figs I , 2, 3 , 4 and 6, RGR varies more than three-fold. The carbon concentration (Section VIII) is lower for fast-growing species, which is partly due to their higher mineral concentration. The lower

232

H. LAMBERS AND H. POORTER

Fig. 7. The carbon budget of a slow-growing species (Corynephorus canescens, left) and a fast-growing species (Galinsogaparvipora, right). In the upper line the RGR of these species is given (expressed in mg g - ' day-'). The second line gives the daily gross COz fixation (mmol (g plant)-I). (After Poorter et al., 1990.)

carbon concentration contributes to a higher RGR, but is only of minor importance in explaining the observed variation in RGR. Quantifications of exudation and volatile losses are scarce but these processes are unlikely to determine a large part of the variation in RGR either (Section IX). Hence, the major differences are due to variation in photosynthetic gain and respiratory losses of carbon. Indeed, carbon gain of the fastest-growing of these 9 species is about 3.1 times that of the slowest-growing one (Fig. 7). This is caused by a much higher SLA (Section V) and a slightly higher LWR (Section VI), rather than by a higher rate of photosynthesis per unit leaf area (Section VIII). Species also differ in the way they utilize the fixed carbon. Although fast-growing species have higher shoot and root respiration rates, the proportion of fixed carlion used in respiration is less. This is due largely to a higher rate of photosynthesis per unit plant weight and, as far as root respiration is concerned, also to the lower RWR of fast-growing species. It is to be expected that the lower RWR contributes to the higher RGR, provided the smaller root size is compensated for by a higher specific activity. Indeed, fast-growing species do have a much higher rate of net ion uptake (Section VII) and water absorption (Poorter et al., 1990) than slow-growing ones. Compared to slow-growing species, their rate of root respiration is not as high as would be expected from their four times higher rate of nitrate uptake and their three times higher relative growth rate. This relatively low respiration rate contributes to the rapid growth of fast-growing species.

PHYSIOLOGY AND ECOLOGY OF GROWTH RATE VARIATION

233

B. Interrelations Up till now we have treated the different physiological and morphological aspects as being independent of each other. This is a simplified view, as there are numerous interrelations. For example, a shift in biomass allocation from leaves to roots implies a decrease in the photosynthetically active area and an increase in the respiratory burden. Consequently, a decrease in the rate of carbon fixation and growth of such a plant is expected. At the same time, a high allocation to roots may ensure better access to nutrients and water, which may result in an increased plant nitrogen concentration, a higher photosynthetic capacity per unit leaf area and also a higher stomata1 conductance. Moreover, there may be a decrease in self-shading. Consequently, the rate of photosynthesis per unit leaf area will increase. Also, root biomass is generally energetically cheaper to construct (Poorter and Bergkotte, 1992), so that growth rate will not decrease to the same extent as LWR decreases. Another simplification we have made so far is to compare different types of plants over a limited time course. Investment of biomass in compounds that reduce herbivory or increase a plant’s stress tolerance then inevitably leads to a decrease in the rate of photosynthesis per unit plant weight (Section IXD). However, due to these investments, the life expectancy of a leaf increases, the net result being a possibly similar or even higher rate of photosynthesis integrated over the entire life span of a leaf (Schulze, 1982). Strong correlations between different plant traits do not necessarily imply causal relations. For example, the correlation of RGR with SLA is probably partly fortuitous. Based on the conventional growth analysis (eqns (lH5)), we conclude that variation in SLA is the main cause for inherent variation in RGR. Simplified, it means that an increase in leaf area relative to leaf weight increases the growth rate. From a mechanistic point of view this statement is not entirely correct. Although photosynthesis indeed is an area-related process, the actual quantum capture and C0,-fixation require light-harvesting complexes, the coupling factor, Rubisco and other photosynthetically active components (Section IX). As this photosynthetic machinery incorporates the major part of a leafs organic nitrogen, PNUE and leaf organic nitrogen may be used as a good approximation of the efficiency and size of the photosynthetic machinery (Section IX). Fast-growing species have both a higher PNUE and a higher organic nitrogen concentration. These two parameters appear to be closely correlated with SLA and, combined, show the same 2.2-fold variation for the nine species presented here, as found for SLA. This correlation partly reflects a causal relation, in so far as a high SLA coincides with a low concentration of chlorophyll per unit leaf area (Poorter et af., 1990), which reduces internal shading and therefore increases PNUE (Section IXB). However, is seems likely that the strong positive correlation

234

H. LAMBERS A N D H. POORTER

between RGR and SLA is also partly fortuitous, in that SLA correlates merely in a non-causal manner with a number of other factors which really determine RGR. Clearly, the parameters of eqn (5) cannot provide the full answer to the causes of inherent variation in RGR.

XIV. SPECIES-SPECIFIC PERFORMANCE UNDER SUBOPTIMAL CONDITIONS Up till now we have paid most attention to plants grown under conditions favourable for plant growth. But how do fast- and slow-growing species perform under suboptimal conditions? When grown at a low nutrient concentration in the environment, the RGR of potentially fast-growing species is reduced more than that of slow-growing ones (e.g. Christie and Moorby, 1975; Robinson and Rorison, 1987; Boot and Mensink, I99 I). However, the inherently fast-growing species are still growing faster than slow-growing ones. This seems to be the general trend (Mahmoud and Grime, 1976; Chapin, 1983; Chapin et af., 1983; Berendse and Elberse, 1989; C.A.D.M. van de Vijver, R.G.A. Boot and H. Poorter, unpublished; but see Hommels et a f . , 1990; Muller and Gamier, 1990). This raises the question whether similar results would have been obtained in a situation where a fast-growing species competes with a slow-growing one under nutrient stress. This question will be addressed in Section XVA. The higher RGR of the inherently fast-growing Hofcus fanafus,in comparison with the slow-growing Deschampsiaflexuosa, at a low nutrient supply is explained by differences in LAR (C.A.D.M. van de Vijver, R.G.A. Boot and H. Poorter, unpublished). This is probably true for most comparisons (Christie and Moorby, 1975; van Andel and Biere, 1989). However, there are very few comparative data khowing the cause of the higher RGR of fastgrowing species at low nutrient supply. When grown at a low quantum flux density, fast-growing sun species have a similar RGR to slow-growing shade species (Pons, 1977). In extreme cases, such as that of the fast-growing tropical pioneer tree Cecropia obtusifolia, the RGR declines to a very low rate when plants are grown in the understorey, rather than large gaps. This is due to the decrease in NAR (Popma and Bongers, 1988). The successful performance of most shade species in shaded habitats is likely to be due to a different response of germination, stem elongation and other processes, to light quality, rather than to quantum flux density (cf. Fitter and Ashmore, 1974; Pons, 1977).

PHYSIOLOGY A N D ECOLOGY OF GROWTH RATE VARIATION

235

XV. THE ECOLOGICAL CONSEQUENCES OF VARIATION IN POTENTIAL GROWTH RATE A. What Ecological Advantage can be Conferred by a Plant’s Growth Potential? The ecological advantage of a high RGR seems straightforward: fast growth results in the rapid occupation of a large space, which is advantageous in a situation of competition for limiting resources. A high RGR may also facilitate rapid completion of the life cycle of a plant, which is essential for ruderals. But what is the survival value of slow growth? Grime and Hunt (1975) and Chapin (1980, 1988) mention several possibilities: (i) Slow-growing species make modest demands and will therefore less likely exhaust the available resources, e.g. nutrients. However, this does not seem to be an evolutionary stable strategy, as a neigbouring individual with a faster nutrient uptake could absorb most nutrients (cf. Schulze and Chapin, 1987). Moreover, these modest demands cannot explain slow growth under, e.g. alpine or saline conditions. (ii) Slow-growing species function closer to their optimum than fastgrowing species in an adverse environment. However, the “ecological” optimum of a plant species often differs from its “physiological” optimum. As the physiological optimum of slow-growing species more or less equals that of fast-growing species (Grime and Hunt, 1975), and all plants, especially fast-growing species, have a great ability to adapt to different environmental conditions (Bradshaw, 1965; Grime et a/., 1986; van der Werf, 1992a),we do not expect fast-growing species to be at a disadvantage in such cases. In fact, both in growth analyses (Section XIV) and in short-term competition experiments (Mahmoud and Grime, 1976; Berendse and Elberse, 1989) carried out at a limiting nutrient supply, potentially fast-growing species grow faster and have a greater competitive ability than slow-growing ones. (iii) Slow-growing species incorporate less photosynthates and nutrients into structural biomass and may thus form reserves for later growth, enabling them to maintain physiological integrity during periods which severely restrict growth, e.g. low nutrient availability. However, under adverse conditions growth is restricted before photosynthesis, causing sugars to accumulate (Chapin e t a / . , 1986b; McDonald et a/., 1986). Hence, it is unlikely that survival during periods of nutrient shortage depends on storage of photosynthates. The presence of stored nutrients may indeed buffer fluctuations in nutrient supply in the field. However, perhaps with the exception of phosphate, there is no convincing evidence that slow-growing species accumulate nutrients to a greater extent (Section VIIIA). None the less, slow-growing

236

H. LAMBERS A N D H. POORTER

species may deplete their smaller resources less rapidly, due to their lower RGR. So far, data on the occurrence of fluctuations in nutrient supply and the plant’s response to them are too scarce and conflicting (Grime et al., 1986; Poorter and Lambers, 1986; Campbell and Grime, 1989) to allow the conclusion that an inherently low RGR has survival value in this context. (iv) In a low-resource environment a high growth rate cannot be realized, so a high potential RGR is a selectively neutral trait. However, as noted before, potentially fast-growing species are still growing faster than potentially slow-growing species. Fast-growing species will then attain a larger size (van Andel and Biere, 1989), which has been shown to be advantageous in terms of competition and fitness (Black, 1958; Stanton, 1984). So, although a very high RGR is not attainable, a slightly higher RGR may still be of advantage. We conclude that a low potential growth rate per se does not confer ecological advantage. Why then do slow-growing species occur more frequently in unfavourable habitats than fast-growing ones? An alternative explanation for the observed differences in potential growth rate is that not RGR itself, but rather one of the components linked with RGR, has been the target of selection (Lambers and Dijkstra, 1987; cf. Grime, 1979).

B. Selection for Traits Associated with a Low SLA The most likely trait selected for is SLA, as variation in this trait is closely correlated with that in RGR (Section V; cf. Poorter, 1989; Poorter and Remkes, 1990). In a situation where water or nutrients are limiting, conservation of the scarce resource is at least as important as its capture. Hence, plants under water stress should decrease their transpiration (von Willert et al., 1992). But also for nutrients it has been shown that unproductive species are more successful due to less leaf turnover, so that nutrient losses are restricted (Monk, 1966; Berendse et al., 1987; Karlsson and Nordell, 1987; Aerts and Berendse, 1989). How can turnover be decreased? This depends on the environmental factor which affects leaf longevity. Herbivory can be reduced by increasing leaf toughness (Coley, 1983; Grubb, 1986), accumulating palatability-reducing compounds (Coley, 1987; Waterman and McKey, 1989) and investment in leaf hairs (Woodman and Fernandes, 1991) or thorns. The abrasive effects of high wind speeds can be reduced by investment in fibre (Woodward, 1983; Pammenter et al., 1986). Trampling resistence may be be the result of a large amount of cell wall material per cell (Dijkstra, 1989). Transpiration can be decreased and water use efficiency can be increased by the construction of leaf hairs or epicuticular

PHYSIOLOGY AND ECOLOGY OF GROWTH RATE VARIATION

237

waxes (Ehleringer, 1983; Richards et al., 1986; Ehleringer and Cook, 1990). Epicuticular waxes may also serve a function by decreasing damage by ultraviolet light, preventing contact between rain water and the interior of a leaf and so restricting leaching of nutrients out of a leaf (Mulroy, 1979). Furthermore, waxes may confer disease resistance (Carver et al., 1991) and diminish deleterious effects of salt spray (McNeilly et al., 1987). Each of these additional investments increases the leafs longevity, but decreases SLA with a concomitant decrease in the rate of photosynthesis per unit leaf weight. Consequently, all of these inherent adaptations to adverse conditions diminish the plant’s growth potential, but positively influence its fitness. Is there any indication that plants without these types of adjustment could survive in unfavourable habitats? This would require introduction of plants that only differ in one specific trait in different environments. However, such isogenic genotypes are not available, and variation in one trait could be expected to affect related traits (Section XIIIB). The best ecological information available is that from introduction of foreign species, e.g. the introduction into Venezuela of two African C, species. The introduced species with a high SLA have outcompeted a native C, species, which possesses a low SLA, in relatively fertile places, but not in more infertile habitats (Baruch et al., 1985). On subantarctic islands the introduced grass Agrostis stolonifera, with a high SLA, is able to survive in the wind-sheltered places but is not found outside these shelters, whereas Agrostis magellanica, characterized by a lower SLA due to more sclerenchyma, occurs in the wind-swept parts of these islands (Pammenter et al., 1986). Similarly, Stephanomeria malheurensis, a species with a relatively low SLA which occurs in the same environment as its progenitor S . exigua ssp. coronaria with a higher SLA, is restricted to sites where it may encounter greater stress. The number of individuals of S. exigua ssp. coronaria by far exceeds that of S. malheurensis, though their RGR is very similar (Gottlieb, 1978). Here again, a high SLA appears associated with competive ability and a low SLA with persistence.

C. Selection for Other Traits Underlying RGR It is likely that other traits underlying RGR have also been the target of natural selection. For example, the relatively high RWR of slow-growing species is a cause of their low RGR (Section VIIA). However, RWR is a rather plastic trait, especially in fast-growing species, so that correlations between plant performance and RWR may reflect phenotypic, rather than inherent variation. In so far as a high RWR reflects a greater ability to compete for soilderived resources (Baan Hofman and Ennik, 1980, 1982; Aerts et al., 1991), it may be of advantage in specific environments. Also, if indeed slow-growing

238

H. LAMBERS A N D H. POORTER

species exude specific compounds which effectively release nutrients from a nutrient-poor soil (Section XIA), this trait which reduces RGR may be selected for at nutrient-poor sites. Exudates, like volatiles, may also have allelopathic effects and aid a plant to thrive in an environment amongst competitors. A low nitrogen concentration in the leaf, though reducing the photosynthetic capacity, may add to the leafs longevity by decreasing herbivory (Section VIIIB), and thus the photosynthetic yield during the leafs entire life span may be higher when the leaf nitrogen concentration is relatively low. These patterns of investment, which inexorably reduce the plant’s growth potential, tend to add to a species’ success in nutrient-poor or arid environments where losses due to herbivory severely reduce the plant’s fitness. Thus, they imply a trade-off between growth potential and adaptation to specific habitat features that limit growth.

D. Consequences of a High Growth Potential for Plant Performance in Specific Environments Fast-growing species often grow in competitive situations in a vegetation which develops a high leaf area index during the season. Optimization of the use of nitrogen for photosynthesis and maximization of production then requires the discharge of the oldest, shaded leaves and retranslocation of a part of the N to the youngest, more exposed leaves (Hirose and Werger, 1987; Hirose et al., 1988; Pons et a[., 1989). Under such circumstances, where leaves function only a relatively short period, protection against adverse conditions, requiring a large investment in quantitative secondary compounds, does not lead to an increased photosynthetic return. Consequently, in fairly dense vegetations where light capture is essential and leaf turnover is high, there is a selection pressure for leaves with a high SLA (Schulze, 1982; Poorter, 1989). Also when grown at a limiting nitrogen supply, when the leaf area index is relatively low, fast-growing species tend to have higher rates of leaf turnover (Williamson, 1976). Here it inevitably leads to greater losses (Berendse and Elberse, 1989) and possibly ultimately to the disappearance of the fastgrowing species from such environments. The rapid leaf turnover of fastgrowing species in a dense canopy might depend on the same, as yet poorly understood, mechanism of N-translocation operating at a limiting nitrogen supply (cf. Horgan and Wareing, 1980; Simpson et al., 1982; Kuiper et al., 1988). If correct, then the poor performance of fast-growing plants in a nutrient-poor environment is the consequence of their adaptation to nutrient-rich situations.

PHYSIOLOGY AND ECOLOGY OF GROWTH RATE VARIATION

239

E. A Low Growth Potential and Plant Performance in Adverse Environments, Other than Nutrient-poor Habitats So far, we have concentrated on nutrient-poor habitats, when referring to adverse soil conditions. Although there is quite convincing information that traits associated with a low RGR confer selective advantage in some unproductive environments, it is by no means certain that this is invariably so. Sites that are rich in heavy metals tend to be inhabited by slow-growing ecotypes (Wilson, 1988; Verkleij and Prast, 1989). However, considering the mechanisms involved in heavy-metal tolerance (Verkleij and Schat, 1990), it appears unlikely that such stress tolerance inexorably reduces an ecotype’s RGR. Moreover, some cadmium-tolerant ecotypes of Silene vulgaris (synonymous for S. cucubalus) have very similar RGRs to a sensitive ecotype (Verkleij and Prast, 1989; Verkleij et al., 1990). Possibly, some heavy-metal tolerant ecotypes have evolved in habitats which are not only rich in heavy metals, but nutrient-poor as well, so that their inherently low RGR is not causally related to their stress tolerance. Similarly, there is some evidence that salt-tolerant ecotypes of Beta vulgaris do not necessarily have a lower RGR than sensitive ones (J. Rozema, personal communication), again suggesting that stress-tolerance and a low RGR are not correlated all that tightly.

F. Conclusions We conclude that there are trade-offs between investment in structures that lead to a high growth potential and in structures associated with conservation of nutrients and biomass, when accumulation of large amounts of secondary plant compounds are involved. In so far as nutrient losses are associated with rapid leaf turnover there may well be consequences of efficient functioning in one environment for the performance in another. A final conclusion awaits further information on the regulation of the nitrogen concentration in leaves of fast- and slow-growing species, both as dependent on the nitrogen supply and on the light climate in the canopy. It is likely that there are trade-offs between tolerance of adverse conditions, other than nutrient-poor conditions, and growth potential, but there is no convincing evidence that stress-tolerance and a low RGR are invariably causally related. More information on the mechanisms underlying RGR and stress-tolerance is required before further generalizations can be made.

XVI. CONCLUDING REMARKS AND PERSPECTIVES Generalizing the above leads to suites of traits of a “typical fast-growing” and a “typical slow-growing” plant species (Table 5). Most of these traits

240

H. LAMBERS AND H. POORTER

refer to slow-growing species from nutrient-poor sites. Species from other adverse habitats may also have a lower RGR (Sections I1 and XVE), but much less comparative data are available. The difference with species from more favourable conditions is probably less pronounced and information on special traits of such slow-growing species is scanty. A typically fast-growing species occurs in productive habitats and, under optimum growth conditions, invests heavily in leaf area (high SLA, high LAR), possibly as a result of higher gibberellin production. Although their rate of photosynthesis per unit leaf area is not necessarily higher, fastgrowing species have higher rates of photosynthesis per unit leaf weight. The higher photosynthesis is partly caused by a higher leaf organic nitrogen concentration, partly by variation in the efficiency with which nitrogen is used for photosynthesis (PNUE). One of the likely causes of the low PNUE of slow-growing species is the investment of nitrogen in compounds and structures associated with the protection of their leaves against both biotic and abiotic adverse conditions, but this is unlikely to offer the full explanation. Fast-growing species have higher rates of shoot and root respiration, expressed per unit shoot and root weight, respectively. But their root respiration is not as much higher as to be expected from their much higher rate of growth and nutrient uptake. We hypothesize that slow-growing species constitutively have a rather costly uptake system which is more effective under nutrient-poor conditions. Fast-growing species have a greater ability to adjust their biomass allocation when exposed to nutrient-poor conditions. There is no evidence that the greater plasticity of allocation of biomas and nitrogen of fastgrowing species per se confers any disadvantage under nutrient-limited conditions. However, the leaf longevity of fast-growing species is also shorter and further reduced under nutrient-poor conditions, leading to relatively large losses of nutrients. This inefficient use of nutrients is considered one of the main reasons for the lack of success of inherently fast-growing species in nutrient-poor environment;. Trade-offs between investment in photosynthetic machinery and the degree to which a plant is defended against herbivory, leaf damage due to strong winds, trampling, drought, salt and/or diseases are likely to have occurred, mainly by adaptations which decrease SLA. A number of topics relating to characteristic differences between fastgrowing and slow-growing species still need continued attention. These include the physiological mechanisms determining inherent variation in specific leaf area, biomass allocation, photosynthetic nitrogen use efficiency, root respiration, the importance of gibberellins and the life-span of the different plant organs. There is also a lack of information on the quantitative importance of losses through exudates and volatiles and their association

PHYSIOLOGY AND ECOLOGY OF GROWTH RATE VARIATION

24 1

Table 5 Typical characteristics of fast-growing and slow-growing C, species, summarizing information presented in the text. Unless stated otherwise, the differences refer to plants grown at optimum nutrient supply. A ? indicates that there are indications but no hard data available in the literature Characteristic

Fast-growing Slow-growing species species

Habitat Nutrient supply Productivity

high high

low low

high high higher lower

low low lower higher

high

low

equal

equal

high higher higher higher

low lower lower lower

low low? high high?

high high? low low?

high high high low

low low low high

Morphology and allocation Leaf area ratio Specific leaf area Leaf weight ratio Root weight ratio Investment of nitrogen in leaves (YOof total plant N) Physiology Photosynthesis/leaf area (when species of similar life form are compared) Photosynthesis/leaf weight Shoot respiration/shoot weight Root respiration/root weight Photosynthetic nitrogen use efficiency Respiratory losses (YOof total C fixed) Exudation rate/root weight Ion uptake rate Gibberellin content Chemical composition Nitrogen concentration Concentration of minerals Water content Carbon concentration Concentration of quantitative secondary compounds Concentration of qualitative secondary compounds

low

high

variable

variable

Plasticity with respect to nutrient supply of SLA of allocation of photosynthesis

equal high high

equal low somewhat lower

Other aspects Leaf turnover Root turnover

high

low

?

?

242

H. LAMBERS AND H. POORTER

with nutrient acquisition and interactions between a plant and other organisms. A cost-benefit analysis for symbiotic associations, as dependent on environmental conditions, is needed to evaluate their significance in the acquisition of nutrients. Further evidence is also required to substantiate our belief that biomass allocation of fast-growing species is more plastic with respect to factors other than nutrient supply. If proven correct, information on the regulation of such plasticity and on the ecological significance thereof is needed. Finally, the physiological and ecological costs and benefits of the various inherent adaptations to adverse environments warrant further research. We have attempted to provide a general background on inherent variation in growth rate and to identify major research areas which need further investigation. Such investigations require a combined approach from ecologists, physiologists, biochemists, phytochemists and theoretical biologists. They are bound to yield information which is of great importance for our understanding of the functioning of plants, both in their natural environment and in a crop situation.

ACKNOWLEDGEMENTS We would like to thank all colleagues who generously allowed us to use some of their unpublished data, and the following colleagues for their constructive criticism on (parts of) earlier drafts of this manuscript: Frank Berendse, Arjen Biere, Rene Boot, Marion Cambridge, Heinjo During, Eric Gamier, Henk Konings, Dick Pegtel, Thijs Pons, Jacques Roy, Adrie van der Werf, Marinus Werger and Chin Wong. We thank Marion Cambridge for her linguistic advice.

REFERENCES Adewusi, S.R.A. (1990). Turnover of dhurrin in green sorghum seedlings. Plant Physiol. 94, 1219-1224. Aerts, R. (1989). Nitrogen use efficiency in relation to nitrogen availability and plant community composition. In: Causes and Consequences of Variation in Growth Rate and Productivity of Higher Plants (Ed. by H. Lambers, M.L. Cambridge, H. Konings and T.L. Pons), pp. 285-297. SPB Academic Publishing, The Hague. Aerts, R. and Berendse, F. (1989). Aboveground nutrient turnover and net primary production of an evergreen and a deciduous species in a heathland ecosystem. J . Ecol. 77, 342-356.

PHYSIOLOGY A N D ECOLOGY OF GROWTH RATE VARIATION

243

Aerts, R., Boot, R.G.A. and van der Aart, P.J.M. (1991). The relation between above- and belowground biomass allocation patterns and competitive ability. Oecologia 87, 551-559. Albuzzio, A. and Ferrari, G. (1989). Modulation of the molecular size of humic substances by organic acids of the root exudates. Plant Soil 113, 237-241. van Andel, J. and Biere, A. (1989). Ecological significance of variability in growth rate and plant productivity. In: Causes and Consequences of Variation in Growth Rate and Productivity of Higher Plants (Ed. by H. Lambers, M.L. Cambridge, H. Konings and T.L. Pons), pp. 257-267. SPB Academic Publishing, The Hague. Apel, K., Bohlmann, H. and Reimann-Philipp, U. (1990). Leaf thionins, a novel class of putative defence factors. Physiol. Plant. 80, 3 15-32 1. Ashenden, T.W., Stewart, W.S. and Williams, W. (1975). Growth responses of sand dune populations of Dactylis glomerata L. to different levels of water stress. J . E c o ~63, . 97-107. Atkin, O.K. and Day, D.A. (1990) A comparison of the respiratory processes and growth rates of selected Australian alpine and related lowland plant species. Aust. J . Plant Physiol. 17, 517-526. Atwell, B.J., Veerkamp, M.T., Stuiver, C.E.E. and Kuiper, P.J.C. (1980). The uptake of phosphate by Carex species from oligotrophic to eutrophic swamp habitats. Physiol. Plant. 49, 487494. Baan Hofman, T. and Ennik, G.C. (1980). Investigation into plant characters affecting the competitive ability of perennial ryegrass (Lolium perenne L.). Neth. J . ugric. Sci. 28, 97-109. Baan Hofman, T. and Ennik, G.C. (1982). The effect of root mass of perennial ryegrass (Lolium perenne L.) on the competitive ability with respect to couchgrass (Elytrigia repens (L.) Dew.). Neth. J . agric. Sci. 30, 275-283. Baas, W.J. (1 989). Secondary plant compounds, their ecological significance and consequences for the carbon budget. Introduction of the carbon/nutrient cycle theory. In: Causes and Consequences of Variation in Growth Rate and Productivity of Higher Plants (Ed. by H. Lambers, M.L. Cambridge, H. Konings and T.L. Pons), pp. 313-340. SPB Academic Publishing, The Hague. Ball, M.C. (1988). Salinity tolerance in the mangroves Aegiceras corniculatum and Avicennia marina. I. Water use in relation to growth, carbon partitioning and salt balance. Aust. J . Plant Physiol. 15, 447464. Barrow, N.J. (1977). Phosphorus uptake and utilization by tree seedlings. Aust. J . Bot. 25, 571-584. Baruch, Z . and Smith, A.P. (1979). Morphological and physiological correlates of niche breadth in two species of Espeletia (Compositae) in the Venezuelan Andes. Oecologia 38, 71-82. Baruch, Z . , Ludlow, M.M. and Davis, R. (1985). Photosynthetic responses of native and introduced C, grasses from Venezuelan savannas. Oecologia 67, 388-393. Bate, N.J., Rood, S.R. and Blake, T.J. (1988). Gibberellins and heterosis in poplar. Can. J . Bot. 66, 1148-1152. Berendse, F. and Elberse, W.T. (1989). Competition and nutrient losses from the plant. In: Causes and Consequences of Variation in Growth Rate and Productivity of Higher Plants (Ed. by H. Lambers, M.L. Cambridge, H. Konings and T.L. Pons), pp. 269-284. SPB Academic Publishing, The Hague. Berendse, F., Oudhof, H. and Bol, J. (1987) A comparative study on nutrient cycling in wet heathland ecosystems. I. Litter production and nutrient losses from the plant. Oecologia 74, 174-184.

244

H. LAMBERS A N D H. POORTER

Berendse, F., Bobbink, R. and Rouwenhorst, G . (1989). A comparative study on nutrient cycling in wet heathland ecosystems. 11. Litter decomposition and nutrient mineralization. Oecologia 78, 338-348. Bernays, E.A., Cooper Driver, G. and Bilgener, M. (1989). Herbivores and plant tannins. In: Advances in Ecological Research Vol19 (Ed. by M. Begon, A.H. Fitter, E.D. Ford and A. Macfadyen), pp. 263-302. Academic Press, London. Bjorkman, 0. (1981). Responses to different quantum flux densities. In: Encyclopedia of Plant Physiology Vol. 12A (Ed. by O.L. Lange, P.S. Nobel, C.B. Osmond and H . Ziegler), pp. 57-1 07. Springer-Verlag. Berlin. Black, J.N. (1958). Competition between plants of different initial seed sizes in swards of subterranean clover (Trifolium subterraneum L.) with particular reference to leaf area and the light microclimate. Aust. J . agric. Res. 9, 299-318. Blom-Zandstra, M. and Lampe, J.E.M. (1985). The role of nitrate in the osmoregulation of lettuce (Lactuca sativa L.) grown at different light intensities. J . exp. Bot. 36, 1043-1052. Blom-Zandstra, M., Lampe, J.E.M. and Ammerlaan, H.M. (1988). C and N utilization of two lettuce genotypes during growth under non-varying light conditions and after changing the light intensity. Physiol. Plant. 74, 147-153. Bloom, A. (1985). Wild and cultivated barleys show similar affinities for mineral nitrogen. Oecologia 65, 555-557. Boot, R.G.A. (1989). The significance of size and morphology of root systems for nutrient acquisition and competition. In: Causes and Consequences of Variation in Growth Rate and Productivity of Higher Plants (Ed. by H . Lambers, M.L. Cambridge, H . Konings and T.L. Pons), pp. 299-3 11. SPB Academic Publishing, The Hague. Boot, R.G.A. (1990). Plant Growth as Affected by Nitrogen Supply. P h D thesis, University of Utrecht. Boot, R.G.A. and Mensink, M. (1990). The effect of nitrogen supply on the size and morphology of root systems of perennial grasses from contrasting habitats. Plant Soil 129, 291-299. Boot, R.G.A. and Mensink, M. (1991). The influence of nitrogen availability on growth parameters of fast- and slow-growing perennial grasses. In: Plant Root Growth. An Ecological Perspective, Special Publication of the British Ecological Society. pp. 161-168. (Ed. by D. Atkinson). Blackwell Scientific Publications, London. Boot, R.G.A., Schildwacht, P.M. and Lambers, H. (1992). Partitioning of nitrogen and biomass at a range of N-addition rates and their consequences for growth and gas exchange in two perennial grasses from inland dunes. Physiol. Plant. (in press). Bradshaw, A.D. (1 965). Evolutionary significance of phenotypic plasticity in plants. Adv. Gen. 13, 115-155. Bradshaw, A.D., Chadwick, M.J., Jowett, D. and Snaydon, R.W.D. (1964). Experimental investigations into the mineral nutrition of several grass species. IV. Nitrogen level. J . Ecol. 52, 665-676. Brouwer, R. (1963). Some aspects of the equilibrium between overground and underground plant parts. Mededelingen van het Instituut voor Biologisch en Scheikundig Onderzoek van Landbouwgewassen 213, 3 1-39. Brouwer, R. (1983). Functional equilibrium: sense or nonsense? Neth. J. ugric. Sci. 31, 335-348. Brugnoli, E. and Lauteri, M. (1991). Effects of salinity o n stomata1 conductance, photosynthetic capacity, and carbon isotope discrimination of salt-tolerant (Gossypium hirsutum L.) and salt-sensitive (Phaseolus vulgaris L.) C, non-halophytes. Plant Physiol. 95, 628-635.

PHYSIOLOGY AND ECOLOGY OF GROWTH RATE VARIATION

245

Bryant, J.P., Chapin, F.S. 111 and Klein, D.R. (1983). Carbon/nutrient balance of boreal plants in relation to vertebrate herbivory. Oikos 40,357-368. Bryant, J.P., Clausen, T.P., Reichardt, P.B., McCarthy, M.C. and Werner, R.A. (1987). Effect of nitrogen fertilization upon the secondary chemistry and nutritional value of quaking aspen (Populus tremuloides Michx.) leaves for the large aspen tortrix (Choristoneura conjictana (Walker)). Oecologia 73, 5 13-5 17. Burbott A.J. and Loomis, W.D. (1969). Evidence for metabolic turnover of monoterpenes in peppermint. Plant Physiol. 44,173-179. Callaghan, T.V. and Lewis, M.C. (1971). The growth of Phleum alpinum L. in contrasting habitats at a sub-antarctic station. New Phytof. 70, 1143-1 154. Campbell, B.D. and Grime, J.P. (1989). A comparative study of plant responsiveness to the duration of episodes of mineral nutrient enrichment. New Phytol. 112,261267. Carver, T.L.W., Thomas, B.J., Ingerson-Morris, S.M. and Roderick, H.W. (1991). The role of the abaxial leaf surface waxes of Lolium spp in resistance to Erysiphe graminis. Plant Pathol. 39, 573-583. Ceulemans, R. (1 989). Genetic variation in functional and structural productivity components in Populus. In: Causes and Consequences of Variation in Growth Rate and Productivity of Higher Plants (Ed. by H. Lambers, M.L. Cambridge, H. Konings and T.L. Pons), pp. 69-85. SPB Academic Publishing, The Hague. Chapin, F.S. I11 (1980). The mineral nutrition of wild plants. Ann. Rev. Ecol. System. 11, 233-260. Chapin, F.S. 111 (1983). Adaptation of selected trees and grasses to low availability of phosphorus. Plant Soil 72, 283-287. Chapin, F.S. 111 (1988). Ecological aspects of plant nutrition. Adv. Min. Nutr. 3, 161191. Chapin, F.S. 111 (1989). The costs of tundra plant structures: Evaluation of concepts and currencies. Am. Nut. 133, 1-19. Chapin, F.S. 111 and Bieleski, R.L. (1982). Mild phosphorus stress in barley and a related low-phosphorus-adapted barleygrass: Phosphorus fractions and phosphate absorption in relation to growth. Physiol. Plant. 54, 309-317. Chapin, F.S. 111, Follet, J.M. and O’Connor, K.F. (1982). Growth, phosphate absorption, and phosphorus chemical fractions in two Chionochloa species. J . EcoI. 70, 305-32 1. Chapin, F.S. 111, Tyron, P.R. and van Cleve, K. (1983). Influence of phosphorus supply on growth and biomass allocation of Alaskan taiga tree seedlings. Can. J. Forestry Res. 13, 1092-1098. Chapin, F.S. 111, van Cleve, K. and Tyron, P.R. (1986). Relationship of ion absorption to growth rate in taiga trees. Oecologia 69, 238-242. Chapin, F.S. 111, Shaver, G.R. and Kedrowski, R.A. (1986b). Environmental controls over carbon, nitrogen and phosphorus fractions in Eriophorum in Alaskan tussock tundra. J . Ecol. 74, 167-195. Cheshire, M.V. and Mundie, C.M. (1990). Organic matter contributed to soil by plant roots during the growth and decomposition of maize. Plant Soil 121, 107-1 14. Chou, C.-H and Kuo, Y.-L. (1986). Alleopathic research of subtropical vegetation in Taiwan. Alleopathic exclusion of understorey by Leucaena leucophylla (Lam.) de Wit. J . chem. Ecol. 12, 1431-1448. Christie, E.K. and Moorby, J. (1975). Physiological responses of semi-arid grasses. I. The influence of phosphorus supply on growth and phosphorus absorption. Aust. J . agric. Res. 26, 423-436. Clarkson, D.T. (1985). Factors affecting mineral nutrient acquisition by plants. Ann. Rev. Plant Physiol. 26, 77-1 15.

246

H. LAMBERS AND H. POORTER

Clarkson, D.T. (1986). Regulation of the absorption and release of nitrate by plant cells: A review of current ideas and methodology. In: Fundamental, Ecological and Agricultural Aspects of Nitrogen Metabolism in Higher Plants (Ed. by H. Lambers, J.J. Neeteson and I . Stulen), pp. 3-27. Martinus Nijhof/Dr W. Junk, The Hague. Clarkson, D.T. and Scattergood, C.B. (1982). Growth and phosphate transport in barley and tomato plants during the development of, and recovery from, phosphate-stress. J . exp. Bot. 33, 865-875. Coley, P.D. (1983). Herbivory and defensive characteristics of tree species in a lowland tropical forest. Ecol. Mon. 53, 209-233. Coley, P.D. (1986). Costs and benefits of defense by tannins in a neotropical tree. Oecologia 70, 238-241. Coley, P.D. ( 1 987). Interspecific variation in plant anti-herbivore properties: The role of habitat quality and rate of disturbance. New Phytol. 106 (suppl.), 251-263. Coley, P.D. and Aide, T.M. (1989). Red coloration of tropical young leaves: a possible antifungal defence? J . trop. Ecol. 5, 293-300. Coley, P.D., Bryant, J.P. and Chapin, F.S. 111. (1985). Resource availability and plant antiherbivory defense. Science 230, 895-899. Collier, D. and Cummins, W.R. (1989). A field study on the respiration rates in the leaves of temperate plants. Can. J . Bot. 67, 3478-3481. Cook, M.G. and Evans, L.T. (1983). Some physiological aspects of the domestication and improvement of rice (Oryza ssp.). Field Crops Res. 6, 219-238. Corre, W.J. (1983a). Growth and morphogenesis of sun and shade plants. I. The influence of light intensity. Acta Bot. Neerlandica 32, 49-62. Corre, W.J. (1983b). Growth and morphogenesis of sun and shade plants. 11. The influence of light quality. Acta Bot. Neerlandica 32, 185-202. Corre, W.J. ( 1 9 8 3 ~ ) Growth . and morphogenesis of sun and shade plants. 111. The combined'effects of light intensity and nutrient supply. Acta Bot. Neerlandica 32, 277-294. Cowan, I.R. (1977). Water use in higher plants. In: Water, Planets, Plants and People (Ed. by A.K. McIntyre), pp. 71-107. Australian Academy of Science, Canberra. Crick, J.C. and Grime, J.P. (1987). Morphological plasticity and mineral nutrients capture in two herbaceous species of contrasting ecology. New Phytol. 107, 403414. Day, D.A. and Lambers, H. (1983). The regulation of glycolysis and electron transport in roots. Physiol. Plant. 58, 155-1 60. Deane-Drummond, C.E. and Glass, A.D.M. (1983). Short term studies of nitrate uptake into barley plants using ion-specific electrodes and 16CIO;I. Control of net uptake by NO; efflux. Plant Physiol. 73, 100-104. Dell, B. and McComb, A.J.' (1978). Plant resins: Their formation, secretion and possible functions. Adv. bot. Res. 6, 277-316. Denslow, J.S., Vitousek, P.M. and Schultz, J.C. (1987). Bioassays of nutrient limitation in a tropical rain forest soil. Oecologia 74, 370-376. Dicke, M. and Sabelis, M.W. (1989). Does it pay to advertise for body guards? In: Causes and Consequences of Variation in Growth Rate and Productivity of Higher Plants (Ed. by H. Lambers, M.L. Cambridge, H. Konings and T.L. Pons), pp. 341-358. SPB Academic Publishing, The Hague. van de Dijk, S.J. (1980). Two ecologically distinct subspecies of Hypochaeris radicata L. 11. Growth response to nitrate and ammonium, growth strategy and formative aspects. Plant Soil 57, 1 1 1-122. Dijkstra, P. (1989). Cause and effect of differences in specific leaf area. In: Causes and Consequences of Variation in Growth Rate and Productivity of Higher Plants (Ed.

PHYSIOLOGY A N D ECOLOGY OF GROWTH RATE VARIATION

247

by H. Lambers, M.L. Cambridge, H. Konings and T.L. Pons), pp. 125-140. SPB Academic Publishing, The Hague. Dijkstra, P. and Kuiper, P.J.C. (1989). Effects of exogenously applied growth regulators on shoot growth of inbred lines of Plantago major differing in relative growth rate: differential response to gibberellic acid and (2-chloroethyl-)trimethylammonium chloride. Physiol. Plant. 77, 5 12-51 8. Dijkstra, P. and Lambers, H . (1989a). Analysis of specific leaf area and photosynthesis of two inbred lines of Plantago major differing in relative growth rate. New Phytol. 113, 283-290. Dijkstra, P. and Lambers, H. (1989b). A physiological analysis of genetic variation in relative growth rate within Plantago major. Func. Ecol. 3, 577-585. Dijkstra, P., ter Reegen, H. and Kuiper, P.J.C. (1990). Relation between relative growth rate, endogenous gibberellins, and the response to applied gibberellic acid for Plantago major. Physiol. Plant. 79, 629-634. Dinkelaker, B., Romheld, V. and Marschner, H . (1989). Citric acid excretion and precipitation of calcium citrate in the rhizosphere of white lupin. Plant Cell Environ. 12, 285-292. Dittrich, W. (1931). Zur Physiologie des Nitratumsatzes in hoheren Pflanzen (unter besonderer Berucksichtigung der Nitratspeicherung). Planta 12, 69-1 19. Doddema, H. and Otten, H. (1979). Uptake of nitrate by mutants of Arabidopsis thaliana, disturbed in uptake or reduction of nitrate. 111. Regulation. Physiol. Plant. 45, 339-346. Drew, M.C. and Saker, L.R. (1978). Nutrient supply and the growth of the seminal root system in barley. 111. Compensatory increase in growth of lateral roots, and in rates of phosphate uptake, in response to a localized supply of phosphate. J . exp. Bot. 29, 435451. Drew, M.C., Saker, L.R. and Ashley, T.W. (1973). Nutrient supply and the growth of the seminal root system in barley. I. The effect of nitrate concentration on the growth of axes and laterals. J . exp. Bot. 24, 1189-1202. Duarte, P., Oscarson, P., Tillberg, J.-E. and Larsson, C.-M. (1988). Nitrogen and carbon utilization in shoots and roots of nitrogen-limited Pisum. Plant Soil 111, 241-244. Ehleringer, J. (1983). Characterization of a glabrate Encelia ,/urinosa mutant: morphology, ecophysiology, and field observations. Oecologia 57, 303-3 10. Ehleringer, J.R. and Cook, C.S. (1984). Photosynthesis in Encelia farinosa Gray in response to decreasing leaf water potential. Plant Physiol. 75, 688-693. Ehleringer, J.R. and Cook, C.S. (1990). Characteristics of Encelia species differing in leaf reflectance and transpiration rate under common garden conditions. Oecologiu 82, 484-489. Ehleringer, J.R. and Cooper, T.A. (1988). Correlations between carbon isotope ratio and microhabitat in desert plants. Oecologia 76, 562-566. Eissenstat, D.M. and Caldwell, M.M. (1987). Characterization of successful competitors: an evaluation of potential growth rate in two cold desert tussock grasses. Oecologia 71, 167-173. Elias, C.O. and Chadwick, M.J. (1979). Growth characteristics of grass and legume cultivars and their potential for land reclamation. J . appl. Ecol. 16, 537-544. Ellenberg, H.E. (1979). Zeigerwerte der Gefasspflanze Mitteleuropas. Scripta Geobotanica 9. Enyi, B.A.C. (1962). Comparative growth-rates of upland and swamp rice varieties. Ann. Bot. 26, 467487. Esau, K. (1977). Anatomy of Seed Plants. John Wiley, New York.

248

H. LAMBERS A N D H . POORTER

Esquerre-Tugave, M.T., Lafitte, C., Mazau, D., Toppan, A. and Touze, A. (1979). Cell surfaces in plant microorganism interactions 11. Evidence for the accumulation of hydroxyproline-rich glycoproteins in the cell wall of diseased plants as a defense mechanism. Plant Physiol. 64,32&326. Evans, G.C. ( 1 972). The Quantitative Analysis of’ Plant Growth. Blackwell Scientific Publications, Oxford. Evans, J.R. (1983). Nitrogen and photosynthesis in the flag leaf of wheat (Triticum aestivum L.). Plant Physiol. 72, 291-302. Evans, J.R. (1989a). Photosynthesis and nitrogen relationships in leaves of C, plants. Oecologia 78, 9-19. Evans, J.R. (1989b). Photosynthesis-the dependence on nitrogen partitioning. In: Causes and Consequences of’ Variation in Growth Rate and Productivity of Higher Plants (Ed. by H. Lambers, M.L. Cambridge, H. Konings and T.L. Pons), pp. 159-174. SPB Academic Publishing, The Hague. Evans, J.R. and Terashima, I . (1988). Photosynthetic characteristics of spinach leaves grown with different nitrogen treatments. Plant Cell Physiol. 29, 157-165. Eze, J.M.O. (1973). The vegetative growth of Helianthus annuus and Phaseolus vulgaris as affected by seasonal factors in Freetown, Sierra Leone. Ann. Bot. 37. 3 15-329. Farquhar, G.D. and Sharkey, T.D. (1982). Stomata and photosynthesis. Ann. Rev. Plant Physiol. 33, 3 17-345. Farquhar, G.D.,O’Leary, M.H. and Berry, J.A. (1982). On the relationship between carbon isotope discrimination and the intercellular carbon dioxide concentration in leaves. Aust. J . Plant Physiol. 11, 539-552. Farrar, J.F. (1989). The carbon balance of fast-growing and slow-growing species. In: Causes and Consequences of Variation in Growth Rate and Productivity of Higher Plants (Ed. by H. Lambers, M.L. Cambridge, H. Konings and T.L. Pons), pp. 241-256. SPB Academic Publishing, The Hague. Fenner. M. (1978). A comparison of the abilities of colonizers and closed-turf species to establish from seed in artificial swards. J . Ecol. 66, 953-963. Fenner, M. (1983). Relationships between seed weight, ash content and seedling growth in twenty-four species of Compositae. New Phytol. 95, 697-706. Field, C. and Mooney, H.A. (1986) The photosynthesis-nitrogen relationship in wild plants. In: On the Economy of Plant Form and Function (Ed. by T.J. Givnish), pp. 25-55. Cambridge University Press, Cambridge. Field, C., Merino, J. and Mooney, H.A. (1983) Compromises between water-use efficiency and nitrogen-use efficiency in five species of California evergreens. Oecologia 60, 384-389. Fitter, A.H. (1985). Functional significance of root morphology and root system architecture. In: Ecological Interactions in Soil. British Ecological Society Special Publication (Ed. by A. H. Fitter, D. Atkinson, D.J. Read and M.B. Usher), pp. 87106. Blackwell Scientific Publications, Oxford. Fitter, A.H. (1987). An architectural approach to the comparative ecology of plant root systems. New Phytol. 106, 61-77. Fitter, A.H. and Ashmore, C.J. (1974). Response of two Veronica species to a simulated woodland light climate. New Phytol. 73, 997-1001. Fitter, A.H., Nichols, R. and Harvey, M.L. (1988). Root system architecture in relation to life history and nutrient supply. Funct. Ecol. 2, 345-351. Fohse, D. and Jungk, A. (1983). Influence of phosphate and nitrate supply on root hair formation of rape, spinach and tomato plants. Plant Soil 74, 359-368. Franceschi. V.R., Wittenbach, V.A. and Giaquinta, R.T. (1983). Paraveinal mesophyll of soybean leaves in relation to assimilate transfer and compartmentation.

PHYSIOLOGY AND ECOLOGY OF GROWTH RATE VARIATION

249

111. Immunohistochemical localization of specific glycopeptides in the vacuole after depodding. Plant Physiol. 72, 586-589. Frehner, M. and Conn, E.E. (1987). The linamarin p-glucosidase in Costa Rican wild lima beans (Phaseolus lunatus L.) is apoplastic. Plant Physiol. 84, 129&1300. Freijsen, A.H.J. and Otten, H. (1984). The effect of nitrate concentration in a flowing solution system on growth and nitrate uptake of two Plantago species. Plant Soil 77, 159-169.

Galston, A.W. and Sawhney, R.K. (1990). Polyamines in plant physiology. Plant Physiol. 94, 4 0 M 1 0 . Gardner, W.K., Parbery, D.G. and Barber, D.A. (1982a). The acquisition of phosphorus by Lupinus albus L. I. Some characteristics of the soil/root interface. Plant Soil 68, 19-32. Gardner, W.K., Parbery, D.G. and Barber, D.A. (1982b). The acquisition of phosphorus by Lupinus albus L. 11. The effects of varying phosphors supply and soil type on some characteristics of the soil/root interface. Plant Soil 68, 3341. Gardner, W.K., Parbery, D.G. and Barber, D.A. (1983). The acquisition of phosphorus by Lupinus albus L. 111. The probable mechanism by which phosphorus movement in the soil/root interface is enhanced. Plant Soil 70, 107-124. Garnier, E. (1991). Above and below-ground resource capture in herbaceous plants: Relationships with growth and biomass allocation. Trends Ecol. Evol. 6 , 126-13 1. Garnier, E., Koch, G.W., Roy, J. and Mooney, H.A. (1989). Responses of wild plants to nitrate availability. Relationships between growth rate and nitrate uptake parameters, a case study with two Bromus species, and a survey. Oecologia 79, 542-5 50.

Gershenzon, J. (1984). Changes in the levels of plant secondary metabolites under water and nutrient stress. In: Phytochemical Adaptations to Stress (Ed. by B.N. Timmermann, C. Steelink and F.A. Loewus). Recent Advances in Phytochemistry Vol. 18, pp. 273-320, Plenum, New York. Gershenzon, J., Maffei, M. and Croteau, R. (1989). Biochemical and histochemical localization of monoterpene biosynthesis in the glandular trichomes of spearmint (Mentha spicata). Plant Physiol. 89, 1351-1357. Ghashghaie, J. and Saugier, B. (1989). Effects of nitrogen deficiency on leaf photosynthetic response of tall fescue to water deficit. Plant Celf Environ. 12, 261271.

Givnish, T.J. (1986). Biomechanical constraints on crown geometry in forest herbs. In: On the Economy of Plant Form and Function (Ed. by T.J. Givnish), pp. 525583. Cambridge University Press, Cambridge. Glass, A.D.M. (1983). Regulation of ion transport. Ann. Rev. Plant Physiol. 34, 31 1326.

Glass, A.D.M., Siddiqi, M.Y., Ruth, T.J. and Rufry, T.W., Jr (1990). Studies of the uptake of nitrate in barley. 11. Energetics. Plant Physiol. 93, 1585-1589. Gottlieb, L.D. (1978). Allocation, growth rates and gas exchange in seedlings of Stephanomeria exigua ssp. coronaria and its recent derivative S. malheurensis. Am. J . Bot. 65, 97&977. Graham, J.H., Leonard, R.T. and Menge, J.A. (1981). Membrane-mediated decrease in root exudation responsible for phosphorus inhibition of vescicular-arbuscular mycorrhyza formation. Plant Physiol. 68, 548-552. Granato, T.C., Raper, C.D. Jr. and Wilkerson, G.G. (1989). Respiration rate in maize roots is related to concentration of reduced nitrogen and proliferation of lateral roots. Physiol. Plant. 76, 419-424. Gray, J.T. and Schlesinger, W.H. (1983). Nutrient use by evergreen and deciduous shrubs in Southern California. 11. Experimental investigations in the relationship

250

H. LAMBERS AND H. POORTER

between growth, nitrogen uptake and nitrogen availability. J . Ecol. 71, 43-56. Gregory, P.J. and Atwell, B.J. (1991). The fate of carbon in pulse-labelled crops of barley and wheat. Plant and Soil, 126, 205-213. Grime, J.P. ( 1979). Plant Strategies and Vegetation Processes. Wiley, Chichester. Grime, J.P. and Hunt, R. (1975). Relative growth rate: Its range and adaptive significance in a local flora. J . Ecol. 63, 393422. Grime, J.P.. Crick, J.C. and Rincon, J.E. (1986). The ecological significance of plasticity. In: Plasticity in Plants (Ed. by D.H. Jennings and A.J. Trewawas), pp. 5-30. The Company of Biologists, Cambridge. Grime, J.P., Hodgson. J.G. and Hunt, R. (1988). Comparative Plant Ecology. Unwin Hyman, London. Grime, J.P., Hall, W., Hunt, R., Neal, A.M., Ross-Frases, W. and Sutton, F. (1989). A new development of the temperature-gradient tunnel. Ann. Bot. 64, 279-287. Grime, J.P., Campbell, B.D., Mackey, J.M.L. and Crick, J.C. (1991). Root plasticity, nitrogen capture and competitive ability. In: Plant Root Growth. An Ecological Perspective. Special Publication of the British Ecological Society, (Ed. by D. Atkinson). Gross, K.L. (1984). Effects of seed size and growth form on seedling establishment of six monocarpic perennial plants. J . Ecol. 72, 369-387. Grubb, P.J. (1986). Sclerophylls, pachyphylls and pycnophylls: The nature and significance of hard leaf surfaces. In: Insects and Plant Surface (Ed. by B. Juniper and R. Southwood), pp. 137-150. Edward Arnold, London. Haider, K., Mosier, A. and Heinemeyer, 0. (1987). The effect of growing plants on denitrification at high soil nitrate concentrations. Soil Sci. SOC.Am. J . 51, 97-102. Harborne, J.B. (1982). Introduction to Ecological Biochemistry. Academic Press, London. Hardwick, R.C. (1984). Some recent developments in growth analysis-A review. Ann. Bot. 54, 807-812. Harrison, A.F. and Helliwell, D.R. (1979). A bioassay for comparing phosphorus availability in soils. J . appl. Ecol. 16, 497-505. Hartmann, T.. Ehmke, A,, Eilert, U., von Bortsel, K. and Theurig, C. (1989). Sites of synthesis, translocation and accumulation of pyrrolizidine alkaloid N-oxides in Senecio vulgaris L. Planta 177, 98-107. Hartwig, U.A., Maxwell, C.A., Joseph, C.M. and Phillips, D.A. (1990). Chrysoeriol and luteolin released from alfafa seeds induce nod genes in Rhizobium meliloti. Plant Physiol. 92, 116122. Helal, H.M. and Sauerbeck, D. (1986). Effect of plant roots on carbon metabolism of soil microbial biomass. Z . ‘Planzenernuhr. Bodenk 149, 181-188. Hirose. T. and Werger, M.J.A. (1987) Nitrogen use efficiency in instantaneous and daily photosynthesis of leaves in the canopy of a Solidago altissima stand. Physiol. Plant. 70, 2 15-222. Hirose. T., Werger, M.J.A., Pons, T.L. and van Rhenen, J.W.A. (1988). Canopy structure and leaf nitrogen distribution in a stand of Lvsimachia vulgaris L. as influenced by stand density. Oecologia 77, 145-150. Hirose, T., Werger, M.J.A. and van Rhenen, J.W.A. (1989). Canopy development and leaf nitrogen distribution in a stand of Carex acutiformis. Ecology 70, 16101618. Hoffland, E., Findenegg, G.R. and Nelemans, J.A. (1989a). Solubilization of rock phosphate by rape. I. Evaluation of the role of the nutrient uptake pattern. Plant and Soil 113, 155-160. Hoffland. E., Findenegg, G.R. and Nelemans, J.A. (1989b). Solubilization of rock

PHYSIOLOGY AND ECOLOGY OF GROWTH RATE VARIATION 25 1 phosphate by rape. 11. Local root exudation of organic acids as a response to Pstarvation. Plant and Soil 113, 161-165. Hoffland, E., Findenegg, G.R., Leffelaar, P.A. and Nelemans, J.A. (1990). Use of a simulation model to quantify the amount of phosphate released from rock phosphate by rape. Transactions ofthe 14th International Congress qf Soil Science (Kyoto) 11, 170-175. Hommels, C.H., Kuiper, P.J.C. and Telkamp, G.P. (1990). Study on nutrient stress tolerance of nine Taraxarum microspecies with a contrasting mineral ecology by cultivation in a low nutrient flowing solution. Physiol. Plant. 79, 389-399. Horgan, J.M. and Wareing, P.F. (1980). Cytokinins and the growth responses of seedlings of Betula pendula Roth and Acer pseudoplatanus L. to nitrogen and phosphorus deficiency. J. exp. Bot. 31, 525-532. Horsman, D.C., Nicholls, A.O. and Calder, D.M. (1980). Growth response of Dactylis glomerata, Lolium perenne and Phalaris ayuatica to chronic ozone exposure. Aust. J . Plant Physiol. 7, 51 1-517. Hubick, K. and Farquhar, G.D. (1989). Carbon isotope discrimination and the ratio of carbon gained to water lost in barley cultivars. Plant. Cell Environ. 12, 795-804. Hughes, A.P. and Cockshull, K.E. (1971). The variation in response to light intensity and carbon dioxide concentration shown by two cultivars of Chrysanthemum morifolium grown in controlled environments at two times of year. Ann. Bot. 35, 933-945. Hunt, R., Nicholls, A.O. and Fathy, S.A. (1987). Growth and root-shoot partitioning in eighteen British grasses. Oikos 50,53-59. Ingestad, T. (1981). Nutrition and growth of birch and grey alder seedlings in low conductivity solutions and at varied relative rates of nutrient addition. Physiol. Plant. 52, 454-466. Jackson, R.B. and Caldwell, M.M. (1989). The timing and degree of root proliferation in fertile-soil microsites for three cold-desert perennials. Oecologia 81, 149153. Jackson, R.B., Manwaring, J.H. and Caldwell, M.M. (1990). Rapid physiological adjustment of roots to localized soil enrichment. Nature 344, 58-60. Jackson, W.A., Kwik, K.D. and Volk, R.J. (1976). Nitrate uptake during recovery from nitrogen deficiency. Plant Physiol. 36, 174-1 8 1. de Jager, A. (1982). Effects of a localized supply of H,PO,, NO,, SO,, Ca and K on the production and distribution of dry matter in young maize plants. Neth. J . agric. Sci.30, 193-203. de Jager, A. and Posno, M. (1979). A comparison of the reaction to a localized supply of phosphate in Plantago major, Plantago lanceolata and Plantago media. Acta Bot. Neerlandica 28, 479489. Janzen, H.H. and Radder, G.D. (1989). Nitrogen mineralization in a green manureamended soil as influenced by cropping history and subsequent crop. Plant Soil 120, 125-131. Johnson, N.D., Liu, B. and Bentley, B.L. (1987). The effects of nitrogen fixation, soil nitrate, and defoliation on the growth, alkaloids, and nitrogen levels of Lupinus succulentus (Fabaceae). Oecologia 74, 4 2 5 4 3 1 . Jones, R.L. and Robinson, D.G. (1989). Protein secretion in plants. New Phytol. 111, 567-597. Jupe, S.C., Causton, D.R. and Scott, I.M. (1988). Cellular basis of the effects of gibberellin and the pro gene on stem growth in tomato. Planta 174, 106-1 11. Kakes, P. (1987). On the polymorphism for cyanogenesis in natural populations of Trzfolium repens L. in The Netherlands. I . Distribution of the genes Ac and Li. Acta Bot. Neerlandica 36, 59-69.

252

H. LAMBERS AND H. POORTER

Karlsson, P.S. (1991). Intraspecific variation in photosynthetic nitrogen utilization in the mountain birch, Betula pubescens ssp. tortuosa. Oikos 60, 49-54. Karlsson, P.S. and Nordell, K.O. (1987). Growth of Betula pubescens and Pinus sylvestris seedlings in a subarctic environment. Funct. Ecol. 1, 3 7 4 . Khan, M.A. and Tsunoda, S. (1970a). Growth analysis of cultivated wheat species and their wild relatives with special reference to dry matter distribution among different plant organs and to leaf area expansion. Tohoku J . Agric. Res. 21,47-59. Khan, M.A. and Tsunoda, S. (1970b). Growth analysis of six commercially cultivated wheats of West Pakistan with special reference to a semi-dwarf modern wheat variety, mexi-pak. Tohoku J . Agric. Res. 21, 6&72. Kimmerer, T.W. and Potter, D.A. (1987). Nutritional quality of specific leaf tissues and selective feeding by a specialist leafminer. Oecologia 71, 548-551. Kleis-San Francisco, S.M. and Tierney, M.L. (1990). Isolation and characterization of a proline-rich cell wall protein from soybean seedlings. Plant Physiol. 94, 18971902. Konings, H. (1989). Physiological and morphological differences between plants with a high NAR or a high LAR as related to environmental conditions. In: Causes and Consequences of Variation in Growth Rate and Productivity of Higher Plants (Ed. by H. Lambers, M.L. Cambridge, H. Konings and T.L. Pons), pp. 101-123. SPB Academic Publishing, The Hague. Konings, H., Koot, E. and Tijman-de Wolf, A. (1989). Growth characteristics, nutrient allocation and photosynthesis of Carex species from floating fens. Oecologia 80, 1 1 1- 121. Koornneef, M., Bosma, T.D.G., Hanhart, C.J., van der Veen, J.H. and Zeevaart, J.A.D. (1990). Isolation and characterization of gibberellin-deficient mutants in tomato. Theoret. App. Genetics 80, 852-857. Korner, C. and Diemer, M. (1987). In situ photosynthetic responses to light, temperature and carbon dioxide in herbaceous plants from low and high altitude. Funct. Ecol. 1, 179-194. Korner, C. and Pelaez Menendez-Riedl, S. (1989). The significance of developmental aspects in plant growth analysis. In: Causes and Consequences of Variation in Growth Rate and Productivity of Higher Plants (Ed. by H. Lambers, M.L. Cambridge, H. Konings and T.L. Pons), pp. 141-157. SPB Academic Publishing, The Hague. Korner, C., Farquhar, G.D. and Roksandic, Z. (1988). A global survey of carbon isotope discrimination in plants from high altitude. Oecologia 74, 623432. Kuehn, G.D., Rodriguez-Garay, B., Bagga, S. and Phillips, G.C. (1990). Novel occurrence of uncommon polyamines in higher plants. Plant Physiol. 94,855-857. Kuiper, D. and Smid, A. (1985). Genetic differentiation and phenotypic plasticity in Plantago major ssp. major: I. The effect of differences in level of irradiance on growth, photosynthesis, respiration and chlorophyll content. Physiol. Plant. 65, 520-528. Kuiper, D., Kuiper, P.J.C., Lambers, H., Schuit, J.T. and Staal, M. (1988). Cytokinin contents in relation to mineral nutrition and benzyladenine addition in Plantago major ssp. pleiosperma. Physiol. Plant. 7 5 , 51 1-517. Kuiters, A.T. (1990). Role of phenolic substances from decomposing forest litter in plant-soil interactions. Acta Bot. Neerlandica 39, 329-348. Lambers, H. (1983). The “functional equilibrium”: Nibbling on the edges of a paradigm. Neth. J . agric. Sci. 31, 305-31 1. Lambers, H. (1985). Respiration in intact plants and tissues: Its regulation and dependence on environmental factors, metabolism and invaded organisms. In:

PHYSIOLOGY A N D ECOLOGY OF GROWTH RATE VARIATION

253

Encyclopedia ?/Plant Physiology. Vol. I8 (Ed. by R. Douce and D.A. Day), pp. 41 8 4 7 3 . Springer-Verlag, Berlin. Lambers, H. ( 1 987). Growth, respiration, exudation and symbiotic associations: The fate of carbon translocated to the roots. In: Root Development and FunctionEfects of the Physical Environment (Ed. by P.J. Gregory. J.V. Lake and D.A. Rose), pp. 125-145. Cambridge University Press, Cambridge. Lambers, H. and Dijkstra, P. (1987). A physiological analysis of genetic variation in relative growth rate: Can growth rate confer ecological advantage? In: Disturbance in Grasslands. Causes, Effects, and Processes (Ed. by J. van Andel, J.P. Bakker and R.W. Snaydon), pp. 237-252. Dr W Junk Publishers, Dordrecht. Lambers, H. and Posthumus, F. (1980). The effects of light intensity and relative humidity on growth rate and root respiration of Plantago lanceolata and Zea mays. J . exp. Bot. 31, 1621-1630. Lambers, H. and Rychter, A. (1989). The biochemical background of variation in respiration rate: respiratory pathways and chemical composition. In: Causes and Consequences of Variation in Growth Rate and Productivity of Higher Plants (Ed. by H. Lambers, M.L. Cambridge, H. Konings, and T.L. Pons), pp. 199-225. SPB Academic Publishing, The Hague. Lambers, H., Posthumus, F., Stulen, I., Lanting, L., van de Dijk, S.J. and Hofstra, R. (1981a). Energy metabolism of Plantago major major as dependent on the supply of nutrients. Physiol. Plant. 51, 245-252. Lambers, H.. Blacquiere, T. and Stuiver, C.E.E. (I981 b). Interactions between osmoregulation and the alternative respiratory pathway in Plantago coronopus as affected by salinity. Physiol. Plant. 51, 63-68. Lambers. H., Simpson. R.J., Beilharz, V.C. and Dalling, M.J. (1982). Translocation and utilization of carbon in wheat (Triticum aestivum). Physiol. Plant. 56, 18-22. Lambers, H.. Freijsen, N., Poorter, H., Hirose, T. and van der Werf, A. (1989). Analyses of growth based on net assimilation rate and nitrogen productivity. In: Causes and Consequences of Variation in Growth Rate and Productivitv of Higher Plants (Ed. by H . Lambers, M.L. Cambridge, H. Konings and T.L. Pons), pp. 117. SPB Academic Publishing, The Hague. Lamont, B. (1982). Mechanisms for enhancing nutrient uptake in plants, with particular reference to mediterranean South Africa and Western Australia. Bot. Rev. 48, 597-689. Lamport, D.T. (1980). Structure and function of plant glycoproteins. In: The Biochemistry of Plants, Vol. 3. Carbohydrates (Ed. by J. Preis), pp. 501-541. Academic Press, New York. Lamport, D.T. and Catt, J.W. (1981). Glycoproteins and enzymes of the cell wall. In: Encyclopedia of Plant Physiology, V o l I 3 B (Ed. by W. Tanner and F.A. Loewus), pp. 133-165. Springer-Verlag, Berlin. Langenheim, J.H., Osmond, C.B., Brooks, A. and Ferrar, P.J. (1984). Photosynthetic responses to light in seedlings of selected Amazonian and Australian rainforest tree species. Oecologia 63, 2 15-224. Lee, R.B. (1982). Selectivity and kinetics of ion uptake by barley plants following nutrient deficiency. Ann. Bot. 50, 4 2 9 4 9 . Liljeroth, E., Baath, E.. Mathiasson, I. and Lundborg, T. (1990). Root exudation and rhizoplane bacterial abundance of barley (Hordeum vulgare L.) in relatior. to nitrogen fertilization and root growth. Plant Soil 127, 81-89. Lipton, D.S., Blanchar, R.W. and Blevins, D.G. (1987). Citrate, malate and succinate concentrations in exudates from P-sufficient and P-stressed Medicago sativa L. seedlings. Plant Physiol. 85, 3 15-3 17.

254

H. LAMBERS AND H. POORTER

Lolkema, P.C., Donker, M.H., Schouten, A.J. and Ernst, W.H.O. (1984). The possible role of metallothioneins in copper tolerance of Silene cucubales. Planta 162, 174-179. Lynch, J.M. and Whipps, J.M. (1990). Substrate flow in the rhizosphere. Plant Soil 129, 1-10. McClure, P.R., Kochian, L.V., Spanswick, R.M. and Shaff, J.E. (1990). Evidence for cotransport of nitrate and protons in maize roots. 11. Measurements of NO; and H fluxes with ion-selective microelectrodes. Plant Physiol. 93, 290-294. McCully, M.E. and Canny. M.J. (1985). Localisation of translocated I4C roots of field-grown maize. Physiol. Plant. 65, 380-392. McDonald, A.J.S., Lohammer, T. and Ericsson, A. (1986). Growth response to step decrease in nutrient availability in small birch (Betula pendula Roth). Plant, Cell Environ. 9, 427432. McDougall, B.M. (1970). Movement of 14C-photoassimilateinto the roots of wheat seedlings and exudation from the roots. New Phytol. 69, 3 7 4 6 . McNeilly, T., Ashraf, M. and Veltkamp, C. (1987). Leaf micromorphology of sea cliff and inland plants of Agrostis stolonifera L., Dactylis glomerata L. and Holcus lunatus L. New Phytol. 106, 261-269. Mahmoud, A. and Grime, J.P. (1976). An analysis of competitive ability in three perennial grasses. New Phytol. 77, 431435. Makino, A., Mae, T. and Ohira, K. (1988). Differences between wheat and rice in the enzymic properties of ribulose- 1, 5-bisphosphate carboxylase/oxygenase and the relationship to photosynthetic gas exchange. Planta 174, 3&38. Margna, U., Margna, E. and Vainjarv, T. (1989). Influence of nitrogen nutrition on the utilization of L-phenylalanine for building flavonoids in buckwheat seedlings. J . Plant Physiol. 134, 697-702. Marschner, H., Treeby, M. and Romheld, V. (1989). Role of root-induced changes in the rhizosphere for iron acquisition in higher plants. Z . Pflanzenernahr. Bodenk. 152, 197-204. Merino, J., Field, C. and Mooney, H.A. (1984). Construction and maintenance costs of Mediterranean-climate evergreen and deciduous leaves. Acta Oecol./Oecol. Plant. 5, 21 1-229. Metraux, J.-P. (1987). Gibberellins and plant cell elongation. In: Plant Hormones and Their Role in Plant Growth and Development (Ed. by P.J. Davies), pp. 296-317. Martinus Nijhoff, Dordrecht. Metzger, J.D. and Hassebrock, A.T. (1990). Selection and characterization of a gibberellin-deficient mutant of Thlaspi arvense L. Plant Physiol. 94, 1655-1662. Mole, S. and Waterman, P.G. (1988). Light-induced variation in phenolic levels in foliage of rain-forest plants. 11. Potential significance to herbivores. J . chem. Ecol. 14, 23-34. Monk, C.D. 1966. An ecological significance of evergreenness. Ecology 47, 504-505. Monson, R.K. and Fall, R. (1989). Isoprene emission from aspen leaves. Influence of environment and relation to photosynthesis and photorespiration. Plant Physiol. 90,267-274. Mooney, H.A. and Chu, C. (1983). Stomata1 responses to humidity of coastal and interior populations of a Californian shrub. Oecologia 57, 148-1 50. Mooney, H.A., Ferrar, P.J. and Slatyer, R.O. (1978). Photosynthetic capacity and carbon allocation patterns in diverse growth forms of Eucalyptus. Oecologia 36, 103-1 11. Mooney, H.A., Field, C., Gulmon, S.L., Rundel, P. and Kruger, F.J. (1983). Photosynthetic characteristics of South African sclerophylls. Oecologia 58, 398401. +

PHYSIOLOGY AND ECOLOGY OF GROWTH RATE VARIATION

255

Morgan, J.A. (1986) The effects of N-nutrition on the water relations and gas exchange characteristics of wheat (Triticum aestivum L.). Plant Physiol. 80, 52-58. Morrow, P.A. and Fox, L.R. (1980). Effects of variation in Eucalyptus essential oil yield on insect growth and grazing damage. Oecologia 45, 209-219. Muller, B. and Gamier, E. (1990). Components of relative growth rate and sensitivity to nitrogen availability in annual and perennial species of Bromus. Oecologia 84, 5 1 3-5 18. Mulroy, T.W. (1979). Spectral properties of heavily glaucous and non-glaucous leaves of a succulent rosette plant. Oecologia 38, 349-357. Nassery, H. (1970). Phosphate absorption by plants from habitats of different phosphate status. New Phytol. 69, 197-203. Newman, E.I. (1985). The rhizosphere: carbon sources and microbial populations. In: Ecological Interactions in Soil. Special Publication No. 4 of the British Ecological Society (Ed. by A.H. Fitter), pp. 107-121. Blackwell Scientific Publications, Oxford. Nicolai, V. (1988). Phenolic and mineral content of leaves influences decomposition in European forest ecosystems. Oecologia 75, 575-579. Niemann, G.J., Pureveen, J.B.M., Eijkel, G.B., Poorter, H. and Boon, J.J. (1992). Differences in relative growth rate in 11 grasses correlate with differences in chemical composition as determined by mass spectrometry. Oecologia 89, 567573. Oberbauer. S.F., Strain, B.R. and Fetcher, N. (1985). Effect of C0,-enrichment on seedling physiology and growth of two tropical tree species. Physiol. Plant. 65, 352-356. Oscarson, P., Ingemarsson, B., af Ugglas, M and Larsson, C.-M. (1987). Short-term studies of NO; uptake in Pisum using "NO;. Planta 170, 55&555. Oscarson, P., Ingemarsson, B. and Larsson, C.-M. (1989). Growth and nitrate uptake properties of plants grown at different relative addition rates of nitrogen supply. 11. Activity and affinity of the nitrate uptake system in Pisum and Lemna in relation to nitrogen availability and nitrogen demand. Plant, Cell Environ. 12, 787-794. Ovington, J.D. and Olson, J.S. (1970). Biomass and chemical content of El Verde lower montane rain forest plants. In: A Tropical Rainforest (Ed. by H.T. Odum and R.F. Pigeon), pp. h53-h77. USEAC, Oak Ridge. Pammenter, N.W., Drennan, P.M. and Smith, V.R. (1986). Physiological and anatomical aspects of photosynthesis of two Agrostis species at a sub-antarctic island. New Phytol. 102, 143-160. Parkhurst, D.F. and Mott, K.A. (1990). Intercellular diffusion limits to CO, uptake in leaves. Studies in air and helox. Plant Physiol. 94, 1024-1032. Parkhurst, D.F., Wong, S.C., Farquhar, G.D. and Cowan, I.R. (1988). Gradients of intercellular CO, levels across the leaf mesophyll. Plant Physiol. 86, 1032-1037. Pavlik, B.M. (1983). Nutrient and productivity relations of the dune grasses Ammophila arenaria and Elymus mollis. I. Blade photosynthesis and nitrogen use efficiency in the laboratory and field. Oecologia 57, 227-232. Pearson, C.J., Volk, R.J. and Jackson, W.A. (1981). Daily changes in nitrate influx, efflux and metabolism in maize and pearl millet. Planta 152, 319-324. Penning de Vries, F.W.T. (1975). The cost of maintenance processes in plant cells. Ann. Bot. 39, 77-92. Penning de Vries, F.W.T., Brunsting, A.H.M. and van Laar, H.H. (1974). Products, requirements and efficiency of biosynthetic processes: a quantitative approach. J . theoret. Biol. 45, 339-377.

256

H. LAMBERS A N D H. POORTER

Penning de Vries, F.W.T., van Laar, H.H. and Chardon, M.C.M. (1983). Bioenergetics of growth of seeds, fruits, and storage organs. In: Proc. Symp. Potential Productivity of Field Crops under Different Environments, pp. 37-59. International Rice Research Institute, Manila. Philipson, J.J. and Coutts, M.P. (1977). The influence of mineral nutrition on the root development of trees. 11. The effect of specific nutrient elements on the growth of individual roots of sitka spruce. J . exp. Bot. 105, 864-871. Plaut, Z., Mayoral, M.L. and Reinhold, L. (1987). Effect of altered sink:source ratio on photosynthetic metabolism of source leaves. Plant Physiol. 85, 786791. Pons, T.L. (1977). An ecophysiological study in the field layer of ash coppice. 11. Experiments with Geum urbanum and Cirsium palustre in different light intensities. Acta Bot. Neerlandica 26, 2942. Pons, T.L., Schieving, F., Hirose, T. and Werger, M.J.A. (1989). Optimization of leaf nitrogen allocation for canopy photosynthesis in Lysimachia vulgaris. In: Causes and Consequences of Variation in Growth Rate and Productivity of Higher Plants (Ed. by H. Lambers, M.L. Cambridge, H. Konings and T.L. Pons), pp. 175-186. SPB Academic Publishing, The Hague. Poorter, H . (1989). Interspecific variation in relative growth rate: On ecological causes and physiological consequences. In: Causes and Consequences of Variation in Growth Rate and Productivity of Higher Plants (Ed. by H. Lambers, M.L. Cambridge, H. Konings and T.L. Pons), pp. 45-68. SPB Academic Publishing, The Hague. Poorter, H. (1991). Interspecijic Variation in the Relative Growth Rate of Plants: The Underlying Mechanisms. PhD thesis, University of Utrecht. Poorter, H. and Bergkotte, M. (1992). Chemical composition of 24 wild species differing in relative growth rate. Plant. Cell Environ. 15, 221-229. Poorter, H. and Lambers, H. (1986). Growth and competitive ability of a highly plastic and a marginally plastic genotype of Plantago major in a fluctuating environment. Physiol. Plant. 67, 21 7-222. Poorter, H. and Lambers, H. (1991). Is interspecific variation in relative growth rate positively correlated with biomass allocation. Am. Nut. 138, 1264-1 268. Poorter, H. and Remkes, C. (1990). Leaf area ratio and net assimilation rate of 24 wild species differing in relative growth rate. Oecologia 83, 553-559. Poorter, H., Remkes, C. and Lambers, H. (1990). Carbon and nitrogen economy of 24 wild species differing in relative growth rate. Plant Physiol. 94, 621-627. Poorter, H., van der Werf, A., Atkin, O.K. and Lambers, H. (1991). Respiratory energy requirements of roots vary with the potential growth rate of a plant species. Physiol. Plant. 83, 46947:. Popma, J. and Bongers, F. (1988). The effects of canopy gaps on growth and morphology of seedlings of rain forest species. Oecologia 75, 625-632. Potter, D.A. and Kimmerer, T.W. (1989). Inhibition of herbivory on young holly leaves: evidence for the defensive role of saponins. Oecologia 78, 322-329. Potter, J.R. and Jones, J.W. (1977). Leaf area partitioning as an important factor in plant growth. Plant Physiol. 59, 1&14. Poulton, J.E. (1990). Cyanogenesis in plants. Plant Physiol. 94, 401405. Powell, C.L. (1974). Effect of P-fertilizer on root morphology and P-uptake of Carex coriacea. Plant Soil 41, 661-667. Prudhomme, T.I. (1983). Carbon allocation to antiherbivore compounds in a deciduous and an evergreen subarctic shrub species. Oikos 40, 344356. Putnam, A. and Tang, C.-S. (Eds) (1986). The Science of Allelopathy. John Wiley, New York.

PHYSIOLOGY AND ECOLOGY OF GROWTH RATE VARIATION

257

Rawson, H.M., Gardner, P.A. and Long, M.J. (1987). Sources of variation in specific leaf area in wheat grown at high temperature. Aust. J. Plant Physiol. 14,287-298. Ratnayaka, M., Leonard, R.T. and Menge, J.A. (1978). Root exudation in relation to supply of phosphorus and its possible relevance to mycorrhizal formation. New Phytol. 81, 543-552. Reimann-Philipp, U., Schrader, G., Martinoia, E., Barkholt, V. and Apel, K. (1989). Intracellular thionins of barley. A second group of leaf thionins closely related to but distinct from cell wall-bound thionins. J . biol. Chem. 264, 8978-8984. Reinink, K . , Groenwold, R. and Bootsma, A. (1987). Genotypical differences in nitrate content in Lactuca sativa L. and related species and correlation with dry matter content. Euphytica 36, 11-18, Rhoades, D.F. (1985). Offensive4efensive interactions between herbivores and plants: their relevance in herbivore population dynamics and ecological theory. Amer. Natural. 125, 205-238. Richards, R.A., Rawson, H.M. and Johnson, D.A. (1986). Glaucousness in wheat: Its development and effect on water-use efficiency, gas exchange and photosynthetic tissue temperatures. Aust. J. Plant Physiol. 13, 465473. Richardson, A.E., Djordjevic, M.A., Rolfe, B.G. and Simpson, R.J. (1988). Effects of pH, Ca and Al on the exudation from clover seedlings of compounds that induce the expression of nodulation genes in Rhizobium trifolii. Plant Soil 109, 3747. Robinson, D. and Rorison, I.H. (1985). A quantitative analysis of the relationships between root distribution and nitrogen uptake from soil by two grass species. J . Soil Sci. 36, 71-85. Robinson, D. and Rorison, I.H. (1987). Root hairs and plant growth at low nitrogen availabilities. New Phytol. 107, 68 1-693. Robinson, D. and Rorison, I.H. (1988). Plasticity in grass species in relation to nitrogen supply. Funct. Ecol. 2, 249-257. Robinson, N.J. and Jackson, P.J. (1986). “Metallothinein-like” metal complexes in angiosperms; their structure and function. Physiol. Plant. 67, 499-506. Rodgers, C.O. and Barneix, A.J. (1988). Cultivar differences in the rate of nitrate uptake by intact wheat plants as related to growth rate. Physiol. Plant. 72, 121126. Romheld, V. (1987). Different strategies for iron acquisition in higher plants. Physiol. Plant. 70, 231-234. Romheld, V. and Marschner, H. (1986). Evidence for a specific uptake system for iron phytosiderophores in roots of grasses. Plant Physiol. 80, 175-1 80. Romheld, V. and Marschner, H. (1990). Genotypical differences among gramineceous species in release of phytosidereophores and uptake of iron phytosiderophores. Plant Soil 123, 147-153. Rood, S.B. and Pharis, R.P. (1987). Hormones and heterosis in plants. In: Plant Hormones and Their Role in Plant Growth and Development (Ed. by P.J. Davies), pp. 463473. Martinus Nijhoff, Dordrecht. Rood, S.B., Pharis, R.P., Koshioka, M. and Major, D.J. (1983). Gibberellin and heterosis in maize. I. Endogenous gibberellin-like substances. Plant Physiol. 71, 639444. Rood, S.B., Buzzell, R.I., Major, D.J. and Pharis, R.P. (1990a). Gibberellins and heterosis in maize: Quantitative relationships. Crop Sci. 30, 28 1-286. Rood, S.B., Zanewich, K.P. and Bray, D.F. (1990b). Growth and development of Brassica genotypes differing in endogenous gibberellin content. 11. Gibberellin content, growth analyses and cell size. Physiol. Plant. 79, 679-685.

258

H. LAMBERS A N D H. POORTER

Rood, S.B., Pearce, D.P., Williams, P.H. and Pharis, R.P. (1990~).A. Gibberellindeficient Brassica mutant-rosette. Plant Physiol. 89, 482437. Rorison, I.H. (1968). The response of phosphorus of some ecologically distinct plant species. I. Growth rates and phosphorus absorption. New Phytol. 67, 913-923. Rozijn, N.A.M.G. and van der Werf, D.C. (1986). Effect of drought during different stages in the life cycle on the growth and biomass of two Aira species. J . Ecol. 74, 507-523. Rufty, T.W., Huber, S.C. and Volk, R.J. (1988). Alterations in leaf carbohydrate metabolism in response to nitrogen stress. Plant Physiol. 88, 725-730. Saab, I.N., Sharp, R.E., Pritchard, J. and Voetberg, G.S. (1990). Increased endogenous abscisic acid maintains primary root growth and inhibits shoot growth of maize seedlings at low water potentials. Plant Physiol. 93, 1329-1 336. Sage, R.F. and Pearcy, R.W. (1987). The nitrogen use efficiency of C, and C, plants. 11. Leaf nitrogen effects on the gas exchange characteristics of Chenopodium album and Amaranthus retrojexus (L.). Plant Physiol. 84, 959-963. Sage, R.F. and Sharkey, T.D. (1987). The effect of temperature on the occurrence of 0, and CO, insensitive photosynthesis in field grown plant. Plant Physiol. 84,658664. Salvucci, M.E. (1989). Regulation of Rubisco activity in vivo. Physiol. Plant. 77, 164171. Sanadze, G.A. ( 1969). Light-dependent excretion of molecular isoprene. Progress in Photosynthesis Research 2, 701-706. Schmidt, J.E. and Kaiser, W.M. (1987). Response of the succulent leaves of Peperomia magnoliaefolia to dehydration. Water relations and solute movement in chlorenchyma and hydrenchyma. Plant Physiol. 83, 19C194. Schulze, E.-D. (1982). Plant life forms and their carbon, water and nutrient relations. In: Encyclopedia of Plant Physiology Vol. 12B (Ed. by O.L. Lange, P.S. Nobel, C.B. Osmond and H. Ziegler), pp. 61 5-676. Springer-Verlag, Berlin. Schulze, E.-D. and Chapin, F.S. 111. (1987). Plant specialization to environments of different resource availability. In: Potentials and Limitations of Ecosystem Analysis (Ed. by E.-D. Schulze and Zwolfer), pp. 120-148. Springer-Verlag, Berlin. Schulze, E.-D., Turner, N.C., Gollan, T. and Shackel, K.A. (1987). Stomatal responses, to air humidity and to soil drought. In: Stomatal Function (Ed. by E. Zeiger, G.D. Farquhar and I.R. Cowan), pp. 31 1-321. Stanford University Press, Stanford. Seemann, J.R. (1989). Light adaptation/acclimation of photosynthesis and the regulation of ribulose- I ,5-bisphosphate carboxylase activity in sun and shade plants. Plant Physiol. 91, 379-386. Seemann, J.R., Sharkey, T.D., Wang, J.L. and Osmond, C.B. (1987). Environmental effects on photosynthesis, nitrogen-use efficiency, and metabolite pools in leaves of sun and shade plants. Plant Physiol. 84, 796802. Servaites, J.C. (1990). Inhibition of ribulose 1,5-bisphosphate carboxylase/oxygenase by 2-carboxyarabinitol- 1-phosphate. Plant Physiol. 92, 867-870. Sharkey, T.D., Stitt, M., Heineke, D., Gerhardt, R., Raschke, K. and Heldt, H.W. (1986). Limitation of photosynthesis by carbon metabolism. 11. C0,-insensitive CO, uptake results from limitation of triose phosphate utilization. Plant Physiol. 81, 1123-1 129. Shipley, B. and Peters, R.H. (1990). A test of the Tilman model of plant strategies: Relative growth rate and biomass partitioning. Am. Nut. 136, 139-153. Sibly, R.M. and Grime, J.P. (1986). Strategies of resource capture by plantsEvidence for adversity selection. J . theoret. Biol. 118, 247-250.

PHYSIOLOGY AND ECOLOGY OF GROWTH RATE VARIATION

259

Siddiqi, M.Y., Glass, A.D.M., Ruth, T.J. and Fernando, M. (1990). Studies on the regulation of nitrate influx by barley seedlings using I3NO;. Plant Physiol. 5 6 , 8 0 6 813. Simpson, R.J., Lambers, H. and Dalling, M.J. (1982). Kinetin application to roots and its effects on uptake, translocation and distribution of nitrogen in wheat (Triticum aestivum) grown with a split root system. Physiol. Plant. 56, 43W35. Smedley, M.P., Dawson, T.E., Comstock, J.P., Donovan, L.A., Sherrill, D.E., Cook, C.S. and Ehleringer, J.R. (1991). Seasonal carbon isotope discrimination in a grassland community. Oecologia 85, 3 14-320. Smirnoff, N. and Stewart, G.R. (1985). Nitrate assimilation and translocation by higher plants: Comparative physiology and ecological consequences. Physiol. Plant. 64, 133-140. Smith, R.I.L. and Walton, D.W.H. (1975). A growth analysis technique for assessing habitat severity in tundra regions. Ann. Bot. 39, 831-843. Sparling, G.P., Cheshire, M.V. and Mundie, C.M. (1982). Effect of barley plants on the decomposition of I4C-labelledsoil organic matter. J . Soil Sci. 33, 89-100. Spitters, C.J.T. and Kramer, T. (1986). Differences between spring wheat cultivars in early growth. Euphytica 35, 273-292. Stanton, M.L. (1984). Seed variation in wild radish: Effect of seed size on components of seedling and adult fitness. Ecology 65, 1105-1 112. Staswick, P.E. (1988). Soybean vegetative storage protein structure and gene expression. Plant Physiol. 87, 25CL254. Stienstra, A.W. (1986). Does nitrate play a role in osmoregulation? In: Fundamental, Ecological and Agricultural Aspects of Nitrogen Metabolism in Higher Plants (Ed. by H. Lambers, J.J. Neeteson and I. Stulen), pp. 481484. Martinus Nijhof/Dr W. Junk, The Hague. Stulen, I., Lanting, L., Lambers, H., Posthumus, F., van de Dijk, S.J. and Hofstra, R. (1981). Nitrogen metabolism qf Plantago major L. as dependent on the supply of mineral nutrients. Physiol. Plant. 52, 108-1 14. Taper, M.L. and Case, T.J. (1987). Interactions between oak tannins and parasite community structure: Unexpected benefits of tannins to cynipid gall-wasps. Oecologia 71, 254-261. Tester, M., Smith, S.E. and Smith, F.A. (1987). The phenomenon of “nonmycorrhizal” plants. Can. J . Bot. 65, 419431. Thompson, K. (1987). Seeds and seed banks. New Phytol. 106 (suppl), 23-34. Tilman, D. and Cowan, M.L. (1989). Growth of old field herbs on a nitrogen gradient. Funct. Ecol. 3, 425438. Tingey, D.T., Evans, R.C. and Gumpertz, M.L. (1981). Effects of environmental conditions on isoprene emissions from live oak. PIanta 152, 565-570. Tingey, D.T., Evans, R.C., Bates, E.H. and Gumpertz, M.L. (1987). Isoprene emissions and photosynthesis in three ferns-The influence of light and temperature. Physiol. Plant. 69, 609-616. Tognoni, F., Halevy, A.H. and Wittwer, S.H. (1967). Growth of bean and tomato plants as affected by root absorbed growth substances and atmospheric carbon dioxide. Planta 72, 43-52. Tsunoda, S. (1959). A developmental analysis of yielding ability in varieties of field crops. I. Leaf area per plant and leaf area ratio. Jap. J . Breeding 9, 161-168. Tukey, H.B. (1970). The leaching of substances from plants. Ann. Rev. Plant Physiol. 21, 305-324. Veen, B.W. and Kleinendorst, A. (1986). The role of nitrate in osmoregulation of Italian ryegrass. In: Fundamental. Ecological and Agricultural Aspects of Nitrogen

260

H. LAMBERS A N D H. POORTER

Metabolism in Higher Plants (Ed. by H. Lambers, J.J. Neeteson and I. Stulen), pp. 477480. Martinus Nijhof/Dr W. Junk, The Hague. van Veen, J.A., Merckx, R. and van de Geijn, S.C. (1989). Plant- and soil related controls of the flow of carbon from roots through the soil microbial biomass. Plant Soil 115, 179-188. Verkleij, J.A.C. and Prast, J.E. (1989). Cadmium tolerance and co-tolerance in Silene vulgaris (Moench.) Garcke ( = S. cucubalus (L.) Wib.). New Phytol. 111, 637-645. Verkleij, J.A.C. and Schat, H. (1990). Mechanisms of metal tolerance in higher plants. In: Heavy Metal Tolerance in Plants (Ed. by A.J. Shaw), pp. 179-193. CRC Press Inc., Boca Raton. Verkleij, J.A.C., Koevoets, P., van 't Riet, J., Bank, R., Nijdam, Y. and Ernst, W.H.O. (1990). Poly(y-glutamylcysteinyl)glycines or phytochelatines and their role in cadmium tolerance of Silene vulgaris. Plant, Cell Environ. 13, 913-921. Vertregt, N. and Penning de Vries, F.W.T. (1987). A rapid method for determining the efficiency of biosynthesis of plant biomass. J. theoret. Biol. 128, 109-1 19. Vitousek, P.M., Matson, P.A. and van Cleve, K. (1989). Nitrogen availability and nitrification during succession: Primary, secondary, and old field seres. Plant Soil 115, 229-239. Waller, G.R. and Nowacki, E.K. (1978). Alkaloids: Biology and Metabolism. Plenum Press, New York. Waring, R.H., McDonald, A.J.S., Larsson, S., Ericsson, T., Wiren, A,, Arwidsson, E. et al. (1985). Differences in chemical composition of plants grown at constant relative growth rates with stable mineral nutrition. Oecologia 66, 157-160. Warren Wilson, J. (1966). An analysis of plant growth and its control in arctic environments. Ann. Bot. 30, 384-402. Waterman, P.G. and McKey, D. (1989). Herbivory and secondary compounds in rain-forest plants. In: Tropical Rain Forest Ecosystems (Ed. by H. Lieth and M.J.A. Werger), pp. 513-536. Elsevier, Amsterdam. Welch, M.B. (1920). Eucalyptus oil glands. J. Proc. R. SOC.N.S. W. 54, 208-217. van der Werf, A,, Kooijman, A., Welschen, R. and Lambers, H. (1988). Respiratory costs for the maintenance of biomass, for growth and for ion uptake in roots of Carex diandra and Carex acutiformis. Physiol. Plant. 72, 4 8 3 4 9 1. van der Werf, A., Hirose, T. and Lambers, H. (1989). Variation in root respiration; Causes and consequences for growth. In: Causes and Consequences of Variation in Growth Rate and Productivity of Higher Plants (Ed. by H. Lambers, M.L. Cambridge, H. Konings and T.L. Pons), pp. 227-240. SPB Academic Publishing, The Hague. van der Werf, A., Visser, A.J., Schieving, F. and Lambers, H. (1992a). Evidence for optimal partitioning of biomass and nitrogen at a range of nitrogen availabilities for a fast- and slow-growing species. Funct. Ecol. (in press). van der Werf, A., Welschen, R. and Lambers, H. (1992b). Respiratory losses increase with decreasing inherent growth rate of a species and with decreasing nitrate supply: A search for explanations for these observations. In: Molecular, Biochemical and Physiological Aspects of Plant Respiration (Ed. by H. Lambers, H. and L.H.W. Van der Plas). SPB Academic Publishing, The Hague (in press). van der Werf, A., van den Berg, G., Ravenstein, H.J.L., Lambers, H. and Eking, R. (1 992c). Protein turnover: A significant component of maintenance respiration in roots? In: Molecular, Biochemical and Physiological Aspects of Plant Respiration (Ed. by H. Lambers, H. and L.H.W. Van der Plas). SPB Academic Publishing, The Hague (in press). Werner, P., Rebele, F. and Bornkamm, R. (1982). Effects of light intensity and light

PHYSIOLOGY A N D ECOLOGY OF GROWTH RATE VARIATION

26 1

quality on the growth of the shadow plant Lamium galeobdolon (L.) Crantz and the half-shadow plant Stellaria holostea L. Flora 172, 235-249. Whitehouse, D.G., Fricaud, A.-C. and Moore, A.L. (1989). Role of nonohmicity in the regulation of electron transport in plant mitochondria. Plant Physiol. 91,487492. von Willert, D.J., Eller, B.M., Werger, M.J.A., Brinckmann, E. and Ihlenfeldt, H.-D. (1992). Life Strategies of Succulents in Deserts with Special Reference to the Namib Desert. Cambridge University Press, Cambridge. Williamson, P. (1976). Above-ground primary production of chalk grassland allowing for leaf death. J . Ecol. 64, 1059-1075. Wilson, J.B. (1988). The cost of heavy-metal tolerance: An example. Evolution 42, 4 0 8 4 1 3. Woldendorp, J.W. (1963). The influence of living plants on denitrification. Mededelingen Landbouwhogeschool, Wageningen 63, 1-100. Woodman, R.L. and Fernandes, G.W. (1991). Differential mechanical defense: herbivory, evapotranspiration, and leaf-hairs. Oikos 60, 1 1-19. Woodward, F.I. (1979). The differential temperature responses of the growth of certain plant species from different altitudes. I. Growth analysis of Phleum alpinum L., P. bertolonii D.C., Sesleria albicans Kit. and Dactylisglomerata L. New Phytol. 82, 385-395. Woodward, F.I. (1983). The significance of interspecific differences in specific leaf area to the growth of selected herbaceous species from different altitudes. New Phytol. 95, 3 13-323. Wyn Jones, R.G. and Gorham, J. (1983). Osmoregulation. In: Encyclopedia of Plant Physiology, Vol 12C (Ed. by O.L. Lange, P.S. Nobel, C.B. Osmond and H. Ziegler), pp. 35-58. Springer-Verlag, Berlin. Young, D.R. and Smith, W.K. (1980). Influence of sunlight on photosynthesis, water relations and leaf structure in the understory species Arnica cordifolia. Ecology 61, 1380-1 390. Zanewich, K.P., Rood, S.B. and Williams, P.H. (1990). Growth and development of Brassica genotypes differing in endogenous gibberellin content. 11. Leaf and reproductive development. Physiol. Plant. 79, 673-678. Zhang, F., Romheld, V. and Marschner, H. (1989). Effects of zinc deficiency in wheat on the release of zinc and iron mobilizing root exudates. Z. fflanzener. Bodenk. 152, 205-210.

This Page Intentionally Left Blank

Positive-feedback Switches in Plant Communities J . BASTOW WILSON and ANDREW D . Q . AGNEW I . Summary . . . . . . . . . . . . . . . I1. Introduction . . . . . . . . . . . . . A . Switches . . . . . . . . . . . . . B. Types of Switch . . . . . . . . . . C . Boundaries . . . . . . . . . . . . D . Vegetational Situations Produced by Switches E . Agencies . . . . . . . . . . . . . 111. Water-mediated Switches . . . . . . . . . A . Concept . . . . . . . . . . . . . B. Fog Precipitation . . . . . . . . . . C . Infiltration . . . . . . . . . . . . D . Sediment Entrapment: Salt Marsh Pans . . E . Ombrogenous Bog Growth . . . . . . F . Snow Accumulation . . . . . . . . . IV . pH-mediated Switches . . . . . . . . . . V . Soil-element-mediated Switches . . . . . . . A . NPK Increase . . . . . . . . . . . B . NPK Decrease . . . . . . . . . . . C . Heavy Metals . . . . . . . . . . . D . Salt . . . . . . . . . . . . . . . VI . Light-mediated Switches . . . . . . . . . VII . Temperature-mediated Switches . . . . . . A . Concept . . . . . . . . . . . . . B . Treeline . . . . . . . . . . . . . C . Graminoid Tussocks . . . . . . . . . VIII . Wind-mediated Switches . . . . . . . . . A . Concept . . . . . . . . . . . . . B . Soil Erosion and Trapping . . . . . . . C . Wind Damage to Plants . . . . . . . IX . Fire-mediated Switches . . . . . . . . . A . Concept . . . . . . . . . . . . . B . Australian Closed-forest/Savannah . . . . C . African Closed-forestisavannah . . . . . D . Conclusion . . . . . . . . . . . . X . Allelopathy-mediated Switches . . . . . . .

. . . .

. . . . . .

. . . . .

. . . . .

. . . . .

. . . .

. . . . . .

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

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

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

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

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

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

264 265 265 267 267 268 270 271 271 271 275 276 278 280 280 281 281 282 283 286 288 290 290 290 292 292 292 292 294 296 296 291 300 301 301

Copyright 0 1992 A c a a h i c Press Limired A / / rights of reproduction in any form reserved

ADVANCES IN ECOLOGICAL RESEARCH VOL . 23 ISBN &12413923-5

263

264

J. B. WILSON AND A. D. Q. AGNEW

XI. Microbe-mediated Switches . . . . . . . . . A . Oldfield Succession and Nitrogen-fixing Microbes B. Forests and Mycorrhizas . . . . . . . . XII. Termite-mediated Switches . . . . . . . . . XIII. Herbivore-mediated Switches . . . . . . . . A. Concept . . . . . . . . . . . . . . B. Grass/Grass Boundary . . . . . . . . . C. Grass/Woodland Boundary . . . . . . . D. Grazing and Nitrogen Cycling . . . . . . E. Insects in Pine Forests . . . . . . . . . F. Conclusion . . . . . . . . . . . . . XIV. Discussion . . . . . . . . . . . . . . . . Acknowledgements . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

.

.

.

.

.

. . . . . . . . . . .

.

.

.

.

. . . . .

.

.

.

. . . .

.

303 303 305 306 308 308 310 3 11 31 1 312 312 313 321 32 1

I. SUMMARY A vegetation positive-feedback switch (or simply “switch”) is a process in which a community modifies the environment, making it more suitable for that community. The term “switch” emphasizes that, from an initial state, such positive feedback can switch the development of vegetation and environment between alternative pathways. These pathways can diverge into alternative stable states. For example, in arid areas with frequent mist, tall vegetation can cause fog precipitation. The resulting increase in water input can enable tall vegetation to grow, which increases fog precipitation further. We suggest the switch is an important and general vegetational process. Positive-feedback switches can operate by modifying any of several features of the environment, including water, pH, soil elements, light, temperature, wind, fire, allelopathic toxins, microbial or termite population, or herbivore activity. Possible examples are given of each. We distinguish between four types of switch: (1) one-sided switch, where a single community modifies the environment of the patches it occupies; (2) reaction switch, where the community additionally modifies the patches it is not in, but in the opposite direction, e.g. by redirecting wind or intercepting water runoff; (3) symmetric switch, where communities of both alternative states modify the same factor of their environment, but in opposite directions; (4) two-factor switch, where the two communities both modify their environments, but in different factors. We see positive-feedback switches producing four major vegetational effects (A-D): a stable vegetational mosaic may be produced in a previously

POSITIVE-FEEDBACK SWITCHES IN PLANT COMMUNITIES

265

uniform environment (situation A), or a vegetational gradient caused by environmental change can be intensified to give a sharp boundary (situation B). These mosaics and boundaries can occur at a wide variety of spatial scales, from landscape-scale to individual plant-scale. Switches can also sharpen or displace temporal boundaries: succession can be accelerated (situation C) or delayed (situation D). Not all of these effects can be produced by all types of switch; in particular, a one-sided (type 1) switch cannot produce a stable mosaic. We have been able to find many examples of positive-feedback switches, often not recognized as switches by the original authors. Perhaps this lack of recognition has been because of the pervasive effect of Clements’ facilitation theory of succession, and of Watt’s theory of cyclic succession, both suggesting that plants change the environment making it less suitable for themselves. Our examples of switches all involve some degree of speculation, but the evidence for many is at least as good as that for Clements’ facilitation succession. We suggest that switches are probably common.

11. INTRODUCTION

A. Switches It is widely accepted that plants modify their environment (e.g. Miles, 1985). It is often assumed, as in Clements’ theory of facilitation succession, that the effect of this modification is to facilitate their replacement by other species. Logically, plants might as frequently modify their environment to their own advantage. This gives the possibility of positive feedback, by which initial differences in vegetation are magnified and stabilized. We refer to this process as a “switch”, following Odum (1971), because a small change in biota or environment can switch between alternative vegetation/environment states. Sometimes, we use the longer form “positive-feedback switch” for clarity, but the term is really tautological, for all switches must be based on positive feedback. An ecological switch therefore means that two (or more) vegetation/ environment states are stable in time or space, but not the intermediates. The mechanism for this is that the vegetation changes its own physical or biological environment, each vegetation state making its environment more suitable for itself and/or less suitable for the other vegetation state. Thus, from initially similar conditions, sites diverge in vegetation/environment by a process of positive feedback (Fig. 1). Thus, the basic elements of such a switch are: (i) Community X changes the environment,

266

J . B. WILSON AND A. D. Q. AGNEW

(b) wan’s cyclic succession

(a) Clements’ facilitation succession

Key:

-

(c) Switch

=

Community

=

Change through time (in the absence of disturbance or extrinsic environmental change)

Fig. 1. Vegetation changes through time by three suggested processes: (a) Clements’ facilitation succession, (b) Watt’s cyclic succession, (c) a switch.

(ii) This change is relatively more favourable for community X than for community Y. The process is the opposite of Clements’ (e.g. 1936) facilitation/relayfloristics model of succession (Fig. I ) . In Clements’ understanding of succession, a community modifies the environment, making it less favourable for itsevand more favourable for another community (the next sera1 stage), and the vegetation/environment of initially dissimilar sites converge to the same climax. Besides the spatial sharpening effects of positive-feedback switches, temporal effects can also occur: switches can either accelerate or delay the speed of vegetational change, which can lead to temporal sharpening. We suggest switches are the mechanism underlying many vegetation patterns and processes. Switch processes and situations have been referred to, using a variety of terms, by previous workers. Barkman (1990) pointed out that some species can drastically change their environment, a process Braun-Blanquet (1 932) had referred to as “constructiveness”. Others have used the terms “positive feedback” (e.g. Roberts, 1987; Jefferies, 1988), “self-intensifying effect” (Frenzel, 1983), “self-reinforcing trend” (Perry et al., 1987) or “bootstrapping” (Perry et al., 1989). The absence of any agreed term, and the absence of any mention of the process from most textbooks, indicate that the switch has not previously been seen as a general principle in plant ecology.

POSITIVE-FEEDBACK SWITCHES IN PLANT COMMUNITIES

261

B. Types of Switch We suggest there are four possible types of switch: (1) One-sided switch. Community X changes the environment (e.g. in-

creases the water input or lowers the soil pH) in patches where it is present. The boundary of areas where this is occurring cannot be stable, because community X can further invade the unmodified environment. (2) Reaction switch. Community X changes the environment in its patches, and also changes the same environmental factor, in the opposite direction, in the patches where it is not present (bare, or community Y). For example, it reduces wind within itself, but funnels the wind into the other community. The late J.J. Barkman (personal communication) suggested this effect might be termed “external reaction”, using “reaction” sensu Clements (1916). ( 3 ) Symmetric switch. Community X changes an environmental factor in its patches, and Community Y simultaneously changes the same environmental factor in its patches, but in the opposite direction. (4) Two-factor switch. Community X changes an environmental factor in its patches, and Community Y changes a dzferent environmental factor in its patches. Usually, the two states will be two communities of different species composition, though occasionally only one species may be present in a state (we do not use the term “community” here to indicate any particular level of organization). Often, there will be a physiognomic difference between the alternative states. Sometimes there will be a difference in cover. At the extreme, one state can be vegetated but the other bare. In the latter case, only switch types (1) and (2) are possible.

C. Boundaries Sharp boundaries between “communities”, and vegetation mosaics in an apparently uniform environment, have long fascinated ecologists. Many vegetation ecologists doubt the reality of plant communities as discrete, integrated entities. They tend to the Gleason “individualistic” or Whittaker “gradient” theories, and under these theories they can see no mechanism that would produce sharp boundaries. They therefore deny that sharp vegetational boundaries occur naturally, except where the environmental spatial change is equally sharp. However, sharp boundaries are common in the field, the “ecotones” of Clements (1905) and the “limes convergens” of van Leeuwen (1966). Some of these sharp boundaries are caused by human interference, others relate to very sharp geomorphological or geological environmental boundaries, e.g. at

268

J. B. WILSON AND A. D . Q. AGNEW ( a ) SIable mosaic Situation

( b ) Sharpening s i l ~ a l i o n

Patches X and Y lnilially wilh very lillle dillerence In vegelalionlenvironment diverge through time (Time t lo Time 3) because 01 the action 01 a switch

The ~ e g e l a l i o n / e n v i r o n m e n tgradient from W lo Z sharpens between Time 1 and Time 3 because 01 the action 01 a switch

( c ) Acceleialion siluallon

( d ) Delay situation

- wilh switch , >I >

.'without switch

Time

Successional change m vegetationlenvironmenl ~n accelerated b y the action 01 a switch

Successional change in vegelal~on/environmen! 8s delayed b y Ihe a c l m 01 a switch

Fig. 2. Four possible vegetational switches produced by switches (outcomes).

the edge of water bodies or lava flows, or between ultramafic and normal soils. Gleason (1939) denied that there were ever sharp vegetational changes without a causative sharp environmental change (or change in immigration pressure). However, it is not always possible to find environmental changes sharp enough to explain observed sharp vegetational changes. We suggest here that many sharp vegetation changes are due to a boundary between two systems being reinforced or sharpened by a switch. Sharp temporal boundaries are also seen, when one type of vegetation is stable for a long period, and then is relatively quickly replaced by another type. We suggest many sharp temporal boundaries are due to switches.

D. Vegetational Situations Produced by Switches There are four possible ecological outcomes of the operation of positivefeedback switches:

POSITIVE-FEEDBACK SWITCHES IN PLANT COMMUNITIES

269

(A) Stable mosaic situation (Fig. 2A). Where there was previously a uniform environment, switches can produce different communities, separated by sharp and stable boundaries, probably in a mosaic. Community X establishes at some sites by chance, and its modification of the environment ensures it holds the sites. Elsewhere, community Y establishes, and probably because of its different modification of the environment also holds its sites. Thus the mosaic is stable. This is similar to the theoretical concept of “several stable points” (Lewontin, 1969) or “multiple stable points from one initial condition” (Sutherland, 1974), though in such papers the involvement of the environment has not generally been elaborated. Some plant ecologists have discussed such mechanisms. Nicholson et al. ( 1 970) referred to “pasture differentiation”, Jackson (1 968) to an “unstable point” in succession (i.e. a switch between alternative pathways), Westoby (1980), Westoby et al. (1989) and Wedin and Tilman (1990) to “alternative stable states” caused by feedback. Wilson and Fitter (1984) speculated on “an element of unpredictability [in] the successional sequence”, and Bornkamm (1988) suggested that seasonal events could act as a switch, affecting the course of succession for several years. Type 1 switches cannot lead to stable mosaics, because community X can always expand. However, there can be a temporary mosaic when invasion is incomplete. Switch types 2 4 can produce stable mosaic situations, because the environment simultaneously changes in community Y (or bare) patches. (B) Sharpening situation (Fig. 2b). Where there was originally a gradual environmental boundary, a switch can produce a sharp vegetational boundary. All four switch types could have a sharpening effect. ( C ) Acceleration situation (Fig. 2c). If a switch is operated by invading species, their invasion can be accelerated. This operates by invading community (e.g. sera1 stage) Y changing the environment to make it more suitable for community Y, and less suitable for the previous community, X. If there is patchy initial invasion by community Y, a temporary mosaic may result, or the switch effect may allow some lateral spread from random invasion by Y, again giving a temporary mosaic. Any of the four types of switch could produce an acceleration situation. (D) Delay situation (Fig. 2d). Existing species, by operating a switch, could delay vegetational change; for example, a switch could delay succession, prolong the effect of initial patch composition, or delay the response to climate change. This operates by community X changing the environment to keep it more suitable for community

270

J. B. WILSON AND A. D. Q. AGNEW

Table 1 Mediating agencies in switches Physicakhemical switches: Water PH Soil elements Light Temperature Wind Fire Allelopath y Biological switches: Microbes Termites Herbivores

(e.g. seral stage) X, and less suitable for the succeeding community, Y. In a multi-stage succession, a community that modifies its environment to its own benefit could both accelerate its own invasion and delay its replacement by the next seral stage. A delay situation could result in a temporary mosaic if there is some patchiness in community X. Any of the four types of switch could produce a delay situation. The delaying effect of switches has been observed by Nicholson (1970) (“the inertia of Nardetum to successional change”), and on a longer timescale by Cole (1985) (“Vegetation inertia is . . . enhanced because a climax vegetation type . . . can create a microclimate favourable to its own members”).

E. Agencies We have attempted to categorize switches as far as possible by the agency involved. We can distinguish between physicakhemical switches and biological switches (Table 1). However, the distinction is not always clear cut. For example, some allelopathy hypotheses involve direct toxicity, others are suggested to involve effects on soil microbes, and some may do both. We give examples of the four types of switch and the four outcome situations described above, as we discuss switches caused by different agencies. Some examples will be speculative, and these we offer as hypotheses for testing, as a plan for a research programme.

POSITIVE-FEEDBACK SWITCHES IN PLANT COMMUNITIES

27 1

111. WATER-MEDIATED SWITCHES A. Concept Increased water availability, up to an optimum, generally increases vegetation mass, and increased vegetation mass can sometimes increase water availability, producing a positive-feedback switch. Enhancement of water availability by vegetation can take place through two mechanisms: fog precipitation (Section B) or increased infiltration (Section C). A water-mediated switch can also operate through sediment entrapment (Section D), ombrogenous bog growth (Section E), or snow accumulation (Section F).

B. Fog Precipitation 1. Concept Fog precipitation (occult precipitation, fog drip, fog trapping) is the impaction of small droplets from low cloud or fog. This process can enhance water input to tall vegetation, compared to adjacent low plant cover (Vogelmann, 1973). This could be a significant source of water input in arid areas with frequent fog (Vogelmann et al., 1968). If the increased water input favours the growth of taller vegetation, the elements of a switch are present, and a sharp boundary can develop between the two vegetation types (Figs 3, 4). Such a switch might explain the very existence of forest in an arid zone (Kummerow, 1962).

2. Evidence: Vegetation Boundaries The presence of a sharp vegetation boundary often indicates that a switch is operating. There are many landscapes in the tropics with hilltop forest patches amid grassland, often with the boundary abrupt (van Someren, 1939; Miller et af., 1988-Chyulu hills, Kenya; Sugden, 1982-coastal hills, Caribbean coast of South America). Probably, the primary factors are higher rainfall precipitation and lower evapo-transpiration on the hills, but fog precipitation may reinforce the environmental difference, and sharpen the boundary (Fig. 3). Often, the boundary coincides with the lower limit of frequent cloud, as in parts of the Ecuadorian Andes (Grubb and Whitmore, 1966). If the boundary is due to fog precipitation, moisture availability should also show a sharp boundary (Fig. 3). This has sometimes been seen: for example, Means (1927) found the soil of a Californian hillside to be wetter beneath trees than 3 m away from the trees, in grassland.

272

J . B. WILSON AND A . D. Q. AGNEW

Higher precipitation and reduced evapotranspiration higher up the hill allows taller vegetation.

The taller vegetation traps fog precipitation, and further increases water status higher up the hill.

Higher water status allows taller vegetation.

.- .* . . . *

*

Fig. 3. A fog precipitation switch.

3. Evidence: Tall Vegetation Enhances Water Input Evidence for water enhancement comes from observations of gr 3 er water collection in gauges that have plant shoots fixed into them (Phillips, 1926) or those with gauze/wire placed in the gauge to catch fog (Marloth, 1904; de Forest, 1923; Nagel, 1956; Twomey, 1957; Vogelmann et a/., 1968; Vogelmann, 1973; Merriam, 1973; Cavelier and Goldstein, 1989). The water input enhancement can be quite large: for example, 67% in the case of Vogelmann

273

POSITIVE-FEEDBACK SWITCHES IN PLANT COMMUNITIES

0 Water input greater

Tall Vegetation

1.

L<->J vegetation

Fig. 4. A water (fog precipitation) switch on a hillside in the montane tropics (type 1, one-sided).

et al.'s (1968) experiment at 1100 m in mountain cloud forest of Vermont.

Kummerow (1962) suggested an extraordinarily high rainfall enhancement (from 0.9mm to 50mm) under a forest patch in Chile. Often, it has been observed that enhancement is greater in periods of low rainfall (e.g. Vogelmann et al., 1968; Vogelmann, 1973), making it clear that the water enhancement is from fog, not from rain. Lysimeters can also give evidence when those containing plants, or containing taller species, collect more precipitation. For example, in upland tussock grasslands in New Zealand, 2&60% more water is collected under tall Chionochloa rigida tussocks than under lower Poa colensoi cover or bare soil (Mark and Rowley, 1976; Holdsworth and Mark, 1990). Other evidence comes from observations that effective precipitation under trees is greater than that in the open (Oberlander, 1956; Costin and Wimbush, 1961; Ekern, 1964; Harr, 1982; Vis, 1986; Wright and MuellerDombois, 1988). Sugden (1982) observed drips from the vegetation during fog, with wet soil beneath. Where records are detailed enough, it can often be seen that such effects occur even in periods with no rain (Marloth, 1907; Parsons, 1960), again making it clear that the precipitation enhancement is from fog, not from rain. Results with isolated gauges/lysimeters/trees, or on the edge of forest, may be misleading if they are merely catching water that would have fallen nearby anyway. Some studies have suggested this, reporting fog drip to be much greater at the edge of a forest than further in (Rutter, 1975). However,

274

J . B. WILSON AND A. D . Q. AGNEW

Marloth ( I 907) discounted this by showing that some precipitation enhancement was recorded by gauges in the centre of a bush thicket or a reed field, be it lesser enhancement than with gauges in the open. Fog precipitation may also input nutrients (Holdsworth and Mark, 1990) or salt, which could contribute to the switch.

4 . Evidence: Enhanced Water Input Results in Taller Vegetation Phytomass, vegetation height and density are all normally higher under higher rainfall (Beard, 1969; Branson, 1975; Shmida and Burgess, 1988), especially in arid climates, though the link between rainfall and vegetation height declines once the vegetation becomes woody (Webb et al., 1978). It is difficult to obtain more than circumstantial evidence of this in fogliable situations. There is spatial correlative evidence, such as Marloth’s (1 904) observation that the higher mountains, where he obtained his fogcatching results, bore much denser vegetation than at lower altitudes. Oberlander ( 1 956) suggested from casual observation that fog catch enabled orchids and tree seedlings to grow beneath trees. Prat (1953) observed that the dune vegetation was not nearly so luxuriant in southern California as on the Monterey Peninsula, where fogs are more frequent. Temporal correlative evidence was obtained by Means ( 1 927), who found that “pine” and Eucalyptus sp. trees were slow-growing at first, but faster subsequently (even though it was during dry years), and attributed this to the water they collected (the larger the trees were, the better fog collectors they were). It is possible that ontogenetic changes were also involved, or perhaps an accumulation of litter.

5. Rainfall Interception by Impaction In a similar way, plants can intercept near-horizontal rainfall, and thus increase local water input. However, since this rain would have fallen nearby anyway, there can be only a local increase in water input. Such a process might cause a temporary, small-scale mosaic. If forest cover in an area increases total rainfall, the process might operate on a landscape scale.

6. Conclusion We see this as a type 1 (one-sided), switch, because tall vegetation traps fog precipitation and short vegetation does not. The elements of the switch are: (i) the taller community changes the environment by increasing the water input: this has been demonstrated; and (ii) this change is relatively more favourable for the taller community than for the shorter one: this has been surmised.

POSITIVE-FEEDBACK SWITCHES IN PLANT COMMUNITIES

275

The switch seems generally to result in a situation B, sharpening, situation, though on a landscape scale mosaics can result, as in the Chyulu Hills, Kenya (van Someren, 1939; Miller et al., 1988). In theory, the switch could lead to accelerating (C) or delaying (D) situations, but these are harder to observe. To the extent that the fog precipitation caught would have fallen in an adjacent area anyway, the switch can be seen as a reaction (type 2) switch.

C. Infiltration 1. Concept Vegetation may trap surface rainfall runoff on gentle slopes, increase its infiltration, and perhaps reduce evaporation, thus increasing water availability to plants. This has been documented from many arid countries (White, 1971). The process can be seen with a patch of vegetation, or around a single plant base. Increasing infiltration will lead, in arid areas, to increased plant cover.

2. Evidence Belsky (1986) described a possibly “endogenous” mosaic in Serengeri grasslands of East Africa, attributing it mainly to rainfall infiltration effects. The dense Andropogon greenwayi community produces more phytomass, with higher soil organic matter content and infiltration rates, than the open Chloris pycnothrix community. This gives a possibility of a water switch, in which the Andropogon state has higher phytomass, which increases the water input, which favours the Andropogon state-positive feedback. The missing link would be experimental evidence that the wetter soil favours the Andropogon. Belsky suggests that the mosaic is initiated by patchiness in soil leaching. The mosaic disappears if grazing is prevented, so a grazing switch may also be involved. Mosaics apparently caused by an infiltration switch can be in the form of bands/arcs of vegetation alternating with bare ground (e.g. Tongway et al., 1989-semi-arid eastern Australia; Cornet et al., 1988-Chihuahuan Desert, Mexico). In Somalia, Boaler and Hodge (1964) found that the vegetation/ bare boundary in such a mosaic is sharp on the uphill edge of the vegetation band, but more diffuse on the downhill margin. When the soil crust of bare areas is dry, heavy rainfall can run downhill in a sheet and collect against the uphill edge of a vegetated arc. Water infiltration under the vegetation can be five times deeper than in bare areas, further assisted by “potholes” in vegetated areas (Boaler and Hodge, 1964), and perhaps soil cracks and higher soil organic matter (Cornet et al., 1988). However, Glover et al. (1 962) found that water infiltration occurred more readily beneath bare areas, and attributed the higher water status below grass cover to funnelling of rainfall

276

J. B. WILSON AND A. D. Q. AGNEW

by grass leaves, the rain being caught by the leaves and running down them to the base of the plant. This mechanism suggests that expansion of the vegetation should take place at the uphill boundary, and drought-induced dying off at the downhill boundary, so the arcs should advance uphill. There is evidence for advance at rates of 0.3-1 myear-’ (Worrall, 1959; Glover et al., 1962; Boaler and Hodge, 1964; Cornet et af.,1988), though in some areas there seems to be no movement (Boaler and Hodge, 1964). The vegetation mosaic can be of many different shapes, perhaps with different origins, but apparently always caused by an infiltration switch (Wickens and Collier, 1971). The initiation and maintenance of such a mosaic often occurs together with other switch mechanisms, including termites, grazing (Glover et al., 1964), soil structure, aeolian soil/sand accretion (Boaler and Hodge, 1962), nutrient accumulation (Tongway et al., 1989), and salt accumulation (Boaler and Hodge, 1964), though the latter workers suggested initiation by random gaps appearing in vegetation as the climate became drier.

3. Conclusion This can be seen as a type 2 (reaction) switch, in which: (i) dense vegetation increases infiltration, and hence water availability, and also stops runoff, reducing water availability downslope of it; and (ii) increased water availability favours (or enables) the growth of the dense vegetation. There is still only limited water, which is absorbed by the stripes. This prevents the vegetation extending to cover the area, so that in a sense the vegetation makes the bare areas drier, the type 2, reaction effect. The result is usually situation (A), a mosaic, though there could also be situation (B), a sharpening of an altitudinal trend.

D. Sediment Entrapment: Salt Marsh Pans 1. Concept Vegetation can trap water-borne sediment, and sediment deposition can increase vegetation growth by nutrient input, by decreasing waterlogging or by decreasing salinity, giving a switch (Fig. 5).

2. Evidence Yapp and Johns (1917) commented that intertwined shoots of the salt marsh “meadow” sward are very effective in silt binding. Depressions in the bare silt before colonization represent, they suggested, incipient primary bare pans.

Bare pan

Salt meadow

I I I I I Bare pan

meadow spp. favoured

I

conditions for salt meadow spp.

I

Fig. 5. A water (sediment entrapment)/salt switch with salt pans on a salt marsh (type 2, reaction).

278

J . B . WILSON AND A. D. Q. AGNEW

They tend to remain filled with tidal water for longer, the silt in the bottom is mobile, and salinities will tend to be more extreme, all of which inhibit the growth of vegetation, and the pans therefore remain bare. Such pans are effectively permanent unless the drainage system changes. Although Yapp and Johns saw cases of pans filling in with vegetation, in every case it was because a drainage outlet had been established. On the salt meadow, sediment entrapment can raise the soil level and allow other communities to invade, so facilitation succession can occur simultaneously. Pethick (1 974) suggested that secondary pans could also arise, when a patch of turf died and the silt eroded, but this supports the switch concept: when the vegetation can no longer mediate, the environment reverts.

3. Conclusion This is a type 2, reaction switch in which: (i) vegetation traps silt, builds up the soil surface and thus reduces tidal inundation, whilst increasing water ponding and maximum salinity in the pans; and (ii) a higher surface in the salt meadow provides more favourable growth conditions for the salt meadow vegetation there, and more unfavourable conditions in the pans. It is a reaction switch because the silt buildup both reduces inundation on the salt meadow, and by building up the pan walls increases it in the pans. It produces a mosaic (A) situation, or at least sharpens (situation B) and makes permanent small initial substrate differences. There are elements of a salt switch here too.

E. Ombrogenous Bog Growth 1. Bog Growth Ombrogenous bogs are those, often dominated by Sphagnum spp., that are dependent for their water and nutrient inputs on precipitation. Once peat buildup raises the bog surface beyond influence of the water table, the rainleached substrate becomes nutrient-deficient and acid (Tallis, 1983). Sphagnum spp. can tolerate, or even require, such conditions (Clymo, 1973). Their growth and peat accumulation raises the bog surface even further from water table influence, completing a landscape-scale switch. Tallis (1983) suggests that the entry of Sphagnum is the critical stage, because of the ability of Sphagnum species to lower the pH of the substratum, though he cites evidence that other mosses can d o this too, perhaps acting as precursors for Sphagnum species. The cycle is broken only if the climate changes or perhaps

POSITIVE-FEEDBACK SWITCHES IN PLANT COMMUNITIES

279

if the bog “bursts” (Barber, 1981). The Sphagnum also acts as a “sponge”, retaining water. Because of the long time-scale involved, the processes have been inferred rather than observed. Bellamy and Rieley (1967) obtained evidence close to the event by examining a small ombrogenous Sphagnum hummock that had established in a fen. Stratigraphy suggested that two non-Sphagnum mosses had established a low hummock, which “deflected the main flow of ground water”. They may also have started acidification. When the hummock was about 0.05 m high, Sphagnum spp. invaded, leading rapidly to acid peat that was above the influence of the ground water, perhaps within 10 years. The mechanism has elements of water, pH and nutrient switches.

2. Hummocks, Hollows and Pools It is possible that a similar switch sometimes operates on a much smaller scale. Many bogs comprise a mosaic of hummocks, hollows, and perhaps deeper pools. Early suggestions were of cyclic succession between hummocks and hollows (Osvald, 1923; Watt, 1947). However, several workers have found that, at least in the bogs they examined, stratigraphic evidence did not support such a cycle (Walker and Walker, 1961; Barber, 1981); rather the whole bog surface responded to climatic change. Some have found evidence that, directly contrary to the cyclic succession hypothesis, the sites of hummocks continue to be hummocks, likewise those of hollows and pools to be hollows and pools, for thousands of years, in some cases through the life of a bog (Casparie, 1972; Boatman and Tomlinson, 1977; Moore, 1977; Barber, 1981; Svensson, 1988; Foster and Wright, 1990), and perhaps reflecting the underlying topography (Boatman et al., 1981; Boatman, 1983). Thus hummocks, hollows and pools could be alternative stable states, indicating the existence of a switch. (Distinction should be made between small vegetated hollows and large bare pools, but the principle is similar for our purpose, and the distinction is not clear-cut.) The switch could be based on greater productivity by the species of a hummock (Wallen et al., 1988), resulting in a surface buildup on hummocks at least as fast as that in the hollows (Boatman and Tomlinson, 1977, though the evidence from their different methods is equivocal-Barber, 1981; Moore, 1989). Probably more importantly, dead hummock material tends to decompose more slowly (Clymo, 1965; Moore, 1989; Johnson et al., 1990; Rochefort et al., 1990; Johnson and Damman, 1991). In larger pools, bare of vegetation, surface buildup is due to gyttja accumulation, which can be at a lesser rate than peat accumulation on hummocks (Foster and Wright, 1990). The story is confused, with some experimental evidence not supporting either the cyclic succession or the switch interpretations (Clymo, 1973; Hayward and Clymo, 1983). Possibly, several factors are involved, with different results on different bogs.

280

J . B. WILSON AND A . D . Q. AGNEW

(-<->-A

0

I

Soil pH

Calluna etc.

lower

favoured Fig. 6. A pH switch controlled by Calluna vulgaris, CJ1e.u europaeus, etc. (type 1, onesided).

F. Snow Accumulation If the vegetation affects snow lie, and the community which attracts the deepest snow is also the most tolerant of it, a switch is possible. Payette (1988), in northern Quebec, gave stratigraphic evidence for rapid changes in vegetation cover, after long periods during which he suggests a selfmaintaining state of a Picea marina and Sphagnum fuscum cycle. The snow accumulation by the krummholz-type Picea stands increases the moisture available to the Sphagnum, which allows peat to be deposited, with a consequent rise in the surface level. Eventually a “dramatic change” (his words) occurs when the surface is raised level with the central plateau of the ombrotrophic mire system. At this point, the Picea is no longer able to layer, the peat becomes “fossil” without current addition, and the permafrost layer rises. This leads to tundra yegetation dominated by Cladina spp. (mainly C . rangijierina and C . stellaris) and Alectoria ochroleuca. This is a one-sided switch (type I), where the PicealSphagnum maintains itself until a limit imposes fragility. The tundra lichen vegetation does not operate a positive feedback mechanism, except in so far as it does not trap snow.

IV. pH-MEDIATED SWITCHES 1. Concept Species of low base-capacity soils may have litter or roots which acidify the soil, and be tolerant of acidity and its associated conditions, producing a positive-feedback switch (Fig. 6 ) .

POSITIVE-FEEDBACK SWITCHES IN PLANT COMMUNITIES

28 1

2. Evidence Pigott ( 1 970) suggested that Deschampsia flexuosa invades Festuca ovinal Agrostis capillaris grassland by producing an unpalatable litter which eliminates the soil comminution organisms (e.g. earthworms, thrips). This changes the humus type from mull, with large pore spaces, to mor, with small, easily waterlogged spaces. The resulting hypoxic conditions reduces the pH, releasing manganese, to which the Deschampsia may be tolerant (Ernst and Nelissen, 1979), but not the Festuca nor the Agrostis. Grazing checks the invasion, and the resulting balance is a mosaic of grassland types. Calluna vulgaris, Erica cinerea and Ulex europaeus produce litter which can acidify the soil (Wilson, 1960; Grubb et al., 1969; Miles, 1985), probably by removing calcium and other bases (Grubb and Suter, 1971). The former two are acidophilous species, growing well on acid soils (Marrs and Bannister, 1978). Ulex europaeus often occurs on soils of pH 4-5 (Etherington, 1981), is tolerant of the high aluminium availability that is associated with low pH, and is susceptible to lime chlorosis (Grime and Hodgson, 1969). The competitive ability of all three species, compared to more calciphilous species, would therefore be greater after the acidification process. Similarly, Corynephorus canescens acidifies the soil in which it grows, and from its distribution appears to be acidophilic (Rychnovska, 1963). It is therefore confined in some areas to soils of low pH-buffering capacity, the pH of which it can modify. A similar process is seen in ombrogenous bog formation as described above.

3. Conclusion These are apparently type 1 (one-sided) switches, with elements: (i) species X acidifies the soil; and (ii) species X is more tolerant of acid soil than the species previously dominant. Records are sparse of the spatial arrangement of these communities. With a type 1 switch, one would not expect to find a stable mosaic. If a grazing switch were present too, the switch could be of type 4 (two-factor), capable of producing a permanent mosaic.

V. SOIL-ELEMENT-MEDIATED SWITCHES

A. NPK Increase 1 . Concept The possibility exists for a species to increase the nutrient content (especially N) of the soil, and to be favoured by that change because it is more

282

J . B. WILSON AND A. D . Q. AGNEW

responsive in its growth to enhanced nutrient supply-a switch.

positive-feedback

2. Evidence Often the soil below savannah trees is more nutrient-rich than that below the matrix grassland (Parker and Muller, 1982; Belsky et al., 1989). Kellman (1979) ascribed this to differential capture of precipitation NPK inputs. Charley (1972) found that Atriplex vescaria in arid parts of Australia withdrew N from the surrounding soil, adding it to the soil directly beneath its canopy. J.J. Barkman (personal communication) suggested that the invasion of Dutch heathlands by Juniperus communis is accelerated and stabilized by its ability to raise the pH, P and K of the soil. However, for none of these examples is there evidence that the nutrient-accumulating species is more responsive to nutrients. Wedin and Tilman (1990) and Tilman and Wedin (1 991a) found different soil nitrogen availability in plots planted with different species. The two species which produced the lowest N availability were those which had a high competitive ability for N (Tilman and Wedin, 1991b). This seems to be good documentation of an NPK switch, though it is not clear whether the switch is an increase or decrease in NPK.

3. Conclusion There is evidence for a one-sided (type I ) switch in which: (i) a plant species raises the NPK status of the soil; and (ii) the increase in NPK status favours that species. It is not clear what vegetation situations it leads to.

B. N P K Decrease 1 . Concept A species that lowered NPK availability, and could tolerate low availability, could operate a positive-feedback switch.

2. Evidence Heilman (1966, 1968) recorded abrupt boundaries between Alaskan woodland on mineral soil and that on Sphagnum. He suggested that Sphagnum invaded woodland. This led to a reduction in the N, P and K status of the surface layers, largely because the low specific gravity of the Sphagnum peat resulted in a low nutrient content per volume, the nutrients being situated in the lower, colder layers of the soil. Moreover, the insulating properties of Sphagnum peat resulted in permafrost much nearer the surface. Thus, the nutrients present were much less available to the trees. There was a pH

POSITIVE-FEEDBACK SWITCHES IN PLANT COMMUNITIES

283

change too. Lower nutrient availability lowered tree growth and switched the woodland into muskeg bog. The disappearance of the trees increased light levels, favouring the Sphagnum again. Heinselman (1970) suggested the same effect for North Minnesota. Bonan and Korzuhin (1989) modelled such a boreal system. In their simulation the system was very sensitive to interactions between mosses, trees and the environment; the insulating property of the moss layer was particularly important. These effects are switches from minerotrophic to ombrotrophic, from young to old successional types and are often triggered by small changes in climate or geomorphological balance. Wilson and Fitter (1984) proposed an effect of Sphagnum in a bog in northern England. They suggest that, where Sphagnum colonizes, pH is reduced. This reduces P availability, largely by suppressing microbial cycling. Lower P availability favours the Sphagnum. Where LoniceralRubus vegetation colonizes, the higher pH enhances P availability, favouring the species present. Thus, pH, nutrients and microbes are part of the switch.

3. Conclusion These are one-sided (type 1 ) switches, in which: (i) Sphagnum reduces nutrient availability, and lowers pH; and (ii) the lower nutrients and pH favour Sphagnum. There are elements of NPK, pH, temperature and microbial switches. In the North American examples, the switch is suggested to lead to situation ( C ) , acceleration of (“retrograde”) succession. There are some indications of an (A) mosaic situation, which could not be explained by a one-sided switch, but the shading effect of the trees could possibly make this a two-factor (type 4) switch, theoretically capable of generating a permanent mosaic. The English example could be a symmetric (type 3) switch, capable of producing a permanent mosaic.

C. Heavy Metals 1. Concept Among plants tolerant of heavy metals, e.g. those found on ultramafic soils, some accumulate heavy metals to considerable concentrations. Such plants are termed “hyperaccumulators”. They have the potential to build up toxic elements in surface litter and soil, and thus prevent invasion by an alternative, less-tolerant community, producing a positive-feedback switch (Fig. 7).

2. Evidence Several elements have been found to be hyperaccumulated (Table 2). The concentration of heavy metal in the plant dry material can exceed the total concentration of heavy metal in the parent rock or in the soil (Table 3).

284

J.

B. WILSON AND A. D. Q. AGNEW

accumulated in hence in litter I

0 Available heavy metal in soil greater

Hyperaccumulator dominated patch

species suppressed

Fig. 7. A soil nutrient (heavy metal) switch controlled by plants that hyperaccumulate heavy metals (type 1 , one-sided).

Concentrations in the plant ash are even higher in comparison with rock or soil, and it is the plant ash content that is more relevant once the leaves fall and decay. Even values below 1.0 in Table 3 may represent an increase in plant-available heavy metal if the litter-derived heavy metal is in a simple soluble mineral form. This suggests that litter from heavy metal accumulators may increase the soil levels of available heavy metal, though we know of no investigation of this. (However, litter decomposition may be slow, and lower specific gravity of litter, in comparison with mineral soil, might counteract the effect.) The accumulators are presumably tolerant of high internal levels of the heavy-metal they accumulate, though there may be some internal compartmentalization. Many hyperaccumulators are confined to metalliferous ground (Brooks et a/., 1981). Psychotria dourarrei,which can contain 2.2% nickel in its leaves, can tolerate I % nickel in solution in its root medium (Baker et al., 1985). Hyperaccumulators might therefore be favoured, in competition with other species, by an increase in heavy-metal availability. If such a switch operated, we would expect to find that communities dominated by hyperaccumulators formed pure stands, and they often do (Baker and Brooks, 1989). We would expect sharp edges, and possibly temporary mosaics. We know of little information on this, though Duvigneaud and Denaeyer-de Smet (1963) describe “copper clearings” of herb communities in forest, and their photographs show sharp boundaries. It is possible that this reflects an equally sharp change in the original soil, though

POSITIVE-FEEDBACK SWITCHES IN PLANT COMMUNITIES

285

Table 2 Records of hyperaccumulation. Element concentrations are the maximum found, usually of leaf material (weight:weight basis) Element

Concentration

Source

(%I In dry matter

1 .o

Cobalt Copper Lead Manganese Nickel Selenium Zinc

Brooks et al. (1987) Malaise et al. (1979) Reeves and Brooks (1983) Jaffre (1 979) Jaffre and Schmid (1974) Cannon ( 1960) Reeves and Brooks (1983)

0.6 0.8 5.5 4.7 0.6 4.0

In plant ash Chromium Molybdenum

4.8 1.7

Uranium

2.5

Wild ( 1 974) Warren and Delavault (1 965) Whitehead and Brooks ( 1 969)

Table 3 Values reported for hyperaccumulators of the ratio of the element concentration in the plant to that in the soil in which the plant is growing (weight:weight basis) Element

In dry matter Lead Manganese Nickel Zinc In plant ash Chromium Copper Nickel Uranium

Concentration Concentration in in plant soil W) (Yo) 0.8 3.3 2.3 2.0 4.7 1.7

4.8 1.2 0.02 40.6 15.3 2.5

0.30 1.50

0.46 0.33 0.37 1.70 12.50 c. 1 . 1

0.00 1 0.37 0.55 3.65

Ratio

2.7 2.2 5.0 6.1 12.7 1 .o

0.4 c. 1 . 1 20.0 109.7 27.8 0.7

Source

Reeves and Brooks (1983) Jaffri (1977) Jaffre ( 1 977) Baker et al. (1985) Jaffre and Schmid ( 1 974) Reeves and Brooks (1983) Wild (1974) Wild (1 974) Wild ( 1 974) Jaffre and Schmid (1974) Wild (1974) Whitehead and Brooks ( 1969)

286

J. B. WILSON AND A . D. Q. AGNEW

the process of upward diffusion of copper that they describe makes this less likely.

3. Conclusion This would be a one-sided (type 1) switch in which: (i) the hyperaccumulator produces litter, and hence soil, high in available heavy metals; and (ii) it can tolerate this, but less tolerant species cannot. We do not know what situations it would lead to, but it should tend to sharpen soil boundaries (situation B).

D. Salt 1. Concept Species that salinify the soil, and are tolerant of salinity, could operate a positive-feedback switch.

2. Evidence: Cohune Palm The cohune palm, Orbignya cohune, dominates patches of woodland on lowlying, deep soils in Belize, sharply demarcated from dicotyledonous forest on slopes (Arnason et al., 1984). This palm produces a dense shade. Its abundant leaf litter has a very low nitrogen content but a salinity “approaching values for mangrove forests in brackish water”. This gives a soil sodium content greater than in the “high bush forest”. Some palms are tolerant of salinity (Arnason et al., 1984), and, if this applies to the Orbignya, a switch could be operating. On the rather different grounds of soil turnover and humus accumulation, Furley (1975) suggests that the Orbignya “appears to develop the soil in which it is best suited to grow”-a clear suggestion that there is a positivefeedback switch operated by this palm through the soil environment. Low light at the soil surface, caused by the palm litter, may also be part of the switch mechanism.

3. Evidence: Arid Shrubland Kovda et al. (1979) suggested that in arid regions worldwide the vegetation tends to cause the accumulation of soluble salts in the upper soil layers. He particularly cited an example from the Karakum Desert where Haloxylon aphyllum and H. persicum litter adds up to 80 g m-* year-’ of salts to the soil surface, mostly sodium carbonate and bicarbonate. This gives a pH of 8.59.0. If the Ha/o.xy/on species are more tolerant of such salts than other species nearby, a switch could result, though that has not been demonstrated. Westoby (1980) suggested that in the inter-mountain region of the USA the exotic annual Halogeton glomeratus absorbs salt from lower soil layers,

POSITIVE-FEEDBACK SWITCHES IN PLANT COMMUNITIES

287

and by its litter creates high salinity at the surface that it can tolerate, but that the native shrub Kochia americana cannot. Soils in such areas are indeed salty at depth (Kearney et al., 1914; Cook, 1961). The succulent leaves of the Halogeton are high in minerals, especially sodium oxalate (Morton et al., 1959; Cronin, 1965). Thus it has the potential to add salt to the surface in litter, as shown by comparison of Halogeton areas with non-Halogeton areas, and by experimental addition of Halogeton litter (Eckert and Kinsinger, 1960). However, such an increase in salinity by the Halogeton has not yet been unequivocally demonstrated. Cook (1961) found the percentage of total soluble salts in soils below a community of the Halogeton with scattered Kochia shrubs to be higher than in soils below Artemisia tridentata, but since this difference was more pronounced at the deeper soil depth it was probably due to pre-existing conditions. Charley and West (1975) did demonstrate that plants of Atriplex confertifolia, in the same part of the USA, increased the salinity of soil around their canopies. The demonstration of a switch would also require evidence that Halogeton glomeratum is more salt-tolerant than Kochia americana, etc., and indeed it is, at least at germination (Williams, 1960; Cronin, 1965; Clarke and West, 1969). This has been used to explain why re-invasion of the Kochia is prevented in areas where it could re-establish before the advent of the Halogeton (Kearney et al., 1914; M. Westoby, personal communication). Other species, sown into Halogeton swards, have failed to germinate (Eckert and Kinsinger, 1960). One sign of the operation of a switch is a mosaic (situation A), and this can occur, with islands of dense Halogeton amongst Artemisia tridentata (Cook, 1961).

4 . Evidence: Mesembryanthemum in Grassland In South Australia, Kloot (1983) showed by field and plot experiment that the exotic Mesembryanthemum crystallinum, invading annual pasture, increased the salinity of the soil below its canopy. This reduced the growth of other species in the community. It also reduced its own growth, though to a lesser degree. Vivrette and Muller (1977) provided similar evidence from annual grassland in California, showing that salt was released from dead, dry Mesembryanthemum by rain or fog-precipitation leaching. Again, the Mesembryanthemum was more tolerant of such leachate than other species of the community. They suggested that the Mesembryanthemum invaded when climate and grazing had opened the community, and that the salt switch retarded re-invasion of the grassland species-a delay (D) situation.

5 . Conclusion This seems, if operative, to be a type 1, one-sided, switch, in which:

288

J . B. WILSON AND A. D. Q. AGNEW

Closed canopy

I Light at ground level

Fig. 8. A light switch on a NSW (Australia) salt marsh (type 1, one-sided).

(i) the salt-accumulator absorbs salt from a distance around itself, or from depth, and by its litter increases the salinity of the nearby surface soil; and (ii) the salt-accumulator is better able to tolerate this salinity than other species. A one-sided switch can explain the sharp boundary of the Orbignya forest. It can delay vegetation change (situation D), as in the Mesembryanthemum situations. It can lead to mosaics as in the Halogeton example, but these are only temporary because there is nothing to stop the continued invasion of the salt-accumulator. However, when accumulation of salt is from areas lateral to the plant, salt is reduced in the unoccupied areas, forming a type 2, reaction, switch, capable of producing the mosaics observed.

VI. LIGHT-MEDIATED SWITCHES 1. Concept A theoretical construct for a positive-feedback light switch would be two communities, one open, one closed, where the species belonging to the open community could establish only in full light, and those of the closed community could establish only in shade (Fig. 8).

2. Evidence In the Sierra Madre, Mexico, Goldberg (1982) describes how an open woodland of Quercus spp. with almost evergreen leaves occurs on ridges with

POSITIVE-FEEDBACK SWITCHES IN PLANT COMMUNITIES

289

dry acid soils, separated by a very narrow ecotone (2-18 m) from more mesic deciduous woodland on deeper, more nutrient rich soils. The soil pH changes much more gradually than the vegetation across the boundary. She suggests that the boundary is sharpened by the inability of both the deciduous flora to establish in the poor acid soils below evergreens, and the seedlings of the evergreen trees to establish in the shade of the mesic woodland. Her observations indicate a similar light environment under both canopies, and she hypothesises that a slow growth rate of evergreen seedlings, and possibly higher shade cast by the richer ground flora of herbs in the deciduous woodland, are enough to account for the effect. Red/far-red ratios may also play a part. Another example of a soil-induced gradient, sharpened by a light switch, is seen in the very sharp boundaries between communities along a salt marsh gradient near Sydney, Australia (Clarke and Hannon, 1971). Although the species differ in salinity tolerance, this cannot account for the sharpness of the boundaries. Clarke and Hannon found that the low herb Arthrocnemum australasicum (= Salicornia australis) was very intolerant of shade and this was a major cause of the abrupt change from Arthrocnemum meadow to Avicennia woodland on the seaward side, and from the Arthrocnemum to Juncus maritimus fen on the upper margin (Fig. 8). Thus, a salinity gradient establishes the species’ positions, but the boundary between communities is sharpened by a light switch; the tall species create shade, and can regenerate under it. In other cases, the initial boundary may be formed by chance establishment. Niering and Egler (1955) described a Viburnum lentago thicket which had established in an oldfield about 30 years earlier. Once the Viburnum canopy was established by “fortuitous distribution of . . . propagules”, low subcanopy light levels (and rabbit grazing) prevented the entry of other species. The Viburnum could reproduce by root suckers. However, Niering and Egler admitted that the exclusion was not complete, and that eventually trees would invade, and later found evidence of this (Niering et al., 1986). Westoby (1980) describes a desert grassland in the southwest USA where, without grazing, Prosopis velutina (mesquite) shrubs cannot invade. Grazing opens the grass cover, and Prosopis shrubs establish. When grazing is removed, the Prosopis shrubs prevent re-establishment of the grass cover. Prosopis can regenerate in its own shade, perhaps because of its large seed. The existence under no-grazing conditions of two alternative stable states, depending on previous conditions, suggests that a switch is operating. Westoby suggests competition for light, and fire, as factors, though wind erosion and competition for water could also be involved. Westoby (1980) also reports a very similar situation in the understorey of some Callitris and Eucalyptus woodlands in arid parts of eastern Australia, where grazing of the grass understorey has led to an increase of Cassia and

290

J . B . WILSON AND A. D. Q. AGNEW

Eremophilu shrubs. When stocking rates are reduced, the grass is unable to return, probably because of shading by the shrubs. Harrington and Johns (1990) gave indirect evidence for such suppression. Light switches can also occur via litter. For example, below Mt Hiuchidake, Japan, bamboo (Susa spp.) can inhibit the regeneration of Fugus crenutu (Tanaka, 1988). This is probably due to litter, not the shade of living plants, since light intensity in Susa thickets is higher than in shrub or tree communities at 1.5 m above the ground, but much lower than in these other communities at the soil surface. Root competition, and perhaps allelopathy by the Sum litter, may also be factors. However, the Susa patches can become smaller due to edge invasion by trees, and periodic flowering and death of Sum might disable the switch.

3. Conclusion Goldberg’s example could be a two-factor (type 4)switch, involving pH and light, though leading only to a sharpening (B) situation. The other examples are type 1, one-sided switches, in which: (i) a woody species reduces the light intensity on the ground, (ii) the woody species can tolerate this low light, and regenerate, but other species cannot. This can give a sharpening situation (B) or in the case of the Viburnum and Prosopis examples a delay situation (D).

VII. TEMPERATURE-MEDIATED SWITCHES A. Concept Tall vegetation can reduce fluctuations in air temperature near the soil surface. If this favours gtowth or re-establishment of the same species because they are less cold-tolerant than their neighbours, a positive-feedback switch can occur. We consider this for treeline and for graminoid tussocks.

B. Treeline 1 . Concept The altitude of alpine timberline is correlated in many parts of the world with a mean air temperature of 10°Cduring the warmest month (Schroeter, 192326; Daubenmire, 1954) (though temperature may be partly a proxy for other factors). Sometimes, treeline is sharp, suggesting a switch is operating.

POSITIVE-FEEDBACK SWITCHES IN PLANT COMMUNITIES

29 1

0 Frost less severe

Forest

A

L<->J Tree seedlings

Fig. 9. A temperature switch at treeline (type I , one-sided).

2. Evidence An explanation of the mechanism might be that the tree cover ameliorates the microclimate, so that temperatures on clear calm nights are lower in the open than under a tree canopy, and lower than the limits of cold tolerance for the tree species (Fig. 9). Wardle (1985b) demonstrated this in New Zealand for Nothofagus solundri, which can form a very sharp timberline (Wardle, 1965). In that case the low temperatures in the open are known to be lower than the cold tolerance of the species. Therefore, seedlings survive under trees but not amongst low vegetation at the same altitude, unless they are experimentally sheltered from frost when young (Wardle, 1965, 1985a). One sign of a switch is that a vegetational pattern, once disturbed, is not restored, or only very slowly. The treeline temperature switch would therefore explain why after fire, etc. treelines are often much lower than previously (e.g. Plesnik, 1978).

3. Conclusion This is a type 1 (one-sided) switch, in which: (i) trees ameliorate frosts beneath them, (ii) the tree species’ seedlings are less tolerant of frost than those of abovetimberline species. As a type I, one-sided switch, it cannot produce a mosaic in a uniform environment, but it can sharpen (situation B) the effects of an altitudinal (or other environmental) gradient. Apparent inertia of treeline to climate change

292

J . B. WILSON AND A. D. Q. AGNEW

(Kullman, 1989) suggests a delay (D) situation can arise too, though slow invasion, e.g. limited by mycorrhizal availability, could be important.

C. Graminoid Tussocks Chapin et al. (1979) show that tussocks of Eriophorum vaginatum in the Alaskan Arctic tundra have a higher temperature during late spring and summer than does inter-tussock vegetation, while Oberbauer and Miller (1979) demonstrate that tussocks of this species have a buffering effect against temperature (and water) fluctuations. Both effects lead to a longer growing season. Tussock growth is limited mainly by temperature and light, not nutrients (Fetcher, 1985), and is thus correlated with the number of degree days (Mark et al., 1985). Therefore the plants’ environmental modification can lead to increased plant growth, representing a switch. However, it has not been shown that the Eriophorum is more able to respond to increased temperature than the surrounding species. The temperature buffering would give a type I , one-sided switch, perhaps leading to a (C) acceleration situation if the Eriophorum were invading, and explaining the dense tussock fields reported (e.g. Oberbauer and Miller, 1979). Alternatively, radiation interception would lead to shading of other plants, a type 2 (reaction) switch, capable of producing a vegetation mosaic, though there seem to be no reports of this.

VIII. WIND-MEDIATED SWITCHES A. Concept Vegetation can dissipate, concentrate or redirect wind energy and thus (section B) change the aeolian erosion, transport and trapping of soil materials, or (section C) reduce direct damage to vegetation, in either case potentially producing a positive-feedback switch.

B. Soil Erosion and Trapping 1 . Concept Vegetation can trap soil, and deeper soil encourages plant growth, giving the possibility of a switch.

2. Evidence Marshall (1970) provides a model of the relations between shrub height and width and its effect on wind speed, showing that there is a minimum density for shrubs of each size class below which soil erosion will occur. His field

POSITIVE-FEEDBACK SWITCHES IN PLANT COMMUNITIES

293

stands were in the winter-rainfall saline rangeland of southern Australia where mainly Chenopodiaceous shrubs form stands with patchy openings which are very slow to revegetate due to topsoil loss. The shrubs trap soil, and therefore grow better-a switch. Agnew and Haines (1960) and Agnew (1961) showed that the trapped soil around shrubs in Iraqi desert had a higher nutrient content than the surrounding soil, and that this led to increased production and presumably more entrapment of dust. This was especially the case under Haloxylon salicornicum, a Chenopodiaceous shrub forming patches of vegetation in hollows in an otherwise stony subdesert surface. The switch sharpens the topographic effect, so that scattered shrubs in hollows become small dune systems in the hollows, with sharp edges. The shrubs eventually degenerate and seedlings re-establish better on the deposited soil than on the open desert surface. The patch of dunes bearing the Haloxylon may not spread indefinitely because there is not enough blown material, and the life of a plant of the Haloxylon on the open stony desert surface is too short, but in any case the observed stands were limited by grazing pressure. Westoby (1980) suggests that in some arid perennial grasslands, the root mat can prevent soil erosion by wind and water. If overgrazing occurs, perennials are replaced by ephemerals which cannot hold the soil, which is therefore eroded. It is difficult for the perennials to become re-established. Harrington and Johns (1990) suggest an interaction with fire, in that the reduced herbaceous growth could not produce enough standing crop to sustain a fire. The lack of fire favours shrubs, which suppress the herbaceous plants further. Westoby (1980) points out that a wind-mediated switch occurs without grazing on sand dunes: when vegetation is removed the dunes become mobile, and vegetation cannot readily re-establish. In secondary succession in the New Zealand alpine, the grass Poa colensoi can form mounds by trapping aeolian silt (Roxburgh et al., 1988).

3. Conclusion The elements of the switch are: (i) the plants trap and/or hold the soil; and (ii) this favours plant growth. This is basically a type 1, one-sided, switch. However, the limited amount of fine material may produce a type 2, reaction, switch, capable of producing the mosaic (A) situation seen in the southern Australian shrublands. The Iraqi desert example is a sharpening (B) situation, with the initial differences caused by topography. Westoby’s grassland example seems to be a delay (D) situation, since recovery from grazing is delayed.

294

J . B. WILSON A N D A. D. Q. AGNEW

C. Wind Damage to Plants 1. Concept Wind damage is very significant near alpine tree lines and in Arctic taiga. We may postulate that the presence of shrubs or trees creates an ameliorated microclimate within the canopy which allows those species to grow and reestablish (Fig. 10).

2. Evidence In the alpine tundra, wind-deformed trees (e.g. Picea engelmannii, spruce), occur as compact islands of krummholz (Wardle, 1968). Sometimes, the plants are arranged in lines parallel to storm winds, showing the importance of wind (Minnich, 1984), which is high above treeline, especially in winter (Wardle, 1968). Sometimes other species are associated with the islands (e.g. Ribes montigeum-Marr, 1977). The islands are packed with snow in winter, whereas much of the open tundra is blown free of snow (Wardle, 1968). This snow accumulation has considerable sheltering effect; above the snow there is basipetal death of exposed needles and shoots during the winter, a process highly correlated with wind (Hadley and Smith, 1983). The krummholz is also self-sheltering, with much less damage (stem breakage, needle death and crown dieback) on undercanopy and leeward branches (Minnich, 1984). It is not clear which effect of wind is most important. Temperatures are higher within a krummholz plant, at least in terminal meristems (Grace, 1989), than in the open tundra (Wardle, 1968; Hadley and Smith, 1986). Low winter temperatures could kill needle tissue directly (Wardle, 1981; McCracken et al., 1985; Tranquillini and Plank, 1989), though others have doubted this explanation (Hadley and Smith, 1983). Wardle (1968) hypothesized that low summer needle temperatures resulted in incomplete needle maturation, and hence winter damage. Though there is evidence for this (Baig and Tranquillini, 1980), Hadley and Smith (1986) disproved it, at least in their situation. Tranquillini (1980) suggested the damage is caused primarily by desiccation. Windward needles can have lower winter water content, lower water potentials, lower cuticular resistance and faster transpiration (Hadley and Smith, 1983, 1986). Others have doubted this explanation (Marchand, 1972; Marchand and Chabot, 1978; Kincaid and Lyons, 1981; Hadley and Smith, 1986). There is also a direct mechanical effect of wind, especially of ice crystals on cuticular wax (Hadley and Smith, 1983, 1989). However, Wardle (1968) doubted whether this was a significant cause of needle mortality, and Hadley and Smith’s (1989) results show only weak correlations between wax erosion and leaf death. Inter-branch rubbing and rime ice also cause damage (Marchand and Chabot, 1978).

Krummholz

Tundra

I I I

I I

I

I Fig. 10. A wind switch in alpine krummholz/tundra (type 2, reaction).

296

J . B . WILSON A N D A. D. Q. AGNEW

It seems likely that more than one environmental factor is involved (Hadley and Smith, 1986); treeline is marginal for tree growth, making the plant susceptible to a range of adverse factors that may differ from place to place and from species to species. In any case branches and leaves can ameliorate wind abrasion and/or low temperature and/or desiccation, directly and via snow accumulation, and this enhances plant growth. If this were simply a type I , one-sided, switch, there could be no permanent mosaic, since patches could expand. Increased wind speeds between the shrub/tree patches due to funnelling could produce a type 2, reaction, switch. However, we have found no confirmation of this. If conditions in the open tundra prevent plant growth, there is a question of how tree islands become established. Marr (1977) had evidence for the intriguing suggestion that tree islands form in more sheltered microsites, in which a seed has germinated, but then by vegetative reproduction climb out and move along the ground onto exposed microsites. The islands can move at 1-2 cm year-' (Benedict, 1984).

3. Conclusion The elements are: (i) krummholz islands reduce wind and thus increase temperature and/or reduce desiccation and/or reduce abrasion, (ii) this favours the tree species, rather than the tundra species which are more tolerant of cold and/or desiccation and/or abrasion. The presence of an apparently long-term mosaic (A) situation suggests a reaction (type 3) switch is indeed operating.

IX. FIRE-MEDIATED SWITCHES

A. Concept Vegetation adapted to burning may be very sharply distinct from an adjacent non-flammable community because of an abrupt change in fire regime at the boundary. Boundaries between the two vegetation types could of course shift, depending on fire frequency and rainfall fluctuations, and yet in many places in the dry subtropics seem to be very stable (Lawton, 1964; van Zinderen Bakker, 1973), suggesting the delaying (D) outcome of a positivefeedback switch. A fire boundary often occurs where there is an underlying gradual change in soil nutrient or water status, so that fire is forcing an ecocline (limes divergens of van Leeuwen, 1966) into a linear edge, producing a limes

POSITIVE-FEEDBACK SWITCHES IN PLANT COMMUNITIES

297

convergens, an ecotone. This suggests the sharpening (A) outcome of a switch. Nevertheless, in order to demonstrate a switch, vegetation must be shown to maintain itself by either encouraging or limiting fire (Fig. 11). There is evidence for both these features. Firstly, Mutch (1970) suggests that certain vegetation is adapted to promote and survive fire better than alternative vegetation. Mutch points to the higher calorific content and lower water content of Eucalyptus spp. and Pinus ponderosa, compared with mesophytic vegetation, as evidence of this. However, Bowman and Wilson (1988) failed to find evidence to support Mutch’s hypothesis, and Snyder (1984) and Troumbis and Trabaud (1989) questioned its logic. Jackson (1 968) and Kellman ( 1 984) suggest another mechanism: the progressive loss of nutrients during a long history of fires in vegetation. Moreover, Kellman reviews neotropical savannahs and suggests that nutrientdeficiency can lead to vegetation of higher flammability. The opposing feature, the maintenance of non-flammability, is shown by certain forest types which are difficult to burn except in rare very dry years, due to lack of fuel for the fire because of fast litter humification in a humid microclimate (Ewel, 1976; Kessel, 1976; Bowman and Minchin, 1987). The best evidence for a fire switch comes from the closed-forest/savannah ecotones of Australia and Africa, and they will therefore be discussed in detail.

B. Australian Closed-forest/Savannah 1. Introduction In parts of Australia, closed rainforests and Eucalyptus-dominated sclerophyll savannah occur side-by-side, with boundaries perhaps only a few metres wide (Stocker, 1969; Unwin et al., 1985; Ash, 1988; Unwin, 1989). (We use the term “savannah” for consistency, though “woodland” would also be appropriate. Distinction could also be made between dry and wet sclerophyll communities.) The boundary can be floristically absolute: Smith and Guyer (1983) found that no tree species occurred in both savannah and closed-forest. In some cases, there can be a specific ecotone community also (Unwin, 1989). Such boundaries might remain stable for several thousand years (Mount, 1964; but see Unwin, 1989). Jackson (1968) described this boundary as an “unstable point, with ecological drift”, meaning that vegetation tends to drift to one or other of two stable states, each side of the “unstable point”, a concept very close to that of the switch. Usually, the general position of the closed-forest/savannah boundary is determined edaphically (e.g. with closed-forest on basaltic rocks and deep, fertile, well-drained soils), and/or climatically (Ash, 1988), with the fire

Closed-forest

Savannah I

I

litter

I I I Low light at ground level

Closed-forest

n

T9

0 Savannah

tires

II

tL<->> I

suppressed

I

\

pressed

/

Fig. 11. A fire/light(/nutrient) switch at the closed-rainforestjsavannahboundary in Africa or Australia (type 4,two-factor).

POSITIVE-FEEDBACK SWITCHES IN PLANT COMMUNITIES

299

switch sharpening the boundary (Bowman and Minchin, 1987), though the nature of the boundary and the factors involved may differ across Australia (Bowman, 1988).

2. Evidence: Closed-forest Conditions One element of the switch is that fire does not occur in the closed-forest, or only on the outer edge (Gilbert, 1959; Smith and Guyer, 1983). When fires sweep across the savannah, the dense evergreen vegetation of closed-forest presents a fire barrier (Bowman and Minchin, 1987). Direct evidence for this comes from an experimental fire lit by Unwin et al. (1985): it became slow burning and of low intensity as it approached the closed-forest. Scorch and fire deaths declined across the boundary into the closed-forest. Under the shade of the closed-forest, closed-forest species can presumably regenerate, perhaps in small gaps. In contrast, savannah eucalypts cannot regenerate in low light (Jackson, 1968; Ashton, 1981; Fig. l l ) , Jackson suggesting that clearings of 25 m diameter were necessary before there was enough light for eucalypt seedling survival.

3 Evidence: Savannah Conditions In the savannah, fire is frequent now (Braithwaite and Estbergs, 1985), was frequent during settlement by Aboriginals (Stocker, 1971; Singh et al., 1981), and may have been almost as frequent before that due to natural fires (Kemp, 198 1; Walker and Singh, 198 I ) . Savannah species accumulate fuel faster than closed-forest species, though the fuel mass can be higher in closed-forest (Mount, 1964; Bowman and Wilson, 1988). Savannah species and fuel components also show lower water contents (Bowman and Wilson, 1988). Dickinson and Kirkpatrick (1985) found that the dry sclerophyll savannah species of Tasmania had a greater tendency to propagate fire, though Bowman and Wilson (1 988) found such species from Northern Territory had a lower energy content than closed-forest species, and were less flammable. It has been suggested that closed-forest species cannot regenerate in a regime of frequent fire (Fig. 11). For example, Clayton-Greene and Beard (1985) found more closed-forest and less savannah on Bougainville Peninsula, Western Australia, than on the mainland. The closed forest appeared to be invading the grassland, and they suggested this was because fire was less frequent on the peninsula (no evidence presented). On some offshore islands, not burnt for some time (they state) there were saplings of closed forest species. Stocker (1966) found that an area of closed-forest disturbed and burnt had no regeneration of typically closed-forest species, though he later observed that many closed-forest species can sprout after less severe fire (Stocker, 1981). In contrast, savannah eucalypts are thought to be able to cope with fire, by surviving as adults (Gill, 1981), by vegetative recovery, especially from

300

J. B. WILSON A N D A. D. Q. AGNEW

lignotubers (Bowman and Minchin, 1987), or regenerating from seed (Cremer, 1962). Bowman (1986) suggested Eucalyptus tetrodonta regeneration was stimulated by fire; this may occur in other species too. An additional complication is that fire might reduce the nutrient content of the ecosystem by volatization, leaching and/or runoff, though the effect seems to be small (Norman and Wetselaar, 1960; Harwood and Jackson, 1975; Lacey et al., 1982). Jackson (1968) suggested this would favour eucalypts. In the short term, levels of other nutrients, especially P, can be higher after a fire (Humphreys and Lambert, 1965; Grove, 1977). Fire can also affect the microflora (Florence and Crocker, 1962). An interaction with termites may be important here, since Eucalyptus tetrodonta is very prone to termite attack (H. Gitay, personal communication), and fire might ameliorate this. Grazing can also interact, e.g. reducing fuel accumulation or preventing regeneration (Lacey et al., 1982), though at least Eucalyptus populnea density can be increased by grazing (Moore and Walker, 1972).

C. African Closed-forest/Savannah 1. Introduction For Africa, there is a rich literature on the role of fire in the maintenance of abrupt boundaries between closed-rainforest and savannah (Hopkins, 1983). Disturbance and grazing can reinforce the fire effect.

2. Evidence: Anthropogenic Savannah The present savannahs in Africa are largely derived by agricultural clearance of closed-forest. The sequence of events seems to be as follows: (i) Closed-forest is cleared for agricultural use, and after some years left fallow. (ii) The grassland which then develops is burnt by human agency. Grass is an ideal substrate for fire, especially after flowering, because standing culms form a fuel-air mixture. (iii) Fires at first erode the edge of the closed-forest, and can even enter the closed forest, but burn only the thin dry litter layers, damaging little within the canopy (Hopkins, 1965). (iv) The edge is sharpened firstly by rearrangement of flora to form a specialized, semi fire-resistant closed-forest border zone, particularly of evergreen lianas (e.g. Landolphia spp. -Trapnell, 1959; Hypoestes fastigiata-West, 1972). Secondly the grassland can alter soil water relations, making it more difficult for forest to re-invade. There is decreased infiltration of the high-density rainfall (Lawton, 1964) and thus lower effective precipitation, and also more effective use of surface soil water, denying it to the deeper tree roots (Knoop and Walker, 1985).

POSITIVE-FEEDBACK SWITCHES IN PLANT COMMUNITIES

30 1

(v) The final result is a mosaic of closed-forest and grassland, sharply defined (Keay, 1951; Hopkins, 1983).

3. Evidence: Natural Savannah In view of the largely anthropomorphic origin of the present vegetation of the Sudanian Zone, it is doubtful whether the contention that this was the natural, pre-disturbance state of the closed-forest/grassland boundary can be proven (Clayton, 1958). However, Keay (1951) proposed that the present situation is one which formerly existed at the climatic boundary of closedforest vegetation and gave rise there to a mosaic of forest and grassland. van Zinderen Bakker (1973) describes relict forest patches in South Africa on steep slopes and in ravines, which have affinities with more equatorial montane forest. He demonstrates an ameliorated microclimate in which litter can decompose more quickly into a soil of high organic content. A mosaic could probably develop from grassland, as early or no burning leads, at least in West Africa, to establishment of fire-intolerant trees and a suppression of the fuel grasses, so that it becomes progressively more difficult to burn (Innes, 1972).

D. Conclusion In both the Australian and African examples, this is a type 4 (two-factor) switch, with elements for one factor: (i) the savannah increases the probability of fire by being drier and more flammable, (ii) the savannah is more tolerant of fire, and for the other factor: (i) the closed-forest decreases the light level on the ground, (ii) seedlings of species of the closed-forest are more tolerant of shade than those of the savannah. However, other switches are often involved, especially grazing ones. The switches can lead to mosaic (A) and,sharpening (B) situations. In theory they could produce a delay (D) situation, which the permanence of the boundaries suggests. The buildup of non-flammable litter in closed-forest could be seen as producing a type 3, symmetric, switch, depending on what is taken as the baseline.

X. ALLELOPATHY-MEDIATED SWITCHES 1. Concept Many plant species are suspected of maintaining the integrity of their own stand by allelopathic influence (Rice, 1984). This process represents a

302

J . B. WILSON AND A . D. Q. AGNEW

toxins leached

Toxins temporarily in soil

other species Fig. 12. An allelopathy switch in southern Californian annual grassland (type I , onesided).

positive-feedback switch (Fig. 12). Allelopathy has especially been used to explain mosaics, and almost pure stands of one species, sometimes seen in semi-arid environments. The allelopathy has to be allo-allelopathy, rather than auto-allelopathy (Newman, 1978).

2. Evidence Bell and Muller (1973) describe a situation in southern Californian annual grassland (comprising species such as Avena fatua, Bromus mollis and B. rigidus), with patches of almost pure Brassica nigra reseeding themselves each year. Boundaries between grassland and the Brassica were often sharp. Bell and Muller failed to find explanation for the existence of the patches in soil pH, texture or temperature. Competition for water, nutrients or light seemed unlikely. They suggested that toxins from Brassica nigra were washed from dead standing Brassica by the first rainfall, and the toxins suppressed germination of the grasses at this critical time. Such toxicity was shown in vitro and also in soil. Removal of dead Brassica removed the suppression of grass germination. The existence of a mosaic with no apparent environmental correlation, and the sharp boundaries between communities, are suggestive of a switch. Williamson (1 990) suggests that in the chaparral of the southeastern USA coastal plain, allelopathy by the shrubs inhibits the growth of grasses and young pines, therefore reducing the fire risk to the benefit of the shrubs. This could be an allelopathy/fire switch, capable of producing the sharp ecotone that is seen.

POSITIVE-FEEDBACK SWITCHES IN P L A N T COMMUNITIES

303

However, many examples of allelopathy are suspect because they can be demonstrated in the laboratory but not in the field, where other environmental effects such as herbivory may be more important than any plant vs. plant toxin (Harper, 1977; Heisey and Delwiche, 1985), and where toxins may be broken down by the soil flora or absorbed by soil colloids. Bell and Muller’s (1973) study is not free from such criticisms. However, it seems extreme to dismiss allelopathy because the evidence is partial; evidence in ecology is rarely complete. A comparable switch could operate by one species bearing a virus, pest or disease, to which a second species was more susceptible (van den Bergh and Elberse, 1962; Rice and Westoby, 1982; Price et al., 1988). This moves the mechanism into the category of a biological switch.

3. Conclusion This seems to be a one-sided (type 1) switch: (i) a species produces an allotoxin; and (ii) the toxin inhibits the growth of other species, allowing greater growth of the toxin-producer. The literature often gives the impression of a mosaic (A) situation. However, a one-sided switch cannot give this. Perhaps there is an unobserved underlying environmental difference, and the situation is a sharpening (B) one.

XI. MICROBE-MEDIATED SWITCHES A. Oldfield Succession and Nitrogen-fixing Microbes 1. Concept Rice et a f . (1960) suggested that pioneer species in oldfields might tolerate low soil nitrogen. If they inhibited bacteria which fixed atmospheric nitrogen, soil N would remain low, favouring the pioneers over the later successional, and more N-requiring, species, halting or at least slowing succession (Fig. 13)-a positive-feedback switch.

2. Evidence In Oklahoma oldfields, succession after abandonment starts with an annual weed stage [Pl] (Booth, 1941; Kapustka and Rice, 1976). The annual grass stage [P2] follows, dominated by Aristida oligantha. The latter stage lasts 1124 years; attempts to reduce this time by seeding-in later successional species have not been successful (Booth, 1941). The following perennial bunch-grass

304

J . B. WILSON AND A. D. Q. AGNEW

Inhibitors of N-fixing bacteria released

Soil N

L-

Bunch grasses of P2 stage cannot invade.

Fig. 13. A microbe switch in Oklahoma oldfields (type 1, one-sided).

stage [P3], often dominated by Andropogon scoparius, is eventually replaced by climax prairie [P4]. Succession to climax is incomplete after 40 years (Savage and Runyon, 1937). Rice et al. (1960) and Rice (1971) saw it as an ecological problem that the P2 annual grass stage so quickly displaced the robust species of the PI annual weed stage, and remained so long before invaded by the P3 perennial bunchgrass stage. They suggested that in this system cultivation led to a decrease in the N and P in the soil, so recently abandoned land was low in soil N. (Rice et al. (1960) refers to some evidence for this: a preliminary experiment at his site, and experiments by others at other, comparable sites.) N-fixation was then inhibited by substances from the P2 species. Plant extracts and sometimes leachates, inhibitory of the free-living Nfixer Azotobacter spp., can be found from species of the PI and P2 stages (Bromus japonicus, Digitaria sanguinalis. Chenopodium album, Conyza canadensis), though also in Andropogon scoparius and Erigeron strigosus of the P3 and P4 stages (Rice, 1964a, 1965b). Although the toxins were tested in vitro, the inhibitors were stable against auto-oxidation and decomposition by microbes (Rice, 1964a, 1968). Inhibition of bacteria of this genus may not be important, since Azotobacter generally makes little contribution to soil N (Evans and Barber, 1977), but overall N-fixing ability (measured by acetylene reduction) is indeed low in the field in the PI and especially P2 stages (Kapustka and Rice, 1976). Rice (1 968) also considered inhibition of Fabaceae/Rhizobium N-fixation. Extracts, dried plant material, root exudates and living plants of stage PI species Bromusjaponicus and Digitaria sanguinalis, and also the P2 dominant

POSITIVE-FEEDBACK SWITCHES IN PLANT COMMUNITIES

305

Aristida oligantha, reduced the nodulation of bean and clover plants. Unfortunately for the argument, there is no significant presence by Fabaceae in stages P1 or P2, though that could conceivably be due to Rhizobium inhibition. For some species, partial chemical identification of the inhibitors has been made (Rice, 1964b, 1965a,b; Rice and Parenti, 1967). However, questions must remain as to the reality of these effects in the field. The other link in demonstrating a switch would be to show that species of the early stages are more tolerant of low-nutrient conditions. Rice et ai. (1960) found that Aristida oligantha (P2) was less affected by N- or Pdeficiency than Andropogon scoparius (P3), which was in turn less affected than Panicum virgatum (P3 and P4). The low-nutrient tolerance of Aristida oligantha is supported by its low N- and P-concentration in the field (Harper et al., 1934). Effects of plant exudates in inhibiting nitrification are also possible. The soil concentration of ammonium N is highest in the climax, and nitrate N the lowest (Rice and Pancholy, 1972, 1973). This correlates with low Nitrosomonas and Nitrobacter numbers in climax soil. Extracts of species from P1, P3 and P4 stages are inhibitory to Nitrobacter and Nitrosomonas (Rice, 1964a, 1965b). However, no differential ability by species of different stages to use ammonium vs. nitrate has been shown. Other possible explanations for the speed of succession between the P2 and P3 stages include dispersal limitation (Rice et al., 1960), allo-allelopathy (Rice, 1984) and auto-allelopathy (Rice, 1971).

3. Conclusion This would be a type 1, one-sided, switch, in which: (i) the pioneers produce toxins, which inhibit N-fixing (etc.) bacteria, which keeps the soil N low, (ii) the low soil N favours the pioneer species in comparison to later successional species. One would not expect a mosaic (A) situation from a one-sided switch, and Rice does not mention mosaics. The situation is a delay (D) one.

B. Forests and Mycorrhizas Many plants, including forest trees, are dependent on mycorrhizal symbiosis for satisfactory growth. If the mycorrhizal fungus is not present in the soil at a site, they could not establish. Yet, the mycorrhizal fungi could not establish without a higher-plant host. This represents a positive-feedback switch. Perry et al. (1989) suggest that such a switch may operate when forest is removed;

306

J . B. WILSON AND A . D . Q. AGNEW

enrich soil

increased

trees

Thicket with

I

nests in thickets

Fig. 14. One possibility of a termite switch in African savannah (type 1, one-sided).

the mycorrhizal fungi could be lost too, making it very difficult and slow for the trees to re-establish.

XII. TERMITE-MEDIATED SWITCHES 1. Concept Termites of many genera make a nest or hive in the centre of their foraging area. Sometimes, it is a raised mound. Plant material, litter and soil from the foraging area are brought back to the hive. This, and the termites’ creation of subterranean passages, amend the soil structure. These soil conditions may enhance the growth of woody vegetation. If the termites are themselves favoured by the higher productivity of the woody vegetation, then a positivefeedback switch exists (Fig. 14).

2. Evidence In many parts of dry tropical/subtropical Africa, scattered mounds occur in grassland. They are often referred to as Mima-like mounds, because of their superficial similarity to mounds on the Mima Prairie, Washington State, and in other North American grasslands, which may have been caused by gopher burrowing (Cox, 1984; Lovegrove and Siegfried, 1986). It seems likely that Mima-like mounds in Africa are occupied or abandoned termite (mainly Mucrotermes spp.) mounds (Darlington, 1985). The system has been called Termitensavanna (Darlington, 1985). However, this interpretation is surprisingly controversial (Gakahu and Cox, 1984; Berg, 1990).

POSITIVE-FEEDBACK SWITCHES IN PLANT COMMUNITIES

307

In many areas, there is a mosaic with patches of trees/shrubs, often evergreen, scattered in the grassland, with a sharp ecotone (Morison et al., 1948; Lee and Wood, 1971a). Often, the patches are on Mima-like mounds. This suggests that mound conditions promote either colonization or survival of woody plants (Okali et al., 1973). The difference in mound conditions could be: (i) Nutrients: The soil of termitaria contains material harvested from the surroundings and from the subsoil (Hesse, 1955; Lee and Wood, 1971b). It is often nutrient-enriched (in N, Ca and perhaps P and K) compared to the surrounding grasslands (Watson, 1967; Lee and Wood, 1971a; Trapnell et al., 1976; Nutting et al., 1987; LopezHernandez er al., 1989). (ii) Soil aeration: The soil of termitaria is often finer and better aerated (Lee and Wood, 1971a). (iii) Soil moisture: This is often higher in termitaria, perhaps because of differences in rainfall infiltration (Glover et al., 1964) and in waterholding capacity (Lee and Wood, 1971a). (iv) Drainage: On some sites the intermound surface is seasonally waterlogged. (v) Temperature: Higher temperatures on mounds could encourage woody vegetation (Wild, 1952). (vi) Grazing: The height of the mounds might protect saplings from grazing (Lind and Morrison, 1974). (vii) Fire: The zone of bare soil at the base of the mounds can act as a fire break (Lind and Morrison, 1974). Any or all of these factors might make termitaria better sites for tree establishment and growth, and explain the presence of the thickets. There are two ways in which the woody vegetation could favour termite colonies, thus closing the positive-feedback loop, and constituting a switch: (i) Higher productivity of vegetation on the mound could be a resource of leaf or wood litter food for the termites. This explanation would not hold in communities where termites forage far from their nests out into the grassland, or where their primary food is grass and/or grass litter (Darlington, 1982). However, Darlington (1985) suggests that the mounds of Termitensavanna in the Kenya Highlands are built by Odontotermes spp., for which wood is a major diet component (Wood, 1978). (ii) Termites might be more likely to create new nests in the shade of a thicket, as Morison et al. (1948) suggested. Their evidence was weak, and the suggestion is difficult to investigate because nests are established rarely, persisting for decades or centuries (Watson, 1967; Goudie, 1988), but especially because termites tend to create new nests

308

J . B. WILSON AND A. D . Q. AGNEW

in or near old nests, even in the absence of woody vegetation on the mound (Goudie, 1988). Although the association between mounds and woody plants is well established, and explicable, and the sharp boundaries suggest a switch, mutual relations between the vegetation and the energetics of the termite nest d o not appear to have been investigated. If a switch does operate in the way we envisage, this would benefit the termites, and could have arisen as coevolution of a mutualism. Our suggestion of a switch would fail if the woody vegetation grows on a termite mound only after the colony dies, as Lind and Morrison (1974) suggest, but Morison et af. (1948) reported that one of the thicket-topped termite mounds that they excavated had at least been recently occupied. In Australia, abandoned termite mounds persist only for 3-10 years (depending on the species of termite), but we know of no such estimates for Africa.

3. Conclusion There is a possible one-sided (type 1) switch, in which: (i) once termites form a mound, woody vegetation can invade because of higher nutrient levels (or higher temperature, or freedom from grazing, fire, etc.); and (ii) woody vegetation may favour the formation or success of termite mounds. Superficially, this would be a type 1, one-sided, switch. If, however, there were an overall limitation in food supply for the termites, or in nutrients, etc., the areas between the mounds might become impoverished, producing a type 2, reaction, switch, able to produce the mosaics seen. We must regard the existence of a termite switch as unproven.

XIII. HERBIVORE-MEDIATED SWITCHES A. Concept If a grazer avoids species X, preferentially grazing species Y, and species X is less tolerant of grazing than species Y, a positive-feedback switch is possible in which areas with species X are less grazed, therefore species X has a relative advantage in them, therefore those areas are even less grazed (Fig. 15). Grazers also have the ability, where micturition/defaecation occurs in the same patches as grazing, to create vegetation mosaics by enhancing nutrient availability. This sometimes represents a positive feedback switch

FestudAgrostis patch

I I

I I

Nardusdominated

dominated patch

I

suppressed

I Fig. 15. A herbivore switch in British upland grassland (type 2, reaction; or type 3, symmetric).

POSITIVE-FEEDBACK SWITCHES IN PLANT COMMUNITIES

Nardus patch

2 E!

309

w

0 \o

310

J. B. WILSON AND A. D. Q. AGNEW

because the enriched vegetation is more palatable, at least after the initial smell has disappeared.

B. Grass/Grass Boundary In the British uplands, Nardus stricta is less palatable to most stock than the other grass species it grows with. Cattle and especially sheep avoid Nardus patches almost completely in spring and summer, in favour of Agrostis spp., Festuca spp., etc. (Milton, 1953; Hunter, 1954; Rawes, 1961; Hunter, 1962; Nicholson et al., 1970; Grant et al., 1987). Even when the grasses grow intermixed, sheep avoid the Nardus (Grant et al., 1985). Nardus stricta is also slow to recover from clipping (Rawes, 1961; Atkinson, 1986), and therefore on some soils close or repeated clipping will result in a decrease in the cover of Nardus (Chadwick, 1960a; Nicholson et al., 1970). The net result is, sometimes, for grazing to increase the relative amount of Nardus in the sward because it is unpalatable (Welch and Rawes, 1964; Welch, 1986), and sometimes to decrease it because it is intolerant of defoliation (Nicholson et al., 1970; Rawes, 1981), or the two may balance (Floate et al., 1973; Rawes, 1981). Once patchiness in species composition arises, the potential exists for a switch, in which patches of Nardus are less grazed, therefore Nardus is favoured in them, and therefore they are less grazed (Fig. 15). A switch might result in a mosaic (A) situation, and mosaics have been seen in Nardus communities (Kershaw, 1957; Chadwick, 1960b), though Chadwick interpreted the Nardus patches as colonization. The switch alone would lead to small-scale mosaics, but the flocking behaviour of the sheep can enlarge the mosaic to landscape scale. There is also a difference in the soils on which the two states occur, though that difference is caused par;tly by the nature of the litter produced by the two types. Nicholson et al. (1970) almost recognized this situation as a switch under the term “pasture differentiation” which “maintains the stability of the plant community”, though they misleadingly called it cyclic succession. A similar switch may be operating to help maintain the Andropogon greenwayi mosaic in the Serengeti grasslands of East Africa (Belsky, 1986). Belsky suggests that the mosaic is primarily determined by rainfall infiltration (see above), but the Andropogon is more tolerant of the defoliation caused by antelope grazing than is the alternative-phase Chloris pycnothrix (Banyikwa, 1987). Without grazing the Andropogon disappears. Nutrient enrichment of the Andropogon patches may also be involved.

POSITIVE-FEEDBACK SWITCHES IN PLANT COMMUNITIES

31 I

C. Grass/Woodland Boundary Browsers maintain edges where broad-leaved woody vegetation is presented at an available height, and thus prevent the woody plant community from expanding into grassland. Successional species are said to be more palatable and less chemically protected than climax woodland species (Cates and Orians, 1975), although others doubt this (Crawley, 1983). Vesey-Fitzgerald ( I 972) suggests that grazers can sharpen boundaries between Tanzanian grassland and forest. Fire follows grass cover into and through forest. Fully burnt patches then differentiate into pure grassland which is grazed, adjacent to relict forest which is not. The grazing removes so much litter from forest edges that fire is unable to repeat, or it favours creeping grass such as Pennisetum kikuyorum. Litter of the Pennisetum decomposes fast, so it does not carry fire. The species grows along forest margins, stopping fire damaging trees at the edge, and the forest thickens, with enhancement of the boundary. This scenario seems to be special to high but variable rainfall conditions in montane Africa where the initial burn is associated with exceptionally dry years after a period of undisturbed forest growth and litter accumulation. Lock (1977) gives a similar example in the Ruwenzori National Park, Uganda, in which Capparis tomentosa thickets are ringed by grassland, with an abrupt boundary. Spread of the Capparis is apparently prevented by heavy, and selective, herbivore pressure, perhaps related to the high protein and other nutrient content of the Capparis (Field, 1971). The trampling leads to a bare zone around the thicket, lowering the fire risk for the thicket.

D. Grazing and Nitrogen Cycling Trumble and Woodroffe (1954) suggested that, in arid Australian shrubland, grazing could stimulate shrub growth by increasing nitrogen input. They also suggested that the shrub species responded more to this increase in N than did associated species, which would produce a switch. There is considerable evidence for a similar switch, operated by lesser snow geese, on the salt marshes of Hudson Bay, Canada. Exclosures show that goose grazing increases shoot productivity by 7&80% (Cargill and Jefferies, 1984; Hik and Jefferies, 1990). There is evidence this is due to increased N supply: clipping without the deposition of faeces does not increase productivity, and experimental removal of naturally deposited faeces removes the increase that occurs naturally (Hik and Jefferies, 1990); N contents of Puccinellia phryganodes and Carex subspathacea are higher in grazed patches (Cargill and Jefferies, 1984); and experimental addition of goose faeces stimulates production and raises the N concentration of the herbage (Bazely and Jefferies, 1985).

312

J . B . WILSON AND A . D. Q. AGNEW

There must be some removal of N from the grazed patches, but Cargill and Jefferies (1984) estimated this at less than 2.2 g m-' year-'. It is probably about balanced by greater N fixation by blue-green algae, higher in grazed areas because there is more bare ground. Bazely and Jefferies (1985) attribute the effects of grazing to faster N cycling rather than to an increased quantity of N in the system. For this to operate as a switch, the grazed areas must attract the grazers. Not only are the faster-growing plants likely to be more palatable, but the higher N content will itself attract them (Cargill and Jefferies, 1984). Grazing is certainly intense in grazed patches, geese removing 8&98% of the annual shoot production (Cargill and Jefferies, 1984). Patchiness is heightened by the behaviour of snow geese, nesting in dense colonies and feeding nearby (Cargill and Jefferies, 1984). This is not strictly a vegetation switch unless there is an effect on species composition, and experiments show Carex subspathacea, characteristic of the grazed areas, to be less abundant when grazing is allowed (Bazely and Jefferies, 1986).

E. Insects in Pine Forests In the southeastern USA, where the climax is thought to be hardwood forest, there is a pyric subclimax of pine. After storm disturbance, southern pine beetle invades the damaged pine trees, killing a patch of trees (Coulson, 1979; Schowalter, 1985). The dead wood is easily ignited by lightening fires. Fire destroys any hardwoods present, and pines re-establish.

F. Conclusion The majority of these examples are basically one-sided (type 1) switches, in which:

(i) areas with abundance of the unpalatable and grazing-intolerant species are avoided, ' (ii) reduced grazing increases the abundance of the species in those areas. At least in the Nardus case, where stock numbers are agriculturally manipulated, the tendency for stock to avoid the Nardus patches will concentrate them on the other areas, producing a reaction (type 2 ) switch, or it could be said to be a symmetric (type 3) switch, depending on the baseline. Either has the potential to create a mosaic, which is observed. The same effect could arise in natural ecosystems if animal numbers were limited by a factor other than forage supply. In the other cases, the switch seems to be a simple onesided one, and only a sharpening (B) situation occurs. (The grazing switch differs from the pest pressure mechanism of coexist-

POSITIVE-FEEDBACK SWITCHES IN PLANT COMMUNITIES

313

ence (Connell, 1978; Fowler, 1988; Wilson, 1990) in that the coexistence is between patches, not within, and in that there is no requirement for a herbivore with frequency-dependent preferences or for multiple herbivores with different preferences.)

XIV. DISCUSSION 1. The Concept The switch concept has not been widely recognized as a general principle in plant ecology, but several ecologists have referred to aspects of the process. For example, Daubenmire (1968) said of ecotones: “While sharp differences in the physical environment may be demonstrated on either side of the ecotones, they are all secondary factors wholly attributable to community influence”. Dice (1952), an ecologist with an integrated concept of the plant community similar to Daubenmire’s, wrote: “Community boundaries are likely to be sharper where the community exerts considerable control over the habitat”. Thus, even workers strongly espousing an integrated concept of plant communities, who would expect sharp boundaries for phytosociological reasons (i.e. community assembly rules), admit that sharp boundaries are generally caused by the process we have called the switch. Gleason (1917), a worker with a concept of plant communities very different from Daubenmire’s or Dice’s, said that sometimes modification of the environment by plants caused a sharp boundary between associations where there was originally only a gradual change in the physical environment, or none.

2. Switches and Community Structure Gleason (1939) listed three possible concepts of the plant community, which we can see as combinations of deterministic vs. stochastic organization, and discrete vs. continuous variation: (i) “The association is a quasi-organism”, i.e. the community is deterministic and discrete, or “integrated”. This view is normally associated with Clements (e.g. 1905). (ii) “The association is a series of separate similar units . . . , repeated in numerous examples”, i.e. deterministic and continuous. This coincides with the views of Whittaker (Whittaker and Woodwell, 1970; Whittaker, 1975; Whittaker and Levin, 1975), in which coevolution is important, but acts to spread species out along an environmental gradient, not to aggregate them. (iii) “The vegetation-unit is a temporary and fluctuating phenomenon”, i.e. stochastic and continuous. This is Gleason’s own Individualistic theory.

314

J. B. WILSON AND A. D. Q. ACNEW

I

altitude Key - - -

= =

1

=

gradient in abiotic environment gradient in the environment after biotic modification position of ecotone

Fig. 16. The gradient with a temperature switch at treeline. Below the ecotone, frosts are moderated by the tree canopy.

This leaves open the fourth possible combination, stochastic organization but with discrete boundaries. Gleason may have omitted it because it seems, at first sight, an illogical combination. Yet this combination represents the view of many plant ecologjsts that community structure is loose, combined with their observation that boundaries are sometimes sharp. If communities are not tightly structured, how can they have sharp boundaries? We suggest the answer is that the sharp boundaries arise by switches.

3. Types of Switch We have shown that four types of vegetation switch are possible. For example, fog-precipitation mediated forest (e.g. Kummerow, 1962, in Chile) is bounded by grassland that is relatively passive in its effect on water balance, so this is a one-sided switch (type l), as are hyperaccumulation of heavy-metals (e.g. Wild, 1974) and microbe-mediated allelopathy (e.g. Rice, 1971).

POSITIVE-FEEDBACK SWITCHES IN PLANT COMMUNITIES

I

315

altitude

Fig. 17. The gradient with a fog-precipitation switch on a hillside in the montane tropics. Above the ecotone, the tall tree canopy traps fog as precipitation. Just below the ecotone, precipitation may be somewhat reduced by interception by the nearby trees. Symbols as Fig. 16.

Reaction switches (type 2) can be suggested for East African infiltrationinduced mosaics (e.g. Belsky, 1986), for salt pans (Yapp and Johns, 1917), for alpine krummholz islands (Marr, 1977), and for grazing when overall grazer numbers are externally controlled. We have found only one example that we consider a symmetric (type 3) switch-the savannah/forest boundary. If Kellman’s (1984) hypothesis is correct, the savannah provides flammable fuels, while the forest provides a fast humification environment so that fuel is removed quickly. Our only probable example of a two-factor (type 4) switch has been the same savannah/forest boundary, but with light effects considered. We cannot tell to what extent the apparent rarity of type 3 and 4 switches is due to lack of investigation, but unless switches are very common it is likely that the several switch elements required will come together but rarely.

4 . Boundaries in Space and Time The temperature switch at alpine treeline (Fig. 16) may give a simple onesided (type 1) sharpening of an environmental gradient in temperature. The fog precipitation switch can similarly sharpen a precipitation gradient (Fig.

316

J . B. WILSON AND A. D. Q. AGNEW

I

I Q

Calluna etc. I

distance/mosaic Fig. 18. The gradient with a pH switch. At one side of the ecotone, pH is lowered by the Calluna (etc.). Symbols as Fig. 16.

17), though possible reduction in precipitation in grassland nearby would create almost a reaction (type 2) switch. We envisage the same switch could, from random colonization, produce a mosaic where the underlying physical environment shows no trend. This situation may be the norm for the pH switch with Calluna vulgaris (etc.) (Fig. 18). Alternatively, an underlying environmental gradient may establish species’ positions, but the boundary between communities be sharpened by a switch mechanism in a quite different environmental/resource factor. An example is Clarke and Hannon’s (1971) salt marsh salinity gradient (Fig. 19), where the switch to the Avicennia is mediated through the light resource. In a few cases, such as the closed-forestlsavannah boundary (Fig. 20), the sides of a boundary can be mediated by different environmental features (two-factor, type 4, switches). Often, a difference in physiognomy is seen between the two states of a switch. However, this may to some extent be an artefact, in that ecologists have sought explanations for vegetation patterns where they are most striking.

POSITIVE-FEEDBACK SWITCHES IN PLANT COMMUNITIES

317

Arthrocnemum

distance upshore

distance upshore

Fig. 19. The gradient with a light switch on NSW (Australia) salt marshes. Light at ground level is reduced more by Avicennia tree cover than by Arthrocnemum shrubs. Symbols as Fig. 16.

Environmental/vegetational gradients may occur at various spatial scales and be of various ages. These are correlated. Old patterns are generally spatially large-scale. Thus, processes on a geological time scale generally cause landscape features, a clear example of which is the gradual leaching and impoverishment of soils following a glaciation (Walker, 1965). We suggest a soil element (NPK) switch may hasten the final impoverishment. At the smallest spatial scales, with generally young patterns, resources can be patterned by individual populations of plants or animals, such as by allelopathic dominant plants or by termites. We have described ecological outcomes in terms of the creation of spatial boundaries (situation A) or their sharpening (situation B), and also of the sharpening of time boundaries in succession, either by acceleration (C) or by delay (D). In some cases, the same alternative states may be found across either a spatial boundary, or a time boundary. Indeed, as Smith and Huston ( 1 989) suggest, spatial and temporal zonation can be seen as products of the same process. Drake (1990) and Case (1990) have produced theoretical evidence that alternative stable community states could also arise from direct interactions between species. No environmental change is involved, so the process would not be included in our definition of a switch. However, in the real world, interactions between plant species are mediated by the environment (Clements et al., 1929), so it is likely that a switch would be operating.

1

Savannah

(b)

I

- 3 j

(c) 3osedforest

I

1

Savannah

Closed forest/

Savannah distance

distance

distance

Fig. 20. The gradient with a fire/light(/nutrient) switch at the closed-rainforest/savannah boundary in Africa or Australia. Symbols as Fig. 16. (a) In the closed-forest, wet non-flammable litter rarely carries fire; in the savannah, dry flammable litter leads to frequent fire. (b) In the closed-forest, the dense canopy creates shade at ground level; in the savannah, the open canopy allows much light through. (c) In the closed-forest, nutrients accumulate; in the savannah, frequent fire may lead to some nutrient loss by volatilization and runoff.

POSITIVE-FEEDBACK SWITCHES IN PLANT COMMUNITIES

319

5 . Switches and Landscape Landscapes where large-scale spatial switches are operating are often mosaics. Either there are multiple stable conditions from one starting condition (Sutherland, 1974), or an initial environmental heterogeneity is stabilized and magnified by the plant communities occupying each state. The effect is seen as a mosaic of vegetation types, most frequently a mosaic of physiognomic types. Because a mosaic landscape is one in which diversity is perceived to be great, it is of value in recreation areas, and to conservation. Geomorphological, geological and climatically imposed mosaics can look very similar to those caused or enhanced by switches. For instance, a krummholz/alpine grassland mosaic can be caused either by an erosion pattern, by a snow cover pattern, or by fire. Any of these may have their ultimate origin in the underlying geology. Like many other features of plant ecology, human influence is a major current cause of ecotones, but the switch principle presents the ecotone as a potentially ancient feature of landscapes. It is therefore important to extend our knowledge of the processes involved, and to recognize the importance of the switch principle in mosaic initiation and maintenance.

6. Keystone Species In many cases a single species of plant may be the prime initiator of a switch mechanism, for instance Orbignya cohune (salinity switch), Calluna vulgaris (soil pH switch) or Picea engelmannii (wind switch). In these cases the plant species involved is a “keystone species” (Paine, 1969; Grabherr, 1989). In the grazing or termite-mediated switches we have discussed, the animal can be seen as a keystone species. If the concept is enlarged by adding the idea of “keystone factors” (Williams, 1980), it would include many of the switch mechanisms described here.

7. Switches vs. Succession It is not always easy from observation of vegetation to distinguish a switch from flora replacement due to some other process, e.g. facilitation succession. Conway (1949), working in the area where Heilman (1968) and Heinselman (1970) described Sphagnum as a switch species from forest to bog, suggested that succession was stepwise, a series of punctuated equilibria in ecological time, with transitions between them triggered by rainfall swings, or raising or lowering of the water table. Perhaps a closer investigation of the timing of events in facilitation successions would often show punctuated equilibria, with each equilibrium made temporarily stable by the delaying (D) outcome of a switch. Facilitation succession has traditionally been viewed as due to autogenic environmental change making conditions less suitable for the present inhabitants and more suitable for their successors. Perhaps it is the latter aspect

320

J. B. WILSON AND A. D . Q. AGNEW

that is important in most cases, and the present inhabitants decline only because of competition from their successors (Connell and Slatyer, 1977). One way or another, Connell and Slatyer’s (1977) tolerance and inhibition models follow the facilitation model in that “early occupants modify the environment so that it is unsuitable for further recruitment of these earlysuccession species”. However, in the inhibition model, recruitment of all species is .equally inhibited, late-successional species dominating only because of their longevity. The concept of a plant holding its site, presumably because competition is mainly for light and therefore cumulative (Wilson, 1988), is superficially similar to a switch, but differs in that the plant inhibits its own species equally. An indicator of the operation of a switch, rather than facilitation succession, is the presence of boundaries between vegetation types that are sharp in space or time, since switches can produce sharp boundaries, but facilitation succession cannot. We could extend the punctuated equilibrium idea to spatial arrangements. This is another way of looking at a mosaic of vegetation types. The operation of a switch in succession could lead to a temporary mosaic if succession is accelerated in some patches. Yarranton and Morrison (1974) described such a mosaic. They attributed it to facilitation succession, though this leaves open the origin for the patches. It is possible that a switch occurs at the level of bare ground vs. vegetation, but that facilitation succession occurs within vegetated patches. Cyclic succession is also a process contrasting with switches. In cyclic succession, the species/communities modify the environment to make it less suitable for themselves, the opposite of switches. Yet cyclic succession, like switches, can generate mosaics (Watt, 1947). An interesting case is the “regeneration complex” (e.g. Osvald, 1923), previously interpreted as cyclic succession, but now believed by many mire ecologists to be a switch. We might expect cyclic succession mosaics to have fuzzier boundaries between patches than switch mosaics, but neither have been quantified. The scale of switch mosaics will probably tend to coarsen through time, but it is not clear what Watt envisaged for cyclic succession mosaics. The crucial difference is that the states of a switch mosaic will be stable, those of a cyclic succession mosaic will oscillate. Westoby et al. (1989) envisaged alternative stable states, but with oscillation between them. Our concept of switches does not involve such oscillation; often patches will be stable almost indefinitely. However, the difference is one of degree. In our concept, the vegetation of a patch will be changed by drastic disturbance, depending on the magnitude of the disturbance and the degree and reversibility of environmental modification effected by the switch. A test of whether a switch is operating should be that when a switchcaused boundary or mosaic is disturbed, it should re-impose itself only very

32 I slowly, and then probably not by direct replacement of the original species. For example, if our suggested termite switch operates, and the termite mounds were destroyed, the thicket should disappear in time too. POSITIVE-FEEDBACK SWITCHES IN PLANT COMMUNITIES

8. Importance of Switches We cannot estimate how common switches are, but we have been able to find many examples, in spite of the concept often not being recognized by the original authors. It is generally accepted that plants modify their environment (e.g. Miles, 1985), and it would be strange if this were always in the direction that disfavoured their persistence. We therefore expect switches are quite widespread. If natural selection is important in structuring plant communities, one would expect that genotypes that modified their environment in their favour would be selected, and switches would therefore become common. In most of the examples given above, a link is missing in the chain of evidence, probably because the switch principle has not been recognized, and the evidence has not been sought. In fact, the evidence for switches is firmer than that for facilitation succession or for cyclic succession. Egler (1977) dramatized, with his $10000 challenge, the paucity of any evidence for facilitation succession. And although Watt (1 947) offered evidence for cyclic succession, it was circumstantial, and further investigation has shown some of Watt’s own examples to be more complex than he suggested (de Hullu and Gimingham, 1984; Marrs and Hicks, 1986; Gimingham, 1988; Svensson, 1988). In contrast, although some of our examples of switches are speculative, there is firm evidence for the elements of many, and there are cases of mosaics and sharp boundaries that are hard to explain any other way.

ACKNOWLEDGEMENTS For comments on drafts we thank A.J.M. Baker, the late J.J. Barkman, P. Bannister, K.J.M. Dickinson, A.H. Fitter, C.H. Gimingham, H. Gitay, S.A. Grant, W.McG. King, W.G. Lee, A.F. Mark, T.R. Partridge, S.H. Roxburgh, R.S. Tangney, P. Wardle and M. Westoby.

REFERENCES Agnew, A.D.Q. (1961). Studies on the plant ecology of the Jezira of central Iraq. 11. Hummock formation by Haloxylon salicornicum (Moq.) Bge. Bull. Coll. Sci., Baghdad 6, 42-60. Agnew, A.D.Q. and Haines, R.W. (1960). Studies on the plant ecology of the Jezira of central Iraq. I. The influence of perennials on annuals in the Jezira. Bull. Coll. Sci., Baghdad 541-60.

322

J. B. WILSON AND A. D. Q. AGNEW

Arnason, J.T., Lambert, J.D. and Gale, J. (1984). Mineral cycling in a tropical palm forest. PI. Soil 79, 21 1-225. Ash, J. (1988). The location and stability of rainforest boundaries in north-eastern Queensland, Australia. J . Biogeogr. 15, 619430. Ashton, D.H. (1981). Fire in tall open-forests (wet sclerophyll forests). In: Fire and the Australian Biota (Ed. by A.M. Gill, R.H. Groves and I.R. Noble), pp. 339-366. Australian Academy of Science, Canberra. Atkinson, C.J. (1986). The effect of clipping on net photosynthesis and dark respiration rates of plants from an upland grassland, with reference to carbon partitioning in Festuca ovina. Ann. Bot. N.S. 58, 61-72. Baig, M.N. and Tranquillini, W. (1980). The effect of wind and temperature on cuticular transpiration of Picea abies and Pinus cembra and their significance in desiccation damage at alpine treeline. Oecologia, Bed. 47, 252-256. Baker, A.J.M. and Brooks, R.R. (1989). Terrestrial higher plants which hyperaccumulate metallic elements-A review of their distribution, ecology and phytochemistry. Biorecovery 1, 81-126. Baker, A.J.M., Brooks, R.R. and Kersten, W.J. (1985). Accumulation of nickel by Psychotria species from the pacific basin. Taxon 34, 89-95. Banyikwa, F.F. (1987). The effect of simulated grazing, nitrogen and density of competition between two Serengeti planing grasses (Andropogon greenwayi Napper and Chloris pycnothrix Trin). Tanzania J . Sci. 13, 15-28. Barber, K.E. (I98 1). Peat Stratigraphy and Climatic Change. Balkema, Rotterdam. Barkman, J.J. (1990). Controversies and perspectives in plant ecology and vegetation science. Phytocoenologia 18, 565-589. Bazely, D.R. and Jefferies, R.L. (1985). Goose faeces: a source of nitrogen for plant growth in a grazed salt marsh. J. appl. Ecol. 22, 693-703. Bazely, D.R. and Jefferies, R.L. (1986). Changes in the composition and standing crop of salt-marsh communities in response to the removal of a grazer. J . Ecol. 74, 693-706. Beard, J.S. (1969). The natural regions of the deserts of Western Australia. J . Ecol. 57, 677-7 1 I . Bell, D.T. and Muller, C.H. (1973). Dominance of California annual grasslands by Brassica nigra. Am. Midl. Nut. 90, 277-299. Bellamy, D.J. and Rieley, J. (1967). Some ecological statistics of a “miniature bog”. Oikos 18, 3 3 4 0 . Belsky, A.J. (1986). Population and community processes in a mosaic grassland in the Serengeti, Tanzania. J . Ecol. 74, 841-856. Belsky, A.J., Amundson, R.G., Duxbury, J.M., Riha, S.J., Ali, A.R. and Mwonga, S.M. (1989). The effects of trees on their physical, chemical, and biological environments in a semi-arid savanna in Kenya. J . appl. Ecol. 26, 1005-1024. Benedict, J.B. (1984). Rates of tree-island migration, Colorado Rocky Mountains, USA. EcoloEy 65, 82&823. Berg, A.W. (1990). Formation of Mima mounds: a seismic hypothesis. Geology 18, 28 1-284. van den Bergh, J.P. and Elberse, W.Th. (1962). Competition between Lolium perenne L. and Anthoxanthum odoratum L. at two levels of phosphate and potash. J . Ecol. 50, 87-95. Boaler, S.B. and Hodge, C.A.H. (1962). Vegetation stripes in Somaliland. J . Ecol. 50, 465474. Boaler, S.B. and Hodge, C.A.H. (1964). Observations on vegetation arcs in the northern region, Somali Republic. J . Ecol. 52, 51 1-544.

POSITIVE-FEEDBACK SWITCHES IN PLANT COMMUNITIES

323

Boatman, D.J. (1983). The Silver Flowe National Nature Reserve, Galloway, Scotland. J. Biogeogr. 10, 163-274. Boatman, D.J., Goode, D.A. and Hulme, P.D. (1981). The Silver Flowe. 111. Pattern development cn Long Loch B and Craigeazle Mires. J. Ecol. 69, 897-918. Boatman, D.J. and Tomlinson, R.W. (1977). The Silver Flowe. 11. Features of the vegetation and stratigraphy of Brishie Bog, and their bearing on pool formation. J . Ec01. 65, 531-546. Bonan, G.B. and Korzuhin, M.D. (1989). Simulation of moss and tree dynamics in the boreal forests of interior Alaska. Vegetatio 84, 3 1 4 . Booth, W.E. (1941). Revegetation of abandoned fields in Kansas and Oklahoma. Am. J. Bot. 28, 415422. Bornkamm, R. (1988). Mechanisms of succession on fallow lands. Vegetatio 77, 95101. Bowman, D.M.J.S. (1986). Stand characteristics, understorey associates and environmental correlates of Eucalyptus tetrodonta F. Muell. forests on Gunn Point, northern Australia. Vegetatio 65, 105-1 13. Bowman, D.M.J.S. (1988). Stability amid turmoil?: towards an ecology of north Australian eucalypt forests. Proc. ecol. SOC.Aust. 15, 149-158. Bowman, D.M.J.S. and Minchin, P.R. (1987). Environmental relationships of woody vegetation patterns in the Australian monsoon tropics. Aust. J. Bot. 35, 151-169. Bowman, D.M.J.S. and Wilson, B.A. (1988). Fuel characteristics of coastal monsoon forests, Northern Territory, Australia. J. Biogeogr. 15, 807-81 7. Braithwaite, R.W. and Estbergs, J.A. (1985). Fire patterns and woody vegetation trends in the Alligator Rivers Region of Northern Australia. In: Ecology and Management ofthe World’s Savannas (Ed. by J.C. Tothill and J.J. Mott), pp. 359364. Australian Academy of Science, Canberra. Branson, F.A. (1975). Natural and modified plant communities as related to runoff and sediment yields. In: Coupling of Land and Water Systems (Ecological Studies, 10) (Ed. by A.D. Hasler), pp. 157-172. Springer-Verlag, Berlin. Braun-Blanquet, J. (1932). Plant Sociology. USA. Brooks, R.R., Naidu, S.M., Malaisse, F. and Lee, J. (1987). The elemental content of metallophytes from the copper/cobalt deposits of Central Africa. Bull. Soc. R. Bot. Belg. 119, 179-191. Brooks, R.R., Trow, J.M., Veillon, J-M. and Jaffre, T. (1981). Studies on manganeseaccumulating Alyxia species from New Caledonia. Taxon 30, 4 2 W 2 3 . Cannon, H.L. (1960). The development of botanical methods of prospecting for uranium on the Colorado Plateau- Botanical prospecting for uranium on the Colorado Plateau. Geol. Survey Bull. 1085-A. Cargill, S.M. and Jefferies, R.L. (1984). The effects of grazing by lesser snow geese on the vegetation of a sub-arctic salt marsh. J. appl. Ecol. 21, 669-686. Case, T.J. (1 990). Invasion resistance arises in strongly interacting species-rich model competition communities. Proc. natn. Acad. Sci. USA 87, 961g9614. Casparie, W.A. (1972). Bog development in southeastern Drenthe (The Netherlands). Vegetatio 25, 1-27 1. Cates, R.G. and Orians, G.H. (1975). Successional status and the palatability of plants to generalized herbivores. Ecology 56, 4 1 W 1 8 . Cavelier, J. and Goldstein, G. (1989). Mist and fog interception in elfin cloud forests in Colombia and Venezuela. J. trop. Ecol. 5, 309-322. Chadwick, M.J. (1960a). Nardus stricta, a weed of hill grazings. In: The Biology of Weeds (Ed. by J.L. Harper), pp. 246256. Blackwell, Oxford. Chadwick, M.J. (1960b). Nardus stricta L. J. Ecol. 48, 255-267.

324

J . B . WILSON AND A. D. Q. AGNEW

Chapin, F.S., van Cleve, K. and Chapin, M.C. (1979). Soil temperature and nutrient cycling in the tussock growth form of Eriophorum vaginatum. J . Ecol. 67, 169-189. Charley, J.L. (1972). The role of shrubs in nutrient cycling. In: Wildland Shrubstheir Biology and Utilization (Ed. by C.M. McKell, J.P. Blaisdell and J.R. Goodin), pp. 182-203. USDA Forest Service (General Technical Report INT-I), Ogden, Utah. Charley, J.L. and West, N.E. (1975). Plant-induced soil chemical patterns in some shrub-dominated semi-desert ecosystems of Utah. J . Ecol. 63, 945-963. Clarke, L.D. and Hannon, N.J. (1971). The mangrove swamp and salt marsh communities of the Sydney district IV. The significance of species interaction. J. Ecol. 59, 535-553. Clarke, L.D. and West, N.E. (1969). Germination of Kochia americana in relation to salinity. J . Range Mgmt 22, 28G287. Clayton, W.D. (1958). Secondary vegetation and the transition to savanna near Ibadan, Nigeria. J . Ecol. 46, 217-238. Clayton-Greene, K.A. and Beard, J.S. (1985). The fire factor in vine thicket and woodland vegetation of the Admiralty Gulf region, north-west Kimberley, Western Australia. Proc. ecol. SOC.Aust. 13, 225-230. Clements, F.E. (1905). Research Methods in Ecology. University Publishing, Lincoln, Nebraska. Clements, F.E. (I916). Plant Succession. Carnegie Institution of Washington. Clements, F.E. (1936). Nature and structure of the climax. J . Ecol. 24, 252-284. Clements, F.E., Weaver, J.E. and Hanson, H.C. (1929). Plant competition: an analysis of community functions. Carnegie Institute, Washington D.C. Clymo, R.S. (1965). Experiments on breakdown of Sphagnum in two bogs. J . Ecol. 53, 747-758. Clymo, R.S. (1973). The growth of Sphagnum: some effects of environment. J . Ecol. 61, 849-869. Cole, K. (1985). Past rates of change, species richness, and a model of vegetational inertia in the Grand Canyon, Arizona. Am. Nut. 125, 289-303. Connell, J.H. (1978). Diversity in tropical rain forests and coral reefs. Science 199, 1302-1310. Connell, J.H. and Slatyer, R.O. (1977). Mechanisms of succession in natural communities and their role in community stability and organisation. Am. Nut. 111, 1 1 19-1 144. Conway, V.M. (1949). The bogs of central Minnesota. Ecol. Monogr. 19, 173-206. Cook, C.W. (1961). Seeding response and soil characteristics on adjacent sagebrush and desert molly soils. J . Rqnge Mgmt 14, 134-138. Cornet, A.F., Delhoume, J.P. and Montana, C. (1988). Dynamics of striped vegetation patterns and water balance in the Chihuahuan Desert. In: Diversity and Pattern in Plant Communities (Ed. by H.J. During, M.J.A. Werger and J.H. Willems), pp. 221-23 I . SPB Academic Publishing, The Hague. Costin, A.B. and Wimbush, D.J. (1961). Studies in Catchment Hydrology in the Australian Alps. IV. Interception by Trees of Rain, Cloud, and Fog. CSIRO Division of Plant Industry (Technical Paper 16), Melbourne. Coulson, R.N. (1979). Population dynamics of bark beetles. Annu. Rev. Entomol. 24, 41747. Cox, G.W. (1984). The distribution and origin of Mima mound grasslands in San Diego County, California. Ecology 65, 1397-1405. Crawley, M.J. (1 983). Herbivory: The Dynamics of Animal-Plant Interactions. Blackwell Scientific, Oxford.

POSITIVE-FEEDBACK SWITCHES IN PLANT COMMUNITIES

325

Cremer, K.W. (1962). The effect of fire on eucalypts reserved for seeding. Aust. For. 26, 129-154. Cronin, E.H. (1965). Ecological and Physiological Factors Influencing Chemical Control of Halogeton glomeratus. Tech. Bull. 1325, US Dept. Agric. Darlington, J.P.E.C. (1982). The underground passages and storage pits used in foraging by a nest of the termite Macrotermes michaelseni in Kajiado, Kenya. J . ZOO[.198, 237-247. Darlington, J.P.E.C. (1985). Lenticular soil mounds in the Kenya highlands. Oecologia Berl., 66, 116121. Daubenmire, R. (1954). Alpine timberlines in the Americas and their interpretation. Butler Univ. bot. Stud. 11, 119-136. Daubenmire, R. (1968). Plant Communities: A Textbook of Plant Synecology. Harper and Row, New York. Dice, L.R. (1952). Natural Communities. University of Michigan Press, Ann Arbor. Dickinson, K.J.M. and Kirkpatrick, J.B. (1985). The flammability and energy content of some important plant species and fuel components in the forests of southeastern Tasmania. J . Biogeogr. 12, 121-134. Drake, J.A. (1990). The mechanics of community assembly and succession. J. theor. Biol. 147, 213-233. Duvigneaud P. and Denaeyer-de Smet, S. (1963). Cuivre et VCgCtation au Katanga. Bull. SOC.R . Bot. Belg. 96, 93-231. Eckert, R.E. Jr and Kinsinger, F.E. (1960). Effects of Halogeton glomeratus leachate on chemical and physical characteristics of soils. Ecology 41, 764-772. Egler, F.E. (1977). The Nature of Vegetation, its Management and Mismanagement. The author, in cooperation with the Connecticut Conservation Association, Norfolk, Connecticut. Ekern, P.C. (1964). Direct interception of cloud water on Lanaihale, Hawaii. Proc. Soil Sci. SOC.Am. 28, 419421. Ernst, W. and Nelissen, H.J.M. (1979). Growth and mineral nutrition of plant species from clearings on different horizons of an iron-humus podzol profile. Oecologia, Berl. 41, 175-182. Etherington, J.R. (1981). Limestone heaths in south-west Britain: their soils and the maintenance of their calcicole-calcifuge mixtures. J . Ecol. 69, 277-294. Evans, H.J. and Barber, L.E. (1977). Biological nitrogen fixation for food and fiber production. Science 197, 332-339. Ewel, J.J. (1976). Litter fall and leaf decomposition in a tropical forest succession in eastern Guatemala. J. Ecol. 64, 293-308. Fetcher, N. (1985). Effects of removal of neighboring species on growth, nutrients, and microclimate of Eriophorum vaginatum. Arctic Alpine Res. 17, 7-17. Field, C.R. (1971). Elephant ecology in the Queen Elizabeth National Park, Uganda. E. Afr. Wildl. J . 9, 99-123. Floate, M.J.S., Eadie, J., Black, J.S. and Nicholson, LA. (1973). The improvement of Nardus-dominated hill pasture by grazing control and fertilizer treatment and its economic assessment. Potassium Inst. Ltd, Colloq. Proc. 3, 33-39. Florence, R.G. and Crocker, R.L. (1962). Analysis of blackbutt (Eucalyptus pilularis Sm.) seedling growth in a blackbutt forest soil. Ecology 43, 6 7 M 7 9 . Forest, H. de (1923). Rainfall interception by plants: an experimental note. Ecology 4, 417419. Foster, D.R. and Wright, H.E., Jr (1990). Role ofecosystem development and climate change in bog formation in central Sweden. Ecology 71, 4 5 W 6 3 . Fowler, N.L. (1988). The effects of environmental heterogeneity in space and time on

326

J. B. WILSON AND A. D. Q. AGNEW

the regulation of populations and communities. In: Plant Population Ecology (British Ecological Society Symposium 28) (Ed. by A.J. Davy, M.J. Hutchings and A.R. Watkinson), pp. 249-269. Blackwell, Oxford. Frenzel, B. ( I 983). Mires-repositories of climatic information or self-perpetuating ecosystems? In: Mires: Swamp, Bog, Fen and Moor: General Studies (Ecosystems of the World, 4A) (Ed. by A.J.P. Gore), pp. 35-65. Elsevier, Amsterdam. Furley, P. (1975). The significance of the Cohune Palm, Orbignya cohune (Mart.) Dahlgren, on the nature and in the development of the soil profile. Biotropica 7 , 32-36. Gakahu, C.G. and Cox, G.W. (1984). The occurrence and origin of Mima mound terrain in Kenya. Afr. J . Ecol. 22, 3 1 4 2 . Gilbert, J.M. (1959). Forest succession in the Florentine Valley, Tasmania. Pap. Proc. R. SOC.T a m . 93, 129-151. Gill, A.M. (1981). Adaptive responses of Australian vascular plant species to fires. In: Fire and the Australian Biota (Ed. by A.M. Gill, R.H. Groves and I.R. Noble), pp. 243-272. Australian Academy of Science, Canberra. Gimingham, C.H. (1988). A reappraisal of cyclic processes in Calluna heath. Vegetatio 77, 6 1-64. Gleason, H.A. (1917). The structure and development of the plant association. Bull. Torrey bot. Club 44, 4 6 3 4 8 1. Gleason, H.A. (1939). The individualistic concept of the plant association. Am. Midl. Nut. 21, 92-1 10. Glover, P.E., Glover, J. and Gwynne, M.D. (1962). Light rainfall and plant survival in E. Africa 11. Dry grassland vegetation. J. Ecol. 50, 199-206. Glover, P.E., Trump, E.C. and Wateridge, L.E.D. (1964). Termitaria and vegetation patterns on the Loita Plains of Kenya. J . Ecol. 52, 367-377. Goldberg, D.E. (1982). The distribution of evergreen and deciduous trees relative to soil type: an example from the Sierra Madre, Mexico, and a general model. Ecology 63, 942-95 1. Goudie, A.S. (1988). The geomorphological role of termites and earthworms in the tropics. In: Biogeomorphology (Ed. by H.A. Viles), pp. 166192. Basil Blackwell, Oxford. Grabherr, G. (1989). On community structure in high alpine grasslands. Vegetatio 83, 223-227. Grace, J. (1989). Tree lines. Phil. Trans. R. SOC.Lond. B 324, 233-245. Grant, S.A., Suckling, D.E., Smith, H.K., Torvell, L., Forbes, T.D.A. and Hodgson, J. (1985). Comparative studies of diet selection by sheep and cattle: the hill grasslands. J . Ecol. 73, 987;-1004. Grant, S.A., Torvell, L., Armstrong, R.H. and Beattie, M.M. (1987). The manipulation of mat-grass pasture by grazing management. In: Agriculture and Conservation in the Hills and Uplands (Ed. by M. Bell and R.G.H. Bunce), pp. 6 2 4 4 . Institute of Terrestial Ecology, Grange-over-Sands. Grime, J.P. and Hodgson, J.G. (1969). An investigation of the ecological significance of lime-chlorosis by means of large-scale comparative experiments. In: Ecological Aspects of the Mineral Nutrition of Plants (Symp. Br. ecol. SOC.9) (Ed. by I.H. Rorison), pp. 67-99. Blackwell, Oxford. Grove, T.S. (1977). Effects of fire on concentrations of phosphorus and potassium in jarrah (Eucalyptus marginata) forest soils. In: Nutrient Cycling in Indigenous Forest Ecosystems, pp. 119-121. Institute of Forest Research and Protection, Western Australia, Como, WA. Grubb, P.J. and Suter, M.B. (1971). The mechanism of acidification of soil by Calluna

POSITIVE-FEEDBACK SWITCHES IN PLANT COMWNITIES

327

and Ulex and the significance for conservation. In: The Scient8c Management of Animal and Plant Communities for Conservation (Symp. Br. ecol. SOC.11) (Ed. by E. DulTey and AS. Watt), pp. 115-133. Blackwell, Oxford. Grubb, P.J. and Whitmore, T.C. (1966). A comparison of montane and lowland rain forest in Ecuador. 11. The climate and its effect on the distribution and physiognomy of the forests. J. Ecol. 54, 303-333. Grubb, P.J., Green, H.E. and Merrifield, R.C.J. (1969). The ecology of chalk heath: its relevance to the calcicole-calcifuge and soil acidification problems. J. Ecol. 57, 175-2 12.

Hadley, J.L. and Smith, W.K. (1983). Influence of wind exposure on needle desiccation and mortality for timberline conifers in Wyoming, USA. Arctic Alpine Res. 15, 127-135. Hadley, J.L. and Smith, W.K. (1986). Wind effects on needles of timberline conifers: seasonal influence on mortality. Ecology 67, 12-19. Hadley, J.L. and Smith, W.K. (1989). Wind erosion of leaf surface wax in alpine timberline conifers. Arctic Alpine Res. 21, 392-398. Harper, H.J., Daniel, H.A. and Murphy, H.F. (1934). The total nitrogen, phosphorus and calcium content of common weeds and native grasses in Oklahoma. Proc. Okla. Acad. Sci. 14, 3 M . Harper, J.L. (1977). Population Biology of Plants. Academic Press, London. Harr, R.D. (1982). Fog drip in the Bull Run Municipal Watershed, Oregon. Water Resour. Bull. ( U S ) 18, 785-789. Harrington, G.N. and Johns, G.G. (1990). Herbaceous biomass in a Eucalyprus savanna woodland after removing trees and/or shrubs. J. appl. Ecol. 27, 775-787. Harwood, C.E. and Jackson, W.D. (1975). Atmospheric losses of four plant nutrients during a forest fire, Aust. For. 38, 92-99. Hayward, P.M. and Clymo, R.S. (1983). The growth of Sphagnum: experiments on, and simulation of, some effects of light flux and water-table depth. J. Ecol. 71, 845-863.

Heilman, P.E. (1966). Change in distribution and availability of nitrogen with forest succession on north slopes in interior Alaska. Ecology 47, 825-83 1. Heilman, P.E. (1968). Relationship of availability of phosphorus and cations to forest succession and bog formation in interior Alaska. Ecology 49, 331-336. Heinselman, M.L. (1970). Landscape evolution, peatland types, and the environment in the Lake Agassiz Peatlands Natural Area, Minnesota. Ecol. Monogr. 40, 235261.

Heisey, R.M. and Delwiche, C.C. (1985). Allelopathic effects of Trichostema lanceolatum (Labiatae) in the California annual grassland. J. Ecol. 73, 729-742. Hesse, P.R. (1955). A chemical and physical study of the soils of termite mounds in East Africa. J. Ecol. 43, 449461. Hik, D.S. and Jefferies, R.L. (1990). Increases in the net above-ground primary production of a salt-marsh forage grass: a test of the predictions of the herbivoreoptimization model. J. Ecol. 78, 18C195. Holdsworth, D.K. and Mark, A.F. (1990). Water and nutrient input:output budgets: effects of plant cover at seven sites in upland snow tussock grasslands of Eastern and Central Otago, New Zealand. J . R . SOC.N . Z . 20, 1-24. Hopkins, B. (1965). Forest and Savanna: an Introduction to Tropical Plant Ecology with Special Reference to West Africa. Heinemann, London. Hopkins, B. (1 983). Successional processes. In: Tropical Savannas (Ecosystems of the World 13) (Ed. by F. Bourliere), pp. 605-616. Elsevier, Amersterdam. de Hullu, E. and Gimingham, C.H. (1984). Germination and establishment of

328

J . B. WILSON AND A. D. Q. AGNEW

seedlings in different phases of the Calluna life cycle in a Scottish heathland. Vegetatio 58, 1 15-1 2 1. Humphreys, F.R. and Lambert, M.J (1965). An examination of a forest site which has exhibited the ash-bed effect. Aust. J . Soil Res. 3, 81-94. Hunter, R.F. (1954). The grazing of hill pasture sward types. J . Br. Grassld SOC.9, 195-208. Hunter, R.F. (1962). Hill sheep and their pasture; a study of sheep grazing in south east Scotland. J . Ecol. 50, 651-680. Innes, R. R. (1972). Fire in West African vegetation. Proc. a . Tall Timbers Fire Ecol. Con$ 11, 147-173. Jackson, W.D. (1968). Fire, earth and water-an elemental ecology of Tasmania. Proc. ecol. SOC.Aust. 3, 9-16. Jaffre, T. (1977). Accumulation d u manganese par des espkces associees aux terrains ultrabasiques de Nouvelle-Caltdonie. C. r. hebd. SPanc. Acad. Sci., Paris D 284, 1573-1 575. Jaffre, T. (1979). Accumulation du manganese par les Proteacees de NouvelleCaledonie. C . r. hebd. SPanc. Acad. Sci., Paris D 289, 4 2 5 4 2 8 . Jaffre, T. and Schmid, M. (1974). Accumulation du nickel par une Rubiacee de Nouvelle-Caledonie, Psychotria douarrei (G. Beauvisage) Daniker. C . r. hebd. SPanc. Acad. Sci., Paris D 278, 1727-1730. Jefferies, R.L. (1988). Vegetational mosaics, plant-animal interactions and resources for plant growth. In: Plant Evolutionary Biology (Ed. by L.D. Gottlieb and S.K. Jain), pp. 341-369. Chapman and Hall, London. Johnson, L.C., Damman, A.W.H. and Malmer, N. (1990). Sphagnum macrostructure as an indicator of decay and compaction in peat cores from an ombrotrophic south Swedish peat-bog. J . Ecol. 78, 633-647. Johnson, L.C. and Damman, A.W.H. (1991). Species-controlled Sphagnum decay on a South Swedish raised bog. Oikos 61, 234-242. Kapustka, L.A. and Rice, E.L. (1976). Acetylene reduction N,-fixation in soil and old field succession in Central Oklahoma. Soil Biol. Biochem. 8, 497-503. Kearney, T.H., Briggs, L.J., Shantz, H.L., McLane, J.W. and Piemeisel, R.L. (1914). Indicator significance of vegetation in Tooele Valley, Utah. J . agric. Res. I , 365417. Keay, R.W.J. (1951). Some notes on the ecological status of savanna vegetation in Nigeria. Bull. Commonw. Bur. Past. Fld Crops 41, C h 5 . Kellman, M. (1979). Soil enrichment by Neotropical savanna trees. J . Ecol. 67, 565577. Kellman, M. (1984). Synergistic relationships between fire and low soil fertility in neotropical savannas: a hypothesis. Biotropica 16, 158-160. Kemp, E.M. (1981). Pre-quaternary fire in Australia. In: Fire and the Australian Biota (Ed. by A.M. Gill, R.H. Groves and I.R. Noble), pp. 3-21. Australian Academy of Science, Canberra. Kershaw, K.A. (1957). The use of cover and frequency in the detection of pattern in plant communities. Ecology 38, 29 1-299. Kessell, S.R. (1976). Wildland inventories and fire modeling by gradient analysis in Glacier National Park, Montana, USA. Proc. a . Tall Timbers Fire Ecol. Confi 14, 115-162. Kincaid, D.T. and Lyons, E.E. (1981). Winter water relations of red spruce on Mount Monadnock, New Hampshire. Ecology 62, 1155-1 161. Kloot, P.M. (1983). The role of common iceplant (Mesembryanthemum crystallinum) in the deterioration of medic pastures. Aust. J . Ecol. 8, 301-306.

POSITIVE-FEEDBACK SWITCHES IN PLANT COMMUNITIES

329

Knoop, W.T. and Walker, B.H. (1985). Interactions of woody and herbaceous vegetation in a southern African savanna. J. Ecol. 73, 235-253. Kovda, V.A., Samoilova, E.M., Charley, J.L. and Skujins, J.J. (1979). Soil processes in arid lands. In: Arid-land Ecosystems: Structure, Functioning and Management, Vol. 1 (Ed. by D.W Goodall, R.A. Perry and K.M. Howes), pp. 4 3 H 7 0 . Cambridge University Press, Cambridge. Kullman, L. (1989). Recent retrogression of the forest-alpine tundra ecotone (Betula pubescens Ehrh. ssp. tortuosa (Ledeb.) Nyman) in the Scandes Mountains, Sweden. J. Biogeogr. 16, 83-90. Kummerow, J. (1962). Mediciones cuantitativas de la neblina en el parque nacional Fray Jorge. Boletin de la Universidad de Chile, Santiago 28, 3637. Lacey, C.J., Walker, J. and Noble, I.R. (1982). Fire in Australian tropical savannas. In: Ecology of Tropical Savannas (Ed. by B.J. Huntley and B.H. Walker), pp. 2 4 6 272. Springer-Verlag, Berlin. Lawton, R.M. (1964). The ecology of the Marquesia acuminata (Gigl) R.E.Fr. evergreen forests and the related Chipya vegetation types of north-eastern Rhodesia. J . Ecol. 52, 467479. Lee, K.E. and Wood, T.G. (1971a). Termites and Soils. Academic Press, London. Lee, K.E. and Wood, T.G. (1971b). Physical and chemical effects on soils of some Australian termites, and their pedological significance. Pedobiologia 11, 376409. van Leeuwen, C.G. (1966). A relation theoretical approach to pattern and process in vegetation. Wentia 15, 25-46. Lewontin, R.C. (1969). The meaning of stability. Brookhaven Symp. Biof. 22 (Diversity and Stability in Ecological Systems), 13-23. Lind, E.M. and Morrison, M.E.S. (1974). East African Vegetation. Longmans, London. Lock, J.M. (1977). The vegetation of Rwenzori National Park, Uganda. Bot. Jahrb. Syst. 98, 3 7 2 4 8 . Lopez-Hernandez, D., Fardeau, J.C., Nino, M., Nannipier, P. and Chacon, P. (1989). Phosphorus accumulation in savanna termite mound in Venezuela. J. Soil Sci. 40, 635-640.

Lovegrove, B.G. and Siegfried, W.R. (1986). Distribution and formation of Mimalike earth mounds in the western Cape Province of South Africa. S. Afr. J. Sci. 82, 432436.

McCracken, I.J., Wardle, P., Benecke, U. and Buxton, R.P. (1985). Winter water relations of tree foliage at timberline in New Zealand and Switzerland. In: Establishment and Tending of Subalpine Forest: Research and Management (Proc. 3rd IUFRO Workshop) (Ed. by H. Turner and W. Tranquillini), pp. 85-93. Swiss Federal Institute of Forestry Research. Malaise, F., Gregoire, J., Morrison, R.S., Brooks, R.R. and Reeves, R.D. (1979). Copper and cobalt in vegetation of Fungurume, Shaba Province, Zaire. Oikos 33, 472478.

Marchand, P.J. (1972). Wind and the winter-exposed plant. Rhodora 74, 528-531. Marchand, P.J. and Chabot, B.F. (1978). Winter water relations of tree-line plant species on Mt. Washington, New Hampshire. Arctic Alpine Res. 10, 10S116. Mark, A.F. and Rowley, J. (1976). Water yield of low-alpine snow tussock grassland in Central Otago. J. Hydrol. (NZ) 15, 59-79. Mark, A.F., Fetcher, N., Shaver, G.R. and Chapin, F.S. I11 (1985). Estimated ages of mature tussocks of Eriophorum vaginatum along a latitudinal gradient in central Alaska, USA. Arctic Alpine Res. 17, 1-5. Marloth, R. (1904). Results of experiments on Table Mountain for ascertaining the

330

J. B. WILSON AND A . D . Q. AGNEW

amount of moisture deposited from the south-east clouds. Trans. S . Afr. phil. SOC. 14,403408. Marloth, R. (1907). Results of further experiments on Table Mountain for ascertaining the amount of moisture deposited from the south-east clouds. Trans. S. A,fr. Phil. SOC.16, 97-105. Marr, J.W. (1977). The development and movement of tree islands near the upper limit of tree growth in the southern Rocky Mountains. Ecology 58, 1599-1 164. Marrs, R.H. and Bannister, P. (1978). Response of several members of the Ericaceae to soils of contrasting pH and base-status. J. Ecol. 66, 829-834. Marrs, R.H. and Hicks, M.J. (1986). Study of vegetation change at Lakenheath Warren: a re-examination of A.S. Watt’s theories of bracken dynamics in relation to succession and vegetation management. J . appl. Ecol. 23, 1029-1046. Marshall, J.K. (1970). Assessing the protective role of shrub-dominated rangeland vegetation against soil erosion by wind. Proc. XI Int. Grassld Cong. 19-23. Means, T.H. (1927). Fog precipitated by trees. Science 66, 402403. Merriam, R.A. (1973). Fog drip from artificial leaves in a fog wind tunnel. W a f . Resour. Res. 9, 1591-1598. Miles, J. (1985). The pedogenic effects of different species and vegetation types and the implications of succession. J. Soil Sci. 36, 571-584. Miller, A.G., Morris, M. and Stuart-Smith, S. (1988). Plants of Dhofar, the Southern Region of Oman. Office of the Adviser for Conservation of the Environment, Diwan of Royal Court, Sultanate of Oman. Milton, W.E.J. (1953). The palatability of herbage on undeveloped grasslands in west-central Wales. Emp. J . exp. Agric. 21, 116-122. Minnich, R.A. (1984). Snow drifting and timberline dynamics on Mount San Gorgonio, California, USA. Arctic Alpine Res. 16, 395412. Moore, P.D. (1977). Stratigraphy and pollen analysis of Claish Moss, North-west Scotland: significance for the origin of surface-pools and forest history. J . Ecol. 65, 375-397. Moore, R.M. and Walker, J. (1972). Eucalyptus populnea shrub woodlands: control of regenerating trees and shrubs. Aust. J. exp. Agric. anim. Husb. 12, 4 3 7 4 0 . Moore, T.R. (1989). Plant production, decomposition, and carbon efflux in a subarctic patterned fen. Arctic Alpine Res. 21, 156162. Morison, C.G.T., Hoyle, A.C. and Hope-Simpson, J.F. (1948). Tropical soilvegetation catenas and mosaics; a study in the south-western part of the AngloEgyptian Sudan. J. Ecol. 36, 1-84. Morton, H.L., Haas, R.H. and Erickson, L.C. (1959). Oxalate and mineral contents of Halogeton glomeratus. Wqeds 7 , 255-264. Mount, A.B. (1964). The interdependence of the eucalypts and forest fires in Southern Australia. Aust. For. 28, 166-172. Mutch, R.W. (1970). Wildland fires and ecosystems-a hypothesis. Ecology 51, 1 0 4 6 1051. Nagel, J.F. (1956). Fog precipitation on Table Mountain. Q. JI R . met. Soc. 82,452460. Newman, E.I. (1978). Allelopathy: adaptation or accident? In: Biochemical Aspects of Plant and Animal Coevolution (Ed. by J.B. Harborne), pp. 327-342. Academic Press, London. Nicholson, I.A., Paterson, I.S. and Currie, A. (1970). A study of vegetational dynamics: selection by sheep and cattle in Nardus pasture. In: Animal Populations in Relation to their Food Resources (Ed. by A. Watson), pp. 129-143. Blackwell, Oxford.

POSITIVE-FEEDBACK SWITCHES IN PLANT COMMUNITIES

33 1

Niering, W.A. and Egler, F.E. (1955). A shrub community of Viburnum lentago, stable for twenty-five years. Ecology 36, 356-360. Niering, W.A., Dreyer, G.D., Egler, F.E. and Anderson, J.P. Jr (1986). Stability of a Viburnum lentago shrub community after 30 years. Bull. Torrey bot. Club 113, 23-27. Norman, M.J.T. and Wetselaar (1960). Losses of nitrogen on burning native pasture at Katherine, N.T. J . Aust. Inst. agric. Sci. 26, 272-273. Nutting, W.L., Haverty, M.I. and LaFage, J.P. (1987). Physical and chemical alteration of soil by two subterranean termite species in Sonoran Desert grassland. J. Arid Environments 12, 233-239. Oberbauer, S . and Miller, P.C. (1979). Plant water relations in montane and tussock vegetation types in Alaska. Arctic Alpine Res. 11, 69-81. Oberlander, G.T. (1956). Summer fog precipitation in San Francisco peninsula. Ecology 37, 85 1-852. Odum, H.T. (1971). Environment, Power and Society. Wiley-Interscience, New York. Okali, D.U.U., Hall, J.B. and Lawson, G.W. (1973). Root distribution under a thicket clump on the Accra Plains, Ghana: its relevance to clump localization and water relations. J. Ecol. 61, 439454. Osvald, H. (1923). Die Vegetation des Hochmoores Komosse. Sven. Vaxtsociol. Sallsk. Handl. 1, 1436. Paine, R.T. (1969). A note on trophic complexity and community stability. Am. Nut. 103, 91-93. Parker, V.T. and Muller, C.H. (1982). Vegetational and environmental changes beneath isolated live oak trees (Quercus agrifoliu) in a California annual grassland. Am. Midl. Nut. 107, 69-8 I . Parsons, J.J. (1960). “Fog drip” from coastal stratus, with special reference to California. Weather, Lond. 15, 58-62. Payette, S. (1988). Late-holocene development of subarctic ombrotrophic peatlands: allogenic and autogenic succession. Ecology 69, 5 16-53 I . Perry, D.A., Molina, R. and Amaranthus, M.P. (1987). Mycorrhizae, mycorrhizospheres, and reforestation: current knowledge and research needs. Can. J. For. Res. 17, 929-940. Perry, D.A., Amaranthus, M.P., Borchers, J.G., Borchers, S.L. and Brainerd, R.E. ( I 989). Bootstrapping in ecosystems. Bioscience 39, 23&237. Pethick, J.S. (1974). The distribution of salt pans on tidal salt marshes. J. Biogeogr. 1, 57-62. Phillips, J. (1926). Rainfall interception by plants. Nature, Lond. 118, 837-838. Pigott, C.D. (1970). Soil formation and development on the carboniferous limestone of Derbyshire. 11. The relation of soil development to vegetation on the plateau near Coombs Dale. J . Ecol. 58, 529-541. Plesnik, P. (1978). Man’s influence on the timberline in the West Carpathian Mountains, Czechoslovakia. Arctic Alpine Res. 10, 491-504. Prat, H. (1953). Fog as a dominant factor for plant life on some parts of the Pacific coast of North America. Proc. 7th Pa@ Sci. Congr. New Zealand, 1949 (5) (Botany), 152-1 56. Price, P.W., Westoby, M. and Rice, B. (1988). Parasite-mediated competition: some predictions and tests. Am. Nut. 131, 544-555. Rawes, M. (1961). The problem of Nardus and its productivity in relation to sheep grazing at Moor House, Westmorland. J. Br. Grassl. SOC.16, 19@-193. Rawes, M. (1981). Further results of excluding sheep from high-level grasslands in the north Pennines. J . Ecol. 69, 651-669.

332

J . B . WILSON AND A . D. Q. AGNEW

Reeves, R.D. and Brooks, R.R. (1983). Hyperaccumulation of lead and zinc by two metallophytes from mining areas of Central Europe. Envir. Pollut. A, 31, 277-287. Rice, B. and Westoby, M. (1982). Heteroecious rusts as agents of interference competition. Evol. Theory 6, 43-52. Rice, E.L. (1964a). Inhibition of nitrogen-fixing and nitrifying bacteria by seed plants (I.) Ecology 45, 824-837. Rice, E.L. ( 1 964b). Inhibition of nitrogen-fixing and nitrifying bacteria by seed plants. 111. Comparison of three species of Euphorbia. Proc. Okla. Acad. Sci. 45, 4344. Rice, E.L. (1965a). Inhibition of nitrogen-fixing and nitrifying bacteria by seed plants. 11. Characterization and identification of inhibitors. Physiologia PI. 18, 255-268. Rice, E.L. (1965b). Inhibition of nitrogen-fixing and nitrifying bacteria by seed plants IV. The inhibitors produced by Ambrosia elatior and A . psilostachya. SW Nut. 10, 248-255. Rice, E.L. (1968). Inhibition of nodulation of inoculated legumes by pioneer plant species from abandoned fields. Bull. Torrey bot. Club 95, 346-358. Rice, E.L. (1971). Some possible roles of inhibitors in old-field succession. In: Biochemical Interactions Among Plants, pp. 128-1 32. National Academy of Sciences, Washington, DC. Rice, E.L. (1984). Aflelopathy, 2nd Edn. Orlando, London. Rice, E.L. and Pancholy, S.K. (1972). Inhibition of nitrification by climax ecosystems. Am. J . Bot. 59, 1033-1040. Rice, E.L. and Pancholy, S.K. (1973). Inhibition of nitrification by climax ecosystems. 11. Additional evidence and possible role of tannins. Am. J . Bot. 60,691-702. Rice, E.L. and Parenti, R.L. (1967). Inhibition of nitrogen-fixing and nitrifying bacteria by seed plants. V. Inhibitors produced by Bromus japonicus Thunb. SW Nut. 12, 97-103. Rice, E.L., Penfound, W.T. and Rohrbaugh, L.M. (1960). Seed dispersal and mineral nutrition in succession in abandoned fields in Central Oklahoma. Ecology 41,224228. Roberts, D.W. (1987). A dynamical systems perspective on vegetation theory. Vegetatio 69, 27-33. Rochefort, L., Vitt, D.H. and Bayley, S.E. (1990). Growth, production, and decomposition dynamics of Sphagnum under natural and experimentally acidified conditions. Ecology 71, 198&2000. Roxburgh, S.H., Wilson, J.B. and Mark, A.F. (1988). Succession after disturbance of a New Zealand high-alpine cushionfield. Arctic Alpine Res. 20, 23C236. Rutter, A.J. (1975). The hyd;ological cycle in vegetation. In: Vegetation and the Atmosphere, Vol. I , Principles (Ed. by J.L. Monteith), pp. 1 1 1-154. Academic Press, London. Rychnovska, M. (1963). An outpost site of Corynephorus canescens in the region between the Danube and the Tisza and its causal explanation. Acta biol. hung. 14, 5746. Savage, D.A. and Runyon, H.E. (1937). Natural revegetation of abandoned farm land in the Central and Southern Great Plains. Rep. 4th Int. Grassld Congr. Aberystwyth, 178-182. Schowalter, T.D. (1985). Adaptations of insects to disturbance. In: The Ecology of Natural Disturbance and Patch Dynamics (Ed. by S.T.A. Pickett and P.S. White), pp. 235-252. Academic Press, Orlando. Schroeter, C. (1923-26). Das Pfanzenleben der Alpen. Albert Raustein, Zurich. Singh, G., Kershaw, A.P. and Clark, R. (1981). Quaternary vegetation and fire

POSITIVE-FEEDBACK SWITCHES IN PLANT COMMUNITIES

333

history in Australia. In: Fire and the Australian Biota (Ed. by A.M. Gill, R.H. Groves and I.R. Noble), pp. 23-54. Australian Academy of Science, Canberra. Shmida, A. and Burgess, T.L. (1988). Plant growth-form strategies and vegetation types in arid environments. In: Plant Form and Vegetation Structure (Ed. by M.J.A. Werger, P.J.M. van der Aart, H.J. During and J.T.A. Verhoeven), pp. 21 1-241. SPB Academic Publishing, The Hague. Smith, J.M.B. and Guyer, I.J. (1983). Rainforest-eucalypt forest interrelations and the relevance of the biological nomad concept. Aust. J . Ecol. 8, 5 5 4 0 . Smith, T. and Huston, M. (1989). A theory of the spatial and temporal dynamics of plant communities. Vegetatio 83, 49-69. Snyder, J.R. (1984). The role of fire: Mutch ado about nothing. Oikos 43, 404-405. van Someren, V.G.L. (1939). Coryndon Memorial Museum expedition to the Chyulu Hills, April-July, 1938. Part 1. General Narrative. JI E. Africa Uganda nut. Hist.

SOC.14, 1-14. Stocker, G.C. (1966). Effects of fires on vegetation in the Northern Territory. Aust. For. 30, 223-230. Stocker, G.C. (1969). Fertility differences between the surface soils of monsoon and eucalypt forests in the Northern Territory. Aust. For. Res. 4, 31-38. Stocker, G.C. (1971). The age of charcoal from old jungle fowl nests and vegetation change on Melville Island. Search 2, 28-30. Stocker, G.C. (I98 I). Regeneration of a north Queensland rainforest following felling and burning. Biotropica 13, 8G92. Sugden, A.M. (1982). The vegetation of the Serrania de Macuira, Guajira, Colombia: a contrast of arid lowlands and an isolated cloud forest. J. Arnold Arbor. 63, 1-30. Sutherland, J.P. (1974). Multiple stable points in natural communities. Am. Nut. 108, 859-873. Svensson, G . (1988). Fossil plant communities and regeneration patterns on a raised bog in south Sweden. J. Ecol. 76, 41-59. Tallis, J.H. (1983). Changes in wetland communities. In: Mires: Swamp, Bog, Fen and Moor: General Studies (Ecosystems of the World, 4A) (Ed. by A.J.P. Gore), pp. 3 1 1-347. Elsevier, Amsterdam. Tanaka, N. (1988). Tree invasion into patchy dwarf-bamboo thickets within a climax beech-fir forest in Japan. In: Diversity and Pattern in Plant Communities (Ed. by H.J. During, M.J.A. Werger and J.H. Willems), pp. 253-261. SPB Academic Publishing, The Hague. Tilman, D. and Wedin, D.A. (1991a). Plant traits and resource reduction for five grasses growing on a nitrogen gradient. Ecology 72, 685-700. Tilman, D. and Wedin, D.A. (I991 b). Dynamics of nitrogen competition between successional grasses. Ecology 72, 1038-1049. Tongway, D.J., Ludwig, J.A. and Whitford, W.G. (1989). Mulga log mounds: Fertile patches in the semi-arid woodlands of eastern Australia. Aust. J. Ecol. 14, 263268. Tranquillini, W. (1980). Winter desiccation as the cause for alpine timberline. In: Mountain Environments and Subalpine Tree Growth (Ed. by U. Benecke and M.R. Davis), pp. 263-267. Forest Research Institute, Technical Paper 70, NZ Forest Service, Wellington. Tranquillini, W. and Plank, A. (1989). Okophysiologische Untersuchungen an Rotbuchen (Fagus sylvatica L.) in verschiedenen Hohenlagen Nord- und Sudtir01s. Zentbl. ges. Forstw. 106, 225-246. Trapnell, C.G. (1959). Ecological results of woodland burning experiments in Northern Rhodesia. J. Ecol. 47, 129-168.

334

J . B. WILSON A N D A. D . Q. AGNEW

Trapnell, C.G., Friend, M.T., Chamberlain, G.T. and Birch, H.F. (1976). The effects of fire and termites on a Zambian woodland soil. J . Ecol. 64, 577-588. Troumbis, A.Y. and Trabaud, L. (1989). Some questions about flammability in fire ecology. Acta Oecol.. Oecol. Plant. 10, 167-1 75. Trumble, H.C. and Woodroffe, K . (1954). The influence of climatic factors on the reaction of desert shrubs to grazing by sheep. In: Biology of Desert-the Proceedings of a Symposium on the Biology of Hot and Cold Deserts Organized by the Institute of Biology (Ed. by J.L. Cloudsley-Thompson), pp. 129-147. Institute of Biology, London. Twomey, S. (1957). Precipitation by direct interception of cloud water. Weather, Lond. 12, 120-122. Unwin, G.L. (1989). Structure and composition of the abrupt rainforest boundary in the Herberton Highland, north Queensland. Aust. J . Bot. 37, 413428. Unwin, G.L., Stocker, G.C. and Sanderson, K.D. (1985). Fire and the forest ecotone in the Herberton highland, north Queensland. Proc. ecol. SOC.Aust 13, 215-224. Vesey-Fitzgerald, D. (1972). Fire and animal impact on vegetation in Tanzania National Parks. Proc. a. Tall Timbers Fire Ecol. Conf. 11, 297-317. Vis, M. (1986). Interception, drop size distributions and rainfall kinetic energy in four Colombian forest ecosystems. Earth Surf. Processes Landf. 11, 591-603. Vivrette, N.J. and Muller, C.H. (1977). Mechanism of invasion and dominance of coastal grassland by Mesembryanthemum crystallinum. Ecol. Monogr. 47, 30 1318. Vogelmann, H.W. (1973). Fog precipitation in the cloud forests of eastern Mexico. Bioscience 23, 9 6 1 0 0 . Vogelmann, H.W., Siccama, T., Leedy, D. and Ovitt, D.C. (1968). Precipitation from fog moisture in the Green Mountains of Vermont. Ecology 49, 1205-1207. Walker, D. and Singh, G. (1981). Vegetation history. In: Australian Vegetation (Ed. by R.H. Groves), pp. 2-3. Cambridge University Press, Cambridge. Walker, D. and Walker, P.M. (1961). Stratigraphic evidence of regeneration in some Irish bogs. J . Ecol. 49, 169-185. Walker, T.W. (1965). The significance of phosphorus in pedogenesis. In: Experimental Pedology (Proceedings of the Eleventh Easter School in Agricultural Science, University of Nottingham) (Ed. by E.G. Hallsworth and D.V. Crawford), pp. 295-31 5. Butterworths, London. Walltn, B., Falkengren-Grerup, U. and Malmer, N. (1988). Biomass, productivity and relative rate of photosynthesis of Sphagnum at different water levels on a South Swedish peat bog. Holarct. Ecol. 11, 70-76. Wardle, P. (1965). A comparispn of alpine timber lines in New Zealand and North America. N.Z. JI Bot. 3, 113-135. Wardle, P. (1968). Engelmann spruce (Picea engelmannii Engel.) at its upper limits on the Front Range, Colorado. Ecology 49, 483495. Wardle, P. (1981). Winter desiccation of conifer needles simulated by artificial freezing. Arctic Alpine Res. 13, 419423. Wardle, P. (1985a). New Zealand timberlines. 1. Growth and survival of native and introduced species in the Craigeburn Range, Canterbury. N . Z . JI Bot. 23,219-234. Wardle, P. (l985b). New Zealand timberlines. 2. A study of forest limits in the Crow Valley near Arthur’s Pass, Canterbury. N.Z. JI Bot. 23, 235-261. Warren, H.V. and Delavault, R.E. (1965). Further studies of the biogeochemistry of molybdenum. Western Miner 38, 64-68. Watson, J.P. (1967). A termite mound in an Iron Age burial ground in Rhodesia. J . EcoI. 55, 663-669.

POSITIVE-FEEDBACK SWITCHES IN PLANT COMMUNITIES

335

Watt, A S . (1947). Pattern and process in the plant community. J . Ecol. 35, 1-22. Webb, W., Szarek, S., Lauenroth, W., Kinerson, R. and Smith, M. (1978). Primary productivity and water-use in native forest, grassland and desert ecosystems. Ecology 59, 1239-1247. Wedin, D.A. and Tilman, D. (1990). Species effects on nitrogen cycling: a test with perennial grasses. Oecologia, Berl. 84, 4 3 3 4 1 . Welch, D. (1986). Studies in the grazing of heather moorland in north-east Scotland. V. Trends in Nardus stricta and other unpalatable graminoids. J . appl. Ecol. 23, 1047-1 058. Welch, D. and Rawes, M. (1964). The early effects of excluding sheep from high-level grassland in the north Pennines. J . appl. Ecol. 1, 281-300. West. 0. (1972). Fire, man and wildlife as interacting factors limiting the development of vegetation in Rhodesia. Proc. a . Tall Timbers Fire Ecol. Con$ 11, 121-145. Westoby, M. (1980). Elements of a theory of vegetation dynamics in arid rangelands. Israel J . Bot. 28, 169-194. Westoby, M., Walker, B. and Noy-Meir, I. (1989). Opportunistic management for rangelands not at equilibrium. J . Range Mgmt 42, 266274. White, L.P. (1971). Vegetation stripes on sheet wash surfaces. J . Ecol. 59, 615-622. Whitehead, N.E. and Brooks, R.R. (1969). Radioecological observations on plants of the lower Buller Gorge region of New Zealand and their significance for biogeochemical prospecting. J . appl. Ecol. 6 , 30 1-3 10. Whittaker, R.H. (1975). Communities and Ecosystems, 2nd Edn. Macmillan, New York. Whittaker, R.H. and Levin, S.A. (Eds) (1975). Niche: Theory and Application. Dowden, Hutchinson and Ross, Stroudsburg, Pennsylvania. Whittaker, R.H. and Woodwell, G.M. (1970). Evolution of natural communities. In: Ecosystem Structure and Function (Proc. 3 1st Annual Biology Colloquium) (Ed. by J.A. Wiens), pp. 137-155. Oregon State University Press, Corvallis, Oregon. Wickens, G.E. and Collier, F.W. (1971). Some vegetation patterns in the Republic of Sudan. Geoderma 6 , 43-59. Wild, H. (1952). The vegetation of Southern Rhodesian termitaria. Rhodesia agric. J. 49, 280-292. Wild, H. (1974). Indigenous plants and chromium in Rhodesia. Kirkia 9, 233-240. Williams, A.H. (1 980). The threespot damselfish: a noncarnivorous keystone species. Am. Nut. 116, 138-142. Williams, M.C. (1960). Effect of sodium and potassium salts on growth and oxalate content of halogeton. PI. Physiol., Lancaster 35, 500-505. Williamson, G.B. (1990). Allelopathy, Koch’s postulates, and the neck riddle. In: Perspectives on Plant Competition (Ed. by J.B. Grace and D. Tilman), pp. 142-162. Academic Press, San Diego. Wilson, J.B. (1988). The effect of initial advantage on the course of plant competition. Oikos 51, 19-24. Wilson, J.B. (1 990). Mechanisms of species coexistence; twelve explanations for Hutchinson’s “Paradox of the Plankton”: evidence from New Zealand plant communities. N . Z . JI Ecol. 13, 117-142. Wilson, K. (1960). The time factor in the development of dune soils at South Haven Peninsula, Dorset. J . Ecol. 48, 341-359. Wilson, K.A. and Fitter, A.H. (1984). The role of phosphorus in vegetational differentiation in a small valley mire. J . Ecol. 72, 463473. Wood, T.G. (1978). Food and feeding habits of termites. In: Production Ecology of

336

J. B. WILSON AND A . D . Q. AGNEW

Ants and Termites (Ed. by M.V. Brian), pp. 55-80. Cambridge University Press, Cambridge. Worrall, G.A. (1959). The Butana grass patterns. J. Soil Sci. 10, 34-53. Wright, R.A. and Mueller-Dombois, D. (1988). Relationships among shrub population structure, species associations, seedling root form and early volcanic succession, Hawaii. In: Plant Form and Vegetation Structure (Ed. by M.J.A. Werger, P.J.M. van der Aart, H.J. During and J.T.A. Verhoeven), pp. 87-104. SPB Academic Publishing, The Hague. Yapp, R.H. and Johns, D. (1917). The salt marshes of the Dovey Estuary. Part 11. The salt marshes. J . Ecol. 5, 65-103. Yarranton, G.A. and Morrison, R.G. (1974). Spatial dynamics of a primary succession: nucleation. J . Ecol. 62,41 7 4 2 8 . van Zinderen Bakker, E.M., Jr (1973). Ecological investigations of forest communities in the Eastern Orange Free State and the adjacent Natal Drakensberg. Vegetatio 28, 299-334.

Index Above-ground organs and oxygen Allelopathy-mediated switches, 30 1-3 deprivation, 1 18-20 Allium porrum, 22 fruits, 118 Ahus glutinosa, 1254, 128, 154 wood, I 18-20 Alocasia macrorrhiza, 2 17 Abscisic acid, 199 Alternaria, 14, 46 Acacia mearnsii, 1 19 A . tenuis, 61 Acarina, 4, 15 Altitude and positive-feedback switches, Acceleration vegetation situation, 265, 2 7 1 4 , 290-2, 314-15 269, 283, 292, 317 Amino acids, 227 Acer rubrum, 120 and oxygen deprivation, 148-50 Acetaldehyde production and oxygen Ammonium uptake, 154-5, 157 deprivation, 125, 1 4 5 4 , 147-8, 151 Amoebae, 2&1 Acid rain, 167 Andropogon Acidity A . greenwayi, 275, 310 and oxygen deprivation, 137-9, 150 A . scoparius, 304, 305 and positive-feedback switches, 278, Animals 280-3 as dispersal agents see Microbial rot Acorus calamus, 1 14, 144 and dispersal of fruits Acremonium, 8 and wetlands, 136-7, 144, 158 Acrotelm, 162, 165 see also Herbivores Actinomycetes, 126 Anthriscus sylvestris, 190, 197 Actinorhizal nitrogen fixers, 128 Antioxidants, 152 ADH see Alcohol dehydrogenase Aphelenchus avenae, 20 induction Aphids, 23 Aerenchyma distribution and function, Arctostaphylos, 52 94, 116, 120-3, 145 A . uva-ursi, 6&5, 67 Aesculus hippocastanum, 126 Arid and semi-arid areas and Agropyron, 204 positive-feedback switches, 2 8 6 7 , Agrostis, 197, 309, 310 289, 293 A . capillaris, 281 water-mediated, 27 1, 274, 275 A . magellanica, 237 Aristida oligantha, 303, 305 A . stolonfera, 204, 237 Artemisia tridentata, 287 A . vinealis, 226 Arthrocnemon, 288, 3 17 Air A . australasicum, 289 mass movement in aquatic spheres, Arthropods see Macroarthropods; 1234 Microarthropod-microbial see also Oxygen; Respiration interactions in soil Alcohol dehydrogenase activity and Ascomycetes, 7 oxygen deprivation, 112, 114, 119, Aspergillus, 6, 20, 46 1447 A . aculeatus 47 see also Ethanol A.Jlavus, 8, 10, 20 Alectoria ochroleuca, 280 A . niger, 61 Allelopathic interactions, 229 Astigmata, 15, 20

338

INDEX

ATP, 101, 126, 141-2 and growth rate variations in higher plants, 212, 2 2 3 4 A triplex A . confert$olia, 287 A . vesicaria, 282 Avena fatua 302 Avicennia, 137, 288, 3 1 6 1 7 A . nitida, 126 Azotobacter, 304 Baccharis, 53 Bacteria, 41, 304 and microarthropod-microbial interactions in soil, 5-6, 14 as food for microarthropods, 8-10 microarthropods as food for, 8 response to grazing, 10-12, 15-16, 17 in rhizosphere, 18-20, 23 Barley and oxygen deprivation, 102, 137-8, 151, 153 Basidiomycetes, 7, 9, 10, 1 I , 12 Beans, 97, 100 Beauveria bassiana, 8 Beetles, 13 Beta vulgaris, 239 Betula B. nigra, 120 B. pendula, 21 Bidens tripartita, 106 Biomass allocation, 199-200 at optimum nutrient supply, 199 plasticity in, 200 Birds, 144, 31 1 and seed dispersal, 39, 52, 60, 66 Bogs, 283. 319 ombrogenous, 278-9 , raised, 162-70 Boletus wriegatus, 130 Botrytis, 46 B. cinerea, I I , 42, 52, 61 Boundaries of switches, 267-8, 3 15-16 Bouteloua gracilis, 20 Bradysia, 7 Brassica B. nigra, 302 B. r a p , 199, 230 Bromus B. japonicus, 304 B. mollis, 146, 302

B. rigidus, 302 Bruguiera, 137 Bulbs, 20 Bulrush, 112-13, 141, 142, 144, 161 Bunch rot, 47

Cabbage, 102-3 Cadmium, 239 Calcium, 23, 154 Callitriche hamulata, 13I , 133 Callitris, 289 Calluna vulgaris, 162, 280, 28 I , 3 16, 3 I9 Caloglyphus, 15 C. micheali, 20 C . polyphyllae, 8 Capparis tomentosa, 3 1 1 Carabodes, 5 Carbohydrates and sugars in fruit, 43, 44-5, 56, 64 and growth rate variations in higher plants, 196, 224, 2 2 6 7 , 235 chemical composition, 2 0 6 7 , 210-11, 213 photosynthesis, 217-18 and oxygen deprivation, 1 0 1 4 , I 1 I , 130, 142, 144, 149, 154-5 Carbon dioxide and growth rate variations in higher plants concentration within leaf, 218-19, 22 I , 233 release, 213, 224 and oxygen deprivation, 112, 119, 125-16 and climate change, 161-2, 164, 165 consequences of, 13- I , I47 Carbon and growth rate variations in higher plants, 188, 191-2, 208-9, 21 1-13, 221-2, 227, 241 budget, 23 1-2 losses, 228-9 Carex, 202, 203, 223 C . acutiformis. C . elata, C . elongata, C . pseudocarpus and C . remota, 155 C . diandra, 221 C. gracilis, 122 C . riparia, 145 C . rostrata, 1 1 1 C . subspathacea, 31 1-12

INDEX

Cassia, 289 Casuarina, 127 catatelm, 162 Cecropia, 106 C . obtusifolia, 234 C . peltata, 21 1-12 Cellular effects of oxygen deprivation, 1414 immediate, 141-2 long-term, 1 4 2 4 Cellulose and (hemi) cellulose and growth rate variations in higher plants, 206, 207, 211, 213, 216 Celosia argentea, 100 Cereals and microarthropod-microbial interactions in soil, 23 and oxygen deprivation, 100, 101-3, 106, 112, 115-18 consequences of, 134, 139, 141-2, 146, 148, 15&1, 154-5 see also Grasses; Maize; Rice Chaetomium bostrycodes, 7 Chemicals composition and growth rate variations in higher plants, 206-14, 24 1 see also Secondary plant chemicals Chenopodium album, 304 Chickpeas and oxygen deprivation, 97, 102, 137-8, 151, 153 Chilopoda, 4 Chionochloa rigida, 273 Chloris pycnothrix, 275, 3 I0 Chlorophyll concentration and growth, 215, 217, 221 Chloroplast, partitioning of nitrogen within, 216-17 Chorisia speciosa, 100 Chromium, 285 Cicer arietinum see Chickpeas Citrus, 52 Cladina rangiferina and C . stellaris, 280 Cladosporium, 46 C . cladosporioides, 6 1 Climate change and bogs, 161-70, 279 and fruit rot, 45-6 see also Arid and semi-arid areas Closed forestjsavannah switches. 297-301, 315-16, 318

339

Cobalt, 285 Coenococcum geophilum, 23 Coleoptera, 9 Collembola and microarthropod-microbial interactions in soil, 3 4 in rhizosphere, 19-23 in saprophytic system, 6, 7-13, 15 Colletotrichum, 46 C . acutatum, 47 C . gloeosporioides, 6 1 Communities see Positive-feedback switches Competition and growth rate variations in higher plants, 2 3 5 4 plant, and oxygen deprivation, 94-5, 1 58-6 1 in saprophytic system, 6-7 see also Positive-feedback switches Conidia, 15 Conidiobolus coronatus, 8 Coniochaeta nepalica, 9 Constitutive defences of fruit, 48-9 Construction costs of plant materials, 212-14 Conyza canadensis, 304 COP (critical oxygen pressure), 1 3 4 5 Copper, 227, 2 8 4 5 Coprophilous fungi, 7, 12 Coriolus versicolor, 10, 1 1 CornusJorida, 60 Corynephorus canescens, 190, 197, 226, 232, 281 costs construction, of plant materials, 212-14 of energy-requiring processes, variations in, 224-5 Cotton, 20, 1 1 1 Crataegus crusgalli, 60 Critical oxygen pressure, I 3 4 5 Crops cotton, 20, 1 1 1 grass see Grasses and growth rate variations in higher plants, 194, 204, 205, 214 irrigation, 134, 152 susceptibility to disease, 5 5 4 vegetables and root crops, 32, 100, 102-3

340

INDEX

Cucumis sativus, 100 Cultivated plants see Crops Cyclic succession, 320 Cyperus odoratus, 106 Cytochrome oxidase, 94, 135

Defence chemical see Toxins of ripe fruit see Microbial rot delay vegetation situation, 265, 269-70, 317 biological mediation, 305 physicakhemical mediation, 290, 292, 293, 301 Denitrification, 228 Deprivation indifference, 16&1 DeschampsiaJiexuosa, 154, 198, 20 1-2, 226, 234, 281 Deserts see Arid and semi-arid areas Deterrence and fruit rot, 56-8, 6&7 interspecies variation in effectiveness, 57-8, 66-7 microbe-specific defences, 56-7, 64-5 Detoxification of harmful ions in anaerobic soils, 157-8 Digestion of microarthropods, 12-1 3, 16 Digitaria sanguinalis, 304 Diplanthera wrightii, 129 Diplopoda, 4, 9 Dispersal of fruit seeds see Microbial rot of microarthropods, 14-15, 16, 21 Display characteristics and risk of fruit rot, 47 Distribution of plants see Oxygen deprivation; Positive-feedback switches Drosophila, 41, 51, 68 Drought sensitivity, 128 see also Arid and semi-arid areas Earthworms, 23 Echinochloa, 100 Ectomycorrhizal-microarthropod interactions, 22-3 Efficiency of nitrogen use see Photosynthetic nitrogen use efficiency respiratory, variations in, 223-4

Eleocharis palustris, 133 Elymus repens, 137, 140 Empetrum nigrum, 162 Encelia farinosa, 198 Endogone, I3 1 Endophytic microbes, 53 Energy metabolism and hypoxic seed, 1 0 1 4 -requiring processes, variations in costs of, 2 2 4 5 Entomobrya purpurascens, 15 Entornopathogenic species, 8 Environment and fruit rot, 45-6 modification, see Positive-feedback switches signals of oxygen deprivation, 136-7 Epicoccum, 14 Epilobium hirsutum, 158 Eremophila, 290 Ericaceous species Erica cinerea, 28 1 see also Microbial rot and dispersal of fruits Erigeron strigosus, 304 Eriophorum vaginatum, 292 Erythrina caflra. 100 Espeletia, 198 Ethanol production, 42 and oxygen deprivation, 97, 1 12, 11416, 118-20, 138, 145-9, 151 Ethylene (hormone), 40,42, 151 Eucalyptus, 194, 21 1, 274, 289, 297, 299 E. populnea, 300 E. tetrodonta, 300 Eupatorium odoratum, 13I Evolution patterns and microbial rot and dispersal of fruits, 38, 68 see also Natural selection Excreta of microarthropods, 3, 11, 13-14, 16, 18 of rabbits, 6, 12 “Extensin”, 21 6 Exudation and volatile losses, 227-9, 238 rate (EXU and VOL), 192, 194

Facilitation succession, 265-6, 319-21 Fagopyrum esculentum, 97

INDEX

Fagus crenata, 290 Fast-growing and slow-growing species compared see Growth rate variations Feces see Excreta Feedback inhibition of photosynthesis, 218-19 see also Positive-feedback switches Fertility of soil and growth see Growth rate variations in higher plants Festuca, 200, 309, 3 10 F. ovina, 190, 197, 281 Filipendula ulmaria, 145 Fire-mediated switches, 293, 296301, 307, 315-16, 318 Flavonoids, 228 Flax, 102, 103 Flies fruit, 41, 51, 68 larvae, 7 sciarid, 12 Fog precipitation and switches, 271-5, 314, 315 Folsomia F. candida, 10, 1 1 , 12, 13, 21-2 F. jimetaria and F. regularis, 1 1 Foraging, see Herbivores Foreign species, introduction of, 137 Forests see Trees Frankia, 128 Frost, 291 Fruit and oxygen deprivation, 137, 152 Fruit rot defences interspecific variations in effectiveness, 57-8 microbe-specific, 5&7 natural selection for, 48-56 defined, 41 effects on dispersal, 3 9 4 3 factors affecting risk of see Risk of fruit rot general deterrent nature of, 56 see also Microbial rot and dispersal of fruits Fungi and fruit rot see Microbial rot and dispersal of fruits and growth rate variations in higher plants, 205, 228 and microarthropod-microbial

341

interactions in soil, 3, 5 as food for microarthropods, 8-10 microarthropods as food for, 8 response to grazing, 10-12, 15-16, 17 in rhizosphere, 17-24 in saprophytic system, 6 9 , 12, 14-17 and oxygen deprivation, 129-33 Fungivores see Microbial rot Fusarium, 8, 20, 46, 52 F. sporotrichioides, 6 1 Galinsoga parviyora, 190, 197, 232 Gases see Carbon dioxide; Oxygen Gaultheria procumbens, 60, 65 Gaylussacia, 47, 60 G ..frondosa, 6G5, 67 Genetics horizontal gene flow, 53 see also Natural selection Geotrichum, 46 G. candidum, 61 Geranium robertianum, 22 Germination and oxygen deprivation, 100-1 and energy metabolism, 10 1 4 Gibberellins and growth rate variations in higher plants, 199, 229-30, 240, 24 1 Gigaspora G . fasciculatum and G . margarita, 2 1 G . rosea, 22 Gloeosporium, 46 GIomus G . ,fasciculatus, 122 G . macrocarpus, 131 G . mosseae, 2 I , 22 G. occultus, 22 Glucose and toxin production, 21 1-13 Glyceria maxima, 1 1 I , 1 14, 144, 149, 154, 157, 159-60 Glycine max, 2 I Glycolysis, 137-9 end-products of, 147-8 rate and A D H induction, 144-7 see also Carbon dioxide and oxygen deprivation; Ethanol production Glycosides, 2 1 1 Gnat larvae, fungus, 7 Gnornonia leptostyla, 9

342

INDEX

Graminoid tussocks, 292 Grapes and fruit rot, 42, 52 Grasses and pasture plants and growth rate variations in higher plants, 201, 204, 237 and oxygen deprivation, 104, 129, 149, 155, 157, 159 and pasture plants and microarthropod-microbial interactions in soil, 19-20, 22 and positive-feedback switches allelopathy-mediated, 302-3 fire-mediated, 297-301 herbivore-mediated, 308-1 2 light-mediated, 289 microbe-mediated, 303-5 soil-element-mediated, 28 1-2, 287 termite-mediated, 3 0 6 7 water-mediated, 271-3, 315 wind-mediated, 293 see also Cereals Gratiola viscidula, 106 Grazing and microarthropod-microbial interactions in soil, 2, 3, 21-2 in microarthropod-microbial interactions in soilbacterial and fungal response to l(r12, 15-16, 17 see also Grasses; Herbivores Greenhouse gases and climate change, 164, 165 Grenzhorizont, 162 Growth rate variations in higher plants, 187-242 biomass allocation, 199-200 chemical composition, 206-14, 241 ecological consequences of ,variations, 235-9 exudation and volatile losses, 227-9 growth analysis, 1 9 1 4 hormonal differences, 229-30 integration of various aspects, 23 I 4 net assimilation rate and leaf area ratio, 1 9 4 6 roots, growth, morphology and nutrient acquisition, 2OCL6, 241 species-specific performance under sub-optimal conditions, 234 specific leaf area, 196-9 see also under Photosynthesis; Respiration

Growth respiration, 222, 224 Haemoglobin and oxygen deprivation, 127, 136 Halogeton glomeratus, 2 8 6 8 HaloxyIon H. aphyllum and H. persicum, 286 H. salicornicum, 293 Heavy metals and growth rate variations in higher plants, 190, 216, 228, 239 Hebeloma crustuliniforme and H. mesophaeum, 131 Herbaceous plants see in particular Growth rate variations Herbivores, 7 and growth rate variations in higher plants, 210, 217, 229, 233, 236, 238 herbivore-mediated switches, 276, 289, 300, 303, 307, 308-13 protection against see Toxins see also Defence; Grazing Holcus lunatus, 22, 198, 226, 234 Hordeum vulgare, 225, 228 Hormones and growth rate variations in higher plants, 229-30 see also Ethylene; Gibberellins Hydrocharis, 1 10 Hyperaccumulators, 2 8 3 4 Hyphomycetes, 7, 1 I Hypoestes fastigiatu, 300 Hypogastrura tullbergi, 1 I , 13 H ypoxia hypoxic seed and oxygen deprivation, 9 5 9 7 , 100-10 aquatic seed, 104-10 energy metabolism and oxygen availability for germination, 1014 germination, 100-1 root apex, 114-18 see also Oxygen deprivation Hysterangium setchellii, 23 Igapo forests, 106, 107 IIex opaca, 60, 210 Induced defences of fruit, 49-51 Infiltration switches, 275-6, 315 Inhibition model of succession, 3 19-20 Insects and dispersal of yeasts, 41

INDEX

and growth rate variations in higher plants, 210 herbivorous, 52 in pine forests, 3 12 Interspecies variation in effectiveness of fruit defences, 57-8, 66-7 Introduction of foreign species, 137 Ions in anaerobic soils, detoxification of 157-8 uptake see Nutrients Iris I . germanica, 147, 15&1 I . pseudacorus, 137, 143, 147, 151, 154, 156 Iron and growth rate variations, 227 and oxygen deprivation, 157-8 Irrigation, 134, 152 Isoetes lacustris, 133 Isopoda, 3, 4, 8, 11 Isoprene, 229 Juglans nigra, 9 Juncus conglomeratus and J. effusus, 111 Juniperus communis, 282 Kampfzonen, 161 Keystone species and switches, 319 Knudsen diffusion, 124 Kochia americana, 287 Krummholz, 2 9 4 5 , 3 15 Laccaria laccata and L . proxima, 130 Lactate dehydrogenase activity (LDH), 145 Lactuca saliva, 207 Landolphia, 300 LAR see under Leaves Lead, 285 Leaves and growth rate variations in higher plants, 230-1 area ratio (LAR), 191-6, 200, 230, 234, 238, 240, 241 and net assimilation rate, 1 9 4 6 see also Specific leaf area respiration rate (LR), 191, 192, 194, 223, 226 toxins, 21 1-12

343

weight ratio (LWR), 191-7, 199-200, 226, 232-3, 241 Ledum groenlandicum, 229 Leeks, 22 Leghaemoglobin, 127 Legumes, 106, 126-7, 228 see also Beans; Peas; Nitrogen fixing Lenticels, 1 19-20, 128 Lettuce, 100, 102, 103 Light-mediated switches, 288-90, 3 15, 318 Lignin and growth rate variations in higher plants, 196, 206, 207, 210, 213, 216 Limnocharis, 1 10 Lindera benzoin, 60 Lipids and fruit rot, 59, 64 and growth rate variations in higher plants, 206, 212, 213 and oxygen deprivation, 1 0 2 4 Litorella unijlora, 131 Lobelia dortmanna, 131 Lonicera, 283 LR see under Leaves “Luxury consumption”, 207 LWR see under Leaves Lycopala epidendrum, 5 Lycopersicon esculentum, 52, 199, 230 Lycopus europaeus, 106 Lycoriella mali, 7 Lythrum salicaria, 106

Macroarthropods, 3, 4, 8, 11 Macrotermes, 306 Maintenance respiration, 222, 224 Maize and oxygen deprivation, 100, 102-3, 115-18 consequences of, 139, 141-2, 146 Manganese, 227, 281, 285 manganous ions and oxygen deprivation, 157-8 Mangroves, 126, 137, 288, 31617 Marasmius androsaceus, 10 Mass movement of air in aquatic spheres, 123-6 Mauritia jlexuosa, 98 Melanerpes formicivorous, 52 Mentha spicata, 2 I 1 Mesembryanthemum crystallinum, 287-8

344

INDEX

Metabolism/metabolic adaptation to oxygen deprivation, 144-50 effects of root anoxia on whole plant physiology, 148-50 end-products of glycolysis, 147-8 glycolytic rate and A D H induction, 144-7 consequences of oxygen deprivation see under Oxygen deprivation energy, and germination, 1 0 1 4 oxygen sensing, 135-6 Metarrhizium anisopliae 8 Microarthropod-microbial interactions in soil, 1-25 historical and biological reasons for, 3-6 in rhizosphere, 2, 17-24 in saprophytic system, 6-1 7 see also in particular Collembola; Mites Microbe-mediated switches, 303-6, 3 14 Microbe-specific defences of fruit, 5 6 7 , 646 Microbial rot and dispersal of fruits: selection for secondary defence, 35-68 fruit rot and effects on dispersal, 39-56 hypotheses and predictions, 5&8 predictions for temperate seed dispersal, 58-67 variations in characteristics of fruits, 37-8 variations in secondary defence chemistry, 38-9 Microsclerotia, I5 Millipedes, 3, 1 1 Mineralization of soil, 17, 20, 2 I see also Nutrients Mites and microarthropod-microbial interactions in soil, 3 in rhizosphere, 20, 23 in saprophytic system, 6, 8-9, 12-13, 15 Molybdenum, 285 Monilinia, 47, 5 I , 67 Monocots as dominant species in wetlands, 96 see also Oxygen deprivation Mortierella isabellina, 7, 1 1

Mosaic vegetation situation, stable, 265. 269, 3 17-20 biological mediation, 307 physicakhemical mediation fire, 301 soil-element, 283, 287, 288 water, 275-6, 278 wind, 293, 296 Mosses, 162, 278-280, 282-3, 319 Mummy-berry disease, 47, 67 Mycelia sterilia, 46 Mycena galopus, 10 Mycorrhizal fungi, 17, 21-2 and growth rate variations in higher plants, 205, 228 and oxygen deprivation, 129-33 Myrosphaerella juglandis, 9 Mycotoxin-producing fungi, 53 Myrica, 127 M . gale, 128 Myriophyllum, 1 I0 M . alterniforum, 133 Myrothecium, 53 myxomycete, 5 Nujas marina, 109 NAR see Net assimilation rate Nardus stricta, 1 54, 309- 10, 3 1 2 Natural selection for defences against fruit rot, 48-56 antimicrobial, 54 biotic, 5 3 4 secondary chemicals as agents, 48-53 structural and chemical defences retained, 55-6 and growth rate variations in higher plants, 2 3 6 8 Nelumbo nucrfera, 100 Nematoda in soil, 3, 4, 19, 20-1, 23 Net assimilation rate, 191-5, 200, 234 and leaf area ratio, 1 9 4 6 Net nitrogen uptake rate, 192, 194, 20&1, 223, 226 Nickel, 284-5 NIR see Net nitrogen uptake rate Nitrates, 154 and growth rate variations in higher plants, 201-3, 206-7, 213, 222, 225 Nitrobacter, 305

INDEX

Nitrogen in anaerobic environments, 152-7 fixation, 228 in oldfields, 303-5 and oxygen deprivation, 126-8, 129 and growth rate variations in higher plants, 189-90, 194, 201-2, 206, 208-9, 21 I , 238, 241 and biomass allocation, 199-200 and respiration, 224-6 see also Photosynthetic nitrogen use efficiency and microarthropod-microbial interactions in soil, 17, 20 and positive-feedback switches, 281-3, 303-5, 11-12 Nitrosomonas, 305 “Noble” rot, 42 Nodulation see Nitrogen fixation Nothofagus solandri, 29 1 Nuphar lutea, 123, 124-5 Nutrients and flooding tolerance and oxygen deprivation, 152-8 detoxification of harmful ions in anaerobic soils, 157-8 nitrogen nutrition in anaerobic environments, 152-7 and growth rate variations in higher plants, 188-90, 194, 207-9, 21 1-14, 233, 238, 240-1 biomass allocation, 199-200 exudation, 227-9 and photosynthesis, 214-15, 2 19-22 and respiration, 222-6 and roots, 200-6 and positive-feedback switches, 274, 276, 281-8, 307 see also in particular Nitrogen Nvmphoides, 1 10 Nyssa sylvatica, 60, 104, 149 Oceanic climate, changes in, 161-70 Odontotermes, 307 Oil glands in plants, 21 1 Oldfield succession and nitrogen-fixing microbes, 303-5 Ombrogenous bog switches, 278-9

345

One-sided switch, 257, 264, 314 biological, 303-1 3 physicakhemical, 271-5, 280-8, 290-3, 301-3 Oniscus asellus, 8 Ony ch iurus 0. ambulans, 21 0. armatus, 9, 1 I , 22 0. encarpatus, 20, 22 0.jirmatus, 20 0.folsomi, 22 0. latus, 10 0. quadrocellatus, 1 I , 15 0. subtenuis, 9, 12, 15 Orbignya cohune, 286, 288, 319 Organic acids as fruit defence, 42, 51-2 and growth rate variations in higher plants, 206, 207, 213, 227-8 and oxygen deprivation, 146, 148 Oribatids see Mites Osmosis, thermal, 124-5 Over-grazing, 167 Oxygen content of atmosphere, fluctuation in, 95-6 deprivation as ecological limit to plant distribution, 93-171 and climate change, 161-70 consequences for survival and metabolism, 133-58 cellular effects, 1 4 1 4 metabolic adaptation to, 144-50 mineral nutrition and flooding tolerance, 152-8 post-anoxic injury, 15&2 sensing oxygen deficiency, 134-41 and plant competition, 94-5, 15841 plant organs liable to, 97-120 see also Above-ground organs; Hypoxia; Underground organs and plant structure, 120-6 and symbiosis, 1 2 6 3 3 see also Respiration Panicum P . laxum, 104 P . virgatum, 305 Parkia auriculata, 106

346

INDEX

Pastinaca sativa, 52 Pasture plants see Grasses Pathogens: saprophyte-pathogenmicroarthropod interactions, 17-21 Pauropods, 3, 4 Paxillus involutus, 130-1 Peanuts, 20 Peas and oxygen deprivation, 97, 100, 102, 103, 114 consequences of, 127, 137-8, 149, 151, 153 see also Chickpeas Peat, 162-70, 280, 282 Pelopoidea, 23 Penicillium, 1 4 1 5 , 46, 61 P. citrinum, 14 P . rubrum, 61 P. spinulosum, 7, 1 1 Pennisetum kikuyorum, 3 1 1 Penthorum sedoides, 106 Perennating organs, anoxia-tolerance variation in, 1 10-1 1 Persistence of fruit, variation in, 39-40 Pestalotiopsis maculans, 61 pH-mediated switches, 280-1, 316 see also Acids; Salts Phaseolus, I27 P . lunatus, 21 1 P . vulgaris, 97 Phasic theory of bog growth, 164 Phenols and fruit rot, 48-9, 51-2 and growth rate variations in higher plants, 196, 21 I , 228 Philoscia muscorum, 1 1 Phleum pratense, 190, 197 Phoma, 46 P. vaccinii, 61 Phomopsis, 46, 61 Phosphates and growth rate variations in higher plants, 201-2, 204, 207, 227 Phosphorus, 199, 206, 281-3, 317 Photosynthesis and growth rate variations in higher plants, 191-2, 21 I , 232-3, 238, 241 described, 214-22 species-specific variation in rate of, 214-15 under suboptimal conditions, 221-2 see also Photosynthetic nitrogen

Photosynthetic nitrogen use efficiency, 192, 215-22, 233, 240 carbon dioxide concentration within leaf, 218-19 feedback inhibition of, 218-19 partitioning of nitrogen within chloroplast, 21 6 1 7 Rubisco activation and variations, 217-18 and water use efficiency, 219-21 Phragmites australis, 106, 1 1 I , 1 12, 1 14, 125, 133, 159-60 Phthiracarus, 12 Phylloplane microbes, 53 Phytosiderophores, 227 Picea, 295 P . abies, 19, 119 P. engelmannii, 294, 3 19 P . marina, 280 P . sitchensis, 9, 130, 132 Pimpinella saxifraga, 190, 1': Pinus P. contorta, 17, 120, 121 P . nigra, 17 P. ponderosa, 207 P. sylvestris, 119 Pisum sativum, 97, 127 Plant nitrogen concentration, 192, 194, 200 Plantago major, 204, 206, 21 5 Plants communities see Positive-feedback switches distribution see Oxygen deprivation fruit see Fruit rot growth see Growth rate variations Plasticity and growth rate variations in higher plants biomass allocation, 22 root growth and nutrient acquisition, 202-5, 241 specific leaf area, 198-9 P N C see Plant nitrogen concentration Poa P . annua, 190, 197, 231 P. colensoi, 273, 293 Podophyllurn peltatum, 198 Polyamines, 150-1, 216 Populus P . deltoides, 119, 231 P. trernuloides, 229 P. trichocarpa, 231

INDEX

Positive-feedback switches in plant communities, 263-321 agencies listed, 271 allelopathy-mediated, 301-3 boundaries, 267-8, 3 15-1 6 community structure, 31 3-14 defined, 265-6 fire-mediated, 293, 296301, 307, 315-16, 318 herbivore-mediated, 276, 289, 300, 303, 307, 308-13 importance of, 321 keystone species, 3 19 and landscape, 317 light-mediated, 288-90, 3 15, 318 microbe-mediated, 303-6, 3 14 pH-mediated, 280-1, 3 I6 and soil see Soils, soil-element-mediated temperature-mediated, 290-2, 307 termite-mediated, 276, 300, 3 0 6 8 types of, 267, 3 1 4 1 5 vegetation situations produced by, 268-70 see also Acceleration; Delay; Mosaic; Sharpening versus succession, 3 19-20 and water see Water, water-mediated wind-mediated, 276, 292-6 Post-anoxic injury, 15&2 Potamogetonjliformis, 1 10, 129 Potassium, 28 1-3, 3 17 Potential growth, 238-9 Potentilla pafustris, 122 Precipitation, 166, 167 change see Climate change rainfall and positive-feedback switches, 271, 2 7 4 5 , 278 snow accumulation, 280 see also Fog Predictions of temperate seed dispersal systems, 5 6 8 , 6&7 Primary chemical compounds and growth rate variations in higher plants, 20&7 Proisotoma minuta, 13, 20, 22 Prosopis velutina, 289-90 Prostigmatids, see Mites Proteaceae, 205 Protein and growth rate variations in higher plants, 206, 212-13, 21617, 221-2, 224, 233

341

Protozoa, 19 Protura, 3, 4, 6 Prunus, 60 PS see Photosynthesis Pseudoscorpionidae, 4 Pseudosinella alba, 15 Psychotria dourarrei, 284 Puccinellia phryganodes, 3 1 1 pulp nutrient chemistry, 43-5 putrescine, 150-1 Pythium P. myriotylum, 15, 20 P. ultimum. 20 Qualitative defence compounds, 207, 209- 10, 24 1 Quality of fruit, variation in, 39-40 Quantitative defence compounds, 207, 24 1 Quantum flux density and growth rate variations in higher plants high, 214, 226 low, 198-200, 2 0 6 7 , 211, 215, 222, 225-6, 234 Quercus, 288-9 Rabbits, excreta of, 6, 12 Rainfall see Precipitation Raised peat bogs, 162-70 Ranunculus sceleratus, 148 Reaction switch, 264, 267, 315 biological, 3068, 3 I2 physicakhemical, 275-8, 293 Recurrence surfaces, 162 Reeds and oxygen deprivation, 106, 111, 112, 114, 125, 133, 159-60 Relative growth rate see Growth rate variations Relative-risk model of interspecific variation in defence effectiveness, 57-8, 66 Removal-rate model of interspecific variation in defence effectiveness, 58, 6 6 7 Resource denial, 160-1 Respiration and growth rate variations in higher plants, 188, 213 at suboptimal nitrogen supply or quantum flux density, 2 2 5 6 and growth of plants, 2 2 2 4

348

INDEX

Respiration and growth rate (cont.) root, 192, 232 see also under Species-specific RGR (relative growth rate) see Growth rate variations Rhagoletis, 5 1 Rhizobium, 128, 228, 304, 305 Rhizoctonia solani, 20 Rhizoglyphus echinopus, 15, 20 Rhizophora, 137 Rhizopus, 46 R. stolonifer, 61 Rhizosphere microarthropod-microbial interactions in, 2, 17-24 and oxygen deprivation, 128-9 consequences of, 140, 1 4 2 4 , 147, 150-1, 157 see also Underground organs Rhus typhina, 60 Rhynia, 3 Rhvtisma acerinum, 8 Ribes montigeum, 294 Rice and oxygen deprivation, 100, 101-3, 112, 116 consequences of, 134, 139, 148, 150-1, 1 5 4 5 Ripening, fruit, 40 Risk of fruit rot, factors affecting, 43-8 ambient environmental conditions, 45-6 pulp nutrient chemistry, 43-5 removal rate of time of exposure, 48 spore inoculum, identity and quantity of. 46-7 synchrony and display characteristics, 47 Roots crops, 102-3 and growth rate variations in higher plants, 200-6, 240-1 exudation, 227-8 nutrient acquisition, 200-5 respiration, 2 2 2 4 , 226 respiration rate (RR), 191, 192, 232 root weight ratio (RWR), 1 9 2 4 , 199, 200, 202, 205, 230, 232, 237, 24 1 specific root length (SRL), 192, 201,202

nodules and symbiosis and oxygen supply, 126-8 root apex hypoxia, 114-18 Rot see Fruit rot Rubisco activation and variations in higher plants, 217-18, 222, 233 Rubus, 60, 283 Ruderals, 223 Rumex, 120 R. acetosa, 154 R. crispus, 190, 197 RWR see under Roots and growth rate

Saccharomyces, 4 I , 46 S. cerevisiae, 6 1 Sagittaria, 110 Salicornia australis, 289 Salt and growth rate variations in higher plants, 190, 239 and switches, 286-8 salt marsh pans, 2768, 289, 315, 316 Saponins, 52, 210 Saprophytic system, 41 microarthropod-microbial interactions in, 617 saprophyte-pathogenmicroarthropod interactions, 17-21 Sasa, 290 Sassafras albidum, 60 Scheuchzeria palustris, 162 Schoenoplectus lacustris, 112-13, 141, 142, 144, 161 Sciarid fly, 12 Scirpus S. americanus, 112, 137, 144 S. lineatus, 106 S . maritimus, 1 12, 1 15 S. sylvaticus, 204 Sclerotinia sclerotiorum, 10 Scrophularia nodosa, 190, 197 Secondary plant chemicals and fruit rot, 35-6, 37-9, 56-7, 59 and dispersal, 42, 43-5, 48-53 pulp nutrient chemistry, 43-5 and growth rate variations in higher plants, 207-10, 21 1-12, 241

INDEX

Sediment entrapment and switches, 2 7 6 8 , 289, 3 15, 3 16 Seeds dispersal see Microbial rot and dispersal of fruits see also under Hypoxia Selection see Natural selection Selenium, 285 Senecio aquaticus, 148 Sensing oxygen deficiency in plant tissues, 1 3 4 4 1 environmental signals, 136-7 metabolic, 135-6 signal molecules, 136 Shade and sun species compared set> Growth rate variations Sharpening vegetation situation, 265. 269. 317 biological mediation, 3 12 physicakhemical mediation, 290, 291, 293, 301, 303 soil-element, 286, 288 water, 271-6 Signal molecules for oxygen deprivation, 136 Silene vulgarisicucubalus, 239 Sinella, 23 Skins of fruit, 50-1, 52 SLA see Specific leaf area Slow-growing and fast-growing species compared see Growth rate variations Smilax rotundijolia, 60 Snow accumulation, 280 SOD (superoxide dismutase), 151, I58 Sodium, 2 8 6 7 Soils, 307 anaerobic, detoxification of harmful ions in, 157-8 erosion and trapping, 292-3 fertility and growth see Growth rate variations soil-element-mediated switches, 276, 281-8, 307 heavy metals, 283-6, 3 14 NPK decrease, 282-3 NPK increase, 281-2, 317 salt, 2 8 6 8 see also Microarthropod-microbial interactions in soil Sordaria jimicola, 9

349

Sorghum bicolor, 21 Soybeans, 21 and oxygen deprivation, 103, 1 1I , 127-8 Sparganium, 1 10 Spartina alterniflora, 129, 142 Species-specific variations in growth rate of photosynthesis, 21415 respiration rate, 222-5 costs of energy-requiring processes, 224-5 efficiency, 2 2 3 4 under sub-optimal conditions, 234 Specific leaf area and growth rate variations in higher plants, 189, 199, 230, 2 3 2 4 components of, 196-8 high, 238, 240-1 low, 236-7 plasticity, 198-9 Specific root length, 192, 201, 202 Sphagnum, 162, 278-9 282-3, 319 S. jiuscum, 280 Spirillum lipoferum, 129 SRL see Specific root length Stable vegetation see Mosaic vegetation Staphylinidae, 9 Starch see Carbohydrates Steccherinum jimbriatum, 7 Stems and growth rate variations in higher plants respiration rate (SR), 191, 192, 232 weight ratio (SWR), 1 9 2 4 , 199 Stephanomeria malheurensis and S. exigua, 237 Stratiotes, 110 Structure of plants and oxygen deprivation, 120-6 aerenchyma distribution and function, 94, 116, 120-3, 145 mass movement of air in aquatic spheres, 123-6 Suboptimal conditions and plant growth, 210-1 I , 221-2, 2 2 5 4 Succession versus positive-feedbak switches, 3 19-20 Sucroseisugars see Carbohydrates Sun species see Shade and sun Sunflower, 103, 146 Superoxide dismutase, 15 I , 158

350

INDEX

Survival consequences of oxygen deprivation see under Oxygen deprivation Switches see Positive-feedback switches SWR see under Stems and growth rate Symbiosis and oxygen supply, 1 2 6 3 3 and mycorrhizas, 129-33 nitrogen fixation in rhizosphere of aquatic plants, 129 root nodules, 1 2 6 8 Symmetric switch, 264, 267, 315 biological, 312 physical-chemical, 283, 294-6 Synchrony and risk of fruit rot, 47 Syringodium ,filijorme, 1 29 Tannins and fruit rot, 49, 52 and growth rate variations in higher plants, 196, 21&-12 Taxodium distichum, 104-5 Temperate seed dispersal systems, 58-67 fruiting classes, 58-60 predictions, 6&7 Temperate vertebrate-dispersed species see Microbial rot and dispersal of fruits Temperature change see Climate change -mediated switches, 29&2, 307 Termite-mediated switches, 276, 300, 3068 Terpenoids, volatile, 2 10, 2 13 Tetrahymena pyrijormis, 135 Thalassia testudinium, 129 Thelophora terrestris, 132 Thermal transpiration (-osmosis), 124-5 Thlaspi arvense, 230 Thorns, 236 Tomocerus T. longicornis, 12 T . minor, 1 I , 13, 17 Toxins in anaerbic soil, 157-8 and fruit rot, 45, 48, 5 I , 55, 5 6 7 and growth rate variations in higher plants, 196, 2 0 S I 3, 21 7, 228 and positive-feedback switches, 283-6, 302-5, 314 I

Trampling, 236 Transpiration, thermal, 124-5 Trapa, 110 Trees and growth rate variations in higher plants, 193, 194, 196, 210, 214, 216, 219, 230-1 and microarthropod-microbial interactions in soil, 6, 8-10, 14-15, 17, 19, 21 and oxygen deprivation, 100, 104-8, I 18-20 and climate change, 162, 1 6 6 9 consequences of, 121, 125-6, 129-30, 134, 149, 154-5 removed before peat formation, 163, 1 6 6 9 tree mycorrhizas, 129-30 and positive-feedback switches fire-mediated, 297-301, 318 herbivore-mediated, 31 2 light-mediated, 288-9, 318 microbe-mediated, 305-6 soil-element-mediated, 282-3, 286 temperature-mediated, 29&2 termite-mediated, 307-31 I tree-line, 29&1, 294, 314, 315 water-mediated, 2 7 1 4 , 3 14, 3 15 wind-mediated, 294-6 Trichoderma, 6, 53 T . harzianum, 20 Trifolium, 127 T . pratense, 22, 146 Triticum aestivum, 142, I98 Tubers see Underground organs Tundra, 280, 292, 294-6 Tussilagofarfara, 137, 139 Two-factor switch, 264, 267, 315 physicakhemical, 281, 283, 288-90, 29630 I Typha, 110 T. angustijolia, 112, 142 T . latifolia, 106, 112, 114, 142, 16&1

Ulex europaeus, 280, 28 1 Underground organs and oxygen deprivation, 95, 11&18 anoxia-tolerance variation in perennating organs, 1 1 0 - 1 1

INDEX

re-emergence fom anaerobic habitats, 112-14 root apex hypoxia, 114-18 Uranium, 285 Urtica dioica, 201-2 Utricularia, 110

Vaccinium, 47, 51, 52, 60 V. corymbosum, 60-5, 67 V . macrocarpon, 42, 60-5, 67 V. vacillans, 60, 65 Vallisneria, 1I0 VAM see Vesicular-arbuscular m ycorrhiza Vapour-pressure deficit, 2 19 Variations in growth, see Growth rate variations Varzea forests, 106, 109 Vegetables, 32, 100, 102-3 see also Beans; Peas Veronica montana and V. persica, 198 Vertebrates and seed dispersal see Microbial rot and dispersal of fruits Vert icillium, 20 V. albo-atrum, 15, 20 V. bulbillosum, 1 1 V . lecanii, 8 Vesicular-arbuscular mycorrhizas, 130-3 microarthropod interactions, 17, 21-2, 25 Viburnum lentago, 289-90

35 1

Victoria amazonica, 109 Vigna unguiculata, 127, 128 Vitis vinifera, 52 VOL see under Exudation Volatile losses, see Exudation

Water and growth rate variations in higher plants, 199, 218-21, 241 water use efficiency (WUE), 219-21 water-mediated switches, 271-80 fog precipitation, 271-5, 314, 315 infiltration, 2754, 3 15 ombrogenous bogs, 278-9 salt marsh, 276-8, 289, 315, 316 snow accumulation, 280 Weight of roots see under Roots and growth rate variations Whet and oxygen deprivation 102-3, 116, 118, 142 Wild rice, 100, 106 Wind and growth rate variations in higher plants, 236, 237 -mediated switches, 276, 292-6 Wood and woodland, see Trees WUE see under Water and growth rate Yeasts and fruit rot, 41, 42, 68 Zea mays, 141, 199, 230 Zinc, 227, 285 Zizania aquatica, 100, 106 Zostera marina, 129, 149, 155, 157

This Page Intentionally Left Blank

Advances in Ecological Research Volumes 1-23 Cumulative List of Titles Aerial heavy metal pollution and terrestrial ecosystems, 11, 218 Analysis of processes involved in the natural control of insects, 2, 1 Ant-plant-homopteran interactions, 16, 53 Biological strategies of nutrient cycling in soil systems, 13, 1 Bray-Curtis ordination: an effective strategy for analysis of multivariate ecological data, 14, 1 Can a general hypothesis explain population cycles of forest lepidoptera? 18, 179 A century of evolution in Sparrina anglica, 21, 1 The climatic response to greenhouse gases, 22, 1 Communities of parasitoids associated with leafhoppers and planthoppers in Europe, 17, 282 Community structure and interaction webs in shallow marine hard-bottom communities: Tests of an environmental stress model, 19, 189 The decomposition of emergent macrophytes in fresh water, 14, 115 Dendroecology: A tool for evaluating variations in past and present forest environments, 19, 1 1 1 The development of regional climate scenarios and the ecological impact of greenhouse gas warming, 22, 33 Developments in ecophysiological research on soil invertebrates, 16, 175 The direct effects of increase in the global atmospheric CO, concentration on natural and commercial temperate trees and forests, 19, 2 The distribution and abundance of lake-dwelling Triclads - towards a hypothesis, 3, 1 The dynamics of aquatic ecosystems, 6, 1 The dynamics of field population of the pine looper, Bupalus piniarius L. (Lep., Geom.), 3, 207 Earthworm biotechnology, and global biogeochemistry, 15, 379 Ecological aspects of fishery research, 7, 114 Ecological conditions affecting the production of wild herbivorous mammals on grasslands, 6, 137 Ecological implications of dividing plants into groups with distinct photosynthetic production capabilities, 7, 87 Ecological implications of specificity between plants and rhizosphere micro-organisms, 21, 122 Ecological studies at Lough h e , 4, 198 Ecological studies at Lough Hyne, 17, 11 5 The ecology of the Cinnabar moth, 12, 1 Ecology of the coarse woody debris in temperate ecosystems, 15, 133 Ecology, evolution and energetics: a study in metabolic adaptation, 10, 1 Ecology of fire in grasslands, 5 , 209 Ecology of mushroom-feeding Drosophilidae, 20, 225 The ecology of pierid butterflies: dynamics and interactions, 15, 51

354

CUMULATIVE LIST OF TITLES

The ecology of serpentine soils, 9, 255 Ecology, systematics and evolution of Australian frogs, 5, 37 Effects of climatic change on the population dynamics of crop pests, 22, I 17 The effects of modern agriculture, nest predation and game management on the population ecology of partridges (Perdix perdix and Alectoris rufa), 11, 2 El Nifio effects on Southern California kelp forest communities, 17, 243 Energetics, terestrial field studies and animal productivity, 3, 73 Energy in animal ecology, 1, 69 Estimating forest growth and efficiency in relation to canopy leaf area, 13, 327 Evolutionary and ecophysiological responses of mountain plants to the growing season environment, 20, 60 The evolutionary consequences of interspecific competition, 12, 127 Forty years of genecology, 2, 159 The general biology and thermal balance of penguins, 4, 131 General ecological principles which are illustrated by population studies of Uropod mites, 19, 304 Genetic and phenotypic aspects of life-history evolution in animals, 21, 63 Geochemical monitoring of atmospheric heavy metal pollution: theory and applications, 18, 65 Heavy metal tolerance in plants, 7,2 Herbivores and plant tannins, 19, 263 Human ecology as an interdisciplinary concept: a critical inquiry, 8, 2 Industrial melanism and the urban environment, 11, 373 Inherent variation in growth rate between higher plants: a search for physiological causes and ecological consequences, 23, 187 Insect herbivory below ground, 20, 1 Integration, identity and stability in the plant association, 6, 84 Isopods and their terrestrial environment, 17, 188 Landscape ecology as an emerging branch of human ecosystems science, 12, 189 Litter production in forests of the world, 2, 101 Mathematical model building with an application to determine the distribution of Dursbane insecticide added to a simulated ecosystem, 9, 133 Mechanisms of microarthropod-microbial interactions in soil, 23, 1 The method of successive approximation in descriptive ecology, 1, 35 Modeling the potential response of vegetation to global climate change, 22, 93 Mutualistic interactions in freshwater modular systems with molluscan components, 20, 126 Mycorrhizal links between plants: their functioning and ecological significance, 18, 243 Mycorrhizas in natural ecosystems, 21, 171 Nutrient cycles and H' budgets of forest ecosystems, 16, 1 On the evolutionary pathways resulting in C, photosynthesis and crassulacean acid metabolism (CAM), 19, 58 Oxygen availability as an ecological limit to plant distribution, 23, 93 The past as a key to the future: The use of palaeoenvironmental understanding to predict the effects of man on the biosphere, 22, 257 Pattern and process in competition, 4, 1 Phytophages of xylem and phloem: a comparison of animal and plant sap-feeders, 13, 135 The population biology and turbellaria with special reference to the freshwater triclads of the British Isles, 13, 235

CUMULATIVE LIST OF TITLES

355

Population cycles in small mammals, 8, 268 Population regulation in animals with complex life-histories: formulation and analysis of a damselfly model, 17, I Positive-feedback switches in plant communities, 23, 263 The potential effect of climate changes on agriculture and land use, 22, 63 Predation and population stability, 9, 1 Predicting the responses of the coastal zone to global change, 22, 212 The pressure chamber as an instrument for ecological research, 9, 165 Principles of predator-prey interaction in theoretical experimental and natural population systems, 16, 249 The production of marine plankton, 3, 117 Production, turnover, and nutrient dynamics of above- and below-ground detritus of world forests, 15, 303 Quantitative ecology and the woodland ecosystem concept, 1, 103 Realistic models in population ecology, 8, 200 Relative risks of microbial rot for fleshy fruits: significance with respect to dispersal and selection for secondary defense, 23, 35 Renewable energy from plants: bypassing fossilization, 14, 57 Responses of soils to climate change, 22, 163 Rodent long distance orientation (“homing”), 10, 63 Secondary production in inland waters, 10, 91 The self-thinning rule, 14, 167 A simulation model of animal movement patterns, 6, 185 Soil arthropod sampling, 1, 1 Soil diversity in the tropics, 21, 316 Stomata1 control of transpiration: Scaling up from leaf to region, 15, 1 Structure and function of microphytic soil crusts in wildland ecosystems of arid to semi-arid regions, 20, 180 Studies on the cereal ecosystem, 8, 108 Studies on grassland leafhoppers (Auchenorrhyncha, Homoptera) and their natural enemies, 11, 82 Studies on the insect fauna on Scotch Broom Sarothamnus scoparius (L.) Wimmer, 5, 88 Sunflecks and their importance to forest understorey plants, 18, 1 A synopsis of the pesticide problem, 4, 75 Theories dealing with the ecology of landbirds on islands, 11, 329 A theory of gradient analysis, 18, 271 Throughfall and stemflow in the forest nutrient cycle, 13, 57 Towards understanding ecosystems, 5, 1 The use of statistics in phytosociology, 2, 59 Vegetation, fire and herbivore interactions in heathland, 16, 87 Vegetational distribution, tree growth and crop success in relation to recent climate change, 7, 177 The zonation of plants in freshwater lakes, 12, 37

This Page Intentionally Left Blank