Advances in Agronomy, Volume 36

ADVANCES IN

AGRONOMY

VOLUME 36

CONTRIBUTORS TO THIS VOLUME C . AZCON-AGUILAR

J. M. BAREA P. BEMIS WILLIAM E. B. A. BISDOM JEAN-MARC BOLLAG C. M. DONALD

IANB. EDWARDS

KEITHW. T. GOULDING J. HAMBLIN

KRITONK. HATZIOS LEMOYNEHOGAN J. LOLL MICHAEL J. NEILRUTGER

K. L. SAHRAWAT S. S. VIRMANI

ADVANCES IN

AGRONOMY Prepared in cooperation with the AMERICAN SOCIETY OF AGRONOMY

VOLUME 36 Edited by N. C. BRADY Science and Technology Agency for International Development Department of Srate Washington, D . C .

ADVISORY BOARD H . J. GORZ.CHAIRMAN

E. J . KAMPRATH T. M. STARLING

J. B. POWELL J . W. BIGGAR M. A . TABATABAI M . STELLY. EX

OFFICIO,

ASA Headquarters I983

ACADEMIC PRESS, INC. (Harcourt Brace Jovanovich, Publishers)

Orlando San Diego San Francisco New York London Toronto Montreal Sydney Tokyo Sao Paulo

COPYRIGHT @ 1983, BY ACADEMIC PRESS, INC. ALL RIGHTS RESERVED. NO PART OF THIS PUBLICATION MAY BE REPRODUCED OR TRANSMITTED IN ANY FORM OR BY ANY MEANS, ELECTRONIC OR MECHANICAL, INCLUDING PHOTOCOPY, RECORDING, OR ANY INFORMATION STORAGE AND RETRIEVAL SYSTEM, WITHOUT PERMISSION IN WRITING FROM THE PUBLISHER.

ACADEMIC PRESS,INC.

11 1 Fifth Avenue, New York,New York 10003

United Kin dom Edition ublished by ACADEM~CPRESS,I&. (LONDON) LTD. 24/28 Oval Road,London NWI

7DX

LIBRARY OF CONGRESS CATALOG CARD N U M B E R

ISBN 0-12-000736-3 PRINTED IN THE UNITED STATES OF AMERICA 83848586

9 8 7 6 5 4 3 2 1

50-5598

This volume is dedicated to Dr. Arthur Geoffrey Norman, editor of the first 20 volumes of Advances in Agronomy.

This Page Intentionally Left Blank

CONTENTS CONTRIBUTORS .................................................

xi

PREFACE ....................................................... IN MEMORIAM.................................................

Xlll

...

xv

MYCORRHIZAS AND THEIR SIGNIFICANCE IN NODULATING NITROGEN-FIXING PLANTS

J . M . Barea and C. Azc6n-Aguilar I . Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . I1. Mycorrhizas .................................................

1 4 23

111. Mycorrhizas in Legumes ...................................... IV . Mycorrhizas in Nodulating Nitrogen-Fixing Nonlegume Plants . . . . . . . V. Conclusions and Perspectives .................................. References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

44 45

46

SUBMICROSCOPIC EXAMINATION OF SOILS

E . B . A . Bisdom I. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . II. Submicroscopic Techniques .................................... 111. Applications of Electron Microscopy ............................ IV . Applications of Ion Microscopy ................................ V . Applications of Other Forms of Submicroscopy .................... VI . Conclusions ................................................. References .................................................

55 57

65 88 89 90 91

THE CONVERGENT EVOLUTION OF ANNUAL SEED CROPS IN AGRICULTURE

C. M . Donald and J . Hamblin I. Introduction .................................................

II. Selection in Domesticated Crops ............................... I11. IV . V. VI .

Ekotypic Parallelism in Crop Plants ............................ Selection, Evolution, and Crop Yield ........................... Progress and Prospects in the Development of Annual Seed Crops . . . A Basic Ideotype for All Annual Seed Crops ..................... References ................................................

vii

97 100 111 112 121 134 139

...

Vlll

CONTENTS CURRENT STATUS AND FUTURE PROSPECTS FOR BREEDING HYBRID RICE AND WHEAT

S. S . Virmani and Ian B . Edwards I . Introduction ................................................ Heterosis in Rice and Wheat .................................. Advantages of Hybrids over Conventionally Bred Varieties . . . . . . . . . Cytoplasmic-Genetic Male Sterility Systems in Rice and Wheat . . . . . Fertility Restoration ......................................... Use of Chemical Pollen Suppressants in Hybrid Production ......... Factors Affecting Cross-Fertilization ............................ Seed Production ............................................ Ix. Quality of Hybrids .......................................... X . Economic Considerations ..................................... XI . Problems .................................................. XI1. Conclusion ................................................ References ................................................

11. I11. IV . V. VI. VII. VIII.

146 147 155 157 169 180 183 191 196 198 200 202 206

THERMODYNAMICS AND POTASSIUM EXCHANGE IN SOILS AND CLAY MATERIALS

Keith W . T. Goulding

I. Introduction ................................................ 11. The Thermodynamics of Ion-Exchange Equilibria . . . . . . . . . . . . . . . . . 111. Calorimetry in Ion-Exchange Studies ........................... IV . Thermodynamics Applied to Potassium Exchange in Soils and Clay Minerals ............................................ V. Exchange Equilibrium and the Kinetics of Potassium Exchange . . . . . . VI . Summary and Conclusions .................................... VII. Appendix: List of Symbols ................................... References ................................................

215 217 228 233 256 258 259 260

HERBICIDE ANTIDOTES: DEVELOPMENT. CHEMISTRY. AND MODE OF ACTION

Kriton K . Hatzios I . Introduction ................................................ 11. Development of Herbicide Antidotes ........................... III. Chemistry of Herbicide Antidotes .............................. IV . Field Performance of Herbicide Antidotes ....................... V . Mode of Action of Herbicide Antidotes .........................

265 270 292 296 301

ix

CONTENTS

VI . Degradation of Herbicide Antidotes in Plants . . . . . . . . . . . . . . . . . . . . . VII . Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

309 310 310

BUFFALO GOURD AND JOJOBA: POTENTIAL NEW CROPS FOR ARID LANDS

LeMoyne Hogan and William P. Bemis I. 11. 111. IV .

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Buffalo Gourd: Cucurbitu foeridissirnu HBK ..................... Jojoba: Simmondsiu chinensis (Link) Schneider . . . . . . . . . . . . . . . . . . . Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

317 319 332 346 347

PROTEIN TRANSFORMATION IN SOIL

Michael J . Loll and Jean-Marc Bollag I. I1. 111. IV . V. VI . VII .

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Protein Sources . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Proteolytic Microorganisms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Characteristics of Proteolytic Enzymes in Soils . . . . . . . . . . . . . . . . . . . Environmental Factors Affecting Proteolysis ..................... Transformation and Binding of Protein in Soil .................... Ecological and Agronomic Importance of Protein Transformation . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

352 352 354 361 364 370 376 377

APPLICATIONS OF INDUCED AND SPONTANEOUS MUTATION IN RICE BREEDING AND GENETICS

J . Neil Rutger I . Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . I1. Breeding Applications of Semidwarf Mutants . . . . . . . . . . . . . . . . . . . . 111. Breeding Applications of Early Maturity Mutants . . . . . . . . . . . . . . . . . IV . Breeding Applications of Other Types of Mutants . . . . . . . . . . . . . . . . . V . Genetic Applications of Mutants . . . . . . . . . . . .................... VI . Future Uses of Mutation in Rice Improvement . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

383 385 396 399 404

408 410

X

CONTENTS

NITROGEN AVAILABILITY INDEXES FOR SUBMERGED RICE SOILS

K . L . Sahrawat I. I1. 111. IV .

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Factors Affecting Mineralization of Organic Nitrogen . . . . . . . . . . . . . . Biological Indexes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Chemical Indexes ........................................... V . Simple Models of Nitrogen-Supplying Capacity Based on Biological and Chemical Indexes ......................................... VI . A Values . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . VII . Electro-Ultrafiltration ........................................ VIII . Plant Analyses ............................................. IX . Nitrogen-Supplying Capacity and Fertilizer Recommendations . . . . . . . X . Perspectives ............................................... References ................................................

415 417 421 428

INDEX . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

453

435 439 441 442 443 445 447

CONTRIBUTORS Numbers in parentheses indicate the pages on which the authors’ contributions begin

C. AZCON-AGUILAR (l), Unidad de Microbiologia, Estacibn Experimental del Zaidin, Granuda, Spain J. M. BAREA (l), Unidad de Microbiologia, Estacibn Experimental del Zaidin, Granada, Spain WILLIAM P. BEMIS (3 17), Plant Sciences Department, University of Arizona, Tucson, Arizona 85721 E. B. A. BISDOM ( 5 3 , Netherlands Soil Survey Institute, 6700 AB Wageningen, The Netherlands JEAN-MARC BOLLAG (35 1 ) , Department of Agronomy, Pennsylvania State University, University Park, Pennsylvania I6802 C. M. DONALD (97), Waite Agricultural Research Institute, The University of Adelaide, Glen Osmond, South Australia 5064 IAN B. EDWARDS (145), Pioneer Hi-Bred Institute, Inc., Glyndon, Minnesota 56547 KEITH W. T. GOULDING (215), Soils and Plant Nutrition Department, Rothamsted Experimental Station, Harpenden, Herifordshire AL5 2JQ, United Kingdom J . HAMBLIN (97), Department of Agriculture, Geraldton District Office, Marine Terrace, Geraldton, West Australia KRITON K . HATZIOS (265), Department of Plant Pathology and Physiology, Virginia Polytechnic Institute and State University, Blacksburg, Virginia 24061 LEMOYNE HOGAN (3 17), Plant Sciences Department, University of Arizona, Tucson, Arizona 85721 MICHAEL J. LOLL (35 l), Department of Agronomy, Pennsylvania State University, University Park, Pennsylvania 16802 J . NEIL RUTGER (383), U.S. Department of Agriculture, Agricultural Research Service, and Department of Agronomy and Range Science, University of California, Davis, California 95616 K. L. SAHRAWAT* (4 1 3 , Soil Science Department, ICRISAT, Patancheru P. O., Andhra Pradesh 502324, India S. S . VIRMANI ( 1 4 3 , International Rice Research Institute, Manila, Philippines

*Present address: Soil Science Department, University of Wisconsin, Madison, Wisconsin 53706

xi

This Page Intentionally Left Blank

PREFACE Two events occurred in the past year which have special significance for this review publication. First, the American Society of Agronomy (ASA) celebrated its 75th anniversary. The first 30 volumes of Advances in Agronomy were prepared under the auspices of this scientific society, and the remaining volumes have been developed in cooperation with it. This long association has been most fruitful for the series and has likewise been beneficial to the society. Most of the articles, particularly in the early years, have been authored by ASA members. An advisory committee chosen by the society has provided advice and guidance from the first to the present volume. The American Society of Agronomy has made great progress during the last three-quarters of a century and is to be congratulated on its 75 years of service. It has grown from the handful of dedicated soil and crop scientists, who met in Chicago in 1907 to form the society, to a membership of more than 12,000 today. It publishes four major research and education journals whose articles make up a fair share of those reviewed in Advances in Agronomy. More than 20 major monographs and 45 special publications have been published by the ASA. The second event of the past year which has special significance to Advances in Agronomy is a sad one. Dr. Geoffrey Norman, who was the founding editor and who continued as editor for the first 20 volumes, passed away on November 14, 1982. Crop and soil scientists throughout the world owe a debt of gratitude to Dr. Norman: He was not only a world-renowned soil microbiologist in his own right, but also an intellectual leader who stimulated biological science in general. We are pleased to publish a brief but meaningful tribute to him in this volume. The articles in Volume 36 reflect the advice given to me by Dr. Norman when 1 became editor. They each focus on a timely topic of wide interest to agronomists. They are written by scientists and educators from seven countries, illustrating the growing internationality of crop and soil science. And they repregent balance in subject matter among crops and soils and among basic and applied research. Advances in Agronomy continues to play a important role in keeping crop and soil scientists abreast of major research findings around the world. We are all challenged to maintain the energy and quality which so characterized Dr. Norman. The 15 scientists who prepared reviews for this volume have set excellent examples for us to follow. N. C. BRADY

...

Xlll

This Page Intentionally Left Blank

IN MEMORIAM ARTHURGEOFFREYNORMAN 1905-1982

A. G. Norman, who was the first editor of Advances in Agronomy, and who continued in that capacity for a period of 20 years, died November 14, 1982, after a distinguished career spanning more than half a century. In the Preface to Volume 1, he wrote, “This volume, Advances in Agronomy, has as its objective the survey and review of progress in agronomic research and practice. The editors . . . will be guided in their choice more by what information may be of use to agronomists than by what constitutes agronomy. The central theme must be soil-crop relationships, for soils without crops are barren, and field crops cannot be considered without reference to the soil on which they are produced.” The broad range of subjects covered by Advances over the years and the status which it has attained attest to the wisdom of this policy which he initiated. At the close of his 20 years of editorial service for this publication, he wrote in the preface to Volume 20, “Those who had a part in what seemed to be an uncertain venture in 1948 can take some pride in its acceptance . . .” “In the next 20 years one may confidently expect the accretion of new knowledge about the characteristics of soils and crop plants and of their interactions to proceed at an accelerating rate. These developments will find their way into later volumes and serve the agronomists of the world in their great task of providing sufficient food for all men.” Dr. Norman’s capacity for well-organized expression, both in speaking and writing, served the American Society of Agronomy well in other editorial efforts. Shortly after World War I1 he initiated the Monographs Series of the society and served as editor of the first six volumes. Other service to the society included a term as President in 1957, a year in which its 50th Anniversary was commemorated by an outstanding program in meetings at Atlanta, Georgia. A. G. Norman was born in Birmingham, England in 1905. He received the B.Sc. degree from the University of Birmingham in 1925 and the Ph.D. in Biochemistry from the same institution in 1928. This was at a time when biochemistry was just emerging as a separate discipline. Dr. Norman’s training there kindled a lifelong interest in plant biochemistry. From Birmingham, Norman went to the Rothamsted Experimental Station, where he began biochemical and microbiological studies on the decomposition of plant materials with special emphasis on cell wall substances. He did some of the first quantitative work on nitrogen transformations in the decomposition of plant materials, which has xv

xvi

IN MEMORIAM

subsequently had a major impact on crop residue management in practical agriculture. He came to the United States in 1930 as a Rockefeller Fellow at the University of Wisconsin, where he studied the microbiology of hemicelluloses and the structure of some fungal polysaccharides. Returning to Rothamsted in 1932, he became head of the Biochemistry Section there in 1933. In 1937 Dr. Norman moved to Iowa State College at Ames as Professor of Soils, where he directed a broad-ranging research program dealing with microbial thermogenesis, biochemistry of the major plant constituents and their decomposition processes in soil, fundamental studies on the chemistry of soil organic matter, and on carbon-nitrogen transformationsduring decomposition of organic materials. His application of biochemical techniques to studies in soil microbiology represented a significant departure from the older traditional techniques, which had their roots in medical bacteriology. Dr. Norman pioneered the application of stable isotope techniques to soil research and was possibly the first to use both "N- and I3C-labeled plant materials in decomposition studies. As early as 1943 he published an article indicating the potential of stable tracer technology in agronomic research. One of his significant early contributions was the use of "N in greenhouse experiments to measure N2 fixation by legumes. Field application of this type of methodology is now being made on a worldwide scale. During World War I1 Dr. Norman left the Iowa State campus to serve with the Chemical Corps for 2 years, directing a research program at Camp Detrick, Maryland. After a brief return to Iowa State, he left late in 1946 to accept a civilian position with the Chemical Corps dealing with basic studies on plant growth regulators and inhibitors. In 1952 he went to the University of Michigan as Professor of Botany and director of a research project in plant nutrition and root physiology as part of a university program promoting the use of radiation and radionuclides in the biological sciences. During this period he became much interested in rhizosphere microorganisms and in metabolic products which, when excreted, modified root growth and root physiology. Working with a number of graduate and postdoctoral students, he developed a research program concerned also with geotropic responses, factors limiting microbial activities in soils, influence of organisms on nutrient uptake, and artificial microbial environments. Other responsibilities at the University of Michigan included Directorship of the Botanical Gardens, a facility providing support for instruction and research in several university departments. In 1964 he was appointed Vice President for Research at the University of Michigan, in which capacity he continued to serve until his retirement. During 1963-1964 he was a Staff Advisor to the National Academy of Sciences, being involved, among other things, with initial organization of U.S. participation in the International Biological Program. From 1965 to 1969 he was Chairman of the Division of Biology in Agriculture of the National Research Council.

IN MEMORIAM

xvii

One of Dr. Norman’s significant contributions to the field of agronomy was through the students who came under his direction. His clear, well-organized lectures and his precise and articulate expression had a lasting influence upon his students. His insistence on clear and concise writing made a valuable contribution to the training of those who took his classes. A. G. Norman was that rare combination of brilliant research scientist, stimulating and effective teacher, and superbly organized and efficient administrator. Advances in Agronomy salutes the memory of a man who not only served this publication well as its editor for 20 years, but who in addition brought great credit to himself and to his profession.

FRANCES BROADBENT

This Page Intentionally Left Blank

ADVANCES IN AGRONOMY. VOL. 36

MYCORRHIZAS AND THEIR SIGNIFICANCE IN NODULATING NITROGEN-FIXING PLANTS J. M. Barea and C. Azcon-Aguilar Unidad de Microbiologia, Estaci6n Experimental del Zaidin Granada, Spain

I. Introduction ............................................ A. Root Microorganisms in the Ecosystem ...................... B. Mycorrhizas and Root Nodules. . . . . . . . . . . . . . . .

............................................ 11. Mycorrhizas . A. General ................................... B. Physiology of Mycorrhizas . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . C. Factors Affecting Mycorrhiza Establishment, Development, and Function.. . . . . . . . . . . . . D. Applications of Mycorrhizal P ...................... Horticulture, and Forestry . . . 111. Mycorrhizas' in Legumes . . . . . . . . . . . . . . . . . . . . . . . . . . A. Introduction ...................................................... B. Occurrence . . .............. ...................... C. Interactions be cies of Rhizobiu and Legumes.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . D. Interactions between Added Fertilizers and Myconhizas in Legume-Rhizobium sp. Systems. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . E. Ecological Significance of Vesicular-Arbuscular Mycorrhizas in Legumes. . , F. Practical Field Application of Mycorrhizal Effects on Legume Production IV. Mycorrhizas in Nodulating Nitrogen-Fixing Nonlegume Plants . . . . . . . . . . . . . . . . , A. Occurrence and Distribution ........................................ ical Features . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . V.

Conclusions

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

1 1 3 4 4 4 10 15 19 23 23 24 25 31 37 41 44 44

44 45 45 46

1. INTRODUCTION A. ROOT MICROORGANISMS IN THE

ECOSYSTEM

Microorganisms, which are known to play vital roles in physiological processes in the ecosystem, are invariably present in the root region, the rhi1

Copyright 8 by Academic F'ress, Inc. All righu of reproduction in any form reserved. ISBN 0-12600736-3

2

J. M. BAREA AND C. AZC6N-AGUILAR

zosphere, of plants growing in soil. Actually, rhizosphere bacteria (including actinomycetes) and fungi carry out a range of activities (e.g., the breakdown of organic matter, nitrogen fixation, secretion of growth substances, increase of the availability of mineral nutrients, and immobilization of those assimilable) of great relevance to plant growth; they also cause plant disease or protect the plant from pathogens. The extent of microbial activity depends, in most cases, on the supply of organic substrates from the root. Hence, the abundance and activity of soil microorganisms in general diminish with increasing distance from the root (Newman, 1979). From the point of view of their relationships with the plant, microorganisms can be classified into three groups: (1) saprophytes, usually opportunists but benefactors in some situations; (2) parasitic syrnbionts or pathogens, potentially harmful to the plant; and (3) mutualistic symbionts, usually called symbionts in the literature, which develop activities beneficial to plant growth (for reviews, see Brown, 1975; Dommergues, 1978; Newman, 1979). It is widely assumed that one of the most beneficial contributions of soil microorganisms to plant development is the supply of nutrients essential to plant growth, particularly those involved in nitrogen (N) and phosphorus (P) cycling. Among these, the organisms concerned with N fixation and the enhancement of P uptake by the plant are especially relevant. As it is well known, N and P are two major elements in plant nutrition that commonly limit plant growth; thus, they are usually added to soil as industrial fertilizers. However, in addition to the energy-intensive technology, implied in the synthesis of chemical fertilizers, most of these compounds are lost when they are added to the soil because they are not readily used by the plant. Actually, no more than 30% of the N fertilizer (Postgate and Hill, 1979) and only about 25% of the P fertilizer (Hayman, 1975a) are taken up by the crop in the year of its application. The rest of the N is lost either in the soil water, causing pollution problems (Bolin and Arrhenius, 1977), or to the atmosphere as a result of denitrification; most of the P fertilizer added is quickly fixed by some soil components and converted into forms which are not readily available to plants. Consequently, N fixation, which cycles N to the biosphere from the atmosphere, is an important factor in biological productivity; it is accepted that more than 60% of the N input to the plant community through fixation has a biological origin (Postgate and Hill, 1979; Brill, 1979). The activities of the N-fixing bacteria either convert N into bacterial proteins (in free-living systems) or make it directly available to plants as NH, in symbiotic associations which occur in root nodules. Many common soil microorganisms can release soluble phosphate from sparingly soluble inorganic and/or organic phosphates known to occur in soil. Several problems inherent with the lack of energy sources in the rhizosphere, micro-

MYCORRHIZAS IN NODULATING N-FIXING PLANTS

3

bial antangonism, difficulties in the translocation of the phosphate ions to the absorption places at root surface, and other factors make the microbial solubilization of phosphates a minor contribution to the P nutrition of plants (Hayman, 1975a). However, mycorrhizas, mutualistic symbioses between plant roots and certain soil fungi, play an unquestionable role in P cycling and in the uptake of phosphate by the plant. Because the known world reserves of P could be depleted in a few decades (Rhodes, 1980), the contribution of this symbiosis to the reduction of fertilizer requirements is of increasing interest. B. MYCORRHUAS AND ROOT NODULES

All but a few vascular plants are able to form mycorrhizas. Under natural circumstances, the mycorrhizal condition is the norm for most of the higher plants. The mycorrhizal fungus has an ecologically protected niche inside the plant root; the products of photosynthesis arrive here, furnishing abundant energetic substrate for the fungi which by means of their network of hyphae or mycelial strands extend the mycelium to the surrounding soil, take up nutrients (mainly phosphate) from the soil solution, and translocate these ions to the host plant (Tinker, 1975; Hayman, 1978). Mycorrhizas therefore have a worldwide recognized value for plant survival and nutrient cycling in the ecosystem. They contribute significantly to plant productivity both in arable and in plantation crops. Several types of mycorrhizas occur; their characteristics will be described later. Three different types of microorganisms (bacteria) are able to induce nodules on roots of higher plants and to inhabit them by establishing mutualistic symbioses. As a consequence of these associations, the microsymbionts are able to fix N. The energy requirements for these processes are satisfied by the photosynthate which is directly received by the bacteria at the plant roots (Hardy and Havelka, 1976). The microorganism exports NH4+ to the plant, avoiding transport and dispersal problems. The bacterial genera and the corresponding host plants involved are (1) Rhizobium, which nodulates, with one exception, on legume roots; (2) Frunkiu, actinomycetes that fix N in nodules they form on nonlegume, often woody, angiosperms; and (3) Nosfoc and Anubuenu, cyanobacteria (formerly blue-green algae), which form N-fixing nodules on the roots of plants of the family Cycadaceae (gymnosperms). Legume-Rhizobium sp. associations are the most important for the incorporation of N into pasture and agricultural ecosystems, whereas the nodulated angiosperms are similarly important in forest ecosystems. Plants bearing N-fixing nodules are usually mycorrhizal when grown in soil. This fact has great ecological relevance because nodulation and nitrogen fixation depend on a balanced mineral nutrition of the host plant (in particular, plants

4

J.

M.BAREA AND C. AZC6N-AGUILAR

have high phosphate requirements), and the mycorrhiza can satisfy these demands. Thus, mycorrhizal fungi not only help the plant itself but also aid the bacterial symbiont to fix N in the nodular tissues. Nodulate and mycorrhizal plants are therefore adapted to cope with nutrient-deficient situations (Harley, 1973).

The intent of this article is the comprehensive study of the role of mycorrhizas in the growth and nutrition of N-fixing nodulated plants. As an introduction for a better understanding of mycorrhizal effects, we will present a brief review of some general, well-established principles on mycorrhizal types, morphology, physiology, and function. Current information will be condensed to achieve an up-to-date presentation of this sllbject and to create a conceptual background for nonspecialist readers. This will constitute a quantitatively and qualitatively important part of the article. Then, the interactions between nodular and mycorrhizal endophytes related to the formation and effects of these dual symbioses, which greatly enhance the development of the common host plant, will be discussed. This part of the article will be concerned not only with conceptual principles but also with the rationally stated hypotheses and the current trends in basic and applied research on this subject. Attention will be given to the ecological significance of plants bearing the two types of symbioses, with emphasis on the possibilities of harnessing them to increase crop yield.

II. MYCORRHIZAS A. GENERAL

The previous statements on the concept and function of mycorrhizas, although concise, may allow us to envisage these widespread associations as the most metabolically active parts of the absorbing organs of almost all land plants. Both the autotrophic host plant and the heterotrophic fungal associate derive, in most cases, physiological and ecological benefits from one another. Furthermore, the “mycorrhiza-dependent’’ plants cannot develop adequately without their mycorrhizal partner. However, the general term mycorrhiza, broadly considered, is of little significance. The taxonomic diversity in the fungi and plants involved and the differences in the morphological, structural, and nutritional features of mycorrhizal associations require a subdivision to reflect the different physiological relationships that are now recognized.

MYCORRHIZAS IN NODULATING N-FIXING PLANTS

5

I . Mycorrhizal Types and Their Structural and Nutritional Features Five types of mycorrhizas can be recognized. These and the main groups of host plants on whose roots they are formed are recorded in Table I, as summarized from Smith (1980) and Azc6n-Aguilar and Barea (1980). The first type, ectotrophic mycorrhizas (ECM),is characterized by a lack of intracellular penetration of the fungus into the cortical cells of the root. A network of fungal mycelia, the Hartig net, is formed by hyphal growth among the host cells. This in turn establishs a close contact between fungus and root-cell plasmalemma, which is critical for nutrient exchange in mycorrhizal associations. In most cases the fungus will develop a mantle or sheath of interwoven hyphae growing around the feeder roots. The fungal mantle is extended some distance into the surrounding soil by mycelial strands or rhizomorphs (only rarely by extramatrical hyphae) (Harley, 1978). The fungi involved are mostly higher basidiomycetes (Boletus, Suillus, Amanita, Lactarius, Tricholoma, Pisolithus, Scleroderma, Rhizopogon, etc.), some ascomycetes (Tuber), and zygomycetes (Mam and Krupa, 1978). The second group, vesicular-arbuscular mycorrhizas (VAM), is by far the most widespread type of mycorrhiza. The nomenclature refers to the formation of vesicles and arbuscules, typical morphological structures that will be considered later. As with ericoid, arbutoid, and orchidaceous mycorrhizas, the VA fungus penetrates into the cortical host cells, but the invading mycelium usually lives only a short time intracellularly (Smith, 1980); lysis of intracellular struc-

Table I Mycorrhizal Types and the Main Groups of Host Plants Involved Nomenclature Traditional

Actual

Ectotrophic

Ectotrophic or sheathing

Endotrophic

Vesicular-arbuscular

Ericoid Arbutoid Orchidaceous

Typical host plants Pinaceae, Fagaceae, Betulaceae,*Eucalyptus, Rosaceae,a Leguminosae" (woody), Cupressaceae Four-fifths of all land plants including agronomically important crops such as woody and herbaceous legumes" (pasture, forage, and grain) and Gramineae Calluna, Vaccinium, Erica, Epacris Arbutus, Monotropa Orchidaceae

"Groups of plants also bearing nitrogen-fixing root nodules.

6

J. M. BAREA AND

c. AZCON-AGUILAR

tures (the arbuscules in VAM) then occurs, but the host cell survives and can be colonized again by the fungus. Vesicular-arbuscular fungi do not form sheaths around the root, but a network of extramatrical hyphae usually develops. This grows into the soil and can extend the mycelium several centimeters beyond the root surface. The total hyphal length can reach more than 1 m of hyphae per centimeter of infected root (see Smith, 1980; Hayman, 1982). These VA fungi are members of the family Endogonaceae that are placed in the genera Glomus, Sclerocystis, Gigaspora, and Acaulospora (Gerdemann and Trappe, 1974). Because they cannot be successfully subcultured axenically, they must be considered ecologically obligate symbionts (i.e., they do not complete their life cycle unless they can colonize a suitable host plant) (Lewis, 1973). An ascomycete (Pezizella ericae) has proved to be a fungal partner of the third type of mycorrhiza, namely, the ericoid, which occurs on roots of some autotrophic shrubs in the families Ericaceae, Epacridaceae, and Empetraceae (Read and Stribley, 1975; Read, 1983). Intracellular coils and extramatrical hyphae are typical structures of these mycorrhizas. The structure of arbutoid mycorrhizas, the fourth type, is characterized by the formation of a sheath but not a Hartig net, and they also form intracellular haustoria. Their nutritional features are not yet fully understood. Confined to the family Orchidaceae, the fifth group of mycorrhizas shows unique characteristics; they infect protocorms and rhizomes, but rarely the terrestrial roots. Their hosts are temporarily or permanently achlorophyllic, and the mycorrhizal fungi (Rhizoctonia spp. and Armillaria melea), which are pathogens for nonorchidaceous hosts, aid the heterotrophic orchid in assimilation of carbohydrates, probably from a simultaneous association with another true autotrophic host plant (Mosse, 1978). The major types of mycorrhizas and the groups of plants on which they occur having been described, the discussion may now be limited to ECM and VAM, the only mycorrhizal types formed on plant families also bearing N-fixing root nodules (Table I). Emphasis will be placed on VA mycorrhizas because these are the commonest type occurring on nodulated plants and also because these mycorrhizas, as deduced from their near omnipresence, play an integral role in most crop-production systems. 2 . Occurrence and Distribution Mycorrhizas, mainly VAM, can be found in most plant species growing in most plant habitats under tropical, temperate, and even arctic conditions (Hayman, 1982). To understand the worldwide distribution and ecological implications of this symbiosis, it is interesting to go back 400 million years and consider the role played by a fairly similar mutualistic association-the “ancestral mycorrhiza”-in the evolution of terrestrial plants (Pirozynski and Malloch, 1975). As

MYCORRHIZAS IN NODULATING N-FIXING PLANTS

7

pointed out by these authors and by Malloch et al. (1980), the SilurianDevonian colonization of the land by “plants” seems to have been facilitated by the development of a mutualistic partnership between a semiaquatic ancestral green alga and a certain aquatic fungus. The ancestral mycorrhiza probably equipped the plants to cope with the problems of starvation and desiccation resulting from the colonization of a nonaquatic habitat, the soil. The earliest of the land plants preserved in a petrified form is the Rhynie fossil dated to 370 million years ago, and this possessed in its “roots” a form of mycorrhiza remarkably similar to the modern VAM (Nicolson, 1975). In this context it can be assumed that mycotrophy (Lewis, 1973) and mycorrhizas are as old as plants that seem to have depended on such mycorrhizas to thrive early in their evolution. The symbiosis followed the course of evolution as a component of the plants and, in such a way, it has been perpetuated as an adaptation for the more efficient absorption of phosphorus. The claim of Pirozynski and Malloch (1975) is that “land plants never had any independence (from mycorrhizal fungi), for if they had, they could never have colonized the land.” On these bases it can be concluded that mycorrhizas have occupied, from the Middle Cretaceous on, a crucial role in the evolution, ecology, growth, and nutrition of the plant cover of the surface of the Earth. They can be found in tropical rain forests, open woodlands, grasslands, savannas, heaths, sand dunes, and other habitats (Safii, 1980; Hayman, 1982). In spite of some descriptions (Sondergaard and Laegaard, 1977), plants growing in aquatic habitats appear to lack mycorrhizas, and they also seem to be rare in the families Cruciferae, Polygonaceae, Chenopodiaceae, Cyperaceae, and others (Gerdemann, 1975). According to Meyer (1973), only about 3% of the higher plants have sheathing mycorrhizas (see Table I). These occur mostly in temperate timber trees (Marx and Krupa, 1978; Fogel, 1980; Trappe, 1981). Vesicular-arbuscular mycorrhizas are formed in most angiosperms, some gymnosperms, and in pteridophytes and briophytes. These are the mycorrhizas of most of the economically important crops [e.g., legumes, maize, wheat, barley, rice, temperate fruit trees, many tropical timber trees, woody shrubs, tropical plantation crops (cocoa, coffea, tea, rubber, etc.), cotton, tobacco, olive, citrus, and grapevine] (Hayman, 1982). Finally, some plants may form both sheathing and VA mycorrhizas [e.g., apple, oak, alder, hazel, juniper, certain woody legumes, and members of the family Populaceae (Trappe, 1977), and members of the genus Eucalyptus (Malajczuk et al., 1981)l. 3 . The Process of Mycorrhiza Formation

The establishment of mycorrhizal status occurs in a sequence of phases involving interactions between the host, fungus, and environment. Because ECM can

8

J. M. BAREA AND C. AZC6N-AGUILAR

be synthesized in vitro, the details of the process of their formation can be accurately studied. Nylund and Unestam (1982) have well illustrated the interactions that are occurring. Taking into consideration many previous descriptions, these authors extrapolate the findings to provide a generalizable sequence of events. According to them, the process is controlled mainly by the host, but this does not preclude an active participation of the fungus (i.e., different mycorrhizal structurescan be originated in the same host by different endophytes). The process is initiated by the germination and development of propagules (spores or hyphae) of the fungi living in proximity to the feeder roots of the host. The host releases certain substances that produce a remote and selective stimulation of the tentative mycosymbionts. This enhances the growth of these fungi to an extent that is dependent on the species. Only mycorrhizal fungi have the ability to respond, because only they recognize the host signal that is meaningless to the other rhizosphere inhabitants. Fungal growth is stimulated, hyphae aggregate around the root establishing a close contact between both mycorrhizal partners, and a hyphal envelope forms, induced by host substances. This envelope structure apparently is essential for the infection. The penetration between root epidermal cells seems to be mechanical, and the host apparently does not resist although it controls the fungal lytic enzymes. During further development of the mycorrhiza, the fungus is more protagonistic and more interactions occur. At the labyrinthine Hartig net formation phase, the morphogenetic changes of the fungus are again a response to host factors although released by fungal induction. The formation of the mantle then takes place, the fungus induces some morphological changes in the host, and the colonization of all suitable root tissue by the fungus completes the mycorrhiza. The mycelial strands, sclerotia, and fruiting structures are developed later (PichC and Fortin, 1982). Because VA fungi have not yet been successfully cultured axenically, studies on the development of VA infections are difficult. Some hyphal growth can be obtained in vitro from germinated spores or infected root pieces, but the growth ceases when the hyphae are excised from the parent spore or when the root piece dies. It is therefore said that these fungi do not grow saprophytically. Nevertheless, the assays carried out by Hepper (1979) on the germination and growth of surface-sterilized Glomus sp. spores indicated that the protein synthesis required for germination is programmed by stored mRNA, but that the spores also have the ability to synthesize the new mRNA that is required for hyphal growth. The fungi possess an Embden- Meyerhof-Pamas system, a tricarboxylic acid cycle, and a hexose monophosphate shunt (MacDonald and Lewis, 1978). These results suggest that the endophytes resemble saprophytic fungi more than obligate biotrophs. Moreover, a certain independent spread of VAM fungi in soil has been reported (Warner and Mosse, 1980), suggesting some saprophytic ability of these fungi.

MYCORRHIZAS IN NODULATING N-FIXING PLANTS

9

Studies on the development of VAM infection (Powell, 1976a) showed that neither spore germination nor the initial direction of hyphal growth was influenced by the presence of host roots. Hyphae from spores were not attracted to the roots until they approached them closely. First, the stimulated germ tube formed a fanlike structure of mainly septate hyphae, from which the infective aseptate ones developed later. Because hyphae from root segments did not form preinfection structures to infect a new root, it was suggested that these fanlike structures have the function of absorbing nutrients or hormones from root exudates. Studies of mycorrhizal infection in root organ cultures (Mosse and Hepper, 1975) also indicated the lack of apparent attraction of germ tubes to the root until they grow very close to it. Once a fungal hyphae is attached to the root surface, root penetration may or may not occur. Young lateral roots seemed to exert a greater stimulatory effect on the fungus. In summarizing the previous statements we have four key facts in VAM formation: (1) spore germination and mycelia development; (2) a stimulation of the germ tubes when they approach the roots closely; (3) attachment of the infective hyphae to the root surface; and (4) root penetration. With respect to the stimulation of the hyphae in the rhizosphere, it is obvious that root exudates are significant, and a positive correlation between VAM infection and the degree of root exudation, which in turn is correlated with an increased permeability of root membranes, has been found (Ratnayake et al., 1978; Azc6n and Ocampo, 1981). However, there is another important factor that distinguishes rhizosphere from nonrhizosphere soil, namely, the presence of active populations of microorganisms. They probably play a role in the development of VA fungi and VAM infection. This is supported by studies on the influence of free-living rhizosphere microorganisms on VA fungi in pure culture. The preliminary results have shown that several fungi stimulate the “growth” of Glomus mosseae in culture. The rate of spore germination, the length of the hyphae, and the number of vegetative spores per resting spore were increased by the action of common rhizosphere inhabitants (C. Azc6n-Aguilar and J. M. Barea, unpublished). Once the infective hypha arrives at the root surface an appresorium is usually produced on cortical cells or on root hairs, and hyphal penetration occurs into or between these cells. When the first successful entry point is established, the root becomes more prone to further penetration. This behavior could be because the fungus is invigorated and/or because of changes in the root as a consequence of the infection. The fungus then colonizes the root cortex and the hyphae multiply both inter- and intracellularly, although they never invade the endodermis, stele, or root meristems. Shortly after infection, a hypha growing into a single cell may show repeated dichotomic branching, and a treelike structure, the arbuscule, is formed. The function of the arbuscules is the biotrophic bidirectional transfer of nutrients, the mechanism of which requires living fungus (Cox and Tinker, 1976). Fine-struc-

10

J. M.BAREA AND C.AZC6N-AGUILAR

ture studies (Cox and Sanders, 1974; Scannerini and Bonfante, 1983; Scannerini et al., 1975; Dexheimer et al., 1979) revealed that the arbuscules are surrounded by the intact host-cell plasmalemma. The cytological changes that occur during arbuscule formation have been well documented by Rhodes and Gerdemann (1980). There is an increase in host-cell cytoplasm, the starch within the invaded cells disappears, and the nuclei become enlarged and at times divide. The cell organelles (mitochondria, ribosomes, etc.) also increase in number. When an individual arbuscule degenerates (they exist for 4-13 days), the cell and its structures return to their normal stage. This cell is then ready for the formation of a new arbuscule (Hayman, 1982). When the mycorrhiza is well established, the fungi may form vesicles. These are oval-to-spherical structures containing oil droplets that can develop inter- or intracellularly. They may have a temporary storage function, after which they remain thin walled or become thick walled as chlamydospore-like structures. When the internal infection has been consolidated, the penetration hyphae ramify externally. These external hyphae may grow along the root surface forming more penetration points and also grow through the surrounding soil forming an extensive tridimensional network of mycelium. A typical feature of the VAM is the dimorphic nature of the external hyphae: the coarse, thick-walled (20-30 pm in diameter) hyphae bearing resting spores are the permanent basis of the mycelium, and the fine, thin-walled hyphae (2-7 pm) are more ephemeral and have absorption functions. The density, geometry, and size of the external mycelium, and the number of entry points per unit of root length (1-25 per millimeter), are of great relevance in the functioning of VAM. When the mycorrhiza matures the external mycelium usually produces large resting spores and smaller secondary spores, or external vesicles. Some VA fungal species do not form spores, and some develop sporocarps (Gerdemann, 1975). B. PHYSIOLOGY OF MYCORRHIZAS

Current literature on mycorrhizal research records progress toward a better understanding of many physiological features of these symbioses, particularly of the mechanisms that account for the mycorrhizal effects on plant growth and nutrition. Some of these mechanisms, however, remain unexplained or poorly understood. The review by Smith (1980) is a detailed and illustrating study on this subject. Her qualitative model of the interactions between fungus, host, and environment is a comprehensive summary of the available information relating the flow of materials.and the feedback controls in mycorrhizal associations. This review and those by Tinker (1978, 1980), Hayman (1975a, 1978, 1982, 1983), Bowen and Bevege (1976), Mosse (1978), Rhodes and Gerdemann (1980), Safir (1980), and Gianinazzi-Pearson and Gianinazzi (1981) thoroughly cover the published knowledge on the nutritional relationships between the mycorrhizal partners. Hence only the main points will be summarized here.

MYCORRHEAS IN NODULATING N-FIXING PLANTS

11

1 . Effect of Mycorrhizas on Plant Growth There is considerable published information indicating that mycorrhizas enhance plant growth, which can be understood to be the result of an improved mineral nutrition of the host plant, for which evidence has been provided using isotopic tracers. An increased concentration and/or content of phosphorus in plants is by far the response to mycorrhizas most often described. However, the concentration and content of other nutrients can also increase (sometimes as a consequence of a better P uptake), and there might also be some nonnutritional effects. Mycorrhizas not only increase plant biomass but also influence the partitioning of this material between shoot and root. The enhanced nutrient uptake and the subsequent translocation to the aerial part of the plant increases the utilization of photosynthate in the shoot, hence relatively fewer photosynthesis products are transferred to the root. Consequently, the root/shoot ratio is usually lower in mycorrhizal plants than in the corresponding nonmycorrhizal controls (Smith, 1980). A change in the hormonal status as induced by mycorrhizal infection also can be involved (Allen et al., 1980, 1982). In some instances adverse effects on plant growth in response to VAM have been found (Smith, 1980;Buwalda and Goh, 1982). In most cases the depression is merely transitory and caused by a fungus-plant competition for available photosynthate at the early infection stages or under suboptimal photosynthetic conditions (e.g., shading or low temperatures). Persistent depressions take place when supraoptimal P concentrations are reached in the plant tissues or when the soil phosphate concentration is such that fungus maintenance becomes expensive. 2 . Eflect of Mycorrhizas on Phosphorous Nutrition a. Source of Phosphorus for Mycorrhizas. Because approximately 95-99% of soil P occurs in forms that are not directly available to plant roots (Bieleski, 1973), the possibility that the fungus could solubilize unavailable P was a tantalizing hypothesis to explain the mechanism for the increased P supply by mycorrhizas. In addition, certain sparingly soluble P compounds seemed to be utilized by VAM as a source of P. This possibility was investigated by experiments in which the labile phosphate pool was labeled with 32P (Sanders and Tinker, 1971; Mosse and Hayman, 1971; Powell, 1975). The specific activity (32P/31P)of P in plant tissues was similar for mycorrhizal and nonmycorrhizal plants although the former take up more phosphate. If mycorrhizal plants did utilize nonlabile (unlabeled) sources of P, the specific activity in these plants would be expected to be lower than in nonmycorrhizal controls. These facts show that the plants used the same soluble phosphate pool irrespective of whether they were mycorrhizal. Consequently, it is widely assumed that mycor-

12

J. M.BAREA AND C. AZC6N-AGUILAR

rhizal plants draw most of the phosphate from the soluble pool, although more efficiently than nonmycorrhizal plants. This is in agreement with the finding that VA infection does not modify the root surface-bound phosphatase activity. Moreover, there is not an increase in the exudation of hydroxyacids that could solubilize phosphate either by their chelating activity or merely by reducing the pH in the mycorrhizosphere. The apparent solubilization of poorly soluble sources of P, such as rock phosphate, can be attributed to the greater contact between the network of external hyphae and the surfaces of phosphate particles in soil where phosphate is being physicochemically or biologically dissociated (see Hayman 1975a, 1978). However, further research on this subject is needed to clarify some of the points previously discussed. b. Phosphate Uptake by External Mycelium. The rate of nutrient absorption by roots or mycorrhizas is known to depend on the rate of nutrient supply to the rhizosphere, this being influenced by the mobility of the ion and its concentration in the soil solution. These facts are of great relevance in P nutrition (Chapin, 1980). Phosphate ions, which are in low concentration in the labile pool (Bieleski, 1973), move by diffusion very slowly because they are readily adsorbed to the soil colloids. Plants take up phosphate much faster than these ions can diffuse to the root surface; consequently, phosphate-depleted zones normally develop around the absorbing organs of the plant. These zones, which are 1-2 mm wide, coincide with the rhizosphere and can be visualized by autoradiography (Owusu-Bennoah and Wild, 1979). Vesicular-arbuscular mycorrhizas enhance P uptake in two different ways. One mode of fungal action is merely physical and is based on the increased number of sites for absorption achieved by the external mycelium. The hyphae growing through soil pore spaces are able to effect phosphate absorption beyond the depletion zone up to 8 cm from the root (Rhodes and Gerdemann, 1975). Thus, mycorrhizal roots explore a much greater volume of soil to take up phosphate. A correlation has been found between the size of the external mycelium and the flux of phosphate into mycorrhizal roots (Sanders et al., 1977). Obviously, once inside the hyphae, phosphate ions are protected against absorption by soil components. On the other hand, the kinetic analyses carried out by Cress et al. (1979) demonstrated that mycorrhizas have a lower apparent Michaelis constant (K,) of phosphate uptake than nonmycorrhizal roots, suggesting as the second mode of fungal action the existence of a pathway of greater affinity for P in mycorrhizal roots. This reinforces the results of Mosse ef al. (1973) which suggest that VAM reduce the threshold value for effective phosphate absorption from soil. In spite of the activity of surface phosphatases in ECM, the bulk of P gained by them also comes from the labile pool, and in this case the mycelial strands, which may reach 12 cm in length, are responsible for the increased absorption of phosphate (Bowen, 1973; Harley, 1978).

MYCORRHIZAS IN NODULATING N-FIXING PLANTS

13

c. Translocation of Phosphate from External to Internal Mycelia. There is now experimental evidence to show that P is translocated to internal fungal structures as polyphosphate granules contained inside vacuoles (Ling-Lee et. al., 1975; Cox and Tinker, 1976; Callow et al., 1978). These are propelled through the hyphal lumen by cytoplasmic streaming to the arbuscules, although bulk flow may also contribute to the translocation (Cooper and Tinker, 1981). The specific mechanism for polyphosphate loading, translocation, and unloading is active and very efficient, and the calculated inflow of phosphate through external hyphae is approximately 1000-fold faster than the phosphate diffusion rate through soil (Bieleski, 1973). Mycorrhiza-specific phosphatase activity has been described in VA infections (for reviews see Gianinazzi-Pearson and Gianinazzi, 1981; Capaccio and Callow, 1982), suggesting that these phosphatases may play a key role in the active phosphate translocation and/or transfer mechanisms in VAM. d. Phosphate Transferfrom Fungus to Host. The main site of phosphate transfer to the host, which occurs by an active mechanism across the membrane of both partners, seems to be the arbuscule (Cox and Tinker, 1976; Cox et al., 1980). This hypothesis is strengthened by the finding that the plasmalemmabound ATPase activity of the host is concentrated around the arbuscules when the VAM infection develops (Marx et al., 1982). It is now accepted that the breakdown of the arbuscules can account for only 1% of the P inflow to the host cells. Phosphate release by other structures such as hyphae or vesicles might also be involved, but the extensive increase of contact surface area makes the arbuscules the more probable sites for nutrient transfer between mycorrhizal symbionts. In ECM, the host tissues are sealed off from the soil when the fungal mantle is well developed (Nylund and Unestam, 1982); all transport between soil and host therefore must pass through the fungal mycelium. The nutrient transfer between fungus and host takes place across the hyphae of the Hartig net. A characteristic feature for the functioning of ECM is the storage of phosphate in the sheath and its slow release into host cells. Polyphosphate granules contribute to the storage pool in these mycorrhizas (Chilvers and Harley, 1980; Strullu, 1982). 3. Absorption of Other Nutrients by Mycorrhizas

It has often been reported that mycorrhizal infection also increases the concentration of nutrients other than phosphorus in plant tissues, but it is unclear if this enhancement of nutrient uptake is merely a consequence of improved P supply. The general pattern of uptake, translocation, and transfer of a nutrient into host cells is similar to that for phosphate, but the role and the relative contribution of the external hyphae will differ with the nutrient involved. Mycorrhizal hyphae will help the plant to overcome uptake limitations in the case of nutrients that diffuse slowly and give way to depleted zones around roots.

14

J. M.BAREA AND C. AZC6N-AGUILAR

Conversely, mycorrhizas will confer little additional advantage for the uptake of ions such as sulfate which may move through soil by mass flow with no ratelimiting step in their way to the root surface. Increased N concentrations in VAM plants have been reported, and because zones of depletion in nitrate are not usually formed, mycorrhizas would improve uptake only when ammonium, which is relatively immobile, is the source of N (Smith, 1980). Ectotrophic mycorrhizas, as well as ericoids, have the ability to take N from organic matter (Raven et al., 1978). A direct effect of VAM in improving Zn and Cu uptake has also been found (Bodes and Gerdemann, 1980). Some rather limited hyphal translocation of S can account for some increases in S uptake by mycorrhizal roots. Mycorrhizal infection may also decrease resistance to water transport; thus mycorrhizal plants recover faster from water stress than do nonmycorrhizal ones. However, the hyphal translocation of water has not been fully demonstrated yet (Safir, 1981). Hence, the role of VAM in relation to drought is actually a topic with many unclear aspects (see Allen and Boosalis, 1983; Levy et al., 1983). 4 . Nonnutritionul Effects of Mycorrhizas

Mycorrhizas can affect plant growth and vigor by mechanisms other than improved host nutrition. The production of substances with hormonal activity is involved in the effects of ECM fungi on root morphology (Slankis, 1974; Ng et al., 1982). The ability to synthesize compounds like auxins, gibberellins, and cytokinins has also been described for VA fungi “growing” in vitro (Barea and Azc6n-Aguilar, 1982b). These substances can alter plant morphology and physiology, and some mycorrhizal effects may be mediated by changes in the hormonal balance. A role of mycorrhizas in improving soil structure also has been suggested (Nicolson, 1960; Sutton and Sheppard, 1976; Koske and Halvorson, 1981). This is significant in eroded soils, sand dunes, and the like; mycorrhizal hyphae can bind soil particles into more stable aggregates, helped by the cementing action of bacterial polysaccharides (Foster and Nicolson, 1981a,b). Mycorrhizal infection can help plants withstand root diseases either by protecting the root system against the pathogen attack or by compensating for root damage (Schenck and Kellan, 1978; Mam and Krupa, 1978; Schonbeck, 1979). These effects are more clearly understood for ectomycorrhizas in which the fungal sheath provides a mechanical barrier to infection by soil-borne root pathogens, and in which the fungi produce antibiotics. The role of VAM in deterring pathogens is much less defined. Mycorrhiza-induced changes in root exudation, especially increase of the arginine content and in thickening cortical root cell walls, can account for the deterence of bacteria, fungi, and nematodes in some

MYCORRHIZAS IN NODULATING N-FIXING PLANTS

15

cases. However, the resistance to diseases can be largely attributed to improved host-plant nutrition (Graham and Menge, 1982). There are also reports indicating that disease symptoms were worsened in mycorrhizal plants. Thus, because of all of these facts the biological control of plant diseases by mycorrhizas is a problem deserving further research. 5. Carbon Nutrition in Mycorrhizas Under normal conditions, heterotrophic mycorrhizal fungi obtain carbon compounds from their autotrophic hosts by means of biotrophic transfer across the living membranes of both partners (Lewis, 1975; Smith, 1980). Direct evidence for such a transfer was obtained by detecting 14C-labeledcompounds in fungal structures associated with plants that photosynthesized with 14C02 (Ho and Trappe, 1973; Bevege et al., 1975; Cox et al., 1975). In ECM, as well as in ericoid, the photosynthate, mainly sucrose, is rapidly converted to the specifically fungal metabolites mannitol and trehalose, and eventually to glycogen. This is the storage sink which can represent a significant photosynthate diversion (Bevege et al., 1975; Harley, 1975). In contrast, the fungus in VAM does not appear to form trehalose or mannitol (Hayman, 1974), and lipids seem to be the alternative sink in these mycorrhizas (Cox et al., 1975; Cooper and Liisel, 1978; Liisel and Cooper, 1979). A large proportion (43.8%) of fatty material was found in the VA mycelium (Cooper and Liisel, 1978) either as deposits (storage sink) or involved in the extensive formation of membranes (growth sink) especially at the arbuscular phase of the infection (reviewed by Smith, 1980). This could be a circumstantial drain of photosynthate, but in general the relative size of the fungal biomass seems small enough to make such a diversion .unlikely (see Tinker, 1978). C. FACTORS AFFECTING MYCORRHUAESTABLISHMENT, DEVELOPMENT, AND FUNCTION

The establishment of successful entry points on host roots, the internal and external development of the fungus, and the resulting plant responses are all dependent on interactions between prevailing fungal, plant, and environmental factors. Vesicular-arbuscular endophytes are present in virtually all soils, but mycorrhizal population levels may differ greatly under various ecological conditions. Indigenous mycorrhizal populations can be diminished by agricultural practices such as heavy fertilization and pesticide treatments. This section reviews these factors and the ways in which they influence mycorrhiza formation and efficiency.

16

J. M. BAREA AND C. AZC6N-AGUILAR

1 . Plant Species: Mycorrhizal Dependency Certain plant species require mycorrhizas to a much greater extent than do others, and this is usually referred to as mycorrhizal dependency, which is “the degree to which a plant is dependent on the mycorrhizal condition to produce its maximum growth or yield, at a given level of soil fertility” (Gerdemann, 1975). In general, plants having rootlet diameters of more than 0.5 mm and lacking root hairs are highly dependent on VAM. Conversely, plants with a dense cover of long root hairs and root systems that have rootlet diameters less than 0.1 mm respond to mycorrhiza only in P-deficient soil (Baylis, 1970; St. John, 1980). However, differences in the relative mycorrhizal dependency between crop species, or even cultivars ( A z c h and Ocarnpo, 1981), are also related to other inherent plant factors (Warner and Mosse, 1982), such as root structure and metabolism and plant growth rates, which could affect the demand for P (Hall, 1975). Legumes, for example, are more mycotrophic than grasses and thus probably will benefit more from mycorrhizas. Consequently, there is a critical effect of VAM on interspecies (legumes-grasses) competition for P (Crush, 1974; Hall, 1978; Haynes, 1980). Ectomycorrhizal hosts usually require their fungal associates in order to thrive (Trappe and Fogel, 1977).

2. Endophyte Species: Specificity Some ectomycorrhizal fungi are relatively restricted in their hosts, although most of them have a broad host range (Marx and Krupa, 1978). Vesiculararbuscular mycorrhizal fungi have very little host specificity and any of them can infect virtually any potential host plant (Mosse, 1978; Hayman, 1982);however, they differ in their effectiveness which appears to be more dependent on the specific soil-plant system they colonize than on the host plant itself. A factor determining its effectiveness in enhancing plant growth is the ability of the fungus to develop a great amount of external mycelium, a characteristic inherent with the fungal endophyte (Mosse, 1972) that is affected by the plant-soil system in question. This parameter, which appears to be independent of the rate of fungal colonization of the root cortex, is positively correlated with the enhancement of plant growth by the VAM fungus (Graham et al., 1982). Another critical factor for the effectiveness of VAM endophytes may be the speed at which they infect the root and start to function (Hayman, 1982). In this context, Abbott and Robson (1981b) found a close correlation between the effectiveness of fungi species and the percentage of root length infected at early stages of plant growth.

MYCORRHIZAS IN NODULATING N-FIXING PLANTS

17

3 . Rate of Photosynthesis and Contributing Factors

Because the heterotrophic partner in mycorrhizal symbioses is supplied with carbohydrates from the photosynthate of the autotrophic host (Harley, 1975), any factor that modifies the photosynthetic products available for distribution would affect mycorrhizal development. Obviously, light is one such factor. High light intensities appear to enhance arbuscule formation (Hayman, 1974) and spore production (Furlan and Fortin, 1977). It seems that the largest plant-growth stimulation by VAM occurred under conditions of light and temperature optimal for the development of the host plant (Hayman, 1974; Daft and El-Giahmi, 1978; Johnson et al., 1982a,b). Light and temperature also influence ECM formation and function (Shemakhanova, 1972; Slankis, 1974; Picht and Fortin, 1982). Changes in soluble sugar levels in the root is the logical consequence of the effect of these factors on the rate of photosynthesis (Bjorkman, 1970; Hayman, 1974; Johnson et al., 1982a,b), and the concentration of these products in root extracts and exudates is closely related to mycorrhizal development (Ratnayake et al., 1978).

4. Soil Conditions In general, ectomycorrhizal fungi are acidophilic (Mam and Krupa, 1978), but there is no correlation between VAM and soil pH (Read et al., 1976). In spite of this it has been shown that soil pH can influence the predominance of a given type of spore. In fact, some species are better adapted for acid soils and others are better adopted for neutral and alkaline soiis (Mosse, 1973a). Mycorrhizal fungi are sensitive to soil moisture status; Redhead (1975) demonstrated that the optimal water supply for plant growth is also suitable for mycorrhizal infection. With obligate aerobes (see Saif, 1981), flooding is detrimental to mycorrhyzal activity. The mycorrhizal effect in saline soils is receiving current attention, but it requires further research (Allen and Cunningham, 1983; Ojale et al., 1983). Soil temperature affects the preinfection stages of mycorrhizal development in that the number of “entry points” increases as the temperature rises from 12 to 25°C (Smith and Bowen, 1979). The level and type of nutrients affect the formation of and the response to mycorrhizas (Smith, 1980; Hepper, 1983). A general principle is that a low-tomoderate soil fertility enhances the degree of mycorrhizal development and plant response. Hence mycorrhizal infection could be excluded in fertile agricultural soil (Mosse, 1978). In particular, the level of plant-available phosphate appears to be a suitable index for predicting a growth response to VAM in a given soil.

18

J. M.BAREA AND C. AZC6N-AGUILAR

5 . Fertilizers Heavy fertilizer applications, whether of .nitrogen or of phosphate, are often detrimental to mycorrhizal development and function, but field responses to these treatments are unpredictable because their effects are dependent on initial soil fertility (Slankis, 1974; Hayman, 1975b; Kruckelmann, 1975). As some VA species are more tolerant to added fertilizers than others, it follows that these soil amendments can change the species composition of VAM fungi in a given field soil. The degree of tolerance and compatibility of a mycorrhizal species to added phosphate is a relevant criterion for selecting suitable endophyte strains for field inoculation (Powell, 1977b; Hayman and Mosse, 1979). Special attention has been paid to the effect of soluble phosphate on mycorrhizal development (Mosse, 1973b). The general conclusion obtained is that P levels in the plant, rather than those in the soil, control the establishment and functioning of mycorrhizas (Sanders, 1975; Azcbn et al., 1978b; Menge et al., 1978; Allen et al., 1981). It has become increasingly apparent that P inhibition of VAM formation is associated with a membrane-mediated decrease in root exudation (Ratnayake et al., 1978). In summary, it appears that a cause-effect relationship exists; the higher the P content in the plant, the lower the soluble carbohydrate content in the roots and exudates and the lower the frequency of entry points (Jasper et al., 1979).

6. Soil Microorganisms

As soil inhabitants, mycorrhizal fungi are probably immersed in the framework of microbial interactions that take place in soil microhabitats (Stotzky, 1972),but our understanding of these events is still incomplete.When the mycelia reach the rhizosphere the interactions between mycorrhizal fungi and other microorganisms are increasingly apparent. The experimental evidence to support the existence of these interactions in the epidemiology of root colonization and mycorrhiza formation and on the mycorrhizal effect on plant growth has been reviewed by Barea and Azcbn-Aguilar (1982a). A range of situations including depression, neutrality, and stimulation of ectomycorrhizal fungi by soil bacteria in the rhizosphere of Pinus radiafa was demonstrated by Bowen and Theodorou (1979). Hence these authors proposed that compatibility with a wide range of soil microorganisms must be taken into account in selecting mycorrhizal fungi. Current literature also indicates that soil microorganisms can stimulate the mycelial growth of VAM fungi and the infection process. The suggested mechanisms include the production of compounds increasing root-cell permeability and the synthesis of plant hormones or vitamins by the soil microbiota, but changes in the physicochemical properties of the microenvironments might be also involved.

MYCORRHIZAS IN NODULATING N-FIXING PLANTS

19

Interactions related to nutrient cycling have been described for nitrogen-fixing bacteria (Rhizobium spp., root-nodulating actinomycetes, Azospirillum spp., Azotobacter spp.) , nitrifying bacteria, phosphate-solubilizing microorganisms, and others. These beneficial microorganisms can be, in turn, stimulated in the ‘‘mycorrhizosphere.” Interactions with root pathogens have already been discussed. On the other hand, the existence of parasites of mycorrhizal fungi has been also described (Daniels and Menge, 1980). These may reduce the inoculum level in a soil or even destroy the external hyphae of the mycorrhiza, hence disrupting nutrient translocation (Rhodes, 1980).

7 . Pesticides Relatively few studies have been concerned with effects of agricultural pesticides on the establishment and efficiency of mycorrhizal associations. Stunting of seedlings of the genus Citrus following methylbromide treatment of the soil was associated with an inhibition of VAM fungi by the fumigant (KleinSchmidt and Gerdemann, 1972). It is now accepted that pesticides, especially fungicides, are detrimental to mycorrhizal development. They reduce the infection and, in some cases, completely eliminate the plant growth stimulation by mycorrhizas (Safir, 1980). The effects of herbicides, insecticides, nematicides, and others on mycorrhizal symbioses is, however, a topic deserving further study because contradictory results have been found in some instances (Hayman, 1982).

8. Other Factors Afecting Mycorrhizas Mycorrhizal establishment and/or function could also be affected by other factors [e.g., the removal of surface soil horizons, as in excavations and mining practices (Rhodes, 1980; Zak and Parkinson, 1982), crop rotation involving nonhost plants (Ocampo and Hayman, 1981; Ocampo et al., 1980)l. D. APPLICATIONS OF MYCORRHIZAL POTENTIAL IN AGRICULTURE, PASTURE,HORTICULTURE, AND FORESTRY

After the recognition that mycorrhizas improve plant growth and nutrition, the next logical step in mycorrhizal research to investigate the possibilities of their practical exploitation by harnessing the advantages of the symbioses. The first point of interest is to attempt to preserve the natural mycorrhizal potential as it exists in the field; the second is to try to improve this potential. Naturally occurring strains of mycorrhizal fungi need, in some cases, to be replaced with more effective mycosymbionts by means of inoculation. The principles that

20

J. M. BAREA AND C. AZC6N-AGUILAR

govern this are derived from the fact that natural selection has usually not led to the dominance of the most effective strains of mycorrhizal fungi in a given area, since “the trend of evolution has been for survival, not high productivity [Bowen, 1980al.” Hence the indigenous symbionts have probably evolved to be compatible and adapted to their environments but, in most cases, they have solved their survival problems at the price of their effectiveness. If indigenous endophytes are efficient but sparse, they must be multiplied on stock plant cultures to obtain a suitable inoculum to maintain an adequate mycorrhizal level. Artificial inoculation with suitable mycorrhizal fungi is receiving increasing attention; it is a common practice in the case of ECM (Mikola, 1970; Trappe, 1977; Molina, 1977; Mam, 1980; Mexal, 1980). It is widely accepted today that the inoculation of tree seedlings with ECM fungi is a crucial step in reforestation programs in regions where these fungi are sparse or inappropriate. The incorporation of V A M inoculation into crop production systems is less well developed but there is increasing interest in this application (Hayman, 1982). Only when the scientific basis of mycorrhizal ecology and biology is completely established will it be possible to exploit mycorrhiza inoculation on a practical scale (Abbott and Robson, 1982; Gianinazzi et al., 1982; Powell, 1982; St. John and Coleman, 1983; Plenchette et al., 1983a,b).

1 . Factors Determining Plant Benefits from Mycorrhizal Inoculation Four major factors can determine the success of the inoculation (Hayman, 1982): the crop species involved, the size and effectiveness of the native mycorrhizal populations, the fertility of the test soil, and the cultivation practices. a. Crop Species. It is important to know which plant species derive more benefit from mycorrhizas (Pope et al., 1983). Obviously, for each plant there will be a level of available phosphate in the soil to which this plant will respond similarly whether or not it is mycorrhizal. Above this level, inoculation is not necessary. b. Test Soil Fertility. Because positive responses are to be expected mainly in low-phosphate soils and where the native endophytes are sparse or inefficient, the number and effectiveness of propagules of the family Endogonaceae and the phosphate status are usually determined in the soil where inoculation is to be tried. Olsen’s method for evaluating the available phosphate is usually used, but other methods are also applied, mainly for acid soils (Mattingly, 1980). Other factors affecting P availability, such as pH, organic matter, texture, and buffering capacity of the soil, should be also considered. c. The Size and Effectiveness of the Native Mycorrhizal Populations. The assessment of the number of VA spores and of the amount of mycorrhizal infection in roots of preceding hosts has been used as a measure of the natural

MYCORRHIZAS IN NODULATING N-FIXING PLANTS

21

infectivity of a soil, but these data, because they do not give accurate information about the viability of the propagules, have not always correlated with the actual infectivity. For this reason, soil-dilution techniques have been developed to assess the most probable number of mycorrhizal propagules (i.e., spores, hyphae, or root fragments present in viable conditions) (Powell, 1980c; Wilson and Trinick, 1983). Nevertheless, it is widely assumed that is necessary not only to determine the natural infectivity of the test soil but also to estimate the effectiveness of native fungi and the advantages of the inoculation (Dodd et al., 1983; Jensen, 1983). The effectiveness of inoculated VA endophytes could be merely a consequence of a more rapid spread of the infection by increasing the inoculum level (Tinker, 1978). This is important because the demand for phosphate is higher at the early stages of plant growth. d. CulrivarionPractices. Finally, if cultivation practices (heavy fertilizer and pesticide applications, topsoil removal, long fallow periods, inadequate crop rotation, etc.), which are known to reduce the natural mycorrhizal potential of a soil have been practiced, the introduction of fresh mycorrhizal inoculum will be essential to restore the biological potential.

2 . Selection of Fungi for Inoculation Purposes As recommended by Hayman (1981), a research approach to select the best possible strains of VA endophytes should include a screening of local isolated endophytes concurrently with some strains requested from international collections. The behavior of these fungi under several phosphate regimens is relevant for selection purposes, but the interactions of the introduced endophytes with the natural soil microbiota, including other mycorrhizal propagules, must be taken into consideration. As these tests are carried out in pots, they cannot always predict the real growth effect in the field. Hence only small-scale field trials can evaluate the significance of mycorrhizal inoculation with selected endophytes (Mosse and Hayman, 1980). The selection of the best ectomycorrhizal fungus for a particular host or habitat is obviously critical also. The major selection criteria are (1) easy isolation; (2) growth rate in pure culture; (3) infectiveness; (4) effects on host growth; ( 5 ) ecological adaptation; (6) interactions with other microorganisms; and (7) host specificity (Molina, 1977). 3 . Production of Inocula

The production of a high-quality pathogen-free inoculum is actually a limiting factor for large-scale inoculation with VA mycorrhizal fungi. As these depend on association with living plant roots, the inocula used now consist of the following:

22

J. M. BAREA AND C. AZCON-AGUILAR

1. Rhizosphere samples taken from stock plant cultures containing mycorrhizal root fragments, hyphae, and spores, either attached to the root or free in the rhizosphere soil. A practical scheme to produce inoculum of this type was proposed by Menge et al. (1977) and has proved to be successful for citrus inoculation. 2. Infected roots. Because clean material is preferable, stock plants must be grown on sand or liquid cultures. The infectivity of this kind of inoculum is suitable (Mosse and Hayman, 1980), and the advantages of this material are that it is cleaner and less bulky than whole-rhizosphere inoculum. The nutrient film culture technique (Elmes and Mosse, 1980; Howeler et al., 1982b) should prove to be the most useful way to obtain inoculum of this type, and the storage, viability, and longevity of the heavy infected clean roots are topics under investigation. 3. Fungal structures. Because pure fungal cultures cannot be obtained axenically, isolated resting structures produced on stock plant cultures have been tried. Resting spores are difficult to detach from the mycelia; thus they cannot be obtained in quantity and, in spite of the fact that they survive better than hyphae from infected root fragments, the spores have a lower infectivity power. These facts limit their efficiency as commercial inocula. Such problems, however, are obviated in the case of Glomus epigaeus, a VA endophyte that forms sporocarps on the soil surface. These sporocarps, which can be easily scraped off, contain many spores (Daniels and Trappe, 1979). Their commercial potential as an inoculum has been evaluated with promising results (Daniels and Menge, 1981). Four sources of ectomycorrhizal inoculum have been used (Molina, 1977): soil inoculum, mycorrhizal nurse seedlings, spores and sporocarps, and pure cultures. The last seem to be the best. In fact, pure cultures of many ECM fungi can be easily obtained for small-scale experiments, nursery beds, containerized seedlings, and other uses (Molina, 1977, 1979; Trappe, 1977). The production of massive quantities that would allow commercialization is receiving increased support, most of it devoted to the production of Pisolothus tinctorius because this fungus possesses many advantages as an inoculum (Rhodes, 1980). 4 . Inoculation Techniques

An important distinction should be made between annual and perennial crops. Annuals must be sown directly and the inocula must be introduced with the seeds or seedlings. Obviously, from the point of view of agronomy, it is crucial to ensure a subsantial early infection, and a strategic placement of the inoculum is essential. Hayman (1982) gives an exhaustive presentation on this subject. The techniques he enumerates and discusses are (a) preinoculated transplants; (b)

MYCORRHIZAS IN NODULATING N-FIXING PLANTS

23

direct incorporation into seed furrows; (c) fluid drilling; (d) seed pelleting; (e) multiseeded pellets; and (f) highly infective soil. Perennial plants are usually preinoculated in containers, nursery beds, or small field plots; when the root systems are heavily mycorrhizal the seedlings can be transplanted. 5. Field Experiments

It is clear that the problems associated with VA inocula limit field experimentation on a practical scale; this would be facilitated by availability of pure fungal cultures. Field experiments have been developed in the meantime with the inocula now available. These studies are usually carried out in microplots which must be carefully selected in order to be representative of larger areas. These microplot assays are valuable because they allow us to establish some basis for a correct and successful inoculation and to predict future results when suitable inocula will be available. The completed field trials of VAM inoculation have been reviewed by Mosse and Hayman (1980), Plenchette (1982), Hayman (1982), Nemec (1983), and Menge (1983). Experiments in unsterile soil, some of them in arable fields under normal cultivation, have produced encouraging results. Among the assays performed in unsterile soil, some were conducted with cereals (Khan, 1975; Saif and Khan, 1977; Owusu-Bennoah and Mosse, 1979; Powell et al., 1980; Clarke and Mosse, 1981), potatoes (Black and Tinker, 1977), apple trees (Plenchette et al., 1981), and onions (Owusu-Bennoah and Mosse, 1979), cassava (Howeler et al., 1982a), chilli (Bagyaraj and Sreeramulu, 1982). Studies using legumes as the test crop will be treated in detail in the next part of this article when we discuss the role of mycorrhizas in the growth, nodulation, and N 'fixation of legumes.

111. MYCORRHIZAS IN LEGUMES A. INTRODUCTION

The largest contribution of biological N fixation to world agriculture is derived from the symbiosis between legumes and species of Rhizobium (Evans and Barber, 1977). The significance of this symbiosis is strengthened by the interest in legumes for food production, forage, green manure, and horticulture. The relevance of the legume-Rhizobium sp. symbiosis in agricultural and marginal environments has focused the attention of scientists from diverse disciplines

24

J. M. BAREA AND C. AZCON-AGUILAR

(plant and microbial physiologists and genetists, plant breeders, and others) on a common objective, namely, the enhancement of the efficiency of symbiotic N faation [see Phillips (1980) for review]. The processes of infection and nodulation of legumes by their specific species of Rhizobium, and the biochemistry, physiology, and genetics of the N fixation, have been well documented and are clearly beyond the scope of this article (Dazzo and Hubbell, 1975; Nutman, 1977; Bergersen, 1978; Broughton, 1978; Casadesds and Olivares, 1978; DCnari6 and Truchet, 1979; Gibson and Newton, 1981; Bauer, 1981). Nevertheless, it is noteworthy that the infection mechanism of a legume root by its appropriate species of Rhizobium, particularly the process of N fixation, has a high energy requirement. The nitrogenase activity is dependent on ATP for the reduction of atmospheric dinitrogen to ammonia: approximately 21 mol of ATP are converted to ADP per mol N, reduced (Shanmugan et al., 1978). This explains why the scarcity of soluble P in soil is a critical limiting factor in the case of legumes, because it not only affects plant growth but also nodulation and N fixation (van Schreven, 1958; Andrew and Robins, 1969; Gates, 1974; Gates and Wilson, 1974). Mineral nutrients other than P, such as Zn, Cu, and Mo, may limit rhizobial growth, nodulation, or symbiotic N fixation (Demeterio et al., 1972; Robson, 1978; Munns and Mosse, 1980; Shukla and Yadav, 1982). Phosphorus and some of these minor elements, of course, may be supplied by mycorrhizal infection (Rhodes and Gerdemann, 1980). Thus, all of these circumstances account for the key role of mycorrhizas in legume production systems. B. OCCURRENCE

The coexistence of a bacterium and a fungus as root endophytes of legumes, establishing a tripartite symbiotic association, was first described by Janse (1896). Jones (1924) and Samuel (1926) found that most of the nodulated legumes they examined were also infected by mycorrhizal fungi of the VA type, but it was Asai (1944) who first reported that the nodulation of several legumes depended on the formation of mycorrhizas. This effect, however, was not attributed to the improvement of P nutrition by mycorrhizas because it was as yet unknown. Actually we know that most of the nodulated legumes so far examined are also mycorrhizal. The majority of these are of the VA type, which develops on both herbaceous and woody legumes, but in the latter the presence of ECM has also been reported (Ross and Harper, 1973; Trappe, 1979; Malloch et al., 1980; Thomazini-Casagrande, 1980; Moiraud et al., 1981; Malajczuk et al., 1981).

Two of the three subfamilies included in the family Leguminosae (i.e., Papilionoideae and Mimosoideae) are known to have VAM and rhizobial nodules.

MYCORRHIZAS IN NODULATING N-FIXING PLANTS

25

The third subfamily, Cesalpinoideae, rarely forms nodules and, in spite of the fact that some members of this subfamily form endomycorrhizas, the two large groups, Amherstieae and Detariae, commonly have ECM. These two groups, with nonnodulated and ectomycorrhizal species, constitute those legumes that diversified first in the evolution of this family. As ECM have the ability to obtain nitrogen from organic sources, nodules and ECM seem to act as alternative means to supply plants with N (Malloch et al., 1980). Indeed, the nodulation of plant roots is an evolutionarily advanced character (Trappe, 1979). Vesicular-arbuscular mycorrhizas have been found in legumes growing in a wide range of habitats (Possingham et al., 1971; Ross and Harper, 1973; Khan, 1974, 1978; Crush, 1975, 1976; El-Giahmi et al., 1976; Sanni, 1976; Thomazini-Casagrande, 1980; Janos, 1980a; Pfeiffer and Bloss, 1980; Schenck and Smith, 1981; Rose, 1981; Diem e t a l . , 1981). The widespread occurrence of VAM in cultivated legumes is clearly demonstrated in the study by Strzemska (1975). The pioneering paper by Asai (1944) already showed the different degrees of mycorrhizal dependency in legume species and pointed out that species of Vicia and especially Lupinus have little dependence on mycotrophy. The case with lupines is noteworthy because these plants can get P from very low phosphate soils, although no more than 10% of their root system becomes mycorrhizal (Trinick, 1977). Lupines are able to induce abnormal VAM infections in clover (Morley and Mosse, 1976), suggesting that a certain type of antagonism toward mycorrhizal fungi may be involved. This relationship, however, remains unclear and mycorrhizal independence can be explained without the implication of fungus toxicants. An illustrative example relating mycorrhizal dependency in legumes to root morphology has been described by Crush (1974). He compared the development of mycorrhizal infection in four legume species and found that Lotus pedunculatus, which has well-developed root hairs, was able to grow well without mycorrhizal inoculation; the tropical legumes Centrosema pubescens and Stylosanthes guyanensis, which form few root hairs, exibit a strong dependence on mycorrhizas; finally, Trijolium repens holds an intermediate position with regard to both root hair production and response to mycorrhizas. C. INTERACTIONSBETWEEN SPECIESOF RHIZOBILIM, MYCORRHIZAL FUNGI, AND LEGUMES

The experiments by Asai (1944), which suggested that mycorrhizal status was a “precondition” for effective growth and nodulation of legumes, were not fully appreciated for a long time. The role of mycorrhizas in the growth, nodulation, and N fixation of legumes has been a subject of increasing interest. The published information on these topics, almost entirely concerned with VAM, is

26

J. M. BAREA AND C. AZC6N-AGUILAR

summarized in this section with emphasis on some aspects of the physiology and biochemistry of these tripartite associations. The implications of these facts of legume ecology and production will be considered later. In the context of these interactions, it should be stated that the formation of VA fungal entry points and nodules on a legume root occurs simultaneously, usually within a few days after seed or seedling inoculation, and it appears that the two endophytes do not compete for infection sites (Smith and Bowen, 1979). Legume nodules are usually not invaded by the VAM fungus (Lanowska, 1966; Crush, 1974; Mosse, 1975; Smith et af., 1979). I . Signijkance of Dual Symbioses in Legume Growth and Nutrition

From the point of view of plant ecology and crop production, legumes are undoubtly a special case because they can be supplied with the two major nutrients, P and N, by naturally existing biological systems. In fact, as a consequence of the simultaneous infection with Rhizobium spp. and mycorrhizal fungi, legumes can receive growth benefits because of improved P and N supplies and also those resulting from the N-P interactions (Munns and Mosse, 1980). The double symbioses in legumes not only reduce the inputs of synthetic fertilizers, thereby saving energy, but they also appear to reduce the cost of the system itself in terms of photosynthate drain (Bevege et al., 1975; Pang and Paul, 1980; Kucey and Paul, 1982). Studying the distribution of I4CO, carbon fixed by Vicia fuba, Kucey and Paul (1982) showed that mycorrhizal fungi utilized -4% of the carbon, whereas nodules used 6% of the carbon fixed by the nonmycorrhizal beans and 12% of that fixed by the mycorrhizal plants. In spite of this, as the rates of CO, fixation for the symbiotic host were higher than for the nonsymbiotic plants, it appears that the host legume compensates for the carbon drained to the endophytes. At early stages of mycorrhizal and rhizobial infection the carbon drain, not yet compensated for, may induce a transitory negative growth response (see Smith, 1980). Mycorrhizal and nodulate plants usually have a lower root/shoot ratio than plants inoculated with either symbiont alone (for examples see Daft and El-Giahmi, 1974; Asimi et af., 1980; Redente and Reeves, 1981). This is a typical response to improved mineral nutrition, as discussed by Smith (1980, 1982) after the experimental testing of the inflow of phosphate into mycorrhizal clover. Moreover, it appears that dual inoculation with a suitable species of Rhizobium and mycorrhizal fungi not only enhances the nutrient content in the aboveground plant material but also seems to provide a nutrient supply that is well balanced. Consequently, the biosynthetic processes taking place in these adequately established legume-Rhizobium sp.-mycorrhiza associations can lead

MYCORRHIZAS IN NODULATING N-FIXING PLANTS

27

to an improvement in the production of seed and/or foliar proteins. This is the beginning of a trophic chain of great relevance in animal alimentation. The role of mycorrhizal infection in legume production was established in studies carried out in the 1970s and early 1980s. These studies using nodulated soybeans were of particular importance for our understanding of this subject. The experimental assays by Ross and Harper (1970) and Ross (1971) demonstrated that mycorrhizal inoculation increased yield, and that mycorrhizal plants contained a higher N concentration and content in their tops than nonmycorrhizal controls. Further, mycorrhizal infection induced an improvement (53%) in the N content in shoots and seeds of nodulated plants, but the effect of mycorrhizal inoculation on a nonnodulating isoline of soybean was not significant Schenck and Hinson (1973). These results reinforced the connection of mycorrhizal effects with an enhancement of nodule function. 2. Mycorrhizal Effect on Nodulation and Nitrogen Fixation The effect of mycorrhizas in stimulating nodulation was reaffirmed in a study (crush, 1974) that also produced more consistent evidence of the mycorrhizal stimulation of N fixation by legume-Rhizobium sp. systems as measured by the acetylene reduction technique. More advanced information on this subject is available (Daft and El-Giahmi, 1974, 1975, 1976). These authors found that several parameters directly related to N-fixation processes in species of Phaseolus, Medicago, and Arachis were affected by mycorrhizal infection. In fact, the amount of nodular tissue, the concentration of legume hemoglobin, and the rates of acetylene reduction were greater in mycorrhizal and nodulated plants than in the nonmycorrhizal but nodulated controls. Using the 15N, tracer technique, Kucey and Paul (1982) confirmed that mycorrhizal and nodulated plants (faba beans) fix more N than those nodulated but nonmycorrhizal. This was attributed to an increase in nodule biomass as induced by mycorrhizal inoculation. As a consequence of these mycorrhizal effects on nodulated legumes, increases in fruit yield, plant growth, and nutrient content of shoots, roots, and seeds were recorded. Early observations on this subject suggested several approaches to the elucidation of physiological aspects of the legume-Rhizobium sp.-mycorrhiza interactions. One of these tried to ascertain whether mycorrhizas enhance symbiotic N fixation only through the stimulation of host-plant nutrition, or whether they also exert a more direct effect on nodulation and nitrogenase activity. The existence of such a direct availability of P to the nodules by mycorrhizal hyphae does not preclude, however, the importance of a suitable P nutrition, as achieved by mycorrhizal inoculation, of the host as a condition for effective symbiotic N fixation. This is not only because of the role of the host as a partner in the association as concerns the expression of the activity (N fixation), but also

28

J. M.BAREA AND

c. AZCON-AGUILAR

because the nodules are actually part of the plant. If the plant is well nourished, the nodules also will receive a suitable P supply for their function. The logical result, therefore, is the existence of a close relationship between host nutritional status and nodule formation. For example, Mosse et ul. (1976) found that plants did not nodulate unless their P concentrations were at least 0.15%; mycorrhizal infection helped the plants to reach this required level, and nodulation then occurred. The conclusions of Abbott and Robson (1977) and the further elaborations of Robson et ul. (1981) also support the idea that the effect of VAM on nodulation and N fixation in subterranean clover closely parallels responses in growth and nutrition. These authors assume that mycorrhizal effects on nodulation take place through host nutrition and that these occur at the same time as the growth responses. In contrast, Smith and Daft (1977) reported that mycorrhiza-induced increases in N fixation rates in Medicago sufivu preceded any effect on plant growth. This suggested the idea that nodules demand phosphate first. At all phosphate additions tested, Smith and Daft (1977) found that mycorrhizal plants had higher values of %N than the nonmycorrhizal controls and mycorrhizal plants of the same size (dry weight and root-to-shoot ratio) higher than that of the nonmycorrhizal controls. Studies comparing matched plants confirmed that VAM increases the rates of N fixation (Smith and Daft, 1978). Moreover, nitrogenase activity in Pueruriu sp. still increased when the growth phosphate response curve became asymptotic (Waidyanatha et ul., 1979). The latter papers therefore support the suggestion that nodule function may be preferentially stimulated by mycorrhizal infection, which makes phosphate directly available to the nodules. Smith et al. (1979) then used time-course experiments to elucidate the development of interactions between the components of these tripartite symbioses, and c o n f i i e d that the mycorrhizal effect on nodulation, nitrogenase activity, and nodule efficiency occurrs before any positive growth response to VAM in a lownutrient soil but not in soil more fertile. This indicates that the “special” demand for P of nodular tissues apparently results when the P supply is a limiting factor, but this is a common situation in most unamended soils. Smith et al. (1979) also reported the mycorrhizal clover roots had a higher P concentration than did the nonmycorrhizal control plants. Because they did not find mycorrhiza-induced increases of P concentration in the nodules, the enhancement of nodule efficiency could be explained by the increases in root P, which accumulated mainly as polyphosphate. These facts seem of great relevance; as stated by these authors “a steady supply of P to root cells, and to adjacent nodules, as such as would be available from continuous polyphosphate conversion at the fungus-root interface, would be stimulating to the development of effective nodules symbiosis and associated N-fixation’’. The stimulation of nitrogenase activity by VAM in the system soybean-

MYCORRHIZAS IN NODULATING N-FIXING PLANTS

29

Rhizobiumjaponicum was also evident before the plant growth responses (Asimi el al., 1980). This also suggests “a particular sensitivity” of Rhizobium to mycorrhizal effect and confirms the role of VAM to satisfy the high P demand for the processes of nodulation and N-fixation. Moreover, increased phosphate additions first eliminated mycorrhizal effects on growth and then, progressively, those on nodulation and nitrogenase activity (Asimi et al., 1980). This gives additional indirect evidence that plant growth and nodule functioning show differential responses to and demand for P, as derived from their different P dependencies. Because there is no contact between the mycorrhizal fungi and Rhizobium bacteroids, any P supply must pass through the host cells. The role played by the VAM-specific phosphatases in the arbuscles developed inside root cells adjacent to nodules was suggested as being of great significance in phosphate transfer to bacteroids (Asimi et al., 1978; 1980). Time-course experiments using 32P and autoradigraphic techniques would be of interest to clearly define the statement that the phosphate apparently is made directly available in some way, to the bacteroids in legume nodules. Another question on the mycorrhizal effects on symbiotic N fixation is whether these are exclusively phosphate-mediated or, conversely, if mycorrhizas have any implications in addition to those resulting from the improved P supply. Carling et al. (1978) found that soluble phosphate can replace VAM in increasing nitrogenase and nitrate reductase activities in nodulated soybeans. Thus they suggest that P is almost exclusively responsible for the mycorrhizal effects and that neither soybean endophyte interacted directly, at least in their experimental conditions. In spite of this, some responses to mycorrhizal inoculation in legumes reported in other situations are difficult to explain on the basis of improved P supply alone (Munns and Mosse, 1980). The increased mycorrhizal uptake of water and elements other than P, which is related to the infectivity and/or effectiveness of legume-Rhizobium sp. systems, may be also involved (Safir et al., 1972; Gray and Gerdemann, 1973; Mojallah and Weed, 1978; Munns and Mosse, 1980). 3. Nonnutritional Interactions

Although the main reasons for the cooperation between VAM fungi and Rhizobium have a nutritional basis, other secondary, nonnutritional effects of the symbiosis have also been suggested (Mosse, 1977b). Further experimental studies seem to support this idea; both types of microorganisms appear to interact, whether at inhabitation of the rhizosphere, at the formation of the symbioses, in the development of these, or even in their effects on plant physiology. As is commonly accepted, soil microorganisms can stimulate the mycelial growth of mycorrhizal fungi and their penetration into susceptible roots of the

30

J. M. BAREA AND C. AZC6N-AGUILAR

infective hyphae. The basis of such interactions and the suggested mechanisms of the microbial activity are the following: a. Production of CompoundsIncreasing Cell Permeability (CIPC). Common rhizosphere microorganisms are able to increase losses of microbial substrates from living roots, and the local production of CIPC by these microorganisms could result from these effects (Bowen, 1980a). These compounds, by relaxing root cell walls, would increase the rates of root exudation, and this could stimulate mycorrhizal fungi in the rhizosphere and facilitate root penetration. The increased leakage of substrates from pine roots caused by species of Bacillus, has been suggested as a cause of the enhancing effect that these bacteria exert on ECM formation (Bowen and Theodorou, 1979). Because extracellular p l ysaccharides (EPS) from Rhizobium spp. behave as CIPC, their role in VAM infection in legumes has been studied, indicating that EPS from Rhizobium meliloti enhance VAM formation on Medicago sativa (Azc6n-Aguilar et al., 1980). The EPS could act in the same way proposed for Rhizobium spp. according to the polygalacturonase hypothesis (Ljunggren and Fahraeus, 1959), hence improving the formation of VAM symbiosis through the establishment of entry points. Alternatively, EPS could act merely by increasing root exudation (Olivares et al., 1977), thereby favoring the development of the preinfection phase of the VAM infection. These results agree with an early observation by Mosse (1 962) indicating that in axenic conditions the fungus failed to penetrate clover roots unless a soil microorganism, a species of Pseudomonas, was also present. This bacterium possesses pectolytic activity, and the author suggested that the most probable explanation is that the bacterial compounds act on the structure of the cell wall, thereby affecting its plasticity. This also can affect the susceptibility of the plant root to fungal infection. b. Production of Plant-Growth-Regulating Substances. Many microorganisms isolated from the rhizosphere, and particularly species of Rhizobium, are able to produce substances with phytohormonal activity. Because this ability seems of interest in rhizosphere biology (Brown, 1975), considerable attention has been paid to its possible influence on mycorrhizal infection. The role of plant hormones (PH), mainly auxins, in the formation of sheathing mycorrhizas is already well established and, to some extent, the stimulation exerted by certain soil microorganisms on ECM infection appears to be through this mechanism (Slankis, 1974). The effect of plant hormones on the formation of VAM in M . sativa has been the subject of some studies. Infection levels in mycorrhizal plants were compared after treatment with pure substances and preparations from cultures of R. meliloti, which are known to contain auxins, giberellins, and cytokinins. Cell-free supernatants from the R. meliloti cultures tested increased VAM infection in M . sativa to an extent similar to that of the pure plant hormones applied in doses similar to those of the supernatants (Azc6n et al., 1978a; Azc6n-Aguilar and Barea, 1978).

MYCOFSHIZAS IN NODULATING N-FIXING PLANTS

31

The morphological and physiological changes that plant hormones can induce in the host plant may favor the establishment of VA symbiosis and its activity, thus leading to a greater rate of nutrient absorption by the plant (Azc6n et al., 1978a). In fact, it is known that gibberellins increase leaf area and the development of lateral roots, that cytokinins are involved in many basic processes of plant growth, including improvement of photosynthetic rate, and that auxins control root formation and increase the elasticity of the cell wall (Torrey, 1976; Thimann, 1977; Tien et al., 1979). All of these activities can affect the formation or effectiveness of VAM. The hormonal interactions in the rhizospheres of legumes seem more complicated as VAM fungi (Barea and Azc6n-Aguilar, 1980, 1982b) are also able to produce PH. These substances can be involved in the mycorrhizal effects. For example, Mosse (1962) found that VAM stimulated branching of infected roots (a typical hormonal effect), and Allen et al. (1980, 1982) demonstrated that VAM infection increases the hormonal level in the host plants. These facts could be important because PH play a role in the infection mechanism of legume roots by Rhizobium spp. (Nutman, 1977). Because mutualistic symbioses involving plants and microorganisms depend, both for their formation and for their function, on a series of interactions between the constituent partners, PH synthesized either by the host or by the endophytes appears to be involved in the establishment and development of these biotrophic associations. Another point of interest for future research is the fact that the exudates of mycorrhizal roots are probably different both quantitatively and qualitatively from those of nonmycorrhizal roots. This will induce changes in the rhizosphere that might affect the development of Rhizobium spp. (Mosse, 1977b).

D. INTERACTIONS BETWEEN ADDEDFERTILJZERS AND MYCORRHIZAS IN LEGUME-Rhizobium Sp. Systems

Although the nonnutritional interactions just discussed may act in some way on the formation and development of the tripartite symbioses, the protagonism of phosphate-mediated mycorrhizal effects on N fixation in legumes is obvious. As can be deduced from the universally accepted role and mode of action of external hyphae in taking up phosphate ions from solution in the soil, mycorrhizas not only enlarge the zones around roots that are depleted of phosphate, but also cause these zones to be more greatly depleted of this plant nutrient. Consequently this produces an impoverishment of the soil after several harvests. The phosphate stock must then be restored, which can be accomplished either by applying soluble phosphate fertilizers or, in some circumstances, by using less expensive, sparingly soluble forms of P. The interactions of these compounds with mycor-

32

J. M. BAREA AND C. AZC6N-AGUILAR

rhizal fungi in the development of legumes have been the subject of several studies of great interest because of their basic and applied implications. 1. Effect of Soluble P Additions

It has been pointed out that mycorrhizal inoculation experiments should include testing of the interactions of VAM fungi at a series of phosphate levels (Abbott and Robson, 1977; Hall, 1978; Powell, 1980a) in order to select the P doses optimal for the mycorrhizal effects. Several greenhouse experiments on the phosphate response curves of mycorrhizal and nonmycorrhizal Rhizobiwn-inoculated legumes were carried out, mostly using clovers (Trifolium sp.) as the test plant (Crush, 1976; Abott and Robson, 1977; Hall et al., 1977; Sparling and Tinker, 1978; Powell, 1980a; Pairunan et al., 1980). These authors applied ranges of soluble phosphate fertilizers at rates equivalent to about 0-300 kg P/ha and have reached a the general conclusion that mycorrhizas markedly increase P uptake, growth, and nodulation in clover at low and intermediate rates of applied P, although plant growth depressions may occur at high levels of available P (Crush, 1976). Large P additions to the soil are known to decrease mycorrhizal infection in several legumes (Abbott and Robson, 1977; Powell and Daniels, 1978; Barea et al., 1980; Pairunan et al., 1980; Asimi et al., 1980; Powell, 1980b; Nielsen and Jensen, 1983; Bethlenfalvay, 1983). This could lead to host immunity to infection. However, certain phosphate additions which reduce the percentage of root lengh infected by VAM do not affect the length of mycorrhizal root per plant. This situation has been observed by Asimi et al. (1980) (soybean) and by Smith (1982) (clover), and it can be explained on the basis of a simultaneous rapid growth of roots at increasing levels of soluble phosphate. From the practical point of view, however, the results of the interrelationships between phosphate additions and VAM on legume-Rhizobium sp. systems are not always predictable and generalizable, because the responses are modulated by the incidence of several factors. These include the characteristics of the soil, the plant species cultivars or lines, the endophyte species involved, and, finally, the interactions between these factors. a. Characteristics ofthe Soil. It is to be expected that the effect of increasing additions of phosphate will depend not only on the doses applied but also on the ability of the test soil to retain a greater or lesser amount of the added phosphate. The period of contact between the applied fertilizer and the soil before planting also influences the P available in a given soil and the resultant mycorrhizal response (see Barrow et al., 1977). Obviously these factors will affect the size of the labile phosphate pool which is known to limit mycorrhizal activity. For legumes, Powell (1980a) demonstrated that the interactions of increasing phosphate additions with mycorrhizal inoculation of white clover depended on the P

MYCORRHIZAS IN NODULATING N-FIXING PLANTS

33

retention capacity of the soil tested. The magnitude of the mycorrhizal effects in stimulating the growth and nodulation of Leucaema leucocephala was also found to vary with the test soil (Munns and Mosse, 1980). Plant response to mycorrhizas in different soils is not always related to soil P content (see Stribley et al., 1980). This could explain why the mycorrhizal infection level in Phaseolus sp. was independent of the level of N and P in the field soils tested (Hayman et a l . , 1976), indicating the importance of the previous fertilizer history of the soil in affecting the development of adaptative strategies of the extant native endophytes against the soil amendments. b. Legume Species. It is well known that nonmycorrhizal plant species, cultivars in species, and even clones in cultivars can differ in their P uptake capacity. This differential behavior is related to the degree of their mycorrhizal dependency and could condition changes in the patterns of the phosphate response curves of these plants, whether or not mycorrhizal. Some studies with legumes have been devoted to this subject. Differences between species have been clearly shown; for example, lotus plants become nodulated independently of mycorrhizal inoculation in soil where clover plants show low nodulation unless mycorrhizal (Crush, 1974). Because lotus needs less P than clover to nodulate (Gibson et al., 1975), these plant species differ in their phosphate response curves (Hart et al., 1981). The combined effects of increasing soil P levels and VAM on plant growth and nodulation were compared in two systems, M . sativa-R. meliloti and Hedysarum coronarium-Rhizobium sp., developing in the same soil. In the case of M . sativa, mycorrhizal plants grew and nodulated significantly better than nonmycorrhizal ones in unamended soil and at intermediate P additions. Higher doses eliminated the mycorrhizal effect on plant growth and nodulation and became supraoptimal for these processes. In the case of H . coronarium, mycorrhizal inoculation also significantly improved plant growth and nodulation at low phosphate additions. However, the response differed from that of M . sativa in that high phosphate levels did not become supraoptimal for growth and nodulation and the response curve to added phosphate became asymptotic. An explanation for the behavior of H . coronarium could be that this legume forms a sort of modified root that accumulates large amounts of Ca (J. Yaiiez, personal communication). These “roots” can then retain the phosphate surplus so that the shoots do not reach P concentrations supraoptimal for growth. Variations in response to VAM have been reported for cultivars of clover (Hall et al., 1977), soybean (Skipper and Smith, 1979), and alfalfa (Lambert et al., 1980a). However, Hall et al. (1977) only found VAM-clover cultivar interactions at the lowest levels of phosphate. The lack of interaction in clovers at higher P additions was also reported by Crush and Caradus (1980). Conversely, Lambert et al. (1980a) even found significant interactions between lines in the alfalfa cultivars, VAM, and the P level.

34

J. M. BAREA AND

c. AZCON-AGUILAR

c. Endophyte Species. Mycorrhizal endophytes are known to differ in their relative efficiences in stimulating growth nodulation and N fixation in legumes. The variation in the efficiency among mycorrhizal fungi has been studied in an attempt to find correlations among several factors including the rate of establishing the mycorrhiza, the number of entry points per unit of root length, the extent of mycorrhizal infection, the degree of enhancement of phosphate recovery from the soil by the plant, and their condition of native or introduced (Abbott and Robson, 1977, 1978, 1981a,b; Mosse, 1977a; Powell, 1980a; O’Bannon et al., 1980; Smith and Smith, 1981b). Although all of these factors are important, there is no close generalizable correlation between any of these parameters and the relative symbiotic efficiency of the endophyte species (Munns and Mosse, 1980). d. Interactions between the Factors Enumerated Previously. Interactions between the prevailing ecological factors affect mycorrhizal activity more than individual factors alone. For instance, the relative efficiency of several VAM fungal species on legume development was demonstrated to depend, in turn, on interactions with the soil and the amount of available P (Powell, 1977a; Mosse, 1977a; Sparling and Tinker, 1978; Powell and Daniels, 1978; Carling and Brown, 1980; Lambert et al., 1980a). Because of the several types of interactions actually operating, it is difficult to extrapolate conclusions from one study to another, and so Carling and Brown (1980) stated that “each system appears to be unique, and each must be evaluated experimentally before the question pertaining to effectivity of a specific VA mycorrhizal fungus in that system can be answered.”

2. EfSect of Rock Phosphate There is some evidence that legumes benefit from VAM in the presence of insoluble phosphates (Ross and Gilliam, 1973). Accordingly, it was suggested that mycorrhizal soybean plants might be able to utilize nonlabile forms of soil phosphate that nonmycorrhizal plants cannot use. However, assays in 32P-labeled soils indicate that mycorrhizal (five different endophytes, which caused a high, but differing, degree of infection in clover roots) and nonmycorrhizal clover plants take up P from the same source, as expected (Powell, 1975). As is known, VAM achieves a better exploitation of the sparingly soluble phosphate because hyphae make a closer contact than roots can with phosphate particles where the soluble ions are being chemically (or biochemically) dissociated. Therefore, the utilization of a nonlabile phosphate source by a mycorrhiza occurs on the condition that, at least slowly, a liberation of some phosphate ions takes place. Undoubtedly, rock phosphate (RP) has been the most utilized source of sparingly soluble fertilizer for the study of restoring soil phosphate. General conclu-

35

MYCORRHIZAS IN NODULATINC N-FIXING PLANTS

sions indicate that in acid soils RP can improve growth, nodulation, and N fixation in nonmycorrhizal legumes, and that inoculation with appropriate VA endophytes greatly enhances its utilization. On the other hand, in near-neutral and alkaline soils RP remains unavailable for both mycorrhizal and nonmycorrhizal legumes (Mosse et al., 1976; Mosse, 1977a,b; Powell and Daniels, 1978; Sparling and Tinker, 1978; Waidyanatha et al., 1979; Delorenzini et al., 1979; Barea et al., 1980; Islam et al., 1980; Munns and Mosse, 1980). These ideas are summarized in Table 11. VAM infection can increase plant growth and nodulation of legumes even when these are growing on neutral and alkaline soils with added RP. This indicates that this kind of phosphate fertilizer might be suitable to maintain the stock of phosphate in a soil; in addition, it does not reduce the level of mycorrhizal infections as does soluble P (Barea et al., 1980). Because VAM plants also take up their P from the plant-available phosphate fraction, the pool of labile P that is restored by chemical dissociation of phosphate ions from RP, it might be speculated that this could be a useful substrate for P uptake even in high pH soils. Furthermore, using a range of nonacidic soils, a situation was found in which clover plants, either Glomus sp.-inoculated plants or the uninoculated controls, were able to use RP. In this case the soil differed in a biological property; the number of phosphate-solubilizing bacteria (PSB) able to dissolve the RP present in it was significantly higher than in the other soils. It is also noteworthy that the number of PSB was stimulated in the root zone of the Glomus sp.-inoculated plants (Barea et al., 1981). A great number of microorganisms can release phosphate ions from sparingly insoluble inorganic and organic phosphate, as deduced from assays carried out in vitro (Greaves and Webley, 1965; Tardieux-Roche, 1966; Barea et al., 1970). However, the effectiveness of these microorganisms either in the untreated soil or when massively inoculated into the soil is doubtful. There are, in fact, some

Table II Interactions between Rock Phosphate (W) and Mycorrhizal Inoculation (VAM)in the Growth and Nodulation of Legumes as Affected by Soil pH Soil reaction0 Treatment

Acid

RP versus no treatment VAM X RF' versus VAM VAM x RP versus RP

+ + +

"+, Significant positive effects on growth and nodulation; -, nonsignificant effects.

Neutral or alkaline

36

J. M. BAREA AND

c. AZC~N-AGUILAR

problems that make an efficient solubilization of phosphates in soil difficult. These difficultiesare inherent in the scarcity of available energy sources and with problems in the translocation to the root surface of any available solubilized phosphate ions (Hayman, 1975a; Tinker and Sanders, 1975). Nevertheless, it was hyphothesized (Barea et al., 1975) that if inoculated PSB could solubilize some phosphate ions, these would be taken up by the mycorrhizal hyphae, thus avoiding refixation problems in the translocation of phosphate to the absorption places at the root surface. Thus, a synergistic interaction between both types of microorganisms was suggested. A series of experiments using clover and other test plants (Azc6n et al., 1976; Delorenzini et al., 1979; Barea et al., 1981) indicated the feasibility of this hypothesis. As mycorrhizal plants can explore microhabitats in nonrhizosphere soil, these plants would gain more benefit from the presumed activity of PSB on RP, as shown in Fig. 1. This possibility is supported by experiments using 32P(Raj et al., 1981). A synergistic cooperation between these symbiotic and asymbiotic microorganisms was also evidenced by the utilization of organic phosphates in volcanic ash-derived soils (Borie and Barea 1981).

Slow diffusion

.:;IS

FIG. 1. Interaction of Va mycorrhiza and phosphate-solubilizing bacteria (PSB). RP, rock phosphate.

MYCORRHIZAS IN NODULATING N-FIXING PLANTS

37

3. Effect of N Fertilizer Because they are not dependent on combined N, legumes do not usually need N fertilizers when they are adequately nodulated. Moreover, these compounds are deleterious for nodulation and N fixation (Gibson and Newton, 1981; Dazzo and Brill, 1978). There is an obvious scarcity of information about interactions between these fertilizers and other treatments, such as mycorrhizal inoculation, on legume-Rhizobium sp. associations. VAM infection (in addition to nodulation) is suppressed by N fertilizers in Pisum sutivum (Lanoswska, 1966) and Trifoliurn substerruneum (Chambers et ul., 1980a,b). Additions of NH4+ were more deleterious than additions of NO,- (Chambers et al., 1980a), which can be explained on the basis of their different pathways of initial assimilation, cation carboxylate storage, and pH regulation (Raven and Smith, 1976; Smith, 1980). It follows that N fertilizer application must be rather limited for the correct development of the tripartite legume-Rhizobium sp. -VAM symbiosis. 4. Effect of Other Fertilizers

Information about interactions between VAM and additions of other nutrients for which nodulated legumes have special demands (Munns and Mosse, 1980) is still scant. Boron (B), one of these nutrients, was investigated for its interactions with VAM in T. prutense and M. sutivu (Lambert et al., 1980b). This study showed that an adequate B supply increases mycorrhizal activity. Boron deficiency in particular delayed the onset of VAM infection. This can affect the mycorrhizal effect on nodulation; however, the experimental design of these assays did not allow assessment of the direct effect of B supply on rhizobia.

E. ECOLOGICAL SIGNIFICANCE OF VESICULAR-ARBUSCULAR MYCORRHIZAS IN LEGUMES

It has been snown repeatedly that plants bearing dual mutualistic symbioses, such as nodulated and mycorrhizal legumes, may be well adapted to habitats with low availability of both N and P (Harley, 1970, 1973). Consequently these symbioses enable the plants to play an extremely important role as pioneer colonizers of nutrient-deficient soils (Janos, 1980b). The ecological and evolutionary significance of such symbioses in the history of terrestrial plants has been already discussed (Malloch et ul., 1980). Today mycorrhizal legumes have similar advantages for colonizing and thriving in adverse situations of different types.

38

J. M. BAREA AND

c. AZC~N-AGUILAR

I . Establishment and Improvement of Pastures in Marginal Soils As the production of pastures, especially when including legumes, is often limited by the low level of available P in the soil, the potential for managing VAM as a strategy to improve productivity in tropical dryland was demonstrated by Jehne (1980). Similarly, a positive effect of VAM in stimulating clover development related to hill-country soil improvement has been already demonstrated (Powell, 1976b; Hayman, 1977). Mycorrhizal fungi help the introduction of forage legumes in new habitats (see Redente and Reeves, 1981; Azc6nAguilar et al., 1982).

2 . Revegetation of Strip Mines and Other Industrial Wastelands Some mine spoils do not contain topsoil, and because they are not vegetated, they usually lack VAM fungal propagules. The establishment and survival of forage legumes (Lathyrus silvestris, Coronilla varia, and Lotus corniculatus) was aided significantly by mycorrhizal inoculation (Lambert and Cole, 1980), indicating the importance of VAM for revegetation purposes in these soils. Actually, plants colonizing coal wastes are mycorrhizal, and the legume species are also nodulated (Schramm, 1966; Daft et al., 1975; Daft and Hacskaylo, 1976; Khan, 1978). Thus these authors support the significance of mycorrhizal and nodulated plants in the rehabilitation of these wastelands into stable plant communities. In fact, the presence of both symbioses in Acacia hofosericeaand other woody and herbaceous legumes growing in restored areas after surface mining has been demonstrated (Langkamp and Dalling, 1982). These authors suggested that such symbiotic associations are a prerequisite for the successful establishment of long-term vegetation on these sites.

3 . Eroded Soil Reclamation The loss of topsoil and vegetation is an obvious consequence of erosion; therefore, eroded soils tend to be depleted of VAM propagules. The reintroduction of these fungi helped the growth and survival of white clover (Powell, 198Oc) and lotus (Hall and Armstrong, 1979) in such soils. The extensive network of fungal hyphae associated with white clover has been reported to be involved in the improvement of soil structure, thereby reducing soil erosion (Tisdall and Oades, 1979). 4 . Sand Dunes Stabilization Programs

Legumes are usually included in the lists of dominant plant species growing in sand dunes (Koske and Halvorson, 1981; Koske, 1981). Because mycorrhizal

MYCORRHIZAS IN NODULA'MNG N-FIXING PLANTS

39

beans proved to be suitable for sand aggregation by binding sand grains to the extensive VAM mycelium (Koske et al., 1975), the possibility of using legumes for dune stabilization seems attractive.

5 . Legumes in Arid and Semiarid Soils The mycorrhizal condition appears especially relevant for legumes growing in those habitats where pasture productivity is limited to a large extent by water availability (Jehne, 1980; Diem et a f . , 1981; Rose, 1981; Trappe, 1981). Because soil water content also affects P availability, the effect of mycorrhizas as related to water-use efficiency seems to be a key factor in pasture productivity in arid and semiarid regions (Safir, 1981). As prompt revegetation of such sites is needed, the use of mycorrhizal legumes that make such processes independent of N and P inputs is an appropriate choice.

6. Tolerance of Legumes to Other Stress Situations Published reports indicate that mycorrhizal plants have a greater tolerance of salinity, low pH, and high soil temperature than their nonmycorrhizal counterparts (Jehne, 1980; Bowen, 1980b). The yield of white clover in a low-pH (4.5) soil was significantly improved by mycorrhizal inoculation (Lambert and Cole, 1980), whereas the growth of noninoculated controls was poor.

7 . Role of VAM in Mixed Cropping Including Legumes The practice of growing legume and nonlegume (grasses) mixtures for pasture production is important economically. The agronomic events associated with grass-legume interactions have been reviewed (Haynes, 1980). Correct balances between species are important for maintaining pasture productivity, because grasses commonly have a competitive advantage over legumes, and many factors have been studied to obtain equilibria. One of these is mycorrhizal infection. Certainly VAM markedly increase the ability of clover to compete against ryegrass (Crush, 1974; Hall, 1978; Buwalda, 1980), because clover is more mycotrophic than the grass. Mycorrhizas also have other implications for plants growing together in the sward. These derive from the fact that the network of VA mycelium is able to link one plant to another. These mycorrhizal conections play an important role in the transport of P between plants (Heap and Newman, 1980; Whittingham and Read, 1982). As legumes also bring N to companion plants in the sward, the role of dually symbiotic legumes in the recycling of N and P in these ecosystems is clearly relevant.

Table III Greenhouse Experiment to Assess the Feasibility of the Introduction of a V A M Fungus (Gbmus mosseae) into a Legume (Hcdysarum coromuium) Rhizosphere in Two Test S o i a Soil l b Parameter % VA infection NodulationC Shoot weight (g/plant) Root/shoot ratio

C 0 2 2.1 1.3

I 57 4

3.8 0.9

Soil

I

N

N+I

C

20 2 2.5 1 .o

74 4 4.1

0 3 4.1

4.2

0.8

0.5

0.5

“After Azcbn-Aguilar er al. (1982). bVesicular-arbuscular endophytes were either not present (C), inoculated G. msseae (I), or native (N). ‘On a scale of 0 to 4.

3 3

26

N

N+I

52 3 5.0 0.5

60 4 4.9 0.5

MYCORRHIZAS IN NODULATING N-FIXING PLANTS

41

F. PRACTICAL FIELDAPPLICATION OF MYCORRHIZAL EFFECTS ON LEGUME PRODUCTION

It is already clear that the degree of efficiency with which a VAM fungus can improve plant growth and fertilizer use is affected by its ability to establish, spread, and survive when inoculated into a plant rhizosphere in a living soil. Hence, preliminary evaluation of the interactions between indigenous and introduced VA fungi, Rhizobium sp., and phosphate fertilizers will be required initially. These effects must be assessed under conditions that are as similar as possible to those prevailing in natural habitats where the field study will be performed. Tests to determine the feasibility of a field inoculation have been proposed, studies to estimate the rate of spread and persistence of mycorrhizal effects are being developed, and a series of small-scale field trials have been carried out. These utilize a range of VA endophytes and legume-Rhizobium sp. systems.

1, Tests for the Assessment of Field Situations in Which VAM inoculation of Legume Crops May Be Feasible A rapid method for assessing the effectiveness of native endophytes under conditions most similar to those governing their natural environments was conceived by Mosse (1977a) to predict whether field inoculation would be rewarding. Powell (1977a) studied a technique using Trifoliurn repens as the test plant and soil cores taken from field plots and kept in pots in the greenhouse. The use of such soil cores has many advantages as they represent undisturbed soil profiles. Another technique with a similar purpose is being developed and seems to be useful for defining the type of interaction (additive, synergistic, or antagonistic) occurring between native and introduced endophytes (Barea et al., 1980) and to predict the feasibility of field inoculation (Azc6n-Aguilar et al., 1982). This technique compares the effect of inoculation in unsterile and steamed aliquots of test soil. The microbiota (except mycorrhizal propagules) was reinoculated into the treated soil in order to get a suitable control (see Smith and Smith, 1981a). Data summarized in Table I11 suggest that VAM inoculation may be rewarding in Soil 1 (5 ppm Olsen P) but not in Soil 2 (33 ppm Olsen P). In the former, it appears that an additive cooperation between native and introduced VA endophytes occurs. The subsequent field experiment on Soil 1 proved to be successful (Azc6n-Aguilar et al., 1982).

2 . Studies to Determine the Spread and Persistence of Mycorrhizal Effects Because the production of sufficient inoculum limits large-scale inoculation, it is of interest to ensure a more efficient use of the available inocula and prevent

42

J. M. BAREA AND C. AZC6N-AGUILAR

the application of excessive amounts. Thus it is important to follow a suitable strategy for an adequate placement of inocula and to know their rate of spread and the residual growth effects in the following years (Mosse and Hayman, 1980). Studies by Powell (1979) and Mosse et ul. (1982) indicate that VA fungi are able to spread to about 4 m from the inoculation points in unsterile field soils. Mosse et u1. (1982) further pointed out a residual growth effect on M . sutivu used as the following year’s crop. Persistence of the mycorrhizal effect on alfalfa was also found in serial cutting of this crop under field conditions (Azc6n-Aguilar and Barea, 1981). 3 . Field Inoculation Experiments

The reports available on field trials of VAM inoculation of legumes for agricultural and revegetation purposes are recorded in Table IV. The published experiments show that efficient mycorrhizal fungi can be introduced into the rhizosphere of legumes growing in the field. This results in an improvement in Table IV Field Experiments on Inoculation of Legumes with VAM Fungi under Natural Conditions in Nollrumigated Soils Crop White clover

Red clover Faba beans Lucerne (alfalfa)

Inoculation technique“ Preinoculation transplants and direct incorporation into seed furrows Preinoculation transplants Seed pelleting Preinoculation transplants Preinoculation transplants Preinoculation transplants Direct incorporation into seed furrows Several Highly infective soil Preinoculation transplants Direct incorporation into seed furrows

Direct incorporation into seed furrows Direct incorporation into seed furrows Peas Direct incorporation into seed furrows soybean Direct incorporation into seed furrows Cowpea Preinoculation transplants Hedysarum coronarium Direct incorporation into seed furrows Lorn sp. Seed pelleting Highly infective soil

Reference Powell (1977b) Powell and Daniels (1978) Powell (1979) Powell (1982) Hayman and Mosse (1979) Hayman er al. (1979) Rangeley er al. (1982) Hayman ef al. (1981) Kucey and Paul (1983) Azc6n-Aguilar et al. (1979) Owusu-Bennoah and Mosse ( 1979) Azc6n-Aguilar and Barea (1981) Mosse et al. (1982) Jakobsen and Nielsen (1983) Bagyaraj et al. (1979) Islam et al. (1980) Azcbn-Aguilar et al. (1982) Hall (1980) Lambert and Cole ( 1980p

“See Section II,D,4 for details on inoculation techniques. forage legumes (of the genera Latyrus and Coronilla) were also inoculated in this study.

MYCORRHIZAS IN NODULATING N-FIXING PLANTS

43

the growth and nodulation of these plants. Grain on forage legumes growing under conditions of tropical and temperate regions that have been assayed indicate that they benefit from mycorrhizal inoculation. Most of the tests recorded in Table IV gave positive responses to VAM. There are indications that the introduced VAM fungi swiftly become established, as deduced from the plant response, but sometimes the establishment was actually ascertained because of differences in the anatomical characteristics of the infections produced by introduced and indigenous endophytes. Inoculation experiments with M . sativa and its symbiotic partners R. melitoti and the VA fungus G. mosseae, carried out under agricultural conditions in untreated arable soil, illustrate the interactions in the tripartite symbiosis (Azc6nAguilar et al., 1979; Azc6n-Aguilar and Barea, 1981). These experiments were designed for the same soil but in plots that had supported different agronomic practices and which differed from one another in some characteristics; plot B possessed twice as much available phosphate and three times as much VA propagules as plot A. In the latter, inoculation with G. mosseae was always effective in promoting plant growth, but R. meliloti was only able to enhance the growth of G. mosseae-inoculated plants. Probably P was the limiting factor for R. meliloti activity, because of the scarcity of available nutrient and the low number of spores of the family Endogonaceae in this test soil (A). Hence, plants did not repond to the R. meliloti inoculation unless they were also inoculated with VA fungi. In contrast, R. meliloti was effective when inoculated alone in plants growing in plot B. The efficiency of indigenous VA fungi, together with the higher concentration of available P in the soil, could cause plants to respond to R. melitoti inoculated alone. In both cases, however, inoculation of R. meliloti plus G . mosseae more than doubled the yield compared to the uninoculated controls. In general, published field trials demonstrate that mycorrhizal inoculations improve the N content in leguminous plants, but they do not indicate whether the extra N results from increased N, fixation or increased N uptake from the soil. The use of 15N to label soil-assimilable N is a suitable way to quantitatively estimate the amount of N in the plant coming from symbiotic dinitrogen fixation by legumes growing under field conditions. This should be a subject of future research. Although positive inoculation responses are to be expected mainly in lowphosphate soils, it seems that the best responses have been obtained in soils of moderate fertility. For example, a twofold increase in growth after inoculation was obtained in plots of white clover given 90 kg P/ha (Hayman and Mosse, 1979), agreeing with other observations (Powell, 1977b; Owusu-Bennoah and Mosse, 1979). The compatibility of certain phosphate levels and mycorrhizal effects is, obviously, an interesting principle in legume production systems, as is the exploitation of interactions between rock phosphate and VAM. The use of

44

J. M. BAREA AND C. AZC6N-AGUILAR

this sparingly soluble form of phosphate seems of interest, especially in phosphate-retentive soils. The incorporation of Rp into pellets together with the seeds and the rhizobial inocula is a tantalizing possibility among several areas needing further research concerning the field inoculation of VAM fungi plus Rhizobium sp. for increasing legume productivity.

IV. MYCORRHIZAS IN NODULATING NITROGENFIXING NONLEGUME PLANTS The nonlegume plant species that are able to form N-fixing root nodules, as induced by soil bacteria, are also known to form mycorrhizal associations. The formation, role, and function of the root nodules and the significance of mycorrhizas in these plants are similar to those just discussed for legume-Rhizobium sp. symbioses. Consequently the subject will be only briefly reviewed, condensing information from the few available reports. A. OCCURRENCE AND DISTRIBUTION

The main group of these plants consists of the actinomycete-nodulated, socalled actinorrhizal plants. These are shrubs or small trees distributed t b u g h many ecosystems of temperate regions. Torrey (1978) reported actinorrhizas in 160 species of 15 genera belonging to 8 families. The genus Datisca must now also be listed (Rose, 1980). According to Trappe (1979) and Rose (1980) the actinorrhizal genera bearing VAM are Casuarina, Eleagnus, Hyppophae, Ceanothus, Colletia, Discaria, Purshia, Robus, and Datisca; VAM, ECM, and actinorrhizas coexist in Alnus, Myrica, Comptonia, Dryas, and Coriaria. The mycosymbionts involved were recorded by Rose (1980), Rose and Trappe (1980), and Molina (1981). Members of the order Cycadales (gymnosperms) also form typical N fixing root nodules. The endosymbionts are the blue-green algae (cyanobacteria) Nostoc or Anabaena (see Akkermans, 1978). Vesicular-arbuscular mycorrhizas have been reported to coexist with the nodules in these plants (Trappe, 1979). Species of Parasponia (formerly Trema) in the family Ulmaceae are nodulated by a species of Rhizobium (Akkermans et al., 1978) and appears to possess VAM (Trappe, 1979). B. S T R U ~ R AAND L . PHYSIOLOGICAL FEATURES

An unusual characteristic of mycorrhizal development in actinorrhizas is that VA hyphae have often been found in the nodular tissues (Rose, 1980; Rose and

MYCORRHIZAS IN NODULATING N-FIXING PLANTS

45

Youngberg, 1981). The VAM structures can be surrounded by ECM in species of Ahus and Ceanothus. Thus it seems that the typical exclusion by nodular and mycorrhizal endophytes does not occur in actinorrhizal plants (Rose, 1980). As in legumes, these mycorrhizas act mainly by their well-known P-mediated mechanism (Mejstrik and Benecke, 1969; Le Tacon and Diagne, 1982), thereby increasing the number of nodules, the weight of nodular tissue, nitrogenase activity, and N, Ca, and P shoot content (Rose and Youngberg, 1981). However, hormone-mediated interactions between endosymbionts can be also involved as these substances seem important in the formation of actinorrhizal symbioses (Miguel et al., 1978). The possible interactions between mycorrhizas and actinorrhizas derived from the former produce calcium oxalate, which is needed by the latter (Trappe, 1979). C . ECOLOGICAL ASPECTS

Actinorrhizal plants are usually involved in the early successional stages of plant communities at low nutrient sites (Harley, 1973). These plants are therefore found colonizing disturbed and marginal habitats such as sand dunes, volcanic ash-derived soils, coal wastes, peat and sphagnum bogs, (Daft and Hacskaylo, 1976; Khan, 1978; Rose, 1980). Rose (1980) reported that 23 of the 25 actinorrhizal plants he tested were VA mycorrhizal. This is, therefore, similar to the situation found with legumes, suggesting the suitability of applying the mycorrhizal effects to actinorrhizal plants (typically woody and perennial) for the successful reforestation of stressed ecosystems.

V. CONCLUSIONS AND PERSPECTIVES Mycorrhizal associations play an important role in the growth and nutrition of higher plants. This results primarily from their more efficient use of soil P. Mycorrhizas appear to have an ecological and evolutionary relevance in the history of terrestrial plants. It is increasingly recognized that this symbiosis can be harnessed in order to improve nutrient cycling and crop productivity by reducing industrial fertilizer inputs, thereby conserving and reducing environmental costs. In addition, mycorrhizal infection can help plants to become reestablished in eroded or degraded habitats, to thrive in arid conditions, to deter pathogens, and to cope with various stress situations. The commonest mycorrhizal types, the VAM, are nearly omnipresent and are now being studied intensively throughout the world. Their contribution to the more efficient use of added P fertilizers, whether soluble (as applied at suboptimal rates) or sparingly soluble (rock phosphate), is being widely appreciated.

46

J. M. BAREA AND C. AZC6N-AGUILAR

The infectivity and the operativity of VAbi aic self regulated by a mechanism that assures that the P supply to the host will be optimal over a wide range of soil P levels. Vesicular-arbuscular mycorrhizas enhace growth, nodulation, and N fixation in grain and forage legumes, crops of the greatest interest for food production in the biosphere. They have a similar function in N-fixing nonlegumes, largely actinorrhizal plants, mostly of interest in forestry. Nodulating, N-fixing plants are usually mycorrhizal in the ecosystem, but the responsible VA fungi are not always the most suitable and can be replaced with more effective strains by means of inoculation. In general, the inoculation of VAM fungi has problems that limit its extensive use on a field scale. Field experiments with VAM must be supported by ecological and physiological studies including (1) the evaluation of the dependency of the test plant on the mycorrhiza; (2) the assessment of field sites where mycorrhizal inoculation with preselected (efficient and ecologically adapted) endophytes may be worth trying; (3) the production of high-quality inocula and the development of suitable inoculation techniques. Further research on mycorrhizas is needed with regard to several topics pointed out in this article. Current research mentioned has included (1) studies of the population ecology and epidemiology of mycorrhizal fungi; (2) studies investigating the causes of the host dependency of VAM fungi for carrying out their life cycle axenically; (3) physiological studies of mycorrhizal symbioses to discover new mechanisms of action; (4) the application of isotope and radiation techniques to make possible the effective management of mycorrhizas in increasing food-crop production; and (5) field inoculation experiments in small plots to establish bases for future mycorrhizal programs. Mycorrhizas therefore can be regarded as an alternative strategy for a more rational agricultural program. However, because the mycorrhizal condition is nearly universal, the natural mycorrhizal potential of a soil needs first to be preserved (avoiding detrimental practices), second to be optimized (manipulating soil conditions to be conducive to the symbiosis), and third, finally, to be considered when inoculation is required.

REFERENCES Abbott, L. K . , and Robson, A. D. 1977. Aust. J . Agric. Res. 28, 639-640. Abbott, L. K., and Robson, A. D. 1978. New Phytol. 81, 575-585. Abbott, L. K . , and Robson, A. D. 1981a. Aust. J . Agric. Res. 32, 621-630. Abbott, L. K . , and Robson, A. D. 1981b. Aust. J . Agric. Res. 32, 631-639. Abbott, L. K . , and Robson, A. D. 1982. Aust. J . Agric. Rex 33, 389-408. Allen, M. F., and Boosalis, M. G . 1983. New Phytol. 93, 67-76. Allen, E. B., and Cunningham, G . L. 1983. New Phytol. 93, 227-236. Akkermans, A. D. 1978. I n “Interactions between Non-Pathogenic Soil Microorganisms and Plants” (Y. R. Dommergues and S. V. Krupa, eds.), pp. 335-372. Elsevier, Amsterdam.

MYCORRHIZAS IN NODULATING N-FIXING PLANTS

41

Akkermans, A. D., Abdulkadir, S., and Tnnick, M. J. 1978. Nature (London) 274, 190. Allen, M. F., Moore, T. S . , and Christensen, M. 1980. Can. J . Bot. 58, 371-374. Allen, M. F., Sexton, J. C., Moore, T. S . , and Christensen, M. 1981. New Phytol. 87, 687-694. Allen, M. F., Moore, T. S., and Christensen, M. 1982. Can. J . Bot. 60, 468-471. Andrew, C. S., and Robins, M. F. 1969. Aust. J . Agric. Res. 20,675-685. Asai, T. 1944. Jpn. J . Bot. 13, 463-485. Asimi, S., Gianinazzi-Pearson, V., and Gianinazzi, S. 1980. Can. J. Bot. 58, 2200-2205. Azcbn, R., and Ocampo, J. A. 1981. New Phyrol. 87, 677-685. Azcbn, R., Barea, J. M., and Hayman, D. S. 1976. Soil Biol. Biochem. 8, 135-138. Azcbn, R., Azcbn-G. de Aguilar, C., and Barea, 3. M. 1978a. New Phytol. 80, 359-364. Azcbn, R., Mm’n A. D., and Barea, J. M. 1978b. Plant Soil 49, 561-567. Azcbn-Aguilar, C., and Barea, J. M. 1978. Can J . Microbiol. 40, 520-524. Azcbn-Aguilar, C., and Barea, J. M. 1980. Invest. Ciencia 47, 8-16. Azcbn-Aguilar, C., and Barea, J. M. 1981. Soil Biol. Biochem. 13, 19-22. Azcbn-Aguilar, C., A z c h , R., and Barea, J. M. 1979. Narure (London) 279, 325-327. Azcbn-Aguilar, C., Barea, J. M.,and Olivares, J. 1980. Trans. Int. Symp. Microbial Ecol. 2nd, p. 129. Azcbn-Aguilar, C., Barea, J. M., Azcbn, R., and Olivares, J. 1982. Agric. Environ. 7, 199-206. Bagyaraj, D. J., and Sreeramulu, K. N. 1982. Plant Soil 69, 375-381. Bagyaraj, D. J . , Manjunath, A., and Patil, R. B. 1979. New Phytol. 82, 141-145. Barea, J. M., and Azcbn-Aguilar, C. 1980. Trans. Int. Symp. Microbial Ecology 2nd. p. 130. Barea, J. M., and Azcbn-Aguilar. C. 1982a. Trans. Int. Col. Mycorrhizes Biol. Perspectives Urilization, Dijon. Barea, J. M., and Azcbn-Aguilar, C. 1982b. Appl. Environ. Microbiol. 43, 810-813. Barea, J. M., Ramos, A., and Callao, V. 1970. Microbiol. ESP.23, 257-270. Barea, J. M., Azcbn, R., and Hayman, D. S. 1975. In “Endomycorrhizas” (F. E. Sanders, B. Mosse, and P. B. Tinker, eds.), p. 417. Academic Press, New York and London. Barea, J. M., Escudero, J. L., and Azcbn-Aguilar, C. 1980. Plant Soil 54, 283-2%. Barea, J. M., Azcbn, R., and Azc6n-Aguilar, C. 1981. FAOIIAEA Consult, Meet. Use Isotopes Stud. Nutrient Availability Food Crops Endomycowhizal Vienna. Barrow, N. J., Malajczuk, N., and Shaw, T. C. 1977. New Phytol. 78, 269-276. Bauer, W. D. 1981. Annu. Rev. Plantfhysiol. 32, 407-449. Baylis, G. T. S. 1970. Plant Soil 33, 713-716. Bergersen, F. J . 1978. In “Interactions between NonpathogenicMicroorganismsand Plants” (Y. R. Dommergues and S . V. Krupa, eds.), pp. 305-337. Elsevier, Amsterdam. Bethlenfalvay, G. J. 1983. Physiol. Plant. 57, 543-548. Bevege, D. I., Bowen, G. D., and Skinner, M. F. 1975. In “Endomycorrhizas” (F. E. Sanders, B. Mosse, and P. B. Tinker, eds.), pp. 149-174. Academic Press, New York and London. Bieleski, R. L. 1973. Annu. Rev. Plant Physiol. 24, 225-252. Bjorkman, E. 1970. Plant Soil 32, 589-610. Black, R. L., and Tinker, P. B. 1977. Nature (London) 267, 510-51 1 . Bolin, B., and Anhenius, E. 1977. Ambio 6, 96-105. Bone, F., and Barea, J. M. 1981. Anal Edafol. Agrobiol. 40, 2351-2381. Bowen, G. D. 1973. In “Ectomyconhizae” (G. C. Marks and T. T. Kozlowski, eds.), pp. 151-205. Academic Press, New York. Bowen, G. D. 1980a. In “Contemporary Microbial Ecology” (D. C. Ellwood, J. N. Hedger, M. J. Latham, J. H. Slater, and J. M. Lynch, eds.), pp. 283-304. Academic F’ress, New Yo& and London. Bowen, G. D. 1980b. In “Tropical Mycorrhiza Research” (P. Mikola, ed.), pp. 165-190. Oxford Univ. Press, London. Bowen, G. D., and Bevege, D. I. 1976. Rev. Rural Sci. 3, 103-112.

48

J. M. BAREA AND C. AZC6N-AGUILAR

Bowen, G . D., and Theodorou,C. 1979. Soil Biol. Biochem. 10, 119-126. Brill, W. J. 1979. Am. Sci. 67,458-466. Bmughton, W. J. 1978. J . Appl. Bucreriol. 45, 165-194. Brown. M. E. 1975. In “Soil Microbiology” (N. Walker, ed.), pp. 21-38. Butterworth, London. Buwalda, J. G. 1980. N.Z. J . Agric. Res. 23, 379-383. Buwdda, J. G., and Goh, K. M. 1982. Soil Biol. Biochem. 14, 103-106. Callow, J. A., Capaccio, L. C. M., Parish, G., and Tinker, P. B. 1978. New Phytol. 80, 125-134. Capaccio, L. C. M., andcallow, J. A. 1982. NewPhyrol. 91, 81-91. Carling, D. E., and Brown, M. F. 1980. Soil Sci. SOC. Am. J . 44, 528-532. Carling, D. E., Riehle, W. G., Brown, M. F., and Johnson, D. R. 1978. Phyroparhology 68, 1590- 1596. Casadestis, J., and Olivares, J. 1978. Anal. Edafol. Agrobiol. 37, 915-948. Chambers, C. A., Smith, S. E., and Smith, F. A. 1980a. New Phyrol. 85,47-62. Chambers,C. A., Smith, S. E., Smith, F. A., Ramsey,M. D., and Nicholas, D. J. 1980b. Soil Biol. Biochem. 12, 93-100. Chapin, F. S., III 1980. Annu. Rev. Ecol. Sysr. 11, 233-260. Chilvers, G. A., and Harley, J. L. 1980. New Phyrol. 84, 319-326. Clarke, C., and Mosse, B. 1981. New Phyrol. 87, 695-703. Cooper, K. M., and Liisel, D. 1978. New Phyrol. 80, 143-151. Cooper, K. M., and Tinker, P. B. 1981. New Phyrol. 88, 327-339. Cox, G., and Sanders, F. E. 1974. New Phyrol. 73, 901-912. Cox, G., and Tinker,P. B. 1976. New Phyrol. 77, 371-378. Cox, G., Sanders, F. E., Tinker, P. B., and Wild, J. A. 1975. In “Endomycorrhizas” (F. E. Sanders, B. Mosse, and P. B. Tinker, eds.), pp. 297-312. Academic Press, New York and London. Cox, G., Moran, K. J., Sanders, F., Nockolds, C., and Tinker, P. B. 1980. New Phyrol. 84, 649-659. Cress, W. A., Thronebemy, G. O., and Lindsey, D. L. 1979. Planr Physiol. 64, 484-487. Crush, J. R. 1974. New Phyrol. 73, 743-752. crush, I . R. 1975. N.Z. J . Agric. Res. 18, 361-364. crush, J. R. 1976. N.Z. J . Agric Res. 19, 473-476. Crush, J. R . , and Caradus, J. R. 1980. N.Z.J . Agric. Res. 23, 233-237. Daft, M. J., and El-Giahmi, A. A. 1974. New Phytol. 73, 1139-1147. Daft, M. J., and El-Giahmi, A. A. 1975. In “Endomycorrhizas” (F. E. Sanders, B. Mosse, and P. B. Tinker, eds.), pp. 581-592. Academic Press, New York and London. Daft, M. J., and El-Giahmi, A. A. 1976. Ann. Appl. Biol. 83, 273-276. Daft, M. J., and EI-Giahmi, A. A. 1978. New Phyrol. 80, 365-372. Daft, M. J., and Hacskaylo, E. 1976. J . Appl. Ecol. 13, 523-531. Daft, M. J., Hacskaylo, E., and Nicolson, T. H. 1975. In “Endomycorrhizas” (F. E. Sanders, B. Mosse, and P. B. Tinker, eds.), pp. 561-580. Academic Press, New York and London. Daniels, B. A., and Menge, J. A. 1980. Phytopurhology 70, 584-588. Daniels, B. A., and Menge, J. A. 1981. New Phytol. 87, 345-354. Daniels, B. A., and Trappe, J. M. 1979. Can. J . Bot. 57, 539-542. Dazzo, F. B., and Brill, W. J. 1978. Plant Physiol. 62, 18-21. Dazzo, F. B . , and Hubbell, D. H. 1975. Proc. Soil Crop Sci. SOC. Flu. 34, 71-74. Delorenzini, C., Barea, J. M., and Olivares, J. 1979. Rev. Larinoarn. Microbiol. 21, 129-134. Demeterio, J. L., Ellis, R., and Paulsen, G. M. 1972. Agron. J . 64, 566-570. Mnari6, J . , and Truchet, G. 1979. Physiol. Veg. 17, 643-667. Dexheimer, J., Gianinazzi, S.,and Gianinazzi-Pearson. V. 1979.2. Pjlunzenphysiol. 92, 191-206. Diem, H. G . , Gueye, I., Gianinazzi-Pearson, V., Fortin, J. A., and Dommergues, Y. R. 1981. Acta Ecol. Ecol. Plant. 2, 53-62.

MYCORRHIZAS IN NODULATING N-FIXING PLANTS

49

Dodd, J., Krikun, J., and Haas, J. 1983. Isr. J. Bot. 32, 10-21. Dommeregues, Y. R. 1978. In “Interactions between Nonpathogenic Soil Microorganisms and Plants’’ (Y. R. Dommergues and S. V. Krupa, eds.), pp. 1-38. Elsevier, Amsterdam. Elmes, R., and Mosse, B. 1980. “Rothamsted Report 1979.” El-Giahmi, A. A., Nicolson, T. H., and Daft, M. J. 1976. Trans. Br. Mycol. SOC. 67, 273-274. Evans, H. J., and Barber, L. E. 1977. Science (Washington. D . C . ) 197, 332-339. Fogel, R. 1980. New Phytol. 86, 199-212. Foster, S. M., and Nicolson, T. H. 1981a. Soil Biol. Biochem. 13, 199-203. Foster, S. M., and Nicolson, T. H. 1981b. Soil Biol. Biochem. 13, 205-208. Furlan, V., and Fortin, J. A. 1977. New Phytol. 79, 335-340. Gates, C. T. 1974. Aust. J. Bot. 22, 45-55. Gates, C. T., and Wilson, J. R. 1974. Plant Soil 41, 325-333. Gerdemann, J. W. 1975. In “The Development and Function of Roots” (J. G. Torrey and D. T. Clarkson, eds.), pp. 575-591. Academic Press, New York. Gerdemann, J. W., and Trappe, J. M. 1974. Mycol. Mem. 5, 1-76. Gianinazzi-Pearson, V., and Gianinazzi, S. 1981. In “The Fungal Community; Its organization and Role in the Ecosystem” (D. T.Wicklow and G. C. Carrol, eds.), pp. 637-652. Dekker, New York. Gianinazzi, S., Gianinazzi-Pearson, V., and Trouvelot, A. (eds.) 1982. “Les Mycorrhizes partie int6grante de la plante: biologie et perspectives d’utilization”. Colloq. INRA No. 13. Gibson, A. H., and Newton, W. E. (eds.) 1981. “Current Perspectives in Nitrogen Fixation.” Aust. Acad. of Sciences, Canberra. Gibson, D. I., Hayes, P., and Laidlaw, A. S. 1975. J. Br. Grassl. SOC. 30, 295-301. Graham, J. H., and Menge, J. A. 1982. Phytopathology 72, 95-98. Graham, J. H., Linderman, R. G., and Menge, 3. A. 1982. New Phytol. 91, 183-189. Gray, L. E., and Gerdemann, J. W. 1973. Plant Soil 39, 687-689. Greaves, M. P., and Webley, D. M. 1965. J . Appl. Bacteroil. 28, 454-465. Hall, I. R. 1975. N.Z. J. Bot. 13, 463-472. Hall, I. R. 1978. N.Z. J. Agric. Res. 21, 509-515. Hall, I. R. 1980. N.Z. J. Agric. Res. 23, 103-105. Hall, I. R., and Armstrong, P. 1979. N.Z. J . Agric. Res. 22, 479-484. Hall, I. R., Scott, R. S., and Johnstone, P. D. 1977. N.Z. J. Agric. Res. 20, 349-355. Hardy, R. W. F., and Haveka, U. D. 1976. In “Symbiotic Nitrogen Fixation in Plants” (P. S. Nutman, ed.), pp. 421-439. Cambridge Univ. Press, Cambridge. Harley, J. L. 1970. Trans. Bor. SOC.Edinburgh 41, 65-70. Harley, J. L. 1973. J . Nut. Sci. Coun. Sri L a n k 1, 31-48. Harley, J. L. 1975. In “Endomycorrhizas” (F. E. Sanders, B. Mosse, and P. B. Tinker, eds.), pp. 1-24. Academic Press, New York and London. Harley, J. L. 1978. Proc. R . SOC. London Ser. B 203, 1-21. Hart, A. L., Jessop, D. J., and Galpin, J. 1981. N.Z. J . Agric. Res. 24, 27-32. Hayman, D. S . 1974. New Phytol. 73, 71-80. Hayman, D. S. 1975a. In ‘‘Soil Microbiology” (N. Walker, ed.), pp. 67-92. Buttersorth, London. Hayman, D. S. 1975b. In “Endomycorrhizas” (F. E. Sanders, B. Mosse, and P. B. Tinker, eds.), pp. 495-509. Academic Press, New York and London. Hayman, D. S. 1977. ARC Res. Rev. (U.K.) 3, 82-85. Hayman, D. S. 1978. In “Interactions between Nonpathogenic Soil Microorganismsand Plants” (Y. R. Dommergues and S. V. Krupa, eds.), pp. 401-442. Elsevier, Amsterdam. Hayman, D. S. 1981. FAOIIAEA Consult. Meet. Use Isotopes Stud. Nutrient Availability Food Crops Edomycorrhizae, Vienna. Hayman, D. S . 1982. In “Advances in Agricultural Microbiology” (N. S. Subba Rao ed.). INH New Delhi.

50

J. M. BAREA AND C. AZCdN-AGUILAR

Hayman, D. S. 1983. Can. J . Bor. 61, 944-963. Hayman, D. S., and Mosse, B. 1979. Ann. Appl. Biol. 93, 141-148. Hayman, D. S., Barea, J. M., and Azc6n. R. 1976. Phyropathol. Medirerr. 15, 1-6. Hayman, D. S., Moms, E. J., and Page, R. J. 1981. Ann. Appl. Biol. 99, 247-253. Haynes, R. J. 1980. Adv. Agron. 33, 227-261. Heap, A., and Newman, E. I. 1980. New Phyrol. 85, 173-179. Hepper, C. 1979. Soil Biol. Biochem. 11, 269-277. Hepper, C. M. 1983. New Phyfol. 93, 389-399. Ho, I., and Trappe, J. M. 1973. Nature New Biol. 244, 30-31. Howeler, R. H., Cadavid, L. F., and Burckhard, E. 1982a. Plant Soil 69, 327-339. Howeler, R. H., Asher, C. J., and Edwards, D. G. 1982b. New Phytol. 90, 229-238. Islam, R., Ayanaba, A., and Sanders, F. E. 1980. Planr Soil 54, 107-117. Jakobsen, I., and Nielsen, N. E. 1983. New Phyrol. 93, 401-413. Janos, D. P. 1980a. Ecology 61, 151-162. Janos, D. P. 1980b. “Tropical Succession 1980.” Janse, J. M. 1896. An. Jard. Bor. Buirenz. 14, 53-212. Jasper, D. A., Robson, A. D., and Abbott, L. K. 1979. Soil Biol. Biochem. 11, 501-505. Jehne, W. 1980. Trop. Grassl. 14, 202-209. Jensen, A. 1983. Plant Soil 70, 155-163. Johnson, C. R., Menge, J. A., Schwab, S . , and Ting, I. P. 1982a. New Phyrol. 90, 665-669. Johnson, C. R.,Graham,J. H., Leonard, R. T.,and Menge, J. A. 1982b. New Phytol. 90,671-675. Jones, F. R. 1924. J . Agric. Res. 29, 459-470. Khan, A. G. 1974. J. GenMicrobiol. 81, 7-14. Khan, A. G. 1975. In “Endomycorrhizas” (F. E. Sanders, B. Mosse, and P. B. Tinker, eds.), pp. 419-435. Academic Press, New York and London. Khan, G . G. 1978. New Phytol. 81, 53-63. Kleinschmidt, G . D., and Gerdemann, J. W. 1972. Phyropafhology 62, 1447-1453. Koske, R. E. 1981. Trans. Br. Mycol. SOC. 76, 411-416. Koske, R. E., and Halvorson, W. L. 1981. Can. J . Bor. 59, 1413-1422. Koske, R. E., Sutton, J. C., and Sheppard, B. R. 1975. Can. J . Bor. 53, 87-93. Kruckelmann, H. W. 1975. I n “Endomycorrhizas” (F. E. Sanders, B. Mosse, and P. B. Tinker, eds.), pp. 511-525. Academic Press, New York and London. Kucey, R. M. N., and Paul, E. A. 1982. Soil Biol. Biochem. 14, 407-412. Kucey, R. M. N., and Paul, E. A. 1983. Can. J. Soil Sci. 63, 87-95. Lambert, D. H., and Cole, H. 1980. Agron. J. 72, 257-260. Lambert, D. H . , Cole, H., and Baker, D. E. 1980a. Crop. Sci. 20, 615-618. Lambert, D. H . , Cole, H., and Baker, D. E. 1980b. Planr Soil 57, 431-438. Langkamp, P. J., and Dalling, M. J. 1982. Ausr. J. Bor. 30, 107-119. Lanowska, J. 1966. Pam. Pulowski. 21, 365-386. Le Tacon, F., and Diagne, 0. 1982. Trans. Int. Col. Mycorrhizes Biol. Perspecrives Urilisarion, Dijon. Levy, Y., Sylvertse, J. P., and Nemec, S. 1983. New Phyrol. 93, 61-66. Lewis, D. H. 1973. Biol. Rev. 48, 261-278. Lewis, D. H. 1975. In “Endomycorrhizas” (F. E. Sanders, B. Mosse, and P. B. Tinker, eds.), pp. 119-148. Academic Press, New York and London. Ling-Lee, M., Chilvers, G. A., and Ashford, A. E. 1975. New Phyrol. 75, 551-554. Ljunggren, H., and Fahraeus, G . 1959. Nature (London) 184, 1578-1579. Lijsel, D. M., and Cooper, K. M. 1979. New Phyrol. 83, 415-426. MacDonald, R. M., and Lewis, M. 1978. New Phyrol. 80, 135-141. Malajczuk, N., Linderman, R. G . , Kough, J., and Trappe, J. M. 1981. New Phyrol. 87,567-572.

MYCORRHIZAS IN NODULATING N-FIXING PLANTS

51

Malloch, D. W., Pirozynski, K. A., and Raven, P. H. 1980. Proc. Narl. Acud. Sci. USA 77, 21 13-21 18. Marx, C., Dexheimer, J., Gianinazzi-Pearson, V.,and Gianinazzi, S. 1982. New Phyrol. 90,37-43. Marx, D. H. 1980. In “Tropical Mycorrhiza Research” (P. Mikola, ed.). Oxford Univ. Press, London. Marx, D. H., and Krupa, S. U. 1978. In “Interactions between Non-pathogenic Soil Microorganisms and Plants” (Y. R. Dommergues and S. V. Krupa, eds.), pp. 373-400. Elsevier, Amsterdam. Mattingly, G. E. G. 1980. Chem. Ind. (London) 1980, 690-693. Mejstrik, V., and Benecke, U. 1969. New Phyrol. 68, 141-149. Menge, J. A. 1983. Can. J. Bor. 61, 1015-1024. Menge, J. A., Lembright, H., and Johnson, E. L. V. 1977. Proc. Int. SOC. Citric. 1, 129-132. Menge, J. A., Steirle, D., Bagyaraj, D. J., Johnson, E. L. V., and Leonard, R. T. 1978. New Phytol. 80, 575-578. Mexal, J. G. 1980. N.Z. J . For. Sci. 10, 208-217. Meyer, F. H. 1973. In “Ectomycorrhizae” (G. C. Marks and T. T. Kozlowski, eds.), pp. 79-106. Academic Press, New York. Miguel, C., Caiiizo, A., Costa, A., and Rodriguez-Barmeco, C. 1978. In “Limitations and Potentials for Biological Nitrogen Fixation in the Tropics” (J. Nbereiner ezal., eds.), pp. 121-134. Plenum, New York. Mikola, P. 1970. Inr. Rev. For. Res. 3, 123-196. Mojallali, H., and Weed, S. B. 1978. Soil Sci. Soc. Am. J . 42, 367-372. Moiraud, A., Capellano, A., and Birtschi, H. 1981. Can. J . Bot. 59,481-490. Molina, R. 1977. In “Mushrooms and Man: An Interdisciplinary Approach to Mycology” (T. Walters, ed.), pp. 147-161. Molina, R. 1979. For. Sci. 25, 585-590. Molina, R. 1981. Can. J. Bor. 59, 325-334. Morley, C. D., and Mosse, B. 1976. Trans. Br. Mycol. SOC. 67, 510-513. Mosse, B. 1962. J. Gen. Microbiol. 27, 509-520. Mosse, B. 1972. Rev. Ecol. B i d . Sol 9, 529-537. Mosse, B. 1973a. Annu. Rev. Phyropazhol. 11, 171-1%. Mosse, B. 1973b. New Phyrol. 72, 127-136. Mosse, B. 1975. In “Soil Microbiology” (N. Walker, ed.), pp. 39-66. Butterworth, London. Mosse, B. 1977a. New Phyzol. 78, 277-288. Mosse, B. 1977b. Coll. Trop. Agric. Univ.Hawaii Misc. Publ. 145, 275-292. Mosse, B. 1978. In “Structure and Functioning of Plant Populations” (A. H. T. Freysen and B. W. Woldendorp, eds.), pp. 269-298. Mosse, B., and Hayman, D. S. 1971. New Phytol. 70, 29-34. Mosse, B., and Hayman, D. S. 1980. In “Tropical Mycorrhiza Research” (P. Mikola, ed.), pp. 213-230. Oxford Univ. Press, London. Mosse, B., and Hepper, C. 1975. Physiol. P h Purhol. 5 , 215-223. Mosse, B., Hayman, D. S., and Arnold, D. J. 1973. New Phytol. 72, 809-815. Mosse, B., Powell, C. L., and Hayman, D. S. 1976. New Phyrol. 76, 331-342. Mosse, B., Warner, A., and Clarke, C. A. 1982. New Phyzol. 90, 521-528. Munns, D. N . , and Mosse, B. 1980. In “Advances& Legume Science” (R.J. Summerfild and A. H. Bunting, eds.), pp. 115-125. HM Stationery Office,London. Nemec, S. 1983. Trop. Agric. 60, 97-101. Newman, E. 1979. Biol. Rev. 53, 511-554. Ng, P. P., Cole, A. L. J., Jameson, P. E., and Mcwha, J. A. 1982. New Phyrol. 91, 57-62. Nicolson, T. H. 1960. Trans. Br. Mycol. SOC. 43, 132-145.

52

J. M. BAREA AND C. AZC6N-AGUILAR

Nicolson, T. H. 1975. In “Endomycomhizas” (F. E. Sanders, B. Mosse, and P. B. Tinker, eds.), pp. 25-34. Academic Press, New York and London. Nielsen, J. D., and Jensen, A. 1983. Plum Soil 70, 165-172. Nutman, P. S. 1977. In “Recent Developments in Nitrogen Fixation” (W. Newton, J. R., Postgate, and C. Rodriguez-Barmeco, eds.), pp. 443-447. Academic Press, New York and London. Nylund, J. E., and Unestam, T. 1982. New fhyrol. 91, 63-79. O’Bannon,J. H., Evans, D. W., and Peaden, R. N. 1980. Can. J. flanr Sci. 60, 859-863. Ocampo, J. A., and Hayman, D. S . 1981. New fhyrol. 87, 333-343. Ocampo, J. A., Martin, J., and Hayman, D. S . 1980. New fhytol. 84, 27-35. Ojala, J. C., Jarrel, W. M., Menge, J. A., and Johnson, E. L. V. 1983. Agron. J. 75, 255-259. Olivares, J., Montoya, E., and Palomares, A. 1977. In “Recent Developments in Nitrogen Fixation” (W. Newton, J. R. Postgate and C. Rodriguez-Barmeco, eds.), pp. 375-386. Academic Press, New York and London. Owusu-Bennoah, E., and Mosse, B. 1979. New Phyrol. 83, 671-679. Owusu-Bennoah, E., and Wild, A. 1979. New fhyrol. 82, 133-140. PairUnan, A. K., Robson, A. D., and Abbott, L. K. 1980. New fhytol. 84, 327-338. Pang, P. C., and Paul, E. A. 1980. Can. J. Soil Sci. 60, 241-250. Pfeiffer, C. M., and Bloss, H. E. 1980. Mycologia 72, 1038-1041. Phillips, D. A. 1980. Annu. Rev. Plant fhysiol. 31, 29-49. Pichk, Y., and Fortin, J. A. 1982. New fhyrol. 91, 211-220. Pirozynski, K. A,, and Malloch, D. W. 1975. BioSystems 6, 153-164. Plenchette, C., Furlan, V., and Fortin, J. A. 1981. Can. J . Bot. 59, 2003-2008. Plenchette, C. 1982. fytoprocr. 63, 86-108. Plenchette, C., Fortin, 1. A., and Furlan, V. 1983a. Plant Soil 70, 199-209. Plenchette, C., Fortin, J. A., and Furlan, V. 1983b. Plant Soil 70, 211-217. Pope, P. E., Chaney, W. R., Rhodes, J. D., and Woodhead, S . H. 1983. Can. J. Bot. 61,412-417. Possingham, J. V.,Groot Obbink, J., and Jones, R. K. 1971. J. Aust. Insf. Agric. Sci. 37, 160-161. Postgate, J. R., and Hill, S. 1979. In “Microbial Ecology” (J. M. Lynch and N. J. Poole, eds.), pp. 191-2 13. Blackwell, Oxford. Powell, C. L. 1975. Newfhyrol. 75, 563-566. Powell, C. L. 1976a. Trans. Br. Mycol. Soc. 66, 439-445. Powell, C. L. 1976b. Nature (London)264,436-438. Powell, C. L. 1977a. N.Z. J. Agric. Res. 20, 59-62. Powell, C. L. 1977b. N.Z. J. Agric. Res. 20, 343-348. Powell, C. L. 1979. N e w f h y t o l . 83, 81-85. Powell, C. L. 1980a. N . Z . J. Agric. Res. 23, 225-231. Powell, C. L. 1980b. N.Z. J. Agric. Res. 23, 219-223. Powell, C. L. 1980~.Soil Biol. Biochem. 12, 247-250. Powell, C. L. 1982. flanr Soil 68, 3-9. Powell, C. L., and Daniels, J. 1978. New fhytol. 80, 351-358. Powell, C. L., Groters, M.,and Metcalfe, D. 1980. N.Z. J . Agric. Res. 23, 107-109. Raj, J., Bagyaraj, D. J., and Manjunath, A. 1981. Soil Biol. Biochem. 13, 105-108. Rangeley, A., Daft, M. J., and Newbould, P. 1982. New Phyrol. 92, 89-102. Ratnayake, M., Leonard, R. T., and Menge, J. A. 1978. New fhyrol. 81, 543-552. Raven, J. A., and Smith, F. A. 1976. New Phyrol. 76, 415-431. Raven, J. A,, Smith, S. E., and Smith, F. A. 1978. Trans. Bor. SOC.Edinburgh 43, 27-35. Read, D. J. 1983. Can. J. Bor. 61, 985-1004. Read, D. J., and Stribley, D. P. 1975. In “Endomycomhizas” (F. E. Sanders, B. Mosse, and P. B. Tinker, eds.), pp. 105-117. Academic Press, New York and London. Read, D. J., Koucheki, H. K., and Hcdgson, J. 1976. New Phyrol. 77, 641-653.

MYCORRHIZAS IN NODULATING N-FIXING PLANTS

53

Redente. E. F., and Reeves, F. B. 1981. Soil Sci. 132, 410-415. Redhead, J. F. 1975. In “Endomycorrhizas” (F. E. Sanders, B. Mosse, and P. B. Tinker, eds.), pp. 469-484. Academic Press, New York and London. Rhodes, L. H. 1980. Outlook Agric. 10, 275-281. Rhodes, L. H., and Gerdemann, J. W. 1975. New Phytol. 75, 555-561. Rhodes, L.H., and Gerdemann, J. W. 1980. In “Cellular Interactions in Symbiosis and Parasitism” (C. B. Cook, P. W. Pappas, and E. D. Rudolph, eds.), pp. 173-198. Ohio State Univ. Press, Columbus. Robson, A. D. 1978. In “The Mineral Nutrition of Legumes on Tropical and Subtropical Soils” (C. S. Andrew and E. J. Kamprath, eds.), pp. 277-293. CSIRO, Melbourne. Robson, A. D., O’Hara, G. W., and Abbott, L. K. 1981. Aust. J. Plant Physiol. 8, 427-436. Rose, S. L. 1980. Can. J . Bot. 58, 1449-1454. Rose, S. L. 1981. Can. J. Bor. 59, 1056-1060. Rose, S. L., and Trappe, J. M. 1980. Mycotaxon. 10, 413-420. Rose, S. L., and Youngberg, C. T. 1981. Can. J. Bot. 59, 34-39. ROSS,J. P. 1971. P h y t o ~ a t h o l o61, ~ ~ 1400-1403. Ross, J. P., and Gilliam, J. W. 1973. Soil Sci. SOC. Am. Proc. 37, 237-239. Ross, J. P., and Harper, J. A. 1970. Phytopathology 60, 1552-1556. Ross, J. P., and Harper, J. A. 1973. J. Elishu Michell Sci. SOC. 89, 1-3. Safir, G. R. 1980. In “The Biology of Crop Productivity” (P. S. Carlson, ed.), pp. 231-252. Academic Press, New York. Safu, G. R. 1981. FAOIIAEA Consult. Meet. Use Isotopes Stud. Nutrient Availability Food Crops Edomycorrhizae, Vienna. Safir, G. R., Boyer, J. S., and Gerdemann, J. W. 1972. Plant Physiol. 49, 700-703. Saif, S. R. 1981. New Phytol. 88, 649-659. Saif, S. R., and Khan, A. G. 1977. Plant Soil 47, 17-26. Samuel, G. 1926. Trans. Proc. R. SOC.South Aust. 50, 245-246. Sanders, F. E. 1975. In “Endomycomzas” (F. E. Sanders, B. Mosse, and P. B. Tinker, eds.), pp. 261-276. Academic Press, New York and London. Sanders, F. E., and Tinker, P. B. 1971. Nature (London) 233, 278-279. Sanders, F. E., Tinker, P. B., Black, R. L. B., and Palmerley, S. M. 1977. New Phyrol. 78, 257-268. Sanni, S. 0. 1976. New Phytol. 77, 667-671. Scannerini, S., and Bonfante, P. 1983. Can. J. Bot. 61, 917-943. Scannerini, S., Bonfante, P. F., and Fontana, A. 1975. In “Endomycorrhizas” (F. E. Sanders, B. Mosse, and P. B. Tinker, eds.), pp. 213-324. Academic Press, New York and London. Schenck, N. C., and Hinson, K. 1973. Agron. J. 65, 849-850. Schenck, N. C., and Kellan, M. K. 1978. Agric. Exp. Stn. Tech. Bull. No. 798. Schenck, N. C., and Smith, G. S. 1981. Proc. Soil Crop Sci. SOC. Flu. 40, 171-175. Schonbeck, F. 1979. In “Soil-Borne Plant Pathogens” (B. Schippers and W. Gams, eds.), pp. 271-280. Academic Press, New York and London. Schramm, J. R. 1966. Trans. Am. Phil. SOC. 56, 1-194. Shanmugam, K. T., O’Gara, F., Anderson, K., and Valentine, R. C. 1978. Annu. Rev. Plant Physiol. 29, 263-276. Shemakhanova, N. M. 1972. “Mycotrophy of Woody Plants.” US Dep. Agric. Natl. Sci. Found., Washington, D.C. Shukla, U. C., and Yadav, 0. P. 1982. Plant Soil 65, 239-248. Skipper, H. D., and Smith, G. W. 1979. Plant Soil 53, 559-563. Slankis, V. 1974. Annu. Rev. Phytopathol. 12, 437-457. Smith, F. A., and Smith, S. E. 1981a. New Phytol. 88, 311-325.

54

J. M. BAREA AND C. AZC6N-AGUILAR

Smith, F. A., and Smith, S. E. 1981b. New Phytol. 88, 299-309. Smith, S. S. E. 1980. Biol. Rev. 55,475-510. Smith, S . E. 1982. New Phytol. 90, 293-302. Smith, S. E., and Bowen, G. D. 1979. Soil Eiol. Biochem. 11, 469-473. Smith, S. E., and Daft,M. J. 1977. Aust. J . PInnr Physiol. 4,403-413. Smith, S . E., and Daft,M. J. 1978. In “Microbial Ecology” (M. W. Loutit and J. A. R. Miles, eds.), pp. 312-319. Springer-Verlag, New Yo&. Smith, S. E., Nicholas, D. J. D., and Smith, F. A. 1979. Aust. J . Plant Physiol. 6, 305-311. Sondergaard, M., and Laegaard, S. 1977. Nature (London) 268, 232-233. Sparling, G. P., and Tinker. P. B. 1978. J. Appl. Ecol. 15, 959-964. St. John, T. V. 1980. N ~ wPhytol. 84, 483-487. St. John, T. V., andColeman, D. C. 1983. C u n J . Eor. 61, 1005-1014. Stotzky, G. 1972. CRC Crit. Rev. Microbiol. 2, 59-137. Stribley, D. P., Tinker, P. B., and Rayner, J. H. 1980. New Phytol. 86, 261-266. Strullu, D. G. 1982. Trans. Int. Col. Mycorrhizes Eiol. Perspectives Utilisation, Dijon. Stnemska, J. 1975. In “Endomyconbizas” (F. E. Sanders, B. Mosse,and P. B. Tinker, eds.), pp. 537-543. Academic Press, New York and London. Sutton, J. C., and Sheppard, B. R. 1976. Cun. J. Eor. 54, 326-333. Tardieux-Roche, A. 1966. Ann. Agron. 17,403-479. Thimann, K. V. 1977. “Hormone Action in the Whole Life of Plants.” Univ. of Massachusetts Press, Amherst. Thomazini-Casagrande, L. I. 1980. Fyton. 39, 113-120. Tien, T. M., Gaskins, M. H.,and Hubbell, D. H. 1979. Appl. Environ. Microbiol. 37, 1016-1024. Tinker, P. B. 1975. In “Endomycomhizas” (F. E. Sanders, B. Mosse, and P. B. Tinker, eds.), pp. 353-371. Academic Press, New York and London. Tinker, P. B. 1978. Physiol. Veg. 16, 743-751. Tinker, P. B. 1980. In “The Role of Phosphorus in Agriculture,” pp. 617-653. Amer. Soc. Agron., Madison, Wisconsin. Tinker, P. B., and Sanders, F. E. 1975. Soil Sci. 119, 363-368. Tisdall, J. M., and Oades, J. M. 1979. Aust. J. Soil Res. 17, 429-441. Torrey, J. G. 1976. Annu. Rev. Plant Physiol. 27,435-459. Torrey, J. G. 1978. EioScience 28, 586-592. Trappe, J. M. 1977. Annu. Rev. Phytoputhol. 15, 205-222. Trappe, J. M. 1979. In “Symbiotic Nitrogen Fixation in the Management of Temperate Forests” (J. C. Gordon., C. T. Weeler, and D. A. Perry, eds.), pp. 276-286. Oregon State Univ. For. Res. Lab., Corvallis. Trappe, J. M. 1981. In “Advances in Food Producing Systems for Arid and Semi-Arid Lands” (J. T. Manassah and E. J. Briskey, eds.), pp. 581-599. Academic Press, New York. Trappe, J. M., and Fogel, R. C. 1977. Range Sci. Dep. Sci. Ser. (Color. State Univ.) 26,205-214. Trinick, M. J. 1977. New Phytol. 78, 297-304. van Schreven, D. A. 1958. In “Nutrition of the Legumes” (E. Hallsworth, ed.), pp. 137-163. Buttenvorth, London. Waidyanatha, U. P., Yogaramam, N., and Ariyaratne, W. A. 1979. New Phytol. 82, 147-152. Warner, A., and Mosse, B. 1980. Trans. Er. Mycol SOC. 74,407-410. Warner, A., andMosse, B. 1982. New Phytol. 90, 529-536. Whittingham, J., and Read, D. J. 1982. New Phytol. 90, 277-284. Wilson, J. M., and Trinick, M.J. 1983. Aust. J . Soil. 21, 73-81. Zak, J. C., and Parkinson, D. 1983. Can. J . Eot. 61, 798-803.

ADVANCES IN AGRONOMY, VOL. 36

SUBMICROSCOPIC EXAMINATION OF SOILS E. B. A. Bisdom Netherlands Soil Survey Institute Wageningen, The Netherlands

I. Introduction .......................................................... 11. Submicroscopic Techniques ............................................. A. Electron Microscopy.. ................. ........ B . Ion Microscopy.. ..................... ..... C. Additional Submicroscopy. ......................................... 111. Applications of Electron Microscopy ..................... A. Unhardened Samples .............................................. B. Thin Sections .................................................... IV. Applications of Ion Microscopy ................ V. Applications of Other Forms of Submicroscopy ...................... VI. Conclusions .......................................................... References .................. ....................................

55 57 57 61

62 65 66 77 88 89

90 91

1. INTRODUCTION Submicroscopy of soils usually begins after light microscopy has been done, and light microscopy itself often supports field studies. This article primarily describes soil materials studied in thin sections, soil peds, and 1arger.mineral grains. Submicroscopy may be regarded as a young field in soil science, although instruments such as the transmission electron microscope (TEM) and the electron microprobe analyzer (EMA) have been in use for more than a decade. In situ light-microscopic studies of soils are usually performed by micromorphologists representing the field of soil micromorphology. For technical reasons, this is itself a relatively young branch of soil science; soil micromorphology could only develop after thin sections had been prepared (i.e., after development of the technique of the impregnation of samples with plastics necessary for their hardening). This technical difficulty did not exist in geological studies of hardrock and, as a consequence, thin-section studies of cohesive rocks were already being made many years before light-microscopic studies of soils were possible. However, as soon as thin sections of soils were feasible, these 55

Copyrighl 0 by Academic F’ress, Inc. All rights of reproduction in any form reserved.

ISBN 0-12-000736-3

56

E. B. A. BISDOM

techniques were also applied to weathered rocks and uncohesive geological deposits. Thin-section studies of soils with the light microscope indicated that available magmfkations were frequently insufficient to study the finer soil particles. The TEM was initially used to investigate especially clayey material that had been pretreated and was disturbed. In sifu studies using the TEM were first made on replicas, and it has recently become possible to examine ultrathin sections of soft soil constituents. Such in sifu studies became popular with the introduction of the scanning electron microscope (SEM) (i.e., with the potential to study soil constituents in soil peds and to examine mineral grains with or without coatings). “Three-dimensional” pictures were obtained and the morphologies of the soil constituents were studied at various magnifications. The investigation of very small particles in soils requires TEM and scanning transmission electron microscope (STEM) investigation of ultrathin sections that are transparent to the electron beam. Such ultrathin sections of organic matter and clays have been prepared, but those from harder soil materials are still difficult to obtain. Ultrathin sections also offer the possibility of electron diffraction of individual soil particles instead of X-ray diffraction of bulk samples that have been pretreated and disturbed. Submicroscopy with the SEM, especially after the introduction of new detector systems, can be used to study the form of pores and minerals in a thin section if used in combination with equipment for image analysis. So far, however, thin sections have been used more often to investigate chemical elements of soil particles. Such microchemical analyses started with the EMA [also called EPMA (electron probe microanalyzer)]. This instrument, as do most of the other instruments used in submicroscopy, requires a polished surface of a thin section or the surface of a thin polished block. More recently, the SEM has been equipped with an energy dispersive X-ray analyzer (EDXRA). This SEM-EDXRA allows in situ analysis of soil materials in unhardened soil components and in thin sections. A scanning electron microscope equipped with a wavelength dispersive X-ray analyzer (SEM-WDXRA) can only analyze chemical elements in polished surfaces. Two problems remained with EMA and SEM-EDXRA-WDXRA analyses of soil components; the lightest elements of the periodic system of chemical elements and trace elements could not be studied using electron microscopy. These problems were solved with the introduction of ion microscopy. The instruments involved use either primary ions for the excitation of secondary ions from materials in thin sections of soils [e.g., ion microprobe mass analyzer (IMMA) and Cameca IMS 3F (ion microanalyzer)] or a laser for the excitation of primary ions [e.g., LAMMA 500 and LAMMA lo00 (laser microprobe mass analyzers)]. Quantification of all chemical elements in soil constituents of a thin section became possible with the Cameca IMS 3F.

SUBMICROSCOPIC EXAMINATION OF SOILS

57

The technical possibilities of in situ submicroscopic research are significant. This young branch of soil science, together with light microscopy, allows us to obtain knowledge on soil components in their natural environment and at a variety of magnifications. Several examples of submicroscopic studies will be given after discussing the capabilities of a number of machines.'

11. SUBMICROSCOPIC TECHNIQUES A. ELECTRON MICROSCOPY

I . Transmission and Scanning Transmission Electron Microscopy The transmission electron microscope is commonly used for the study of loose and very small particles in soils using a static electron beam. A lateral resolution of 0.2 nm can be achieved (Boekestein et al., 1981). If ultrathin sections can be made with a thickness of about 1 pm, the TEM can be used because the electrons can pass through the specimen. Ultrathin sections of clayey and of organic material were prepared and studied with the TEM by Bresson (198 1) and Foster (1981), respectively. The high-voltage electron microscope (HVEM) uses a much higher kinetic energy than the TEM, about lo00 keV compared to 20-200 keV. The HVEM can examine a specimen somewhat thicker than can the TEM but so far no results have been published in soil micromorphology submicroscopy. The polyester resin used for the embedding of soil material is expected to become brittle with high voltage electron microscopy. Samples somewhat thicker than 1 pm can also be studied with the STEM. The STEM (Fig. 1) can be used as a TEM, SEM, or STEM. The instrument makes it possible to study an ultrathin section by a scanning beam. Microchemical analysis is also possible with a TEM or an STEM that has a 'Abbreviations used in this paper: A E S , Auger electron spectroscopy; BESI, backscattered electron scanning image; EDXRA, energy dispersive X-ray analyzer (analysis); EMA, electron microprobe analyzer (analysis); EPMA, electron probe microanalyzer (analysis); ESCA, electron spectroscopy for chemical analysis; HREM, high resolution electron microscope (microscopy); HVEM high voltage electron microscope (microscopy); IMMA, ion microprobe mass analyzer (analysis);LAMMA, laser microprobe mass analyzer (analysis); LMA, laser microspectral analyzer (analysis); RS, Raman spectroscopy; SEM, scanning electron microscope (microscopy);SIMS, secondary ion mass spectrometer (spectrometry); STEM, scanning transmission electron microscope (microscopy); TEM, transmission electron microscope (microscopy); WDXRA, wavelength dispersive X-ray analyzer (analysis); XRD, X-ray diffraction.

58

E. B. A. BISDOM

FIG. 1. Scanning transmission electron microscope (Philips EM 400T/ST).

lateral resolution of about 5 nm. The instruments can be equipped with an EDXRA. Analysis of the heavier chemical elements in ultrathin or somewhat thicker thin sections thus becomes possible. Because of beam spot diameters that are smaller than those present in conventional SEM instruments, it is possible to analyze at magnifications which are larger than those possible with an SEMEDXRA (i.e., larger than X l0,OOO). Very small particles can be analyzed and identified with the TEM and STEM if equipment for electron diffraction is available. This technique is usually applied to study loose clay minerals but can also be used to study small particles in ultrathin sections of soils. Work has been done with the STEM-EDXRA on thin sections (5 pm thick) and at a maximum magnification of X50,OOO. Work with the TEM-EDXRA and STEM-EDXRA on ultrathin sections, and diffraction

SUBMICROSCOPIC EXAMINATION OF SOILS

59

studies of these specimens, must yet be done. The principal difficulty is that we still must learn how to prepare ultrathin sections of harder soil materials. Ionthinning techniques seem to give the best results at present. If ultrathin sections have been prepared, TEM and STEM instrumenis are available for various types of studies on an ultramicro scale.

2 . Electron Microprobe Analysis and Scanning Electron Microscopy The electron microprobe analyzer is the oldest machine used for microchemical analysis of soil materials in polished thin sections. It is used for microanalysis only (i.e., for semiquantitative and quantitative measurements using a set of standards with which to compare the results of the analyses). Older EMA instruments often caused localization problems for materials in a thin section of soil. Modem machines, however, can be equipped in such a way that one can find soil components in thin sections with relative ease, which is necessary for heterogeneous soils with fine particles and complicated fabrics. The EMA is equipped with a WDXRA system. Wavelength dispersive analysis is done with a WD detector which contains a crystal that is used for Bragg reflection and a gas-filled proportional counter (Boekestein et al., 1981). Only one element can be measured at a time, unless more detectors are used. The WD detector has a high efficiency, because of the thin entrance window of the proportional counter, and a high peak-to-background ratio. This ratio is important because element-characteristic radiation is represented by the peaks and noncharacteristic radiation is represented by the background. The detectable elements are B-U (22 5 ) . The maximum magnification of the EMA is about X500, which can be a problem (Bisdom et al., 1975, 1976); the lateral resolution is about 1 pm. The scanning electron microscope (Figs. 2 and 3) can also be equipped with a WDXRA system, which allows magnifications up to X10,OOO. The SEMWDXRA and the SEM-EDXRA have lateral resolutions of about 1 pm, as does the EMA. This means that the minimum diameter of a spot that is analyzed in a thin section is 1 pm. The SEM-WDXRA, like the EMA, can only work with polished surfaces, whereas both polished and rough surfaces can be examined with the SEM-EDXRA. Consequently, materials in soil peds are now investigated microchemically, not only on the basis of morphology, by using the SEM. An additional advantage of the EDXRA is that the current of the primary electron beam on the specimen is lo-" A, whereas it is lo-' A for the EMA. Consequently, in EMA the polyester resin of the thin section is easier to damage than in SEM. Energy dispersive X-ray analysis utilizes an ED detector which consists of a lithium-drifted silicon crystal. If the detector has a beryllium window, very soft

E. B. A. BISDOM

FIG. 2. Scanning electron microscope (Jeol-JSM-35C).

FIG. 3. Scanning electron microscope (Philips SEM 505).

SUBMICROSCOPIC EXAMINATION OF SOILS

61

X rays are absorbed and the characteristic radiation of elements with low atomic numbers is not detected. Elements Na-U (22 11) are detected in this way. If an ECON detector is used without a beryllium window, the radiation of C, N, 0, and F can also be measured. The ED detector has very small processing times and can give information on a range of elements simultaneously. However, the energy resolution is rather poor, which affects the peak-to-background ratio and the minimal detectable concentration. Ideally, SEM-EDXRA is used for reconnaissance and semiquantitativework and EMA and SEM-WDXRA for quantitative and semiquantitative work. Trace elements are usually not measurable with electron microscopes. However, under ideal conditions, lo-'* g of an element [approximately 0.1%can be measured (Boekestein et al., 1981)]. B. ION MICROSCOPY

Various instruments for the analysis of secondary ions, excited from the sample by primary ions, have been tested on soil samples, including the IMMA of ARL, Cameca IMS 300 (ion microscope), Cameca IMS 3F (Fig. 4), and LAS of Riber [an apparatus in which SIMS, ESCA, and Auger (see later discussion) analysis can be done]. Only polished thin-section material which has been removed from the support glass can be used in these instruments. The IMMA uses a scanning primary ion beam and a mass spectrometer for mass analysis of sputtered ions (Bisdom et al., 1977). The primary electron beam of the electron

FIG. 4, Ion microanalyzer (Cameca IMS 3F).

62

E. B. A. BISDOM

microscope has been replaced in the IMMA by a primary ion beam and secondary ions are produced instead of secondary electrohs. In the IMMA the sample can be viewed during analysis through a binocular microscope (Henstra et af., 1981a), whereas this is not possible with the Cameca instruments. Localization is done in the latter machines with a low-power optical microscope. A viewport is present in the LAS series of instruments. All four instruments are used for secondary ion mass spectrometry (SIMS). Such spectra of secondary ions give information on all chemical elements that are present in a sample including hydrogen. Both trace and major elements can be measured. Background problems, such as are present in electron microscopy, are virtually absent. Background readings are usually below 5 counts/sec, with total count rates on the order of lo8 counts/sec (Liebl, 1975). Trace concentrations can therefore be analyzed, usually down to the range and in many cases even down to the range. Trace amounts (10-I8 g) of sample material are measurable. The sentitivity of SIMS is much better than that of X-ray analytical techniques (i.e., 1OOO-10,OOO times) (Henstra er al., 1981a). All elements can be detected with SIMS but the secondary ion yield differs for various elements. Also, the same element in a different matrix may give a different secondary ion yield. The secondary ion yield of the sample can be strongly enhanced by bombarding with a reactive species such as oxygen or nitrogen. The primary ions used for bombarding the sample can be charged either positively or negatively. The sample is continuously eroded under ion bombardment; consequently, the determination of concentration as a function of depth is important. Depth-concentration profiling with a resolution of about 5 nm is possible. Probe diameters on the surface of the sample range from 500 to 1-2 FmIsotopic analysis is possible in SIMS, and isotopic abundance ratios can be measured with high accuracy. This should allow in siru age dating of materials in thin sections of soils and in horizons of soil profiles. Another possibility is to label chemical elements with stable isotopes, which should allow the study of transport phenomena in soils by measuring the position of these labeled isotopes in thin sections. Secondary ion mass spectrometry offers a large variety of measurement possibilities which can be added or compared to those of electron microscopy. As ion microscopy is a younger field than is electron microscopy, various measurement techniques are still being developed. Experiments have indicated, however, that quantitative and semiquantitative measurements of soil materials in thin sections of soils are now possible (see Section IV). C. A D D ~ O N ASUBMICROSCOPY L

Various additional submicroscopic techniques are available, but only a few (those which seem to be the most promising for research of soil materials) have

SUBMICROSCOPIC EXAMINATION OF SOILS

63

been tested so far, namely, Raman spectroscopy (RS), laser microprobe mass analysis (LAMMA), electron spectroscopy for chemical analysis (ESCA), and Auger electron spectroscopy (AES). Raman spectroscopy has been tested for soil materials by Jeanson (1981). Illumination of the soil sample is done with a laser beam. The Raman spectrum is obtained when monochromatic light is passed through a transparent substance. The light is scattered by the transparent substance and undergoes energy transformations. The frequency of some of the light is therefore changed, resulting in the addition of certain lines to the Raman spectrum. The lines of the spectrum are thus characteristic of the molecular structure of the examined area in a thin section or soil ped. An important aspect of RS is that it is nondestructive. Laser microprobe mass analysis, which is destructive, has been done with the LAMMA 500 of Leybold-Heraeus (Henstra et al., 1980a; Bisdom et al., 1981). An optical microscope is used to focus a high-power pulsed laser onto an area of the thin section or soil ped less than 1 pm in diameter. A microvolume of about 10- '*-lo- l 4 cm3 is evaporated and ionized. These ions are detected by a mass spectrometer, also used in ion microscopy. The difference between the two methods is that excitation of the ions takes place with a laser beam in LAMMA and with primary ions in ion microscopes. The LAMMA 500 was made for the investigation of ultrathin specimens of 0.1- 1 pm, and the light optical and ion detection systems were therefore placed on opposite sides of the piece of ultrathin section. Thicker thin sections of about 15 pm could therefore only be analyzed by applying laser milling, in which the laser shots evaporate soil material from the edges of the piece of thin section inward. So far, in the LAMMA lo00 (Fig. 5), the optical and ion systems are placed on the same side of the sample. Consequently, our common thin sections can now remain on their support glass during analysis. Electron spectroscopy for chemical analysis uses X rays or uv photons to irradiate materials in thin sections (Henstra et al., 1981b). Ultraviolet and X-ray electromagneticradiation can be used to excite outer- or inner-shell electrons and this causes the ejection of electrons (McCrone and Delly, 1973). Ultraviolet radiation, with its long wavelength and lower energy, can eject the outermost electrons, whereas X-ray radiation, which has a short wavelength and higher energy, can eject inner-shell electrons. The energies of the ejected electrons can be used to distinguish pure elements and elements in a bonded state; the difference can be observed as different peaks in an energy spectrum. Electron spectroscopy for chemical analysis is mainly used for chemical bonding studies in situ. Tables by Wagner el al. (1979) are available for ESCA studies; all elements except hydrogen can be studied. The depth of analysis is 2-10 nm, and the minimal detectable concentration is about 0.1%. Electron spectroscopy for chemical analysis will usually succeed for soil materials that are homogeneous over fairly great distances; the lateral resolution of analysis is about 3 mm, which

64

E. B. A. BISWM

FIG.5. Laser microprobe mass analyzer (LAMMA 10o0, bybold-Heraeus).

FIG. 6. AES (Auger electron spectroscopy), ESCA (electron spectroscopy for chemical analysis), and SIMS (secondary ion mass spectrometry) are possible with the LAS 3000 (Riber).

SUBMICROSCOPIC EXAMINATION OF SOILS

65

is too large for heterogeneous samples with small particles. Bemer and Holdren (1977) have used this technique for the study of weathered feldspar. Auger electron spectroscopy is a typical surface analytical technique. Primary electrons are used to obtain Auger electrons from the sample. The energy of the Auger electrons is recorded in an energy spectrum which can be used for quantitative and qualitative chemical analysis of soil materials up to a depth of 1-2 nm below the surface of the thin section. All elements except H and He can be analyzed, and the lateral resolution of analysis is about 0.1 pm. The minimal detectable concentration is about 0.1%(Henstra et al., 1981b). Experiments with LAS (see section II,B) gave results on iron-coated organic material from the Netherlands, but no information was obtained on nonconductive clayey material. The LAS instrument (Fig. 6) was able to perform SIMS and ESCA analyses.

Ill. APPLICATIONS OF ELECTRON MICROSCOPY Applied electron microscopy has been subdivided into studies of unhardened samples (Section II1,A) and thin sections (Section II1,B) because most of the in situ submicroscopic literature describes unhardened materials in soil peds. These studies were usually performed with an SEM that had no equipment for microchemical analysis. The main goal was the study of the morphology of soil particles in the peds and the arrangement of the soil into certain fabric patterns. Thin sections were used for microchemical analysis with the EMA, an instrument which was specifically built for this purpose. Analysis by the SEM-EDXRA and the SEM-WDXRA followed later. Currently, microchemical analyses can also be done in soil peds with the SEM-EDXRA, and the morphology of soil particles and certain types of fabrics can also be studied in thin sections. It remains true, however, that quantitative analysis of the chemical elements in soil constituents requires a polished surface and must be done with an EMA or an SEM-WDXRA. The most impressive three-dimensional morphology of soil components is found in unhardened soil peds using the SEM. Several review articles have been written on electron microscopy as applied to soils. The use of TEM, SEM, and EMA in pedology was discussed by Bocquier and Nalovic (1972). The use of light microscopy, TEM, and SEM in micropedology was treated by Stoops (1974). A number of submicroscopic techniques which can be applied to soil micromorphology were given by Smart (1974), and details of TEM and SEM techniques as applied to soils and sediments are discussed by Smart and Tovey (1981, 1982). Published studies using SEM and EMA were indicated by Bisdom et al. (1976). Two review papers have been published in which TEM, SEM, SEM-EDXRA-WDXRA, EMA, and nonelectron microscopic work on thin sections of soils (i.e., ion microscopy, laser

66

E. B. A. BISWM

analysis, and electron spectroscopy for chemical analysis) are discussed (Bisdom, 1981a,b) .

1. General

The submicroscopic study of unhardened soil samples can be done by TEM,

STEM,and SEM. The TEM and STEM can give magnificationsover X 1,OOO,OOO

depending on the type of soil particles that are studied. The TEM and STEM are usually used for very small soil particles that are present in ultrathin sections or in pretreated and disturbed samples. Ultrathin sections are discussed in Section III,B, but pretreated and disturbed samples form no significant part of this article. The SEM can reach magnifications of more than X100,OOO. The maximum magnification is again dependent on the type of soil particle that is investigated. The SEM is an ideal instrument for three-dimensional studies of soil constituents and therefore it is frequently used for morphological examination. Much attention has also been paid by specialists in soil mechanics and soil microscopy to the spatial relationships between individual constituents in soil peds.

2 . Clay Minerals Individual clay minerals in soil peds or aggregates are usually difficult to recognize with the SEM because they commonly form stacks that can be partly or wholly coated with other fine soil constituents. X-raypowder diffractograms of bulk and disturbed samples and TEM studies of pretreated and disturbed individual clay minerals are usually performed simultaneously with SEM studies of materials in soil peds. Keller (1976a,b,c, 1977a,b, 1978a) and Keller and Haenni (1978) studied kaolinite in various deposits around the world and were able to classify these deposits into transported and residual types on the basis of texture differences found in scanning electron micrographs. Gillott (1974) and Tessier and Berrier (1978) recognized that the in situ investigation of clay minerals in soil peds required ‘special preparation techniques such as freeze-drying or critical-point drying if air-drying does not give the required results. Smart and Tovey (1982) discuss these and other techniques for electron-microscopic work. Spherulitic halloysite in volcanic deposits was examined with the SEM and TEM by Sudo and Yotsumoto (1977) and Violante and Violante (1977). Differences in shape and mineralogical properties were found to exist between the spherulitic halloysite bodies. Sudo and Yotsumoto (1977) called the bodies “chestnut-shell-like” on a morphological basis and “allophane-halloysite-

SUBMICROSCOPIC EXAMINATION OF SOILS

67

spherules” on a genetic basis (i.e., halloysite formed from allophane which had originated from volcanic glasses). Violante and Violante (1977)explained that the spherulitic halloysite possibly may have formed inside vitreous bubbles as a result of processes exerted by surrounding minerals. Palygorskite in acid-etched carbonate nodules was studied with the SEM by Yaalon and Wieder (1976)and in calcareous crusts by Nahon et al. (1975).The morphology of allophane, immogolite, and halloysite in volcanic ash soils was studied with the SEM by Eswaran (1972). Various scanning electron micrographs of clays were presented by Smart and Tovey (1981).Scanning electron microscopy-energy dispersive X-ray analysis has obtained information on chemical elements present in clay of soil peds. The maximum magnification at which this is possible is XlO,OoO, and the analysis can be performed on a l-Fmdiameter spot. 3 . Weathered Minerals

Much SEM work has been done on various types of weathering minerals. Although clay minerals also weather, most such studies concern the larger primary minerals such as feldspar, olivine, mica, and quartz. Minerals like feldspar, mica, and quartz can also form a part of the clay fraction of a soil, but it is the larger particles which are studied because they can be compared with the results obtained from light-microscopic investigation of the same or similar samples. a. Feldspars. Scanning electron micrographs of feldspars weathering to halloysite and kaolinite have been published by Eswaran and de Coninck (1971). The feldspars weathered to halloysite in an Entisol and to kaolinite in an Ultisol. Feldspar altered to kaolinite and gibbsite in a granite profile from Malaysia. No intermediary crystalline or amorphous phase was found during such weathering, whereas the amorphous phase was present during the transformation of feldspar into halloysite (Eswaran and Wong, 1978). Weathered feldspar in decomposing basalt was photographed with the SEM by Benayas and Alonso (1978).In Israel, weathering of plagioclase gave halloysite pseudomorphs in the vesicularly weathered basalt and smectite pseudomorphs in the saprolite profiles (Singer,

1973). Feldspars in Scottish soils showed holes and pits due to continuous dissolution and etching (Wilson, 1975). Such holes and pits were thought to have originated where crystal dislocations met the surface of the feldspars. No residual layer was observed at the boundary between the unweathered feldspar and a void or crack. The existence of such a residual layer was found to be unlikely. Experimental etching of a microcline perthite (Wilson and McHardy, 1980) confirmed that etch marks developed along crystal dislocations emerging on the surface (i.e., dislocations associated with perthitic lamellae). Analyses with ESCA (see Section I1,C) by Berner and Holdren (1977)of surface layers of feldspars c o n f i e d

68

E. B. A. BISDOM

that weathering occurs along dislocations, cleavages, and fractures (i.e., a residual layer, which requires an equal rate of attack on all parts of a feldspar grain, is not necessary and probably does not exist). Pitted feldspar in an altered rock fragment was thought to be the result of dissolution (Taupinard, 1976), whereas Keller (1978b) explained this pitting as the result of uneven dissolution and nonuniformity in composition. Millot et al. (1977) indicated that such pits in feldspar could originate when secondary calcite replaced feldspar, a process which was called epigenesis. b. Quartzes. Many SEM studies concern the morphology of weathered quartz grains. If the degree of weathering of individual quartz grains can be assessed, it is possible to use such information to deduce the developmental history of individual horizons in a soil profile. Legigan and Le Ribault (1974) studied the evolution of quartz in a humic and fermginous podzol in France that was developed in aeolian sands. Surface features of the quartz grains were related to the sedimentary and pedological history of the profile. Well-polished surfaces indicated transport in streams, whereas polished surfaces found with shock imprints indicated a fluviatile or wind-transported origin. Striae with a certain density on the surfaces of quartz grains were interpreted as being formed by the rubbing of quartz grains against each other during glacial activity. If the quartz grain had dissolution figures on its surface or an iron crust with or without organic matter, it was thought to be caused by pedogenesis. This type of approach permitted the indication of various environments in the studied profile and also helped to unravel the history of the sands. Eswaran and Stoops (1979) worked with a zero phase in a weathering sequence of quartz, established in a Xerochrept formed on Keuper marls in Spain. The quartz crystals were idiomorphic to hypidiomorphic. The surface textures of quartzes were studied in a 19-m-deep profile developed on granite in a tropical environment. Weathering of the quartz started a few centimeters above the fresh rock with fragmentation of the quartz grains and the presence of hairline cracks in the weathered quartz. Etching of the quartz grains occurred at a depth of 18.5 m and the quartz showed large dissolution pits that were interconnected by grooves and hairline cracks. Some idiomorphic secondary quartz was precipitated between the depths of 9.5 and 16 m on the surface of heavily etched primary quartz grains. Triangular dissolution pits were developed in primary quartz grains at a depth of about 9.5 m and heavily etched quartzes with linear striations were found at a depth of 1 m. These linear striations differed from the etch grooves found on the surfaces of quartzes at greater depths in the profile. The surfaces of quartz grains from Neogene sands in the Ivory Coast were examined by Leneuf (1972). The quartzes came from depths of 3,30, and 90 m. Two classes of weathering figures were distinguished, one related to the crystal lattice of quartz and the other apparently not related to it. The first class comprised cavities with the same alignments; cavities with tetrahedral, rectangular,

SUBMICROSCOPIC EXAMINATION OF SOILS

69

irregular polyhedral, and wedge-shaped forms; fissures with a concentric outline; cubic figures in a regular network; and lines with a relief and at 30, 60, and 120 degrees. The second class contained irregular cavities, fissures related to desquarnation, vermiform fissures, fine particles on the surface of quartz grains, and newly formed secondary quartz from silicon which had passed through the profile. These surface features on the quartz grains indicated that silicon had been mobilized in the upper part of the strata and that only part of the silicon participated in the formation of kaolinite. Secondary quartz could form from the transported silicon at deeper levels in the profile. Scanning electron micrographs of quartz particles in surface soils of the Hawaiian Islands were studied by Jackson et ul. (1971). These soils developed over quartz-free mafic (basic) rocks. Wind deposition of the quartz was inferred by comparison of the sharp angular, chip- or shard-like morphology of the grains with that of quartzes in aerosolic dust and pelagic sediment. The percentage of quartz varied with the elevation of the soil, the age of the soil, and the amount and source of annual rainfall. Scanning electron microscope and X-ray diffraction (XRD) analyses of airborne particles indicated that the coarser ones, with radii of 10-100 pm, consisted predominantly of quartz, whereas the finer particles, with radii of 1-10 pm, were mainly clay minerals. The clay minerals were found in the air as constituents of aggregates, as coatings on quartz grains, and as individual platelets, and were derived from the soil by sandblasting. Riezebos (1974) studied weakly cemented Miocene sands of deposits from South Limburg, the Netherlands, with the SEM. Secondary quartz was found not only at grain contacts but also around detrital quartz grains. Overgrowth of secondary quartz on the larger grain surfaces formed steps and striations. Such steps and striations were therefore not the result of glacial environments. Douglas and Platt (1977) investigated the surface morphology of quartz and the age of soils in glacial material from Wisconsin. Quartz in late Pleistocene (Wisconsin) deposits was only slightly weathered with a mainly broad, flat or conchoidal breakage surface, and sharp or slightly rounded upturned plates. Quartzes in sands of Illinoian age showed both sharp and rounded upturned plates. Precipitation of secondary silica had occurred on the quartz grains and a modification of the surface morphology was the result. Some solution pits were also present. Corroded surfaces with solution Vs and highly rounded upturned plates were found to be associated with quartz grains of Kansan age. The rounded forms were caused by dissolution and precipitation of silica. Flaking was also found, representing intense chemical weathering. Moss and Green (1975) pointed out that the concept of deformation sheeting (i.e., forming plates, steps, etc. on the surface of quartz grains) probably is more realistic than explanations based on existing cleavages in quartzes. Attention was also paid to microfractures and the laminae of quartzes between them called “sheets.” Such a sheet of quartz, usually 2-20 Fm thick, was considered to be

E. B. A. BISDOM

70

the smallest weathering entity. Microfractures can subdivide the sheet into small-

er particles that are clay sized. In nature, however, quartz is frequently common in the 2- to 20-pm silt fraction and does not occur in a dominant form in the clay fraction. It was also pointed out by Moss and Green (1975) that quartz grains can already be well-rounded when they leave the source rock and that it is therefore unrealistic to always assume angular particles that gradually become more rounded with increasing maturity. Conversely, angular quartz can often be found in soils and sediments. Magaldi (1978) indicated that two contradictory interpretations exist, one which cites the more rounded and another that cites the more angular quartz grains during weathering. A cathodoluminescent (CL) study of quartz sand grains was made by Tovey and Krinsley (1980), who pointed out that the common secondary electron (SE) micrographs (emissive mode micrographs) do reveal surface information on the quartz grains but no subsurface information as seen in cathodoluminescent micrographs. The surfaces of quartz grains, cross sections of quartzes, etched grains, and heated quartzes were studied with the SE and CL modes. Cathodoluminescence is significantly affected by slight changes in the chemical composition of the quartz grain, and cracks that are not visible in the SE mode can often be recognized in the CL mode. Study of the spatial distribution of narrow and broader dark bands, of dark patches, and of other characteristics in the CL micrographs, together with information obtained from SE micrographs, allowed some insight into the various processes which affected the quartz grains. c. Micas. Scanning electron microscope studies of weathered micas are often done in combination with nonsubmicroscopic techniques. Jackson and Sridhar (1974) studied Li exfoliated and freeze-dried phlogopite flakes. Scanning electron micrographs indicated that the osmotic force and swelling created by Li+ resulted in the gliding out of interstratified saponite layers which became twisted and curled during this process. Saponite was formed from phlogopite with vermiculite as an intermediate. Gliding out of layers only occurred when salt was removed from the solution and electric double-layer swelling took place in distilled water during the experiments. Scanning electron microscopy allowed the study and portrayal of tracks and holes in micas (Lee et al., 1974). The tracks were produced by spontaneous fission of 238U under natural conditions (235U must be activated to give thermal neutron bombardment and induced fission particle tracks). Upon splitting of the uranium nucleus, large fragments can move through the micas with considerable energy and leave behind trails of damage called “tracks” which are about 20 pm long and have diameters of about 0.015 pm. These fission tracks play a role during the weathering of micas and also influence cation exchange capacity. Tarzi and Protz (1978) studied the weathering of micas obtained from rocks. Upon the start of weathering, the micas split at their edges and this process proceeds inward along planes. The exfoliated stage is reached when the layers +

SUBMICROSCOPIC EXAMINATION OF SOILS

71

become separated. During the exfoliation process bending may affect the individual layers which may then take various forms. Holes in the micas were thought to have been occupied by quartz and other minerals, rather than to have resulted from spontaneous fission processes as advocated by Lee et al. (1974). Secondary material could accumulate in the spaces provided by the weathering micas. Crusts could form in them and roots could penetrate the mica. Secondary micas are frequently observed in weathering micas. Verheye and Stoops (1975) made a scanning electron micrograph of kaolinite between biotite lamellae in a soil from the Ivory Coast. Illite was distinguished by Taupinard (1976) on weathering biotite flakes of an altering granite in France together with dissolution, new formation, and disaggregation features. Sousa and Eswaran (1975) found that large biotite flakes in a saprolite from Angola were pseudomorphically altered to goethite. Scanning electron microscope observations indicated that microdroplets of goethite covered the surfaces of weathered biotite. d. Other Minerals. Dissolution of olivine to deeply etched and pitted weathered olivine probably occurred at particular sites where structural dislocations emerged in the olivine, similar to weathering feldspars (Wilson, 1975). This weathering mechanism was confirmed during experimental studies by Grandstaff (1978). The initial dissolution of freshly crushed olivine was where lattice imperfections occurred (e.g., dislocations and cleavage planes). Pits and rounded edges were found in altered forsterite. Dissolution was more rapid along surface discontinuities than along the general surface in the initial phases of weathering, whereas surface dissolution could dominate the overall rate of reaction in subsequent phases. Berner et al. (1980) studied the weathering features of augite, hypersthene, diopside, and hornblende. In the initial phase, only part of the surface of the altering pyroxenes and amphiboles was affected, as was the case with olivine and feldspar. Lens-shaped etch pits formed parallel to the long and short axes of the minerals, according to SEM observations, and this gave different alteration patterns of deeply striated surfaces with end-to-end alignment along the long axes and rough-walled cracks with side-by-side alignment along the short axes. Secondary clay could be found in cracks of the weathered minerals. Tooth- or needle-shaped walls were present in the cracks because primary mineral fragments were maintained between expanding lens-shaped pits during weathering. Scanning electron micrographs of weathered amphiboles from Israel also showed tooth- and needle-shaped walls of cracks and pores (Williams and Yaalon, 1977). Detrital garnets from fluviatile, littoral, and aeolian desert sands were studied with the SEM by Magaldi (1977). Furrows, V-shaped pits, triangular pits, quadrangular pits, clusters of polygon-shaped pits, and coalescent etch figures were found. Flicoteaux et al. (1977) studied the alteration of phosphate minerals in phosphate-containing Cretaceous-Tertiary sediments of the Senegalese-

72

E. B. A. BISDOM

Mauritanian basin. Pseudomorphous transformation of wavellite to crandallite was found. Crandallite crystallites could take different orientations with respect fo wavellite. Scanning electron microscopy also demonstrated an increase in porosity during the transformation of wavellite to crandallite. M o m (1978) studied isotropic phosphatic nodules, probably weathered guano fragments, in the A1 horizon of a soil developed on basaltic colluvium on Santa Fe Island of the Galapagos archipelago. Small craters and globules were present in the nodules.

4 . Newly Formed Minerals

A considerable number of newly formed minerals in unhardened samples of soils have been studied by SEM. Submicroscopy has mainly been used to obtain information on the surface morphology of the minerals. Nonsubmicroscopic techniques were used primarily for identification purposes. a. Carbonate, Gypsum, Anhydrite, and Celestite. Needle-shaped calcite from Turkey, called lublinite, was studied with the SEM by Stoops (1976).The individual lublinite crystals were stacked in an echelon with their c-axes in a parallel position. This explained certain optical characteristics as determined in thin sections with the light microscope. Various scanning electron micrographs of lenticular gypsum, weathered lenticular gypsum with a comb structure, gypsum microlites, and a rosette-like aggregate of prismatic gypsum crystals were published by Stoops et al. (1978). The authors also studied anhydrite fibers, which were parallel to each other, on gypsum in soils from Peru. Celestite was found as long square prisms elongated according to (100)and had a well-developed (011)form. Stoops et al. (1978)found celestite in gypsiferous soils from Algeria, Iran, and Iraq. Upon weathering of celestite, grooves could develop normally to the prism faces. b. Halite, Thenardite, Bloedite, Hexahydrite, and Barite. The morphologies of halite, thenardite, and bloedite were studied with the SEM by Driessen (1970) and Driessen and Schoorl (1973).These salts came from the Konya basin in Turkey and were present in salt crusts. Mirabilite was recognized in the field but could not be transported to the lab because of its high water content. The porosity of the salt crusts could also be investigated, and it was found that the needle-shaped thenardite gave more porosity to the crust than the platy bloedite. Halite could seal the surface of the soil. Vergouwen (1981)studied salts from the same basin in Turkey with the SEM-EDXRA. Crystallographic properties and morphologies of individual salt crystals were examined. The relations between different salt crystals in salt assemblages were also studied. It was found that identification on the basis of morphology alone is not always possible; microchemical in situ analysis with the EDXRA is then necessary. Thenardite occurred in two crystal forms, as needles and in another crystal form when

SUBMICROSCOPIC EXAMINATION OF SOILS

73

associated with other salt minerals. Trona, bloedite, and hexahydrite made the salt crust very fluffy. Halite formed a smooth crust and sealed the soil. Several scanning electron micrographs of various morphologies of halite in soils were published by Eswaran et al. (1980). Attention was also given to crust formation by halite. Tursina et al. (1980) published scanning electron micrographs of thenardite in hydromorphous Solonchaks from the Soviet Union. Scanning electron microscopy also permitted the effect of salt crystallization on soil fabric and structure to be studied. Stoops et al. (1978), using the SEM, found hexahydrite on ped surfaces of a salic Gypsiorthid from Iran. Hexahydrite was mixed with gypsum crystals. Barite (microlites consisting mainly of prism) was found by Stoops and Zavaleta (1978) in a typic Haplustalt of Peru. c. Pyrite, Jarosite, and Gypsum. Scanning electron micrographs of pyrite, jarosite, and gypsum in a paleosol of eastern Nigeria were published by Moormann and Eswaran (1978). Pyrite framboids were found associated with organic matter, and fine gypsum needles could protrude from these. van Breemen and Harmsen (1975) photographed jarosite by SEM before and after dialysis with distilled water over a period of 4 months. Miedema et al. (1974) studied pyrite, jarosite, and gypsum in four soils of inland polders of the Netherlands. Paramananthan et al. (1978) investigated the effects of drainage on pyrite-containing marine clays in the coastal area of Malaysia and presented SEM photographs of pyrite, jarosite, gypsum, ferriorganans, fungal mycelia, and diatoms. d. Iron- and Manganese-Containing Minerals. Iron-containing minerals in laterites have been the subject of a number of SEM studies. Schmidt-Lorenz (1974a,b, 1975) studied many laterites of tropical regions and remnants of laterites in paleosols of Europe. Several scanning electron micrographs of hematite and various types of goethite were presented. The process of lateritization was subdivided into primary and secondary ferrallization. Kuhnel et al. (1975) studied goethite in laterite profiles and found that the highest crystallinity of the mineral was found near the surface of the laterite and the lowest at the base of the profile between the soil and bedrock (i.e., at the start of weathering). Poorly crystalline goethite could also contain nickel, chromium, and aluminium. Hematite and goethite crystallites were studied with the SEM in plinthite by Moormann and Eswaran (1978) and Eswaran et al. (1978). Iron-containing minerals have also been studied in nonlateritic soils. Lepidocrocite was found in the upper part of the B horizon of Molkenpodzols in the Vosges of France (Guillet et al., 1976). Lepidocrocite was a weathering product of hematite and occurred as stacks of subparallel platelets, with local intermineral porosity, on scanning electron micrographs. Babanin et al. (1976) indicated that very fine goethite particles with diameters of less than 5-6 nm were dominant in Ortstein. The forms and compositions of iron compounds in various soil concretions in a number of soils from the Soviet Union were investigated.

74

E. B. A. BISDOM

Scanning electron microscope studies of manganese-containing minerals such as lithiophorite, nsutite, birnessite, and feitknechtite were made by Eswaran et al. (1978). The minerals were present in nodules found in tropical soils. e. Other Minerals. Various forms of gibbsite were studied in tropical soils by Eswaran et al. (1977). Gibbsite can be present in very small amounts in tropical soils but may also form gravel-sized aggregates or sheets that are recognizable in the field. Dobrovolsky (1977) studied gibbsite crystals with a diameter of 1-20 pm in peaty soils of the Kilimanjaro area of Africa at an altitude of 2950 m above sea level; he favored a biological origin of the mineral. Biogenic opal has been studied with the SEM in soils of the United States (Wilding and Drees, 1971, 1973, 1974; Wilding and Geissinger, 1973). Opal isolated from trees differed considerably in amount and size and was dependent on the tree species. Only hackberry produced enough opal to be incorporated in the soil. Scanning electron micrographs demonstrated that there was a characteristic difference between tree-leaf opal and grass opal. Opaline isolates of wet soils often contained sponge specules and diatoms. Wilding et al. (1977) presented a review on silica present in soils and the conversion of silica hydrogel to silica polymorphs (opal, chalcedony, quartz, cristobalite, and tridymite). 5 . Organic Matter Humic and fulvic acids (HA and FA, respectively), inclusive of metal and clay complexes, were studied with the SEM by Chen and Schnitzer (1976). Fulvic acid morphologies were investigated at pH 2-10 and those of HA at pH 6-10. Metal-FA and clay-FA complexes were also studied at different pH. The SEM was used by Bruckert et al. (1974) to investigate organomineral complexes in aggregates from Andosols of the Canary Islands and France. The morphology of these aggregates was different from that of aggregates consisting of a clayhumus complex. Benayas et al. (1974) published a scanning electron micrograph of plant remains and small soil components in the upper part of an Andosol in the Canary Islands. Organomineral complexes in alkaline extracts of soil were investigated by Dormaar (1974). Fungal aggregates in sand-dune soil from Canada consisted of threads of branching mycelium from fungi to which sand grains adhered (Clough and Sutton, 1978). It was also found that amorphous material could form a sheet on the hyphae and act as an adherent between fungal hyphae and sand grains. The amorphous material consisted of polysaccharides and was possibly produced by fungi or bacteria. Aggregates formed when the fungal mycelium was in active symbiosis with the host plant.

6. Soil Structure and Fabric Numerous studies have been performed with the SEM to obtain information on various aspects of soil structure and fabric. Specialists in soil mechanics have

SUBMICROSCOPIC EXAMINATION OF SOILS

75

done considerable work to obtain information on the behavior of especially clayey soils under different experimental conditions, whereas soil micromorphologists frequently have had a closer look at soil constituents in soil peds and aggregates. a. Arrangements, Orientations, and Behavior of Soil Components under Various Conditions. Scanning electron micrographs and X-ray diffraction measurements were made of oriented clay samples obtained at different pF values by Tessier and Pedro (1976). Micrographs were made parallel or perpendicular to the orientation plane of the clay platelets of calcium kaolinite, calcium montmorillonite, and calcium illite. It was seen that considerable changes in the clay structures could occur with only minimal changes in the measured ranges of pF values; changes were greatest in the lower pF ranges. Structural changes in soil pore systems induced by Na/Ca exchange were studied with the SEM by Chen et al. (1976). At a low sodium adsorption ratio (SAR), fine material adhered to the sand grains or formed large aggregates. At a higher SAR, the fine material separated from the sand grains and filled pores. Another result was that calcium montmorillonite formed large irregular porous aggregates when a suspension was quickly frozen and dried, whereas sodium montmorillonite gave very thin sheets that were usually folded. An explanation for this phenomenon was presented. Sheeran and Yong (1974) indicated that rearrangement of soil particles in the soil environment is relatively simple as long as the soil is porous, but can only occur by way of individual minerals if only little porosity is left. Experiments indicated that virtually all changes in the orientation of clay particles occurred at lower pF levels, a result which was also obtained by Tessier and Pedro (1976). Much work on the quantification of individual clay particle alignments, including those in scanning electron micrographs, has been done by Tovey (1974, 1980) and Tovey and Wong (1974, 1980). Attention was given to photogrammetric and quantification techniques used in TEM, light microscopy, and XRD. A film measuring technique and digital computer techniques for the quantitative analysis of the orientation of clay particles in scanning electron micrographs of peds and aggregates were discussed. Such techniques can help in the quantification of soil fabric types in such micrographs. Attention was also given to particle alignments in scanning electron micrographs caused by mechanical stresses during experiments and various preparation techniques such as oven-drying, airdrying, substitution-drying, freeze-drying, and critical-point drying. The SEM has also been used to study the broken surfaces of soil fragments from a thin iron pan, the argillic horizon of an alfisol, and the cambic, argillic, and oxic horizons of tropical soils formed on basalt (Eswaran, 1971). Argillans have also been examined with the SEM (Osman and Eswaran, 1974; Callot, 1978; Koppi, 1981). Using the SEM, an impression of the degree of orientation of clay and silt in pores can be gained; whether microlayers are present or absent in the argillan can also be ascertained.

76

E. B. A. BISDOM

b. Aggregates and Crusts. Scanning electron microscope studies have been done by Moreno et al. (1978) of aggregates from black earth in Southern Spain. Clay minerals exhibited platy intermineral pores when observed at higher magnifications with the SEM. The aggregates also contained a few cylindrical pores with diameters of 0.5-2 pm. The aggregates in the soil had a similar microstructure. Buol and Eswaran (1978) investigated aggregates in oxisols and found that inter- and intraaggregateporosity could be considerable. Moura and Buol(l975) studied a Eutrustox in Brazil. The soil originally had a porosity of between 15 and 34%; continuous cultivation over a period of 15 years had decreased the porosity to 10-22%. The type of porosity was studied with the morphology of the minerals on the surfaces of pores, fractures, and aggregates. Only a few fine pores were present in clay balls, which had a higher density than the surrounding soil materials. Aggregates and weathered complexes in Andosols of France were studied in detail, with the inclusion of SEM and TEM, by Hktier (1975). Organomineral complexes were extracted from the aggregates in a step-by-step method, and each of the residues was studied. The aggregates had diameters of 5-50 pm and contained minerals, organic matter, and embedding cement which were apparent at higher magnifications. The minerals were coated with organomineral complexes. Extractions removed virtually all of the coatings, but an insoluble humic debris consisting of humin remained on part of the mineral surfaces. It was also demonstrated that humic acids, which were the most condensed and stable, were situated in clay-humus spherules, the central part of which were often occupied by glomerated halloysite. Toogood (1978) performed studies on aggregate stability in Ap horizons of various soils in Alberta, Canada. Only very weak correlations were found between the stability of the aggregates and their organic-matter content, clay percentage, carbonate content, or specific surface. On a microscale, considerable differences between individual aggregates were indicated by SEM,and the suggestion was made that general rules should be developed to explain aggregation and cementation for each individual soil type in separate regions and under different management systems. This could form a basis to obtain techniques for improving aggregate stability for individual soils. Intergranularcontacts were examined in sands, loesses, and clays by Barden et al. (1973) to study collapse phenomena when wetted under load. The SEM allowed the investigation of the arrangement of individual and of combinations of clay platelets on and between larger soil components at various magnifications. Ducloux and Ranger (1978) examined aggregates in fragipan horizons of French soils with the SEM. Strands and bridges of clay minerals and iron oxides were found between the aggregates. Such a structure can be rigid, and if it is broken will break by brittle failure. Wang et al. (1974) studied a large number of fragipans in Nova Scotia, Canada. Clay bridges were indicated by SEM between

SUBMICROSCOPIC EXAMINATION OF SOILS

77

coarser particles, and these linkages were interpreted to give brittleness under dry conditions. If the bridges become wet they can be deformed and may collapse to give slaking. The clay bridges also exhibited a low permeability to the fragipans because they create a discontinuous pore system between the coarser soil constituents. The SEM was used by Ehlers (1977) to study the morphology and texture of a thin silt crust, usually less than 1 mm thick, in the Ap horizons of “Parabraunerden” from the Federal Republic of Germany. The aggregates in the loess can easily break down because only limited binding substances such as calcium carbonate, organic matter, and clay are present, in addition to considerable quantities of silt. Erosion of the Ap therefore occurs and silt sheets with a thickness of several centimeters may form on the slope of a hill. Miehlich (1978) studied “Tepetate” (a duripan) in central Mexico. The cementing agent was amorphous silica, probably mostly opal, which was derived from the weathering of volcanic glass in ash deposits. Porosity in soil aggregates can be measured with the SEM combined with an image analyzer (Sergeyev et af., 1980a; Sokolov et af., 1980). The soil aggregate or soil ped is broken in half and the conjugated surfaces are photographed by the SEM. Subsequently, the film of the two surfaces can be manipulated to obtain the real pores in the aggregate. These can then be measured using an image analyzer such as the Quantimet. Gillott (1980) used SEM and Fourier methods to obtain information on grain shape and texture of sediments. This information was used to study provenance, correlation, pedogenesis, and environmental processes. Sergeyev et af. (1980b) examined the microstructure of many clay soils with the SEM and were able to identify five main types; turbulent, laminar, honeycomb, skeletal, and matrix. Mathematical morphology methods were used to obtain quantitative structural characteristics.

B. THINSECTIONS I . General

At present most of the submicroscopic work with thin sections concerns microchemical analysis. All such studies offer information on chemical elements, usually the heavier ones, in thin sections of soils 15-30 pm thick. Many of these studies, done with the EMA and SEM-EDXRA, concentrate on measurements of transported materials in soils, on minerals, and on soil constituents which are difficult to determine with the light microscope. With the SEM-EDXRA one can also analyze soil materials in unhardened soil peds, whereas with the SEMWDXRA one can only analyze materials in thin sections. Microanalysis is also possible with the TEM and the STEM if these instru-

78

E. B. A. BISDOM

ments are equipped with an EDXRA. The TEM-EDXRA analysis can be done on materials in ultrathin sections; STEM-EDXRA analysis can be done on both ultrathin (about 1 p,m thick) and thicker-to-common (15-30 pm thick) thin sections. Such analyses can be done at magnifications much larger than are possible with the EMA or the SEM-EDXRA-WDXRA (Le., more than X 10,OOO to more than X 1OO,OOO, depending on the thickness of the thin section). Electron diffraction of individual soil components can also be done on ultrathin sections if they can be prepared.

2 . Clay Minerals a. Thin Sections. The EMA was used by Brinkman et al. (1973) to study the chemical elements of cutans in a thin section of a pseudogley horizon in coversand soil from the Netherlands. The SiO,/Al,O, ratio was 3.5 in unaltered parts of cutans and 4-6.5 in weathered parts. This indicated a relative accumulation of silica in the weathering cutans. X-ray diffraction microcamera work indicated that less clay minerals and more extremely fine-grained quartz were present in the altered cutans. This suggested the new formation of microcrystalline quartz from silica derived from weathering smectite and illite. The elements Fe, Al, Mg, and K were partly removed from the cutans, whereas residual enrichment of mile and minerals of the kaolinite group occurred in the cutans during weathering. The plasma in a tropical groundwater podzol (Tropaquod) from Surinam was examined by Veen and Maaskant (1971) with the EMA. Cutans of a hardpan present at a depth of 120-150 cm exhibited wide extinction bands when viewed with the light microscope. Fine soil material of the matrix between coarser grains in the hardpan could be isotropic, and EMA indicated that SiO,/Al,O, ratios were lowest for this isotropic plasma; this was also the case for the SiO, percentage. Percentages of A1,0, could be lower or higher when compared to the birefringent soil matrix. The conclusion was that the birefringent plasma was broken down to form a gel-like isotropic substance which had less SiO, but was rich in A1,0,. Bajwa and Jenkins (1978) investigated the potential of selected techniques for the identification of clay minerals in thin sections of soils. If relevant standards were available and if relatively pure accumulations occurred in an area of several square micrometers, the identification of individual clay minerals was possible. Under such conditions, major element ratios of individual clay minerals and the differential adsorption of Sr were adequate to distinguish most clay minerals. Jenkins (1981) introduced the technique of low temperature ashing in which features observed in thin section under the light microscope can be exhumed by destruction of the impregnating resin. When the plastic is removed from a pore it becomes possible to study the minerals which are present in the wall of such a

SUBMICROSCOPIC EXAMINATION OF SOILS

79

pore by SEM or SEM-EDXRA. This makes the transition of light microscopy to SEM of three-dimensional entities easier. Thin sections of various soils in Buenos Aires, Argentina were studied with the light microscope by Scoppa (1978- 1979). Scanning electron micrographs were made of acicular calcite and of an argillan with montmorillonite. Bocquier and Nalovic (1972) studied, among others, ferriargillans around a pore in a sesquioxidic nodule from Cameroun with the EMA and the SEM. The argillan had two zones: an outer zone nearest to the pore in which Si, Al, and some Fe were found, and an inner zone with more Fe and less A1 in the X-ray images. It was interpreted that kaolinite was present in the outer zone whereas iron hydroxides were dominant in the inner zone away from the pore and against the matrix of the nodule. Primary and secondary illuviation of “lessivd” soils, formed in silty materials from France, were examined with the SEM and the EMA by Jamagne and Jeanson (1978). Primary illuviation took place in brown leached soils with a Bt horizon (sols bruns lessivks”). Organomineral complexes were formed and a partial desorption of the adsorption complex took place. Ferriargillans are characteristic of these brown and leached soils. Secondary illuviation occurs in leached soils with tongues (sols lessivds glossiques). Cutans, formed by secondary illuviation processes in the leached soils with tongues, were derived from ferriargillans by deferration. An EMA was used to measure the compositional differences in primary and secondary argillans in situ and to compare these microchemical data with those of transition zones and soil matrices. de Oliveira (1981) used an SEM-EDXRA to investigate femargillans in a hydromorphic profile in the state of Bahia, Brazil. The microchemical data obtained were used for pedogenetical interpretation. An SEM-EDXRA was used by Ledin (1975-1976) to test its possibilities in natural and artificial samples, including clays. The SEM-WDXRA was also used to study limed clay soil from Sweden in thin sections and the crystalline masses of calcium carbonate that formed after the addition of CaO (Ledin, 1981). Calcium carbonate can have a cementing effect when it grows between clay packets or when it partly or totally surrounds microaggregates. Argillans from Glossaqualfs in western France were studied with the SEM-EDXRA by Ducloux (1976). Argillans which were in the process of deferration had a highly variable iron content compared with the primary ferriargillans. Measurements by SEM-EDXRA and EMA indicated that the deferration process occurred sirnultaneously with the extraction of potassium (Ducloux, 1978). Rubefaction (or reddening) of soils was studied by Bresson (1974a,b) in fluvioglacial calcareous materials in the Jura Mountains of France; EMA, SEM-EDXRA, and TEM were used. The red color of the B horizons of the reddish soils was caused by accumulation of reddish materials, derived by leaching from the upper horizons, on the outer sides of structural elements and on the

80

E. B. A. BISDOM

walls of root channels. The leaching of reddish materials started in the upper horizons of the profile after decarbonization of the fluvioglacial calcareous material. Rubefaction can therefore also occur today under humid conditions and in a temperate climate. Ducloux (1978) explained that rubefaction in the silty materials of western France occurred in a rather well-drained leached soil, in neutral and weakly acid environments, after leaching by peptization of iron compounds that were present on and between illuviated clay particles. These iron compounds could color the cutans as well as the soil matrix when dehydrated to form cryptocrystalline, red-colored hematite. Cutans from the B horizons of a Haplohumod, a Humaquept, and a Haplaquod in the Netherlands appeared to be homogeneous under the light microscope (Bisdom and Jongerius, 1978). The SEM-EDXRA was used to test whether this homogeneity was also present if the heavier elements were measured in these cutans. The cutans of the Haplaquod showed a nonhomogeneous composition with EDXRA, but the other two soils were homogeneous in their heavier chemical elements. b. Replicas and Ultrathin Sections. Microchemical analyses of 15- to 30-pm-thick thin sections have been discussed in this section on thin sections. However, small soil particles such as clays can also be studied in ultrathin sections or replicas. The replica technique duplicates the surface morphology or topography of a sample and gives a very thin film that is transparent to electrons. Details of this technique have been discussed by Gillott (1974), Smart (1974), Stoops (1974), and Smart and Tovey (1982). A thin film of metal and carbon is usually evaporated onto the surface of the sample. The replica is obtained when the film and the soil are separated. Such a stripped film is usually subjected to shadowing and needs support by a fine-mesh metal grid (McKee and Brown, 1977). Individual particles in the replica can only be recognized if their morphology is characteristic. Replicas do not allow microchemical or roentgen identification. Consequently, the use of replica techniques is limited but can give information on the microfabric of a soil sample at very high magnifications using a TEM (Benayas et al., 1974; McKyes et al., 1974; Singer and Norrish, 1974; Benayas and Alonso, 1978). Biogenic opal and inorganic opal in gems were studied by Wilding et al. (1977) using the replica technique. The bonding or cementing agents of the Champlain Sea clays were investigated using a combination of replica, XRD, and selective dissolution techniques. It was found that amorphous materials played an important role by coating the minerals in the soil. This made the soils extremely sensitive to machines (McKyes et al., 1974). The preparation of ultrathin sections of the harder soil samples is difficult. SO far, only soft soil materials such as clays and organic matter have permitted the preparation of ultrathin sections. These are usually prepared in two ways, either by ultramicrotomy or by ion thinning. Ultramicrotomy makes use of either a

SUBMICROSCOPIC EXAMINATION OF SOILS

81

glass or a diamond knife, whereas ion thinning erodes the sample slowly until a hole occurs in the thin section; the soil material can be studied in the ultrathin wedge which surrounds the hole, allowing the transmission of electrons. Details on the preparation of ultrathin sections have been discussed by McKee and Brown (1977) and Smart and Tovey (1982). Ultrathin sections can be studied with TEM and STEM. High resolution electron microscopy (HREM) has not been used for the study of undisturbed small soil materials in ultrathin sections. Disturbed, laboratory-treated clay samples in ultrathin sections were examined with HREM by Brown and Jackson (1973) and Lee et al. (1975a,b). Transmission electron microscope studies of soil microstructure in ultrathin sections of clays were done by Smart (1975). On this ultramicro scale, various types of open and dense microstructures were found. It was advocated that ultrathin sections should be made of clays under natural and experimental conditions. In this manner one can compare microstructures and obtain an idea of conditions under which microstructures in natural clays were formed. Such information is important for studies in soil mechanics (i.e., consolidation, deformation, failure, and compaction effects in soils). Ultrathin sections and TEM have been used to examine the biodegradation and humification of roots by microorganisms (Kilbertus et al., 1972), the decomposition of leaves (Kilbertus et al., 1973), and the decomposition of plant material (Kilbertus and Reisinger, 1975). The root environment was studied in various soils by Foster (1978). It was found that bacteria could play a role in the formation of crumbs in the soil. Some roots are able to secrete large amounts of polymeric material into the soil; this substance can bind both organic and mineral materials. Foster (198 1) used cytochemical techniques to localize organic materials in ultrathin sections (i.e., polysaccharides derived from roots and soil microorganisms, neutral polysaccharides from cell-wall remnants, and polyphenolrich humic materials). Ion thinning and TEM were used by Bresson (198 1) to study microfabrics of clays (plasmic fabrics). Attention was given to the form, size, and distribution of ultramicropores and to the morphology of fine soil constituents. It was found that even at the high magnifications offered by TEM, microfabrics could be complex. An important objective of TEM studies on ultrathin sections is the investigation of the relation between iron compounds and clay minerals in soil microaggregates. Such studies can only be done in a satisfactory manner, however, if the TEM is able to perform electron diffraction for identification purposes. McHardy and Birnie (1975) found, that electron diffraction and identification of single particles in mixtures of fine-grained minerals can be difficult. Coarser minerals in ultrathin sections give less difficulty (Wenk, 1976), as is also true of individual clay minerals (Sudo and Yotsumoto, 1977). Fulvic acid aggregates at various pH values were studied by Schnitzer and Kodama (1975). Jepson and Rowse (1975) studied individual kaolinite particles with TEM-

82

E. B. A. BISDOM

EDXRA. When ultrathin sections of all soil materials can be prepared, this type of analysis will be used regularly for microchemical analysis of soil particles at magnifications larger than x 10,OOO. Scanning transmission electron microscopy-EDXRA can do the same. However, in the SEM-EDXRA mode, the microchemistry of normal thin sections (with a thickness of 15-30 pm instead of about 1 pm for ultrathin sections) can also be studied at magnifications above X10,OOO. The best way, however, is to analyze ultrathin sections with TEM-EDXRA or STEM-EDXRA and to combine this with electron diffraction measurements. 3 . Weathered Minerals

Seddoh and Pedro (1975) used EMA to investigate the alteration of biotite, plagioclase, and quartz in weathering granite. X-ray images showed that Fe was concentrated along cleavage planes and microcracks in weathering biotite, whereas Si and Al seemed relatively stable. Magnesium and K were leached along microcracks and cleavages. Special attention was paid to the micmhemistry of soil materials present in microcracks of the minerals. Whalley et al. (1982) studied the propagation of such cracks in igneous rocks by SEM. Meunier (1977) used EMA for the study of the weathering of biotite, muscovite, and feldspar in French granite profiles. Evolutionary trends in mineral weathering could be established using submicroscopic, XRD, and wet-chemical analyses. Stoch and Sikora (1976) studied the weathering of a dark-green mica from a Tertiary weathering crust on granites and gneisses of lower Silesia, Poland, with EMA. Curmi (1979) and Curmi and Fayolle (1981) used EMA, SEM-EDXRA-WDXRA, XRD, and wet-chemical techniques to study mineral weathering in granites of Brittany, France. Detailed submicroscopic and quantitative micorchemical analyses were done which allowed part of the alteration processes from primary to secondary minerals to be monitored. It was found, for example, that the transformation of exfoliated micas to hydroxy-aluminum vermiculite involved a significant loss of K, Fe, Mg, Ti, Ca, Na, and Si. This loss occurred at the edges of the exfoliated lamellae, whereas the central part of these lamellae remained unweathered in these profiles. Biotite, however, was completely weathered close to the surface of the profiles. Submicroscopic measurements of chemical elements were also used to obtain some insight into the nature of newly formed minerals in the profiles. The SEM and the EMA were used by Bottino et al. (1976) to examine the weathering of feldspar in gneiss and micaschist from Italy. Feldspar changed into kaolinite with an intermediate gel phase. If this gel lost Si, gibbsite could form. The same combination of instruments gave information on mycorrhizal weathering of biotite flakes (Mojahli and Weed, 1978). Potassium was lost from flake edges and adjacent to cracks in the biotite; muscovite was not weathered.

SUBMICROSCOPIC EXAMINATION OF SOILS

83

Berthelin and Belgy (1979) used organic acids produced by microorganisms (i.e., oxalic acids and other complexing agents) to experimentally weather phyllosilicates. Chlorite of granite was entirely destroyed and vermiculite was only partly altered. Biotite could be weathered to vermiculite or to a white, almost amorphous substance containing predominantly Si and Al, or it was completely broken down. Weathering sequences during the experiments were similar to those found during podzolization. Schwaighofer (1976) used SEM-EDXRA and other techniques to study the weathering of pyroclastic rocks from Tenerife, Canary Islands. Measurements were made on weathering titanaugite, titanbiotite, and anorthoclase. Moinereau (1977) examined the weathering of basalt in an organic and humid environment of a temperate climate. Various primary minerals and glass were dissolved by organic acids forming complexes with Al, Fe, Ti, and, to a lesser extent, Mg and Ca from the minerals. Potassium and Na were completely leached; Ca and Si were partly leached. Amorphous gel-like material on the surface of a basalt fragment consisted of Si, Fe, Mg, Al, some C, and a little Ti and Ca. This composition was too complex for allophane. Transmission electron microscopy indicated that halloysite and beidellite were present in the amorphous material. Delvigne et al. (1979) studied the weathering of olivine. Measurements with EMA were done on fermginous pseudomorphs of olivine from the Ivory Coast, and SEM-EDXRA were done of olivine and different types of iddingsite from the Galapagos Islands. Composite grains, which can be used as provenance indicators for Maas sediments in the southeastern part of the Netherlands, were studied by Bisdom et al. (1978) and Riezebos el al. (1978). The nature of the ore and other minerals in the composite grains was assessed. Weathered dune sands from Fraser Island, Australia were examined with SEM-EDXRA by Little et al. (1978). The original microstructure of the quartz grains was a very important factor during the different stages of alteration and influenced the rate of weathering. 4. Newly Formed Minerals

Hutton et al. (1972) used EMA to examine the chemical elements in silcretes and silcrete skins from soils in South Australia. Massive silcrete was formed when silicon from sources outside the studied profile was deposited between coarser quartz grains. These massive silcretes occurred in the lower part of the landscape. Silcrete skins, present at higher levels in the landscape, formed when quartzes in the profile were subjected to weathering and extensive leaching occurred. Only elements like Ti, Zr, Ce, and P remained because they formed parts of resistant minerals such as rutile, zircon, and cerium phosphates. Brewer et al. (1973) used EMA to study iron-manganese pans from Newfoundland, Canada. The thin pans were present in peaty soils. Most data could be

84

E. B. A. BISDOM

correlated with light-microscopic observations. Some material, however, had the same chemical composition but was different with the light microscope. Manganese and Fe often occurred side by side but without contamination by the other. Childs (1975) studied Fe-Mn concretions in soils from New Zealand. Initial formation of the concretions occurred by precipitation of Fe and Mn oxides in pores between soil constituents. Scanning electron microscopy and EMA were used, with other techniques, to study femginous pans (crusts) and calcareous crusts in Senegal and Mauritania (Nahon, 1976). Two different and opposing geochemical weathering mechanisms were recognized in a toposequence of fenuginous pans. In one, iron oxides and hydroxides remained stable and quartz and kaolinite were dissolved (formation of an iron pan); in a second, iron was dissolved and quartz and kaolinite were stable. Various stages in the development of an iron pan were described. Calcareous crusts were best developed in the upper meter of profiles formed from Eocene marls. The explanation was that once a calcareous crust or an iron pan has formed, it develops downward into the landscape by recrystallization and reconstruction processes which affect the original profile components. This could explain why iron pans and calcareous crusts are often situated directly on rock. Wieder and Yaalon (1974) used SEM and EMA to investigate the formation of carbonate nodules in soils from Israel. Chemical elements of the nodules and of the surrounding soil material were compared. Microcalcite increased in the internal part of the carbonate nodules, whereas clay increased toward their fringes. This process could lead to expulsion of part of the clayey material from the nodule into the surrounding soil matrix. An SEM-EDXRA analysis of newly formed carbonate in a Humaquept and of secondary silica in a red-yellow padzolic Tertiary paleosol of the Netherlands was made by Bisdom et al. (1975). McKee and Brown (1977) analyzed loose, undisturbed soil materials (i.e., a fractured gross soil sample and a fractured barite nodule). B6rdossy et al. (1978) studied gibbsite and other minerals in bauxite samples of different ages and origins; minerals in a laterite paleosol of the Katmandu basin in Nepal were investigated by Miiller (1976). Newly formed pyrite, carbonate, and various forms of iron- and manganesecontaining cutans were examined by Kooistra (1978) in recent marine sediments of the intertidal zone of the southwest Netherlands. Various forms of pyrite were distinguished with the light microscope. When pyrite oxidized in polder soils, black forms changed to dark, red-brown, adhesive nodules. Iron remained in the dark, red-brown rings at the oxidized periphery of pyrite spheres, whereas S disappeared. Measurements by EDXRA were also used to study Fe and Mn accumulations in a neomangan-neoferran compound and in peat and root pseudomorphs of the intertidal zone. It was found that P was also present, and this probably indicated the presence of phosphate in neoferrans. These studies

SUBMICROSCOPIC EXAMINATION OF SOILS

85

were continued by Kooistra (1981) and gave various new data on new formations in dark-to-blackish materials, with or without organic matter, that are difficult or impossible to analyze by light-microscopic techniques alone. 5 . Organic Matter Measurements by EMA of a decaying root indicated that P, Ca, Fe, and A1 were present (Qureshi et al., 1969).The highest concentration of P was present near the root margin together with some Fe. Qureshi et al. (1978)found that P was concentrated in the outer tissues of old and fresh roots together with Ca. Jeanson (1972)used SEM and EMA to investigate the results of earthworm and mycelial activities in silty B-horizon material to which glucose and peptone were added. Humidity was kept sufficiently high that a gley environment could be maintained during the experiment. Evaporation caused a 1-mm crust to form which consisted of three layers containing Fe, Mn, Ca, K, Si, and Al. Concentric rings with different compositions were formed in worm burrows and cracks. Nodules of the B horizon could be weathered, in part by microorganisms. Elements set free during weathering of the nodules (i.e., mainly Fe, Mn, and Ca) could be found again in the crust and in the cutans in the worm burrows. Righi (1975)used SEM and EMA to study the micromorphologicalfeatures of organic matter in spodic B horizons of Podzols. Organometallic complexes were found in the cemented spodic B of the B2h horizons. The complexes consisted of amorphous organic matter, A1 and Fe which illuviated into the B2h horizon, and coated sand grains. A loose spodic B was found in the Bh horizons of the podzols. Organic matter found as aggregates and pellets between sand grains was much less transformed than in the cemented spodic B and contained less fulvic acids according to wet-chemical analysis. The organic matter in the aggregates was mainly derived from decaying roots in the Bh horizon itself. Aluminum and Si were present in the aggregates, which also contained numerous fine quartz grains. Scanning electron microscopy-EDXRA can be used to distinguish heavier chemical elements (Z 2 11). If an ECON(EDAX Carbon Oxygen Nitrogen) detector is available one can also obtain information on these three elements. High A1 concentrations in a Haplaquod of the Netherlands were found primarily in strongly humified root mats in the B horizon (Bisdom and Jongerius, 1978) and were virtually absent in the strongly humified root remnants that occurred in the A2 horizon. Dupont and Jeanson (1978)studied the soil material in earthworm burrows in highly organic fine sandy sediments of the Somme Bay estuary in France, finding that the microchemical composition of the soil material at different distances from the central hole of the burrows varied significantly. Frozen specimens of plant roots were studied by Chino and Hidaka (1977) with SEM-EDXM (i.e., epidermis, cortex, endodermis, and the pericycle

86

E. B. A. BISDOM

layer). Chlorine, Ca, and Fe occurred primarily in the endodennis, whereas K was situated in the endodermis and pericycle layer. Aluminum, Si, P, S, and Ti were also found in the root. Different treatment procedures gave different accumulations at different places in the same type of root tissue. Organans and argillans were studied with SEM and EMA by van Ranst et al. (1980). It was found that organans are not usually layered and are composed of discrete, very fine bodies that are stacked at random. Organans usually contained Al with or without Fe and without Si. All three elements were found in the argillans. McKeague (1981) studied compound cutans, cutans, and nodules of Canadian soils using SEM-EDXRA. Organic matter and Al were present in the isotropic parts of compound cutans in an Ortstein horizon. Pyrite and iron oxide were found in nodules with diameters of 20 pm, which initially were thought to be fecal pellets. Phosphorus was found in a black, Fe-rich nodule of the Aeg horizon of a Luvic Gleysol.

6. Soil Structure and Fabric McHardy and Birnie (1975) used SEM and EMA to study phenomena associated with gleying in a surface water gley developed on red-brown varved lacustrine clay from Scotland. Extremely thin gray coatings on ped surfaces or fractures of the mainly red-brown Bg and Cg horizons were examined. Electron microprobe and chemical analyses indicated that the Fe,O, content of the gray materials was lower than that of the red-brown parent material and that no significant differences in composition occurred between the two as far as the other elements were concerned. Features associated with gley and illuviation processes were studied in soils from the Soviet Union using SEM and EMA (Dobrovolsky et al., 1977). Ferruginous films were present on the surfaces of peds and minerals and on the walls of pores if gleyification was weak to moderate. If the gleyifcation process became stronger, ferrous and ferric films dissolved and EMA indicated that Si, Al, and Fe remained in amorphous floccules. Measurements by EMA were also done in cutans of Solonetzes of the west Caspian coastal lowland. More Si and Al and less Ca and Mg were found in these cutans than in the adjacent clay matrix. Phosphorus in calcite grains from a Haplaquept developed in marine Cretaceous (“Gault”) clay was studied by Qureshi and Jenkins (1978). Phosphorus was just detectable with EMA and was present in small quantities of about 0.5%. The phosphorus was dispersed at relatively constant levels across the calcite grains and accounted for 50-808 of the total soil phosphorus. This phosphorus was released in the upper soil horizons when the calcite dissolved. Desert varnish on pebble and rock surfaces in the Mojave Desert of California (United States) was examined by Potter and Rossman (1977). Over 70% of clays were found in the desert varnish. Iron and clay were found in the orange coat-

SUBMICROSCOPIC EXAMINATION OF SOILS

87

ings, whereas iron and manganese were present in the black coatings. Ugolini et al. (1977) studied particle migration in a soil solution from a subalpine Podzol from the central Cascades, Washington (United States). Tension lysimeters were installed below various horizons of the Podzol and used to extract solutions. An SEM-EDXRA was used to study particles on a 0. I-pm Nuclepore filter through which the collected solutions had passed. Organic particles which were predominant in the migrant material of the A and B horizons were arrested in the B2hir horizon. Traces of Al, Si, P, S, and Fe were found in the organic particles. The B2hir horizon had a dual nature of organic particles illuviated and silicates eluviated. Cutans in cemented podzolic B horizons (Ortsteins) from podzolic soils in Nova Scotia and New Brunswick, Canada were examined with SEM-EDXRA. Organic matter itself could not be detected, because no ECON detector was available, but it was indicated indirectly if an S peak and a minor P peak were found (McKeague and Wang, 1980). An important role in the strengthening of the cement of the B horizons in the podzol was played by Al and Fe organic complexes. Brown and dark-reddish-brown cutans were also examined. A discussion of a possible step-by-step genesis of Ortstein was also included in the paper. McKeague and Pro& (1980) advocated a continuum between Ortstein and duric horizons in Canadian soils because of a similarity in cementing material and the knowledge that the levels of extractable A1 + Fe did not meet the requirement of a podzolic B. Duric horizons with discrete cutans that linked grains were selected for EDXRA. Variation in element distribution was found to be considerable. Aluminum, Fe, Si, and organic matter could act together or in various combinations as cementing agents of duric horizons. Pagliai et al. (1981) used SEM-EDXRA and image analysis to study structural improvements of soils in Italy after the application of sewage sludges, fertilizers, and soil conditioners. Porosities and the distribution of chemical elements were measured in the surface horizons. The paper also discussed the weathering of minerals and the formation of calcareous crusts. Image analysis of pores in thin sections of soils is now done by a number of laboratories with the Quantimet, an image analyzer, and other machines. The instrument can measure the area, perimeter, size, number, and other parameters of pores and makes possible the introduction of a classification graph in which the relationship between a variety of pore systems can be recognized (Jongerius et al., 1972; Jongerius, 1974,1975; Ismail, 1975; Murphy et al., 1977; Bullock and Murphy, 1980). Micromorphometric porosity data was integrated with saturated hydraulic conductivity data of undisturbed pedal clay soils by Bouma er al. (1977, 1979) for pores with diameters larger than 30 pm. Pores smaller than 30 pm in diameter were not measured for a technical reason, that is, transmitted light is used to obtain photographs from the pores and the soil constituents in the thin section; this thin section itself is about 30 pm

88

E. B. A. BISDOM

thick and therefore pores with a diameter of less than 30 pm cannot be measured. The introduction of backscattered electron scanning images (BESI), obtained from a layer of a few micrometers in thickness just underneath the surface of the thii section, makes it possible to obtain micrographs in which the image analyzer can also measure capillary pores (i.e., pores with a diameter smaller than 30 pm) (Bisdom and Thiel, 1981; Jongerius and Bisdom, 1981). The BESI technique is very useful for material contrast studies in thin sections of soils.

IV. APPLICATIONS OF ION MICROSCOPY Ion microscopy is new to soil science and the literature is concerned mainly with the possibility of applying secondary ion mass spectrometry to soil constituents in thin sections (see Section II,B). Trace concentration, trace amount, depth concentration profile, isotope, and compound analyses can be done. Ion microprobe mass analysis of ARL has been used to examine an alder root fragment from a Humaquept in the Netherlands (Bisdom ef al., 1977; Bisdom and Jongerius, 1978). Brownish-colored, rather homogeneous fine material was present together with humified remnants of the root tissue. These remnants were identical in morphology and appearance to the clayified roots described by Parfenova et al. (1964). In siru formation of clayey material was found at some sites of the alder root with the light microscope, but such sites were too small for uncontaminated sampling and subsequent XRD analysis. Ion images and positive and negative ion spectra were made of the materials in the alder root. It was found that the chemical elements, including those found in organic compounds, were rather homogeneously distributed throughout the clayified root. The organic matter occurred together with heavier chemical elements which are also present in clay minerals. It could not be decided, however, whether biogenic clay (Parfenova ef al., 1964) had been formed (i.e., it was not possible to measure whether the organic components played an active or a passive role during the formation of clay minerals in the decaying alder root) (Bisdom ef al., 1977). The Cameca ion microscope type IMS 300 has been used to quantify both trace and major elements in thin sections of soils (Henstra et al., 1980b). The experiments involved bauxite from Surinam, and quantification was possible in an area of the thin section 300 pm in diameter. Frequently, however, such an area is too large for most micromorphological and submicroscopic studies; consequently the Cameca IMS 3F has been tested. This instrument has allowed quantitative analysis of trace and major elements in a spot with a diameter of 1.5 pm. Only the soil material of polished thin sections can be studied.

SUBMICROSCOPIC EXAMINATION OF SOILS

89

V. APPLICATIONS OF OTHER FORMS OF SUBMICROSCOPY A few technical details on Raman spectroscopy, laser microprobe mass analysis, electron spectroscopy for chemical analysis, and Auger electron spectroscopy were discussed in Section I1,C. Jeanson (1981) has done RS on clay-iron coatings in earthworm burrows, aggregates constructed by soil animals, and decomposed straw fragments. It is not necessary to prepare thin sections and samples are not destroyed. Molecular analyses of an area or point in the sample were done and a distribution image of each component was obtained. Laser microprobe mass analysis was done in thin sections of a weathered granite from Spain (Bisdom, 1967a,b) with a LAMMA 500 (Bisdom et al., 1981). Laser microprobe mass analysis and SEM-EDXRA concerned titaniumcontaining clouds derived from weathering biotite [i.e., from Ti contained in the crystal lattice of the original biotite or the rutile (sagenite) inclusions of this mineral]. Light microscopy revealed minute and larger droplets in Ti-containing clouds together with turbid secondary anatase. Both instruments indicated that the Ti content of the clouds increased with the degree of crystallinity. The secondary anatase contained the most Ti. Scanning electron microscopy- EDXRA found Al, Si, Ti, and Fe in the clouds, whereas LAMMA measured these elements and Na, K, Mg, and Mn. Weaver (1976) studied the nature of TiO, in kaolinite from Georgia (United States) using SEM-EDXRA, EMA, and TEM. Titanium was released as Ti(OH), from primary minerals and formed an amorphous hydrous oxide gel upon precipitztion. Secondary anatase was formed upon dehydration of the gel as small crystals in a granular aggregate. The aggregates or pellets had various forms and diameters between 0.05 and 0.1 Fm. The LAMMA 500 usually works with ultrathin sections; the LAMMA lo00 can analyze thin sections and polished blocks and can also be used for the examination of materials in soil peds. Quantification is not yet possible with these machines. Electron spectroscopy for chemical analysis and SEM were used by Berner and Holdren (1977) to study the mechanism of feldspar weathering. Electron spectroscopy for chemical analysis can analyze very thin surface layers and was therefore used to study the surface reaction layer of weathering feldspar. No differences in composition were found between the immediate subsurface of the feldspar and the fresh feldspar itself, which indicates that no surface reaction layer existed. Auger electron spectroscopy with the combination instrument (AES,ESCA, and SIMS) of the LAS series of Riber has succeeded on ironcoated organic material from the Netherlands, possibly because of the conductivity conferred to the sample by the iron; nonconductive clayey material gave no results. Nonquantitative ESCA and SIMS analyses of the same samples as used in AES, however, gave no problems.

90

E. B. A. BISDOM

Auger electron microscopy would be an ideal technique if no charging problems existed for soil materials in thin sections of soils. Electron spectroscopy for chemical analysis is a technique which will very often be applicable, but the lateral resolution (3 mm) is too large for most.micromorphological and submicroscopic studies. Secondary ion mass spectrometry is a good technique (see Section IV) for the analysis of thin sections of soils. However, LAS does not yet have the capability for quantitative analysis.

VI. CONCLUSIONS Submicroscopy makes possible the in situ investigation of soil materials in thin sections, soil peds, soil aggregates, minerals, and other entities. Electron microscopy, with or without equipment for in situ microchemical analysis, has been used for the study of clays, weathered and newly formed minerals, organic matter, and soil structure and fabric. More recently, ion microscopy and laser microprobe mass analysis were introduced. These techniques make it possible to study both trace and major chemical elements, which is not possible with electron microscopy. Raman spectroscopy is another field of submicroscopythat will be of major interest, especially because it is nondestructive. Electron spectroscopy for chemical analysis can only be used for samples which are homogeneous over larger areas of a thin section; auger electron spectroscopy is difficult to apply to materials in thin sections of soils when they are nonconductive. Consequently, electron microscopy, ion microscopy, laser microprobe analysis, and Raman spectroscopy at present seem to be the most promising techniques. However, many instruments that are regularly used in the metals industry have not been tested for possible uses in soil science, although the most promising ones have been. Submicroscopic studies can provide microscale in situ information, whereas X-ray diffraction and wet-chemical analyses are done on bulk and disturbed samples. Information from light microscopy, which usually precedes submicroscopy during the in situ study of soil materials, is a powerful determination technique when correlated with data from XRD and wet chemistry. The same is possible for soil physics. Porosity data obtained by physical measurements can be compared with data on porosity acquired with an image analyzer from photographs made with the transmitted light of a light microscope. Presently, such photographs can also be made with the scanning electron microscope using backscattered electrons. Thus, capillary pores can also be measured with the image analyzer, usually a Quantimet. Soil-physical, light-microscopic, and submicroscopic techniques can thus be combined to solve problems related to soil porosity.

SUBMICROSCOPICEXAMINATION OF SOILS

91

At present there exists the International Working-Group on Submicroscopy of Undisturbed Soil Materials (IWGSUSM), in which specialists in soil micromorphology, soil mechanics, soil physics, soil chemistry, soil biology, and other soil sciences have joined forces. The purpose and objectives of this group have been described by Bisdom and Wells (1981). The group has become part of the Subcommission on Soil Micromorphology of the International Society of Soil Science in 1983.

REFERENCES Babanin, V. F., Karpachevskiy, L. O., Opalenko, A. A., and Shoba, S. A. 1976. Sov. Soil Sci. (Engl. Transl.) 8, 314-320. Bajwa, I., and Jenkins, D. 1978. In ‘‘Soil Micromorphology” (M. Delgado, ed.),pp. 3-17. Univ. of Granada, Spain. Barden, L., McGown, A., and Collins, K. 1973. Eng. Geol. (Amsterdam)7 , 49-60. Biirdossy, Gy., Csanhdy, A., and CsordBs, A. 1978. Clays Clay Miner. 26, 245-262. Benayas, J., and Alonso, J. 1978. In “Soil Micromorphology” (M. Delgade, ed.), pp. 717-739. Univ. of Granada, Spain. Benayas, I., Alonso, J., and Fernandez Caldas, E. 1974. In “Soil Microscopy” (G. K. Rutherford, ed.),pp. 306-319. Limestone Press, Kingston. Berner, R. A,, and Holdren, G. R. 1977. Geology 5, 369-372. Bemer, R. A., Sjoberg, E. L., Velbel, M. A., and Krom, M. D. 1980. Science (Washington,D.C.) 207, 1205-1207.

Berthelin, J., and Belgy, G. 1979. Geoderma 21, 297-310. Bisdom, E. B. A. 1967a. Leidse Geol. Meded. 37, 33-67. Bisdom, E. B. A. 1967b. Geol. Mijnbouw46,333-346. Bisdom, E. B. A. 1981a. In “Submicroscopy of Soils and Weathered Rocks” (E. B. A. Bisdom, ed.), pp. 67- 116. Cent. Agric. Publ. Doc.,Pudoc, Wageningen. Bisdom, E. B. A. 1981b. In “Submicroscopy of Soils and Weathered Rocks” (E. B. A. Bisdom, ed.), pp. 117-162. Cent. Agric. Publ. Doc.,Pudoc, Wageningen. Bisdom, E. B. A., and Jongerius, A.’1978. In “Soil Micromorphology” (M. Delgado, ed.), pp. 741-756. Univ. of Granada, Spain. Bisdom, E. B. A., and Thiel, F. 1981. In “Submicroscopy of Soils and Weathered Rocks” @. B. A. Bisdom, ed.), pp. 191-206. Cent. Agric. Publ. Doc., Pudoc, Wageningen. Bisdom, E. B. A., and Wells, C. B. 1981. In “Submicroscopy of Soils and Weathered Rocks” (E. B. A. Bisdom, ed.), pp. 17-27. Cent. Agric. Publ. Doc., Pudoc, Wageningen. Bisdom, E. B. A., Henstra, S., Jongerius, A., andmiel, F. 1975. Neth. J . Agric. Sci. 23, 113-125. Bisdom, E. B. A., Henstra, S., Hornsveld, E. M., Jongerius, A,, and Letsch, A. C. 1976. Neth. J . Agric. Sci. 24, 209-222. Bisdom, E. B. A., Henstra, S., Jongerius, A., Brown, J. D., von Rosenstiel, A. P., and Gras, D. J. 1977. Neth. J . Agric. Sci. 25, 1-13. Bisdom, E. B. A., Gerlofsma, A., Poelman, J. N. B., and Riezebos, P. A. 1978. Geol. Mijnbouw 57, 407-416. Bisdom, E. B. A., Henstra, S., Jongerius, A., Heinen, H. J., and Meier, S. 1981. Neth. J . Agric. Sci. 29, 23-36. Bocquier, G . , and Nalovic, Lj. 1972. Cah. ORSTOM Ser. Pedol. 10, 411-434. Beekestein, A., Henstra, S., and Bisdom, E. B. A. 1981. In “Submicroscopy of Soils and Weath-

92

E. B. A. BISDOM

ered Rocks” (E.B. A. Bisdom, ed.), pp. 29-44. Cent. A H c . Publ. Doc., Pudoc, Wageningen. Bottino, G., Rosa, M. A., Stafferi, L., and Veniale, F. 1976. Rend. Soc.Ital. Mineral. Petrol. 32, 521-537. Bouma, J., Jongerius, A., Boersma, O., Jager, A., and Schoonderbeek, D. 1977. Soil Sci. Soc.Am. J . 41,945-950. Bouma, J., Jongerius, A., and Schoonderbeek, D. 1979. Soil Sci. Soc.Am. J . 43,261-264. Bresson, L. M. 1974a. I n “Soil Microscopy” (G. K. Rutherford, ed.),pp. 526-541. Limestone Ress, Kingston. Bresson, L. M. 1974b. Thesis, Univ. of Paris, France. Bresson, L. M. 1981. In “Submicroscopy of Soils and Weathered Rocks” (E. B. A. Bisdom, ed.), pp. 173-189. Cent. Agric. Publ. Doc., Pudoc, Wageningen. Brewer, R., Protz,R., and McKeague, J. A. 1973. Can. J. Soil Sci. 53, 349-361. Brinkman, R., Jongmans, A. G., Miedema, R., and Maaskant, P. 1973. Geoderma 10,259-270. Brown, 1. L., and Jackson, M. L. 1973. Clays Clay Miner. 21, 1-7. Bruckert, S., Hetier, J. M., and Gutienez, F. 1974. Sci. Sol 4, 225-245. Bullock, P., and Murphy, C. P. 1980. J. Microsc. (Oxford) 120, 317-328. Buol, S. W., and Eswaran, H. 1978. I n “Soil Micromorphology” (M. Delgado, ed.), pp. 325-347. Univ. of Granada, Spain. m o t , G. 1978. In “Sod Micromorphobgy” (M.Delgado, ed.), pp. 349-368. Univ. of Granada, Spain. Chen, Y.,and Schnitzer, M. 1976. Soil Sci. SOC.Am. J . 40, 682-686. Chen, Y., Banin, A., and Schnitzer, M. 1976. Scanning Electron Microsc., pp. 425-432. Childs, C. W. 1975. Ge&rma 13, 141-152. Chino. M., and Hidaka, H. 1977. Soil Sci. Planr Nu#r. (Tokyo)23, 195-200. Clough, K. S., and Sutton, J. C. 1978. Can.J . Microbiol. 24, 333-335. Curmi, P. 1979. Thesis, Univ. of Rennes, France. Curmi, P., and Fayolle, M. 1981. In “Submicroscopy of Soils and Weathered Rocks” (E. B. A. Bisdom, ed.), pp. 249-270. Cent. Agric. Publ. Doc., Pudoc, Wageningen. Delvigne, J., Bisdom, E. B. A., Sleeman, J., and Stoops, G. 1979. Pedologie 29, 247-309. de Oliveira, J. J. 1981. In “Submicroscopy of Soils and Weathered Rocks” (E. B. A. Bisdom, ed.), pp. 271-276. Cent. Agric. Publ. Doc., Pudoc, Wageningen. Dobmvolsky, G. V., Fedorov, K.N., Balabko, P. N., Stasyuk, N. V., and Shoba, S. A. 1977. In “Problems of Soil Science” (V.A. Kovda, ed.),pp. 446-455. Nauka, Moscow. Dobrovolsky, V. V. 1977. Sov. Soil Sci. (Engl. Transl.) 9, 42-48. Dormaar, J. F. 1974. SoilSci. SOC.Am. Proc. 38, 685-686. Douglas, L. A., and Platt, D. W. 1977. Soil Sci. Soc.Am. J . 41, 641-645. Driessen, P. M. 1970. Agric. Res. Rep. No. 743. Pudoc, Wageningen. Driessen, P. M., and Schoorl, R. 1973. J. Soil Sci. 24,436-442. Ducloux, J. 1976. Sci. Sol 1, 23-36. Ducloux, J. 1978. Thesis, Univ. of Poitiers, France. Ducloux, J., and Ranger, J. 1978. In “Soil Micromqhology” (M. Delgado, ed.), pp. 815-832. Univ. of Granada, Spain. Dupont, J. P.,and Jeanson, C. 1978. In “Soil Micromorphology” (M. Delgado, ed.), pp. 833-850. univ. of Granada, Spain. Ehlers, W. 1977. 2. Pfhzeneraehr. Bodenkd. 140, 79-90. Eswaran, H. 1971. SoilSci. SOC.Am. Proc. 35, 787-790. Eswaran, H. 1972. Clay Miner. 9, 281-285. Eswaran, H., and de Coninck, F. 1971. Pedologie 21, 181-210. Eswaran, H., and Stoops, G. 1979. Soil Sci. Soc.Am. J . 43,420-424.

SUBMICROSCOPIC EXAMINATION OF SOILS

’

93

Eswaran, H., and Wong, C. B. 1978. Soil Sci. Soc.Am. J . 42, 154-158. Eswaran, H., Stoops, G., and Sys, C. 1977. J. Soil Sci. 28, 136-143. Eswaran, H., Lim, C. H., Sooryanarayana, V., and Daud, N. 1978. In “Soil Micromorphology” (M. Delgado, ed.), pp. 851-885. Univ. of Granada, Spain. Eswaran, H., Stoops, G., and Abtahi, A. 1980. J . Microsc. (Oxford)120, 343-352. Flicoteaux, R., Nahon, D., and Paquet, H. 1977. Sci. Geol. Bull. 30, 153-174. Foster, R. C. 1978. In “Modification of Soil Structure” (W. W. Emerson, R. D. Bond, and A. R. Dexter, eds.), pp. 103-109. Chichester, England. Foster, R. C. 1981. In “Submicroscopy of Soils and Weathered Rocks” (E. B. A. Bisdom, ed.), pp. 309-317. Cent. Agric. Publ. Doc., Pudoc, Wageningen. Gillott, J. E. 1974. In “Soil Microscopy” (G. K. Rutherford, ed.), pp. 143-164. Limestone Press, Kingston. Gillott, I. E. 1980. J. Microsc. (Oxford)120, 261-277. Grandstaff, D. E. 1978. Geochim. Cosmochim. Acta 42, 1899-1901. Guillet, B., Rouiller, J., and Souchier, B. 1976. Bull. SOC. Geol. Fr. 18, 55-58. Henstra, S., Bisdom, E. B. A., Jongerius, A., Heinen, H. J., and Meier, S. 1980a. Beirr. Elektronenmikrosk. Direktabb. Oberfl. 13, 63-66. Henstra, S., Bisdom, E. B. A., Jongerius, A,, Morgan, A. E., Werner, H. W., and de Grefte, H. A. M. 1980b. In “Electron Microscopy” (P. Bredemo and V. E. Cosslett, ed.), Vol. 3, pp. 224-225. Eur. Congr. Electron Microsc. Found., Leiden. Henstra, S., Bisdom, E. B. A., and Boekestein, A. 1981a. In “Submicroscopy of Soils and Weathered Rocks” (E. B. A. Bisdom, ed.), pp. 55-65. Cent. Agric. Publ. Doc.,Pudoc, Wageningen. Henstra, S., Boekestein, A., and Bisdom, E. B. A. 1981b. In “Submicroscopy of Soils and Weathered Rocks” (E. B. A. Bisdom, ed.), pp. 45-53. Cent. Agric. Publ. Doc.,Pudoc, Wageningen. Hetier, J. M. 1975. Thesis, Univ. of Nancy, France. Hutton, J. T., Twidale, C. R., Milnes, A. R., andRosser, H. 1972. J. Geol. SOC. Aust. 19,31-39. Ismail, S . N. A. 1975. “Soil Survey Paper,” No. 9. Neth. Soil Surv. Inst., Wageningen. Jackson, M. L., and Sridhar, K. 1974. Soil Sci. Soc.Am. Proc. 38, 843-847. Jackson, M. L., Levelt, T. W. M., Syers, J. K., Rex, R. W., Clayton, R. N., Sherman, G. D., and Uehara, G. 1971. Soil Sci. SOC.Am. Proc. 35, 515-525. Jamagne, J., and Jeanson, C. 1978. In “Soil Micromorphology” (M. Delgado, ed.), pp. 935-965. Univ. of Granada, Spain. Jeanson, C. 1972. Rev. Ecol. Biol. Sol 9, 479-489. Jeanson, C. 1981. In “Submicroscopy of Soils and Weathered Rocks” (E. B. A. Bisdom, ed.), p. 307. Cent. Agric. Publ. Doc., Pudoc, Wageningen. Jenkins, D. 1981. In “Submicroscopy of Soils and Weathered Rocks” (E. B. A. Bisdom, ed.), pp. 163-172. Cent. Agric. Publ. Doc.,Pudoc, Wageningen. Jepson, W. B., and Rowse, J. B. 1975. Clays Clay Miner. 23, 310-317. Jongerius, A. 1974. In “Soil Microscopy” (G. K. Rutherford, ed.), pp. 67-83. Limestone Press, Kingston. Jongerius, A. 1975. Vortr. IMANCO Symp., pp. 161-185. Jongerius, A., and Bisdom, E. B. A. 1981. In “Submicroscopy of Soils and Weathered Rocks” (E. B. A. Bisdom, ed.), pp. 207-216. Cent. Agric. Publ. Doc.,Pudoc, Wageningen. Jongerius, A., Schoonderbeek, D., Jager, A., and Kowalinski, S. 1972. Geoderma 7 , 177-198. Keller, W. D. 1976a. Clays Ckzy Miner. 24, 107-113. Keller, W. D. 1976b. Clays Clay Miner. 24, 114-1 17. Keller, W. D. 1976c. Clays Clay Miner. 24, 262-264. Keller, W. D. 1977a. Clays Ckzy Miner. 25, 311-345.

94

E. B. A. BISDOM

Keller, W. D. 1977b. Clays C h y Miner. 25, 347-364. Keller, W. D. 1978a. Clays Clay Miner. 26, 1-20. KeUer, W.D. 1978b. Geology 6 , 184-188. Keller, W. D., and Haenni, R. P. 1978. Clnys Clay Miner. 26,384-3%. Kilbertus, G . , and Reisinger, 0. 1975. Rev. Ecol. Biol. Sol 12, 363-374. Kilbertus, G., Reisinger, O., and Delon, R. 1972. Rev. Ecol. Biol. Sol 11, 27-46. Kilbertus, G., Mangenot, F., and Reisinger, 0. 1973. Rech. Coop. CNRS Ecol. Sol 3, 151-193. Kooistra, M. J. 1978. “Soil S w e y Paper,” No. 14, Neth. Soil Surv. Inst., Wageningen. Kooistra, M. J. 1981. In “Submicroscopy of Soils and Weathered Rocks” (E. B. A. Bisdom, ed.), pp. 217-236. Cent. Agric. Publ. Doc., Pudoc,Wageningen. Koppi, A. J. 1981. A m . J. Soil Res. 19, 251-254. Kiihnel, R. A., Roorda, H. J., and Steensma, J. J. 1975. Clays Clay Miner. 23, 349-354. Ledin, S . 1975-1976. Grunforbdttring 27, 145-150. Ledin, S. 1981. In “Submicroscopy of Soils and Weathered Rocks” (E. B. A. Bisdom, ed.), pp. 295-306. Cent. Agric. Publ. Doc., Pudoc, Wageningen. Lee, S. Y.,Jackson, M. L., and Sachs, I. B. 1974. Soil Sci. Soc.Am. Proc. 38,423-428. Lee, S . Y.,Jackson, M. L., and Brown, J. L. 1975a. Clays Clay Miner. 23, 125-129. Lee, S. Y.,Jackson, M. L., and Brown, J. L. 1975b. Soil Sci. Soc. Am. Proc. 39,793-800. Legigan, P., and Le Ribault, L. 1974. C.R. Hebd. Seances Acad. Sci. Ser. D 279, 799-802. Leneuf, N. 1972. Cah. ORSTOM Ser. Geol. 4, 53-65. Liebl, H. 1975. J. Phys. 8, 797-808. Little, I. P., Annitage, T.M., and Gilkes, R. J. 1978. Geoderma 20, 225-237. Magaldi, D. 1977. Boll. Sew. Geol. Iral. 98, 117-128. Magaldi, D. 1978. In “Soil Micromorphology” (M. Delgado, ed.),pp. 967-988. Univ. of Granada, Spain. McCrone, W. C., and Delly, J. G. 1973. “The Particle Atlas,” 2nd ed. Ann Arbor Sci. Publ., AM Arbor, Michigan. McHardy, W. J., and Birnie. A. C. 1975. J. Soil Sci. 26, 426-431. McKeague, J. A. 1981. In “Submicroscopy of Soils and Weathered Rocks” (E. B. A. Bisdom, ed.), pp. 277-282. Cent. Agric. Publ. Doc., Pudoc, Wageningen. McKeague, J. A., and h a ,R. 1980. Can. J. Soil Sci. 60,45-52. McKeague, J. A., and Wang, C. 1980. Can. J. Soil Sci. 60, 9-21. McKee, T.R., and Brown, J. L. 1977. In “Minerals in Soil Environments” (R. C. Dinauer, ed.), pp. 809-846. Soil Sci. Soc. Am., Madison, Wisconsin. McKyes, E.,Sethi, A., and Yong, R. N. 1974. Chys Clay Miner. 22,427-433. Meunier, A. 1977. Thesis, Univ. of Poitiers, France. Miedema, R., Jongmans, A. G.,and Slager, S. 1974. In “Soil Microscopy” (G. K. Rutherford, ed.), pp. 772-794. Limestone Press, Kingston. Miehlich, G. 1978. Forsch. Geol. Paliionrol. 44-45, 27-41. Millot, G., Nahon, D., Paquet, H., Ruellan, A., and Tardy, Y. 1977. Maroc Sci. Geol. Bull. 30, 129-152. Moinereau, J. 1977. Cah. ORSTOM Pedol. 15, 157-173. Mojallali, H., and Weed, S. B. 1978. Soil Sci. Soc. Am. J. 42, 367-372. Moormann, F. R., and Eswaran, H. 1978. Pedologie 28, 251-270. Moreno, F., Comejo, J.. and Rives, V. 1978. In “Soil Mimmorphology” (M. Delgado, ed.),pp. 1335-1348. Univ. of Granada, Spain. M o m , H.J. M. 1978. In “Soil Micromorphology” (M. Delgada, ed.), pp. 1007-1018. UNv. of Granada. Spain. Moss, A. J., and Green, P. 1975. J. Geol. Soc. Aust. 22,485-495. Moura, W. F., and Buol, S . W. 1975. Experienria 21, 161-177.

SUBMICROSCOPIC EXAMINATION OF SOILS

95

Miiller, H. 1976. Abh. Math. Naturwiss. K1. A M . Wiss. Lit. Mainz, Abt. I 185, 43-53. Murphy, C. P., Bullock, P., and Turner, R. H. 1977. J . Soil Sci. 28, 498-508. Nahon, D. 1976. Thesis, Univ. of Aix (Marseilles III), France. Nahon, D., Paquet, A., Ruellan, A., and Millot, G. 1975. Sci. Geol. Bull. 28, 29-46. Osman, A., and Eswaran, H. 1974. In “Soil Microscopy” (G. K. Rutherford, ed.), pp. 846-857. Limestone Press, Kingston. Pagliai, M., Ferrari, G. A., and Magaldi, D. 1981. In “Submicroscopy of Soils and Weathered Rocks” (E. B. A. Bisdom, ed.), pp. 283-294. Cent. Agric. Publ. Doc., Pudoc, Wageningen. Paramananthan, S . , Sooryanarayana, V., Syed Sofi, S . O., and Eswaran, H. 1978. In “Soil Micromorphology” (M. Delgado, ed.), pp. 589-609. Univ. of Granada, Spain. Parfenova, E. I., Mochalova, E. F., and Titova, N. A. 1964. In “Soil Micromorphology” (A. Jongerius, ed.), pp. 201-212. Elsevier, Amsterdam. Potter, R. M., and Rossman, G. R. 1977. Science (Washington, D.C.) 196, 1446-1448. Qureshi, R. H., and Jenkins, D. A. 1978. SoilSci. SOC.Am. J . 42, 703-705. Qureshi, R. H., Jenkins, D. A., and Davies, R. I. 1969. Nature (London) 221, 1142-1143. Qureshi, R. H., Jenkins, D. A., and Davies, R. I. 1978. Soil Sci. SOC.Am. J . 42, 698-703. Riezebos, P. A. 1974. Geol. Mijnbouw 53, 109-122. Riezebos, P. A., Bisdom, E. B. A., and Boersma, 0. 1978. Geol. Munbouw 57, 417-431. Righi, D. 1975. Sci. Sol 4, 315-321. Schmidt-Lorenz, R. 1974a. Min. Dtsch. Bodenkd. Ges. 20, 68-79. Schmidt-Lorenz, R. 1974b. Min. Dtsch. Bodenkd. Ges. 20, 114-122. Schmidt-Lorenz, R. 1975. Joint Meet. ISSS G h a I975 ~ (unpublished data). Schnitzer, M., and Kodama, H. 1975. Geoderma 13, 279-287. Schwaighofer, B. 1976. Geoderma 6, 285-315. Scoppa, C. E. 1978-1979. Rev. Invest. Agropec. INTA 14, 37-69. Seddoh, F. K., and Pedro, G. 1975. Cah. ORSTOM Ser. Pedol. 13,7-25. Sergeyev, E. M., Spivak, G. V., Osipov, V. I., Rau, E. I., Sokolov, V. N., and Filippov, M. N. 1980a. Scanning 3, 262-272. Sergeyev, E. M., Grabowska-Olszewska, B., Osipov, V. I., Sokolov, V. I., and Kolomenski, Y.N. 1980b. J . Microsc. (Ogord) 120, 237-260. Sheeran, D. E., and Yong, R. N. 1974. In “Soil Microscopy” (G. K. Rutherford, ed.), pp. 179-189. Limestone Press, Kingston. Singer, A. 1973. Isr. J . Earth Sci. 22, 229-242. Singer, A., and Norrish, K. 1974. Am. Miner. 59, 508-517. Smart, P. 1974. In “Soil Microscopy” ( G . K. Rutherford, ed.), pp. 190-206. Limestone Press, Kingston. Smart, P. 1975. Soil Sci. 119, 385-393. Smart, P.,and Tovey, K. 1981, “Electron Microscopy of Soils and Sediments: Examples.” Oxford UNv. Press (Clarendon), Oxford. Smart, P., and Tovey, K. 1982. “Electron Microscopy of Soils and Sediments: Techniques.” Oxford Univ. Press (Clarendon), Oxford. Sokolov, V. N., Osipov, V. I., and Tolkachev, M. D., 1980. J. Microsc. (Ogord) 120,363-366. Sousa, E. C., and Eswaran, H. 1975. Pedologie 25, 71-79. Stoch, L., and Sikora, W. 1976. Clays Clay Miner. 24, 156-162. Stoops, G. 1974. In “Soil Microscopy” (G. K. Rutherford, ed.), pp. 101-108. Limestone Press, Kingston. Stoops, G. 1976. Am. Miner. 61, 172. Stoops, G., and Zavaleta, A. 1978. Geoderma 20, 63-70. Stoops, G., Eswaran, H., and Abtahi, A. 1978. In “Soil Micromorphology” (M. Delgado, ed.), pp. 1093-1113. Univ. of Granada, Spain.

96

E. B.

A. BISDOM

Sudo, T., and Yotsumoto, H. 1977. Clays Clay Miner. 25, 155-159. Tarzi, J. G.,and Rotz, R. 1978. Clays Clay Miner. 26, 352-360. Taupinard, J. 1976. Bull. Soc.Fr. Mineral. Cristallogr. 99, 246-250. Tessier. D.,and Berrier, J. 1978.In “Soil Micronmphology” (M. Delgado, ed.), pp. 717-739. univ. of Granada, Spain. Tessier, D.,and Pedro, G. 1976. Sci. Sol 2, 85-99. Toogood, J. A. 1978.In “Modification of Soid Structure” (W. W. Emerson, R. D. Bond, and A. R. Dexter, eds.), pp. 21 1-215. Chichester. England. Tovey, N. K. 1974.In “Soil Microscopy” (G. K. Rutherfoni, ed.), pp. 119-142. Limestone Press, Kingston. Tovey, N. K. 1980. J. Microsc. (Oxford)120, 303-315. Tovey, N. K., and Krinsley, D. H. 1980.J. Microsc. (Oxford)120, 279-289. Tovey, N. K., and Wong, K. Y. 1974. In “Soil Microscopy” (G. K. Rutherfoni, ed.), pp. 207-222. Limestone Press, Kingston. Tovey, N. K., and Wong, K. Y. 1980. J. Microsc. (Oxford)120,329-342. Tmina, T. A., Yamnova, I. A., and Shoba, S. A. 1980.Pochvovedenie 2, 30-43. Ugolini. F. C., Dawson, H.,and Zachara, J. 1977. Science (Washington, D.C.) 198, 603-605. van Breemen, N.,and Hannsen, K. 1975. Soil Sci. SOC. Am. Proc. 39, 1140-1148. van Ranst, E., Righi, D., de Coninck, F., Robin, A. M., and Jamagne, M. 1980. J. Microsc. (Oxford)120, 353-361. Veen, A. W. L., and Maaskant, P. 1971. G e o d e m a 6, 101-107. Vergouwen, L. 1981.In “Submicroscopy of Soils and Weathered Rocks” (E. B. A. Bisdom, ed.), pp. 237-248. Cent. Agric. Publ. Doc., Pudoc, Wageningen. Verheye, W., and Stoops, G. 1975.Pedologie 25, 40-55. Violante, P.,and Violante, A. 1977.Agrochirnica 21, 513-522. Wagner, C. D., Riggs, W. M., Davis, L. E., Moulder, J. F., and Muilenberg, G. E. 1979. “Handbook of X-Ray Photoelectron Spectroscopy.” Perkin-Elmer, Eden Prairie. Wang, C., Nowland, J. L., and Kodama, H. 1974. Can. J . Soil Sci. 54, 159-170. Weaver, C. E. 1976. Clays Clay Miner. 6, 101-107. We&, H. R. (ed.) 1976. “Electron Microscopy in Mineralogy.” Springer-Verlag. Berlin. Whalley, W. B., Douglas, G. R., and McGreevy, J. P. 1982.Z. Geomorphol. 26,33-54. Wieder, M., and Yaalon, D. H. 1974. Geodema 11, 95-121. Wilding, L. P., and Drees, L. R. 1971.Soil Sci. Soc.Am. Proc. 35, 1004-1010. Wilding, L. P., and Drees, L. R. 1913. Soil Sci. SOC. Am. Proc. 37, 647-650. Wilding, L. P., and Drees, L. R. 1974. Clays Clay Miner. 22, 295-306. Wilding, L. P., and Geissinger, H.D. 1973. J . Sediment. Petrol. 43,280-286. Wilding, L.P., Smeck, N. E., and Drees, L. R. 1977.In “Minerals in Soil Envrionments” (R. C. Dinauer, ed.), pp. 471-552. Soil Sci. Soc. Am.,Madison, Wisconsin. Williams, C., and Yaalon, D. H. 1977. G e o d e m a 17, 181-191. Wilson, M. J. 1975. Soil Sci. 119, 349-355. Wilson, M. J., and McHardy, W. J. 1980. J. Microsc. (Oxford) 120, 291-302. Yaalon, D. H., and Wieder, M. 1976. Clay Miner. 11, 73-79.

ADVANCES IN AGRONOMY, VOL. 36

THE CONVERGENT EVOLUTION OF ANNUAL SEED CROPS IN AG R ICULTURE C. M. Donald1 and J. Hamblin2*3 ’Waite Agricultural Research Institute, The University of Adelaide, South Australia, and 2Department of Agriculture, Geraldton District Office, Marine Terrace, Geraldton, Western Australia

I. Introduction .......................................................... 11. Selection in Domesticated Crops ......................................... A. Charles Darwin’s Views .................. B. Selection within Annual rops .................................. 111. Ecotypic Parallelism in Crop Plants ....................................... IV. Selection, Evolution, and Crop Yield .................. Biological Yield, Harvest Index, and Grain Yield. ............. Progress and ent of Annual Seed Crops.. ............

B. V.

B. C. D. E. F.

Barley .......................... Rice.. .......................................................... Maize ................................... Sorghum ............................... ... Common or American Bean ........................................

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

References ..............

97 100 100 101 111 112 113 119 121 121 122 123 124 126 127 129 130 130 131 133 134 139

I. INTRODUCTION Crop evolution consists of three phases: the natural evolution of a species to the “roto-crop” stage, domestication, and further evolution within the domesticated species. In the proto-crop stage, the main requirement is for a species to possess some attribute desired by man. Domestication of a seed crop was often a 3All correspondence regarding this article must be addressed to Dr. J. Hamblin. 97

Copyright 0 by Academic Press, Inc. All rights of reproduction in any form reserved. ISBN 0-12-000736-3

98

C . M.DONALD

AND J. HAMBLIN

“once only” event, when certain major mutations ensured that a species was adapted into the agricultural environment and altered unfavorably for survival in the wild (Mangelsdorf, 1965), although this may be an oversimplification in some cases (Harlan, 1971). These mutations include such well-known features of crop plants as indehiscent fruiting bodies, which allow easier and more complete harvesting, and soft seededness, which ensures simultaneous germination of all seeds. The third phase of evolution within the domesticated species is a continuing phase. The crop changes under selection pressures, both natural and manimposed. It is the role of agronomists and plant breeders to optimize the environment and plant type to ensure maximum crop production within this environment. The world’s field crops are commonly grouped into cereals, pulses or grain legumes, fiber crops, oil crops, root crops, and rubber. This classification, based partly on botanical considerations and partly on crop product, has little ecological basis. However, many of these crops, such as the cereals, the grain legumes, and some of the oil crops, are annual seed crops with parallel ecological features. Increased understanding of the factors governing crop photosynthesis and respiration, distribution of assimilates, and seed growth permits us to compare and contrast the performance of annual seed crops. This may be in terms of their branching, leafiness, light profile, photosynthesis, biomass, flowering, seed setting, grain filling, harvest index, and yield, and/or in terms of agronomic factors such as soil fertility, plant density, and plant arrangement. At first sight, it may seem difficult to compare cotton with maize or sunflower with wheat, but such comparisons provide a major challenge to our thoughts. For instance, why are some annual seed crops so much more productive than others? And what can be done to remodel less efficient crops? Annual seed crops provide most of man’s food and, in some countries, a significant part of the animal feed as well as industrial products of great importance (fibers, oil, etc.). They are cultivated from the equator to near the Arctic circle and are adapted to diverse edaphic situations. Some are extremely tall (more than 4 m) whereas others are dwarf, some are climbers with tendrils. When these patterns of adaptation and the morphological and physiological characters are surveyed, Can we perceive common features, either plant or cultural, that may be exploited to increase seed yield per hectare in all the various annual seed crops and environments? We consider that the yield potential of annual crop species will increase at a faster rate than occurs with empirical selection for yield if suitable ideotypes are identified. A considerable list of common features and practices that influence yield in all annual seed crops can indeed be identified, and it may be possible to design a basic ideotype for all these crops, involving principles of crop physiology and associated agronomic practices equally applicable to any annual seed crop. Such a common model or ideotype is formulated here.

CONVERGENT EVOLUTION OF ANNUAL SEED CROPS

99

A sharp distinction is drawn between the ecology of annual field crops grown for their seed and that of most horticultural crops. Horticultural plants are cultivated for their fruits, unripe seeds, roots, stems, or leaves. “We believe,” remarks Schwanitz (1966, p. 29), “that the transition from normal to giant growth is the most important step in the evolution of wild species into cultivated plants.” The modem apple, tomato, celery stalk, or lettuce heart are examples. However, it is important to recognize the large environmental as well as the genetic component in the improvement of horticultural crops. Letluce and many other species may be so widely spaced that the plants are almost noncompetitive. Alternatively, there may be deliberate reduction in the number of harvested parts, as in flower or fruit growing where huge blooms and fruits of some species are produced by deliberate thinning. Schwanitz’s view of gigantism through evolution under domestication is clearly valid with respect to horticultural plants; there has been remarkable progress toward gigantism of the harvested part within markedly more favorable environments. Under horticultural conditions, natural selection is heavily suppressed, but selection by man has been highly effective. The range of varieties within a single species (e.g., within Brussicu oleruceu, the cabbage, cauliflower, brussel sprouts, kohlrabi, kale, etc.) illustrates this point most vividly. For annual seed crops, however, Schwanitz’s views regarding gigantism of plant parts do not hold. The prime need for cultivators of these crops has always been the quantity of seed in the bag or basket, of the crop yield per unit of land, rather than the size of the individual seed or the seed yield per plant. In modem seed crops, a reduction in individual plant yield to as little as 5% of the yield of like plants growing in isolation is usual (Donald, 1963), yet for a long time agricultural scientists failed to recognize the significance of competition within these monocultures. With competition of such intensity, the extent of natural selection is limited only by the genetic variability between the plants and by any inequalities of the immediate environment of the individuals. If a drought occurs, there will be intense competition for water. If water is abundant, there may be equally intense competition for nitrogen. If water and all nutrients are freely available, there will be extreme competition for light. Nothing is farther from reality than the following analysis by Schwanitz, which is a common viewpoint (italics ours): “Cultivated plants are also exposed to the influence of their environment; they too are threatened by frost and drought, pest and disease. But man has been careful to protect the plants that are useful to him from excessive hazard. By tilling and fertilising the soil, by regulating the water supply and eliminating the struggle for life, and by protecting the plants from pests and disease, he has created an artificial environment that favours the plant more and above all exposes it to less rigorous requirements than those met in nature. Hence natural selection in cultivated forms is less harsh than among wildplants” (p. 116). Only in a few horticultural crops is that

100

C. M. DONALD AND J. HAMBLIN

statement generally acceptable. For cultivated seed crops, the seedling environments have been substantially improved; nonetheless, intense competition among plants rapidly develops. This competition more than any other factor has governed their evolution. Many of the generalizations on crop evolution are largely based on horticultural crops and simply do not apply to annual seed crops, the principal component of man’s agriculture. Our views are essentially similar to those of Clements et al. (1929, p. 77), who state that “Competition is keenest when individuals are most similar and . . . make nearly the same demands on the habitat and adjust themselves less readily to their mutual interactions,” and also that “The closeness of competition between plants of different species varies directly with their likeness in vegetation or habitat form.”

II. SELECTION IN DOMESTICATED CROPS A. CHARLESDARWIN’S VIEWS

Darwin (1868) grouped the selection forces operating among domesticated animals and plants into three loosely defined categories: methodical or con-

scious selection by man “according to some pre-determined standard”; unconscious selection by man through retaining ‘better’ animals or plants (‘better’ in the eyes of the herdsman or the cultivator) and natural selection, occurring without any purposeful intervention by man. He drew no particular distinction between the operation of these three forms of selection among animals and plants. Darwin readily illustrated conscious selection with horticultural crops: the development of large gooseberries, double flowers, and early maturing peas, and the selection of high sugar content in beets;hence his tribute to man’s capacity to select methodically from within a varying population those features or attributes he values. However, he used the term unconscious selection when man selected for general superiority of animals or plants without attempting to define the specific factors for which selection occurred. Consequently, some confusion has arisen in the subsequent use of this term. Though the herdsman might choose a ‘better bull’ or the cultivator might choose a ‘better plant’ without any predetermined standard or without evaluating any array of desirable features, he neverthelessdoes make a perfectly conscious and deliberate choice. The term unconscious selection seems scarcely appropriate. Darlington (1%9) uses Darwin’s term unconscious selection but gives it a different meaning. He speaks of “unconscious selection by the cu1tivator”as the transformation of the crop by selection during cultivation, tilling, sowing, reap-

CONVERGENT EVOLUTION OF ANNUAL SEED CROPS

101

ing, and threshing. But this is not selection by man. It is clearly the operation of natural selection within the environment of man’s cultural practices, involving no active selection by man himself, conscious or otherwise. It is proposed here, at least for annual seed crops, that Darwin’s “unconscious selection by man” be termed “nonspecific selection by man,” and that Darlington’s “unconscious selection by the cultivator” be regarded as “natural selection for adaptation to agriculture.’’ Darwin recognized the close adaptation by natural selection of numerous varieties of wheat to various soils and climates even within the same country; “that the whole body of any one sub-variety ever becomes changed into another and distinct sub-variety, there is no reason to believe. What apparently does take place, is that some one sub-variety . . . which may always be detected in the same field, is more prolific than the others and gradually supplants the variety that was first sown” (p. 389). B. SELECTION WITKIN ANNUAL SEED CROPS

We now recognize several categories within Darwin’s general processes of selection by man and natural selection. These are discussed here in relation to annual seed crops.

I . Selection by Man in Annual Seed Crops Conscious selection by man relates especially to fruiting organs and seed; to larger ears of wheat, larger cobs of maize, or heads of sunflowers, all undoubtedly contributing to an improved harvest index and grain yield in the early years of domestication. The choice of seed size, color, and flavor was also a basis for selection, notably in rice and beans. An early and important case is the selection of the dwarf habit in the naturally climbing common bean (Phaseolus vulgaris) by the American Indians (A. M. Evans, 1980). The dwarf mutant, in nature or mixed cultivation, would have been effectively lethal because of suppression by taller plants (Smartt, 1969; Hamblin, 1975); it could not have emerged by natural selection, but man has preserved and propagated it as a key mutant. A similar situation has occurred in rice with the development of short, high-yielding types which are rapidly eliminated in mixtures with tall, low-yielding types (Jennings and Aquino, 1968; Jennings and Herrera, 1968; Jennings and de Jesus, 1968). However, in many instances it is difficult to distinguish selection by man from natural selection within the changed environment that man has provided. How effectively did early cultivators select for better yield and to what extent did better yields arise through the natural selection of genotypes producing more seed? In present-day annual seed crops, it is certainly very difficult to select by

102

C . M.DONALD

AND J. HAMBLIN

subjective judgement those plants capable of higher seed yield in pure culture (Bell, 1963; McGinnis and Shebeski, 1968; Walker, 1969; Hamblin, 1971). The genetic worth of any plant is confounded by the influences of its immediate environment, including its neighbors (Hamblin et al., 1978). Percival (1921) quotes Virgil and other classical writers as advocating the selection of the best ears for use as seed, but we do not know to what extent this influenced yield per unit area. Nevertheless, many effective selections have occurred in the past; we have records for a century or more of successful cereal varieties being developed from single ear or plant selections by farmers, such as wheat cultivars; Chidman, Hunter’s White, and Squarehead (United Kingdom); Canadian Fife (Canada); Fultz (United States), and Purple Straw and Prior barley (Australia). Each was an arbitrary choice of a plant which, in the eye of the observer, looked more productive and indeed proved to have features which were sustained in pure cultures over a range of environments. But the unsuccessful and unrecorded conscious selections for yield in cereal crops doubtless occurred in great numbers. Wherever conscious or methodical selection has been practiced, correlated secondary selection for other characters has been almost inevitable. Thus the selection of the dwarf habit in beans was probably accompanied by unplanned and unrecognized selection for determinate growth, a feature of plant architecture no less important than reduced height itself. Increased cob size in corn initially increased yields over its progenitors (Galinat, 1965; Wilkes, 1977). However, once the modem types had become established, secondary selection probably was adverse to productivity in pure culture, reducing any gain resulting from the conscious component of selection. Large maize cobs, selected in the field, were almost inevitably from tall, broad-leaved, highly competitive plants that had exploited the habitat of their neighbors (see the work of Gardner and coworkers at Nebraska, considered in more detail on p. 115-1 16). A striking example of such secondary selection was reported by Wilcox and Schapaugh (1980), who selected “phenotypically superior single (soybean) plants, based on a visual estimate of seed yield and lodging resistance.” No change in seed yield or lodging resistance occurred, although the selected plants were significantly taller and later than the unselected control plants. It seems the selected plants were chosen only because of competitive advantage resulting from their tallness and longer growth period. When competing against like neighbors, these taller and later plants showed no advantage. However, it must be remembered that only recently has man become obsessed with yieldhnit area. In many situations, these competitive features would provide benefits in terns of animal feed, building materials, and so on (Hamblin and Rosielle, 1983). It is conceivable that nonspecific selection by man may have occurred among horticultural plants, for example the propagation of a fruit tree because of its general form, cropping capacity, fruit quality, and so on, without any precise

CONVERGENT EVOLUTION OF ANNUAL SEED CROPS

103

definition or evaluation in the mind of the observer. It is difficult, however, to identify nonspecific selection of plants within annual seed crops. If the selection was made in the field, some particular feature, whether number or size of fruits or seeds, growth form, vigor, or some other specific factor, probably would have been the basis of selection. Today’s growers of annual seed crops are most interested in yield per unit area. This is a form of conscious selection for yield, not within the crop but between cultivars within the region. This was also the case when hexaploid wheats replaced tetraploids over northern Europe and when tetraploid cottons from the Americas replaced most of the diploid cottons of the Old World. In each instance further selection for adaptation to new environments occurred. 2 . Natural Selection in Annual Seed Crops

Within man’s crops, natural selection is always potentially operative. Darwin emphasized that “natural selection . . . a power incessantly ready for action, is as immeasurably superior to man’s feeble efforts as the works of Nature are to those of Art” (p. 77). Natural selection results from the relationship of each plant to its physical environment and to its neighbors. If the plants within a crop are genetically variable in even the slightest degree some biotypes will increase and others will decrease in ensuing generations. The relative advantage of a particular biotype may change with the environment (e.g., for barley, Harlan and Martini, 1938; for beans, Hamblin, 1975; for rice, Adair and Jones, 1946). Two mechanisms of natural selection may be recognized (Nicholson, 1962): environmental selection and selection through competition. Nicholson considered that the views of Darwin and Wallace toward natural selection were based primarily on one or other of these mecharismeDarwin’s on selection through competition and Wallace’s on environmental selection. a. Environment Selection. Nicholson wrote that environmental selection “removes all individuals which are not sufficiently potent to withstand the severe conditions to which the species is exposed from time to time, and so leads towards the production of a population in which all individuals can survive, even under these severe conditions” (p. 65). Examples of environmental selection abound in man’s crops. It occurs when early flowering individuals in a crop are prevented from setting seed by a late frost, when late flowering plants produce no seed because of hot, dry conditions, when some plants grow poorly on an acid soil or under saline conditions, or in any one of numerous circumstances within the physical environment. Cotton differs from most annual seed crops in that it entered cultivation as a perennial shrub. It thus continued, confined to the tropics, for several millennia. When taken to temperate areas, in less than 600 years natural hybridization and selection for earliness and avoidance of frosts gradually led to a branched shrub

104

C . M. DONALD AND J. HAMBLIN

of strictly annual habit (Hutchinson, 1965). Similarly, maize, originally confined to the subtropics of the Americas, progressively extended northward into areas with shorter seasons and severe winters because of interspecific crossing and environmental selection (for earlier flowering forms having different photoperiodic responses (Galinat, 1965). At the experimental level, a barley composite involving crosses between 31 cultivars (Allardand Jain, 1962) rapidly adapted to the climate at Davis, California. The population shifted strongly to earliness in heading date for 5 generations and then more slowly for the next 15, a directional selection. At the same time there was a steady elimination of the “tails” (either the earliest or latest) and variance decreased, stabilizing selection around the optimum heading time for the locality. Similar environmental selection has been recorded for other climatic features, especially length of day, and to soil features, such as texture, pH, fertility, and salinity. “All those who have closely attended the subject insist on the close adaptation of numerous varieties (of wheat) to various soils and climates even within the same country” (Darwin, 1868, p. 388). Environmental selection within cultivated crops also occurs from the influence of man’s cultural practices (Harlan er al., 1973). Although it may be claimed that these practices are “artificial,” the plant responses and the selection of the “fit” are truly natural selection in the Darwinian sense. Successful plants must be more suited to, or less harmed than other plants by man’s treatment of the crop. In primitive agriculture, emergence from varying depths of planting was critical; in many earlier crops, and still today in some species, resistance to damage during temporary grazing gave selective advantage. There was advantage also for those plants that responded better to man’s cultivation practices, to his soil-water storage, or to his irrigation practices. More recently, those plants that respond to artificial fertilizers or are less harmed by pesticides are at an advantage. Faced by these practices, some plants will show a relative gain and others a reduction in prolificity (number of seeds produced). However, seed number per plant by itself will not ensure that any individual is represented in successive crops. A high proportion of each plant’s seed must go into the bulk seed used for sowing the next crop. This attribute is distinct from prolificity but just as significant in determining the pattern of natural selection. Initially seed must be retained on the plant until the crop is cut, but then must be threshable yet not so light as to be lost during winnowing. The seed must retain viability until the following sowing time, but it then must germinate without problems of hard-seededness, physiological dormancy, or undue delay. It must emerge and establish from varying depths of sowing. This list of selection filters, doubtless far from complete for many seed crops, indicates the powerful selection pressures that inevitably occurred from the earliest years of domestication, and future natural selection will follow with the adoption of new cultural practices. Natural selection in response to many of these features of the crop environ-

CONVERGENT EVOLUTION OF ANNUAL SEED CROPS

105

ment must have occurred within a single, or at most a few, generations; the most ill-fitted plants, such as those that shed all their seed before the harvest, would have no descendants in the next generation of the crop. b. Selection through Competition. If any plant within a crop takes up more water, nutrients, or light than another at the expense of that other plant, it will have the potential to be advantageously represented by its progeny in the ensuing generations; it will be selected through competition. Successful competitors within seed crops have been selected for the following: i . For the Annual Habit. In perennial plants, part of the assimilates are distributed to storage organs, that is, to underground organs or enlarged stem bases (which may themselves be harvested, but are not relevant to this article). Storage competes directly with seed production. Where annual seed crops are descendants of perennial species, selection toward the annual habit has occurred simply because of the capacity of annuals to produce more seeds. However, remnants of perenniality remain in some annual crops, so that “ratooning,” the growth of a second crop from the bases of the first, is possible. Many sorghum cultivars and also some rice varieties, especially those that tiller freely, can be ratooned. But within seed crops evolution toward a strictly annual habit has been continuous, so that seed ripeness and plant death are coincident or nearly so. The avoidance of diversion of resources to vegetative organs combined with the adaptive value of an annual growth pattern in regions of limited season has led to the development of annual seed crops in many families and genera (Chang, 1976). ii. For Tallness. The most universal factor for which natural selection has occurred in crops has been plant height. Even slight superiority, through advantageous competition for light, can give a plant sufficient yield increment to ensure its dominance in a few generations. Fischer (1978) found that each centimeter of superiority in height among spaced wheat genotypes (40 X 40 cm) gave a yield advantage of 0.58%. At normal densities this advantage is considerably enhanced (Jensen and Federer, 1964; Khalifa and Qualset, 1974). Similar results have been obtained in a segregating population of wheat. During bulk breeding, Khalifa and Qualset (1975) found that the mean height of the population increased from the F, to the F, and that the shortest types were eliminated. There was also a negative relationship between F, height and pure culture yield. Data from other cereals have been similar. In a segregating barley population, there was a positive relationship between single plant height and yield in the F, population; this was reversed in the F, plots (Hamblin and Donald, 1974). Perhaps the most striking data are those available for rice. When two tall, leafy varieties of rice were mixed in equal proportions with three semidwarf, erect cultivars (each providing 20% of the seed sown), the low-yielding tall varieties suppressed the high-yielding semidwarf varieties within four generations to less than 0.5% of the population (Jennings and de Jesus, 1968). Similar results occurred with segregating rice populations (Jennings and Herrera, 1968). In

106

C. M. DONALD

AND J. HAMBLIN

maize, dwarf types yielded less when grown in competition with tall types than they did in pure culture (Pendelton and Seif, 1962). Thus over the millennia, and probably quite early in the history of cropping, annual seed crops became tall. Percival (1921) instances Triticum uestivurn to 150cm, T. turgidurn to 180 cm, and even T . compacturn (Club or Dwarf Wheats) to 140 cm. Yet under marginal soil and climatic conditions, even these tall wheats were superseded and replaced by the still taller rye (Secale cereale). Under fertile conditions in the United States, maize attained heights of more than 3 m by the mid-twentieth century; Goldsworthy (1970) records a local sorghum of more than 4 m in height in Nigeria. The advantages of height in competition are not confined, however, to cereals. We have estimated (from the data in Table I and Fig. 1 of Schutz and Brim, 1967) that each centimeter of height advantage in soybeans gave Jackson (a tall variety) a yield increase of 2.5%, and 4% of mean yield, when competing in hills 46 cm apart with the short varieties Hill and Lee. The advantage was 0.7-1% when competing in rows 107 cm apart. In pure culture, Hill and Jackson had similar yields whereas Lee was 13% higher yielding than Jackson, the tall variety. These results are partly confounded by maturity effects, but in all cases the taller lines were more competitive than the shorter lines, and the late lines were more competitive than the early lines. Tallness in seed crops had certain advantages to early cultivators, which continues in the village agriculture of many regions. It gives stem material of value for fuel, bedding, building, and thatching purposes, so that tallness is esteemed. Also tallness is at an advantage when competition with weeds is severe (Pal et al., 1960). It tends also to be linked, in the minds of many growers, with greater yield; as will be discussed later, this is a fundamentally mistaken belief in weed-free situations. But there is little reason to believe that tallness developed through selection by man, because it dominates through natural selection within a few generations of its appearance within a crop. Finally, an equilibrium is reached when the tallest plants suffer grain loss or collapse because of wind damage. Further, these tallest plants tend to have reduced harvest indices (Rosielle and Frey, 1975; Donald, 1981; Hamblin and Rosielle, 1983), so that they yield less grain than their slightly shorter neighbors. iii. For a Leafy Canopy. The third powerful influence of natural selection within seed crops is for a canopy of wide, horizontal or floppy leaves. Such plants are able to intercept light preferentially. The large, subcircular, horizontally disposed leaves of cotton and the large leaves of many sunflower cultivars permit considerable light interception even by crops of low leaf area index (LAI). The influence of canopy structure on competitive ability was demonstrated by the comparative behavior of wheat varieties of different leaf habit (Tanner et al., 1966). In a weedy situation, the floppy-leaved varieties were able to suppress weeds and yield well whereas those of erect leaf habit were depressed in yield. In

CONVERGENT EVOLUTION OF ANNUAL SEED CROPS

107

a weed-free situation, the yields were reversed. Similarly, in an F3 population of barley at strongly competitive spacing (Hamblin and Donald, 1974), there was a positive correlation between leaf length and yield. Reference was made earlier to the almost total suppression of short, erect-leaved rices by tall, floppy-leaved cultivars. The leafhess of successful competitors must also be subject to stabilizing influences. Maize or rice plants cannot continue to grow taller and leafier indefinitely. The disadvantage of tallness and leafhess, especially of tallness beyond that needed for successful competition, is the tendency of the crop to lodge, leading to a disorganized light profile, reduced seed production, and harvest problems. The trend to leafhess and strong competitive ability also tends to be associated with heavy water use, prolonged growth, lateness in maturity, and a decreased harvest index (Donald, 1981). An unexpected feature of the evolution of wheat under domestication has been the much lower photosynthetic rate per unit area of leaf of modem wheats compared to primitive species of Triticum and Aegilops (Evans and Dunstone, 1970; Khan and Tsunoda, 1970; Evans and Wardlaw, 1976). Evans and coworkers suggested that this falling rate is caused by the reduced surface/volume ratio of the mesophyll cells, which Kranz (1966) had shown to have become progressively larger during domestication. However, Evans and Dunstone (1970) found that the leaf size had increased more than the photosynthetic rate had fallen, so that the photosynthesis per leaf was much greater in modem wheats. They also noted a positive relationship between leaf size and grain size and reasoned that selection for yield would lead progressively to increased grain size, increased leaf size, larger cell size, and lower photosynthetic rate. Khan and Tsunoda (1970) take the view that the change in leaf size and photosynthetic rate is caused by an improvement in the environment of plants under agriculture which has selected for a mesophytic habit from a wild xerophytic habit. Similar changes with domestication have occurred in cotton and tomatoes (Stebbins, 1974).

An alternative explanation has been offered for this enigma of falling photosynthetic rates during the evolution of wheat under domestication (Donald, 1981), based on competitive relationships. Within the crop canopy, plants with large, usually wide, floppy, drooping leaves would have had strong competitive ability for light and a clear selective advantage throughout domestication. As long as the photosynthetic rate per leaf was sufficiently maintained, a progressive increase in leaf size ensured natural selection, with a consequent relaxation of selection pressure on photosynthetic rate per unit of leaf area. Less leafy plants were at great selective disadvantage because of shading by their leafy neighbors which, although they might have been “physiologically weaker,” were “ecologically powerful.” Thus there would again be stabilizing selection between directional selection for larger leaves in the competition for light and opposing

108

C. M. DONALD

AND J. HAMBLIN

directional selection for an adequate or advantageous rate of photosynthesis. It is proposed that modem wheats have leaf sizes and photosynthetic rates representing the outcome of this selection. Selection for yield, it is suggested, has not been involved as the initiating factor, and man has played a role in the falling rates of photosynthesis only through his crop production. iv. For Tillering or Brunching. The selective advantage of free tillering is illustrated by the response of wheat to poor establishment, which was doubtless common in early agriculture. Abundant tillering compensates for low plant uumbers. When the stand density of wheat at Adelaide was reduced from 184 to 35 plants/m, the number of tillers per plant increased from 5.5 to 13.7, and the number of grains per plant increased from 46 to 215, with no significant change in yield per square meter (Puckridge and Donald, 1967). Similar results were obtained by Bremner (1969). In both studies, a genetically uniform cultivar was used; it is clear that if a genetically diverse crop were depleted in plant numbers for any reason, free-tillering genotypes would have great selective advantage over less tillered kinds. Under domestication there seems to have been little reduction in the tillering capacity of wheat or barley. Modem cultivars are capable of producing 30 or more fertile tillers (ears) per plant at wide spacing, although only 2 or 3 are produced when competing at crop densities (Puckeridge and Donald, 1967). In contrast to wheat and barley, there has been a strong trend in maize and sorghum toward single-stemmed plants. One may ask why this should be so, because they are all graminaceous plants of basically similar vegetative structure. The probable explanation lies in the culture of these four species by man. Wheat and barley have always been harvested as a plant community, with sickle, scythe, mower, or header, and there is no recognition of the individual plant. In contrast, for many millennia maize and sorghum have been hand-harvested by pulling the cob or by cutting off the inflorescence. Here lay the opportunity to set aside the largest cobs or heads for seed the following season. This would lead indirectly to the preferential selection of sparsely tillered and ultimately of single stemmed plants, an instance of secondary selection by man. The phenomenon of branching in dicotyledonous crops is ecologically parallel to tillering in cereals. The competitive ability of soybean cultivars was assessed in several experiments by their growth in pure cultures and mixtures (Mumaw and Weber, 1957). The most consistent feature linked with the competitive success of a cultivar in mixtures was the branching growth habit, even exceeding the influence of a 12-cm height advantage of some cultivars. In all comparisons over 2 years, branching varieties contributed about 60% of the total yields of the mixtures, and nonbranching varieties contributed about 40%. v. For Seed Size. In most natural communities, small seeds have marked selective advantage over larger seeds. Small-seeded annual plants probably will yield a greater number of seeds than large-seeded plants: small seeds are more

CONVERGENT EVOLUTION OF ANNUAL SEED CROPS

109

easily dispersed and buried; grazing animals can pick up small seeds less easily, are less likely to crush them while chewing, and are more likely to pass some of the seed through the digestive tract undamaged. Most of these selective advantages disappear in the crop situation; there are no grazing animals, and dispersal and resowing do not depend on natural forces. Large seeds have one powerful selective advantage, namely, they produce large seedlings (Evans and Bhatt, 1977) which have strong competitive ability over smaller seedlings. In the wild state this advantage probably is reduced or lost because of defoliation by animals, but within a crop larger seedlings give rise to larger plants with more seed. When wheat seeds of 45 mg were sown alternately with small (27 mg) seeds of the same variety, the large seeds produced plants with seed yields 57% greater than those from the small seeds (Christian and Grey, 1941). However, in mixtures the advantages of large seed size do not necessarily lead to survival. Hamblin (1975) found that the relative competitive advantage of bean varieties (Phaseolus vulgaris) changed with environment; large-seeded types yielded relatively more when competition was most severe (i.e., when yield/plant was small), but in nearly all situations the smallest seeded variety produced the most seeds per plant (was more prolific), although it was never the highest yielding variety and often yielded only moderately. He also showed that as the small-seeded type dominated the mixture so the average yield of the population fell. In another study (Hamblin and Morton, 1977) involving segregating populations, natural selection was always for small seeds, and in three out of four cases it was for increased seed numbers. The results were obtained at crop densities and contrasted markedly with the results obtained at low density, illustrating the importance of not extrapolating from one situation to another with changed competitive relationships (Donald and Hamblin, 1976; Donald, 1981). A similar result was found for a bulk cotton population that was grown without conscious selection for 10 generations; there was a linear increase in seed numbers and a linear decrease in seed size over the generations (Quisenberry et al., 1978). These stabilizing factors for seed size (i.e., the competitive advantage of large seeds and the greeter prolificacy of plants with small seeds) would, despite fluctuations in their significance from season to season, lead to a loose equilibrium for seed size. But early in the history of cropping, another factor was superimposed: the conscious selection by man of larger and plumper grain, features associated in the minds of growers with high yield. Several Greek and Roman writers emphasized the importance of retaining large grain from the harvest to be sown the following year (Percival, 1921). Large grains could readily be separated during winnowing or by shaking grain on a shallow tray. Thus an added and powerful selective advantage, unrelated to field performance, lay with plants producing larger seeds. Although continuing recombination and segregation also would have ensured stabilizing selection for small-seeded

110

C . M.DONALD AND J. HAMBLIN

prolific genotypes, the equilibrium size presumably would have tended to increase. The doubling of seed size from the wild wheats (Triticum rhaoudar, to 20 mg) to the cultivated wheats (T. aestivum, 40-45 mg) was doubtless partly the result of active selection for seed size by man, but the increase in competitive advantage of large seeds and seedlings under wheat cultivation as compared to that in the wild state was probably notable. In some crops there was intense selection of individual seeds, within the harvested crop, for color, shape, and greater size; this was particularly so in the common bean (P. vulgaris) and also in many grain legumes. Not only were distinct and strikingly different races of beans developed in nearby villages, but there was a fivefold range in seed size under domestication (Evans, 1973), probably far beyond the limits of any equilibrium relationship resulting from natural selection. Perhaps man’s concern for selecting large seeds in cereals was reduced, eventually, because it added little to yield. For example, in Christian and Grey’s study (1944), there was no difference in yield between crops established wholly from large seed (45 mg) and those established wholly from small seed (27 mg) of the same genotype. If genetic selection has been made for large seed size, there is usually a compensating decrease in the number of grains per plant (Grafius et al., 1976; Hamblin and Morton, 1977). Indeed the data of Grafhs er al. indicated that the best means of increasing yield was to select for grains per head and to allow seed size to vary more or less randomly. Although man has taken a strong interest in larger wheat seed, the influence of natural selection has been so allpervading and continuous that his direct influence, at least until plant breeding began, may have been quite limited. vi. For Speed of Germination. In the wild, the irregular or protracted germination of wheat or of any other annual grass is a partial protection against uncertain climatic conditions, so-called false starts to the rainfall season. But cultivated wheats germinate more rapidly and evenly than do wild wheats (Evans and Dunstone, 1970). This evolution to speedy germination was wholly because of natural selection. Within a crop sown in a prepared seed bed, plants that emerged first had strong competitive advantage over their neighbors. In a drilled barley crop (cv. Clipper) at Adelaide, plants that emerged 1 day earlier gave rise to seedlings 15% heavier at day 17 (from sowing) and to plants 14% heavier at day 70. There was a reduction of 14% in the mean number of grains per plant for each delay of 1 day in emergence (Soetono and Donald, 1980). In that instance, the differences in day of emergence were caused principally by individual seed environments (depth, soil physical condition, and so on), but the same effect of delay would occur when prolonged germination was a genetic character. Thus there would be a progressive selection under crop conditions toward rapid and simultaneous germination, a feature, albeit imperfect, of most modem annual seed crops.

CONVERGENT EVOLUTION OF ANNUAL SEED CROPS

111

vii. For Root Characters. Knowledge of the role of root systems in natural selection under domestication is seriously lacking. If it is reasonable to extend to the root system our understanding of the features of plant tops giving selective advantage through competition, one would suggest that just as a tall, leafy, tillered plant secures an undue share of the light, so plants with a widely ramifying root system would be at a selective advantage by absorbing water and nutrients more rapidly and more extensively than could neighboring plants with a restricted system. Passioura (1972) has shown that restricting root growth early in a plant’s development reduces preflowing water use, and he has suggested that this characteristic may be important in situations where a crop is maturing grain in dry situations. However, O’Brien (1979) has pointed out that reduced root development in early stages of growth may lead to problems of nutrient uptake. It is not possible, with our current knowledge, to make any generalizations about root growth and crop development (Fischer, 1981).

111. ECOTYPIC PARALLELISM IN CROP PLANTS It is evident that all crops have been subject to many similar selective and evolutionary processes during domestication. They have become adapted to both the natural environment of the region and the manmade environment of local cultural methods. Despite the diversity of environments in the cropped areas of the world and the specific responses by individual crop species, there have been many common trends in all crops. Various writers have pointed to features which they regard as typical of wild plants and which commonly disappear under cultivation. These especially include morphologically wild characters associated with seed dispersal and seed burial (awns, brittle rachis, shattering pods, pointed seeds, wings, spines, etc.). Positive responses have included the development of synchronous ripening, rapid and simultaneous germination, and larger seeds. Selection through competition has been recognized in its more generalized expression; those genotypes that are prolific and produce more seed, and in the next generation more seedlings, contribute an ever-increasing proportion of the population. However, there has not been adequate recognition of the influence of competition in ensuring the success of plants of common architecture in all seed crops, irrespective of species, soil, or climate. Their structure can be clearly designated: tall, free branching (or tillering), a dense canopy of large, horizontally disposed leaves, and indeterminate habit. Although these may be combined and expressed in many variant ways, they integrate to give a plant of strong competitive ability. Turesson’s (1922) concept of an ecotype, “a product arising as a genotypical

112

C.

M.DONALD AND

J. HAMBLIN

response by the ecospecies to the particular habitat,” is applicable to this situation. Within each crop species a cultivation ecotype has developed that is comparable to Turesson’s climatic or soil ecotypes. Turessm further recognized that botanically and morphologicallydiverse species in a common environment may show similar responses. For example, he pointed out that many unrelated species display a stable prostrate form on the exposed west coast of southern Sweden. Philipschenko (1927) called this phenomenon ecotypicparallelism, whereby the identical reaction of related or unrelated organisms to a particular environment resulted in the appearance of a series of similar ecotypes that may be of quite dissimilar genetic structure. There has been ecotypic parallelism among all annual seed crops within the cropping environment through the development of plants of strong Competitive ability having an improved capacity for seed production compared to their wild counterparts and a lack of dependence on burial and dispersal mechanisms for their seeds. A widespread assumption among crop ecologists is that all these changes during domestication have increased productivity; that because an individual plant produces more seed than its neighbors its frequency will increase in the next generation, and as the whole community progressively is made up of more prolific plants it thus becomes every higher yielding. This is false. The cultivation ecotype certainly has advantages over the wild ecotype in its capacity for grain yield, such as prolificacy, strong and rapid establishment, larger seed, and freedom from excessive spines or wings on the seed. But many cultivation ecotypes have serious weaknesses associated with their strong competitive ability; each plant competes against its like neighbors. Only now are these being slowly overcome, usually by empirical means. Recognition of this phenomenon, and of the consequent pattern of progress needed in annual seed species, can contribute toward substantially increased grain yields.

IV. SELECTION, EVOLUTION, AND CROP YIELD

Crop yield is a manmade concept. It does not necessarily relate to natural selection or to crop evolution and it is expressed by the nonbiologicalcriterion of weight of product per unit area. The harvest in some crops is a vegetative part (sterns, leaves, or roots) whereas in others it is a reproductive organ (the fruit or seed). Yet whatever the plant part man has used, natural crop evolution on the one hand and trends in crop yields on the other must be recognized as separate, if interrelated, phenomena. Many believe that in the case of crops grown for their seed, plant evolution and increased seed yield must inevitably proceed hand in hand. This is not so. In

CONVERGENT EVOLUTION OF ANNUAL SEED CROPS

113

some instances, as evolution continues seed yield advances; in other instances, it declines. “Natural selection, it should not be forgotten, can act solely through and for the advantage of each being,” wrote Darwin (1868, p. 184). Within seed crops it ensures the greater abundance of certain genotypes in ensuing crops, without implications for yield by the entire seed-producing community. And therein lies the dichotomy: on the one hand the performance of the individual competing in a mixed community, and on the other the performance of that same individual genotype growing as a pure crop stand, each plant competing against like neighbors. During selection by man, the bases for many of his choices were, by selfdefinition, advantageous. If he preferred and selected large mottled seeds, then any increases of these in the next generation were advantageous, by his standards. A secondary effect, such as a reduction in the number of seeds produced per plant, might reduce his yield per unit area, however. When this occurred, he may have accepted it; more probably he was unaware of it. When man has sought to select for yield deliberately, he has usually based his attempts on the performance of individual plants or individual shoots. Selection of larger wheat ears, corn cobs, sunflower heads, or plants with more inflorescences was considered a route toward higher yields, and no doubt was highly successful during the early years of domestication. But, as considered later, such selection had progressive limitations. Natural selection for adaptation to the farm and to the farmer’s practices also offered fum prospects that the progeny would contribute to increased seed yield. Again we emphasize the distinction between biological prolificacy (many seeddplant) and crop-to-crop survival that ensures representation in the bulk seed sown for the next crop (see Section III,B,2a). Plants that have both high biological prolificacy and strong crop-to-crop survival will dominate because of exacting natural selection for performance within the agricultural environment. It is especially in relation to natural selection by interplant competition that evolution and increased yield do not go hand-in-hand. There is growing experimental and circumstantial evidence that successful competitors within mixtures of biotypes may be poor producers in pure stands. Instances in which these features are either unrelated or negatively related are recorded in many seed crops, particularly in the cereals, and they are discussed in the following section. A. COMPETITIVE ABILITYIN MIXTURESVERSUS SEEDYIELDIN PURESTANDS

The successful plant within a genetically uniform crop growing in a uniform environment will be the plant suffering the least competition from its neighbors.

114

C. M.DONALD AND I. HAMBLIN

It follows then that the crop should be comprised wholly of plants of low competitive ability, interfering with each other’s growth only minimally (Donald, 1968a,b). Such plants are, in Darwinian terms, unfit plants because of their weak capacity to compete or survive against other plant types in natural communities or in crops of mixed genotypes. A further postulation may be made: strong competitors, which in general are tall, leafy, and freely branched, not only suppress short, erect, unbranched plants but when grown in a pure community may so interfere with each other’s growth (as, for example, by strong mutual shading) as to perform relatively poorly. If this is so, then one not only may expect that performance in mixtures will be no guide to performance in a pure stand (a nil relationship), but further that there will be instances of a negative relationship between performance in the two situations. Both these situations have been reported frequently.

1 . Wheat The first recorded case of yield reversal for varieties grown in pure and mixed culture was described by Montgomery (19 12). He found that one variety rapidly dominated the mixture but that it was not necessarily the one that was highest yielding in pure culture. Similar results have often been reported (Engledow, 1925; Klages, 1936; Christian and Grey, 1941; Laude and Swanson, 1942; Khalifa and Qualset, 1974). The results of Khalifa and Qualset (1975) on a segregating wheat population, suggesting that competition was eliminating the short, high-yielding lines, have already been discussed (see Section III,B ,2,b,ii).

2 . Rice Perhaps the most striking case of a negative relationship between competitive ability and yield among crop varieties is that reported between tall and semidrawf rice cultivars in the Philippines (Jennings and de Jesus, 1968). The strong competitive ability of these tall leafy varieties was discussed earlier (Section II,B,2,B,ii and iii), but these successful cultivars in mixtures were poor producers in pure culture, When the dwarf erect types were almost eliminated, the lower yielding of the two tall cultivars reduced the other tall cultivars to a low frequency in the mixture. Similar results were obtained by Sakai (1955) and Akihama (1968). Jennings and Herrera (1968) also demonstrated a negative relationship between competitive ability in mixtures and yield in pure culture for segregating populations. 3. Barley

Harlan and Martini (1938) grew 11 barley varieties at 10 centers across the United States for 4-10 years; the seed harvested at each site was resown at that

CONVERGENT EVOLUTION OF ANNUAL SEED CROPS

115

site the following year. One variety rapidly dominated the mixture, but the particular variety varied with the site. In several instances, the variety most successful in terms of farm use in the region was reduced to a very low frequency. The poor competitive ability of genotypes that are high yielding in pure culture has also been reported by Suneson and Wiebe (1942), Suneson (1949), Wiebe et al. (1963), and Allard and Adams (1969). On the basis of morphology (long leaves, tall plants), Hamblin and Donald (1974) suggested that the high yield of individual F, plants was the result of high competitive ability. This situation was reversed in pure culture F, plots, where high yield was associated with short, small-leaved plants. The negative relationship between competitive ability and pure culture yield for these lines was confirmed by direct measurement (Hamblin and Rowell, 1975). 4 . Oats

Smith et al. (1970) examined all pair combinations of five oat varieties. The tallest variety (Rodney) was the lowest yielding in a pure culture but the most competitive in mixed culture. In pure culture, however, a variety of intermediate height (Brave) had the highest yield; nonetheless, competitive ability was closely related to plant height.

5 . Maize The yield of brachytic maize genotypes grown in alternate rows with normal maize genotypes was less than when this maize was grown as a pure stand (Pendleton and Seif, 1962). The mass selection experiments for yield of Gardner and co-workers (Gardner, 1961, 1968, 1969; Lonnquist et al., 1966, personal communication) can be interpreted in terms of selection for increased competitive ability (our interpretation) rather than for high-yield potential (Gardner’s interpretation). These workers mass-selected single plants on the basis of plant yield, using a grid system to give local control of environmental variation (Gardner, 1961). The selected individuals provided the progeny for the next round of hybridization and selection. Eventually the different generations were tested for yield, and it was found that yield increased linearly with time for several cycles and then reached a plateau. The plots used were either single or double unbordered rows in which the within-row spacing was the same as the between-row spacing. The height of the populations increased in step with the yield. This would mean that the most advanced generations were surrounded by shorter, earlier generations whose yield potential would not have been fully expressed because of interrow and intergeneration competition. Ultimately, the plants became so tall that increased competitive advantage was offset by the disadvantages of increased lodging, and no further yield increase was observed. Increased yield from mass selection is a rare event; therefore it is important

116

C.

M.DONALD AND J. HAMBLJN

that further work be done to determine whether mass selection for yield was effective, or whether selection was in fact for competitive ability. This is particularly important in view of the contrasting CIMMYT results to be considered in Section V,D.

6. Sorghum Averaged over 16 hybrids and two environments, Kern and Atkins (1970) found significant yield depression for short hybrids bordered by tall hybrids and significantyield increases for tall hybrids bordered by short hybrids. On average, a k m difference in height between rows increased or decreased yield by 0.2% for the taller and shorter hybrids, respectively. Kern and Atkins (1970) considered that small yield differences between genotypes of different heights were of doubtful validity unless the data were obtained from bordered plots.

7. Soybeans The performance of soybean cultivars in pure culture was not necessarily related to survival in mixture (Mumaw and Weber, 1957). Survival was related to branching pattern and height. Branching in dicotyledons may be ecologically parallel to tillering in graminaceous crops. Similar results were obtained by Schutz and Brim (1967) and by Hinson and Hanson (1962) in which height and maturity were the dominant factors. Mumaw and Weber (1957) concluded that “a relatively high yield of a variety in pure stands was not necessarily an indication of its ability to survive in mixed populations.”

8. Beans (Phaseolus vulgaris) The experiments of Hamblin (1975) have already been considered (Section II,B,2,b,v). In summary, he found that yield in pure culture and survival in mixtures were not related.

9. Sunflowers Working with eight varieties of sunflower, Fick and Swallers (1975) found that in one-row plots the tallest variety had the highest yield, and that there was a close correlation between height and yield (Fig. 1A). When three-row plots were used the height/yield correlation disappeared (Fig. 1B) and the yield rankings changed markedly. In three-row plots, there is still a suggestion that the shortest genotype suffereda competitive disadvantage. Also, it was probably sown at a density too low for maximum yield.

CONVERGENT EVOLUTION OF ANNUAL SEED CROPS

117

.r

*k

Height (cm)

FIG. 1A. The effect of height on yield of single rows of eight varieties of sunflower. These. rows are not bordered. Note there is a strong correlation between height and yield.

120

150 lecl Height (cm)

FIG. 1B. The effect of height on yield of the same varieties in 3-row plots, where data was obtained from the center TOW only. These rows are bordered. Note there is no correlation between height and yield, although it appears that the center row of the shortest genotype is still suffering interplot competition. (Data from Fick and Swallers, 1975; Table I.)

10. Cotton

Significant genotype-neighbor interactions were found for yield of four cotton varieties grown as single rows (Moran-Val and Miller, 1975). Competitive ability was in part related to height and the authors cautioned that “If competitive effects of the magnitude observed in this study are generally present in yield trials, however, fully bordered plots would be indicated.” 1 1 . Comments on the Results Presented in This Section

The papers just discussed, relating yield in pure culture to competitive ability in mixtures, suggest that there is frequently zero or negative association between these two factors. Although this negative association has not always been found between yield and competitive ability (e.g., Johnston, 1972; Blijenburg and Sneep, 1975), it is so common that it cannot be ignored. In the cases where a positive association has been found between yield and competitive ability, the

118

C.

M.DONALD AND J. HAMBLIN

comparisons are often between adapted and unadapted types so that the observed result is to be expected. It may be argued also that the results obtained using mixtures of varieties are artifacts caused by the small sample of varieties used in many of the experiments discussed. We believe that the latter argument is not correct for several reasons. First, there is the sheer number of experiments involved; the result has been observed so often that it is unlikely to be an artifact. Also the argument that results from mixture experiments using varieties may be artifacts applies equally whether the relationship between yield and competitive ability is positive or negative. Second, and more convincing, there are studies involving segregating populations (Jennings and Herrera, 1968; Hamblin and Donald, 1974, Khalifa and Qualset, 1975; Hamblin and Rowell, 1975) in which yield in pure culture was not associated with competitive ability in mixtures for a whole range of species. In these experiments the lines used were related and chosen at random. Third, the results make evolutionary and biological sense (see the next section and Section II,B,2,b,i-vii). 12. Principles Governing Competitive Success and Yield It follows from the foregoing studies that successful plants for pure-stand grain yield are often poor competitors (however, we do not equate poor competitive ability with physiological deficiency). In the study of wheat by Christian and Grey (1941), seed size was the feature giving competitive success; in the study by Suneson (1949), it was a prolific root system; in those for rice (Jensen and de Jesus, 1968) and barley, (Hamblin and Donald, 1974) it was height, leafhess, and leaf length; and for soybeans (Mumaw and Weber, 1957), it was a branching habit. In each instance of yield reversal, it was the shorter, less leafy, less branched cultivars or segregates that gave higher yields in pure stands. Three propositions may now be stated: 1. That, under domestication, competitive plants gained dominance through the natural selection of “the fit”; in some instances, man’s purposeful selection of successful competitors has maintained the place of these plants. 2. That, in pure crop stands, highly competitive plants give lower seed yields than do less competitive individuals, especially at high plant densities. 3. That, because natural selection has favored competitive plants with reduced capacity for yield in pure culture, various natural selection processes must be reversed by plant breeding and selection.

In the Darwinian sense of fitness, it is the unfit plants that will succeed in the pure culture crop situation. Acceptance of this proposition imposes on the plant breeder the need to avoid using selection on individual plant performance in

CONVERGENT EVOLUTION OF ANNUAL SEED CROPS

119

situations where there is significant competition among plants of differing genotypes. In this circumstance, poor competitors, whatever their potential merit in pure stands, will give depressed yields. The unfit plant must be allowed to express its potential in pure culture; such plants will perform poorly in mixtures with other genotypes, as in a segregating population in rows or hill plots of mixed genotypes. They will also perform poorly in spaced plant situations (Donald and Hamblin, 1976). Individual plant yields in those circumstances may be positively misleading for predicting crop yields (Wiebe et af. (1963). However, a crop comprised of a single genotype of weak competitive ability still depends on effective exploitation of the environment for high yield. If the annual plants to be used in crops are weak competitors, then the number of plants per square meter must be increased to ensure that they compete with one another sufficiently to exploit the environment fully. It may seem a paradox to propose that productive annual crops will be comprised of weakly competitive plants and to say these plants must be sown at a density sufficient to ensure that they will compete intensely with one another. But these are not incompatible thoughts or objectives. Each is aimed at increased crop efficiency, the first by reducing the pressure of each plant on its neighbors through plant form, and the second by increasing the pressure of the whole community on the available resources through an increased population density. Thus, testing should not merely avoid competition between different genotypes. The lines to be evaluated must be tested in pure stands at densities sufficiently high to ensure that there is interplant competition of considerable intensity.

B.

BIOLOGICAL YIELD, HARVEST INDEX,

AND

GRAINYIELD

The relationships between biological yield and seed yield in seed-producing annual crops display important differences, as is illustrated in Fig. 2. The upper graph (Fig. 2A) shows the general relationships found in cereals (Donald, 1963). Biological yield increases with density until it reaches a plateau. This is maintained up to very high densities unless crop failure occurs from lodging or the advent of disease among the attenuated plants. Grain yield increases to a maximum at a density approximating the minimum density giving full biological yield. To the extent that when maximum seed yield is attained there is maximum exploitation of the environment in terms of biological yield, cereals are efficient in ensuring their prolificacy. On the other hand, the maximum yield of seed cotton or lint by cotton crops is achieved at a density at which the biological yield is still rising steeply (Fig. 2B). The biological yield of the highest yielding crop falls far short of the capacity of the cultivar to produce dry matter. In this regard it is an inefficient crop, which will eventually be replaced to advantage by cultivars giving increasing lint yields up to full biological yield. It may be that this contrast between an annual such as

120

C.

M. DONALD AND J. HAMBLIN

CEREALS

-

Dm~ity COTTON

Density

-

FIG. 2. The general relationship between biological yield (BY) and grain yield (GY)for cereals (afterbnald, 1%3)aodforcotton(dataderivedfromKirkeral.. 1%9;Fig. 14,16,19,21 (BY);21 (GY).(GY is the fruit weightlha; seed yield data was not available).

wheat and an annual such as cotton relates to the period for which they have been cultivated (i.e., wheat for some millennia, annual forms of cotton for little more than 500 years) (Hutchinson, 1965). The future trend in the relationship between biological yield and grain yield of annual crops may follow the pattern shown in Fig. 3. If nonbranched plants are used, the density @lants/ha) required to give the full biological yield will be greater. Because of the improved canopy structure, the biological yield attainable will be greater, although perhaps not markedly so (10% in the example shown). The main contribution to potential yield will be an improved harvest index, perhaps 0.35-0.40 to 0.50 in cereals, representing an increase of about 25% in grain yield. The influence of an increased biological yield of 10% and an increased harvest index of 25% would be a 37% increase in grain yield. The extent to which the trend toward increased harvest index is displayed in annual seed crops of diverse growth form is discussed later in this article. The same trends are evident in each crop, and there is no instance in which newer cultivars demonstrate a reversal of the relationship. However, whether harvest index itself is a useful selection criterion, or a useful description of past selection trends, has yet to be critically determined (Hamblin and Rosielle, 1983).

CONVERGENT EVOLUTION OF ANNUAL SEED CROPS

I

BY

121

--

0-

II fi'

Density

*

FIG. 3. Present relationship (solid lines) and possible future relationship (broken lines) between biological yield (BY) and grain yield (GY).

V. PROGRESS AND PROSPECTS IN THE DEVELOPMENT OF ANNUAL SEED CROPS In this section we briefly review crop evolution, including recent plant breeding efforts for a series of annual seed crops. This will allow a basis for proposing a generalized ideotype for all annual seed crops. A. WHEAT

Wild wheats, both diploid and tetraploid, are annuals and have primitive features associated with effective seed dispersal and burial (Bell, 1965). Less easily defined but no less important characteristics of these wild wheats relate to growth form, permitting survival and seed production under close grazing by sheep or goats. Many are fine stemmed and prostrate, so that some ears are borne on a nearly horizontal stem only a few centimeters above the ground and can lie within a grazed sward. Under cultivation, these primitive characters of wheat were at a selective disadvantage and have been lost, except for the brittle rachis and enclosed grain of T. munucuccum (Bell, 1965). Taller, more erect, and more leafy plants had major competitive advantage and probably became dominant quickly. Bruegel's paintings (sixteenth century) depict shoulder-high wheat crops. Although having a tendency to lodge, these tall crops were feasible in premechanized agriculture because most of the grain would be recovered during hand harvesting. In early wheat breeding programs, improved disease resistance was a major

122

C.

M. DONALD AND J. HAMBLIN

contributor to increased yields (Athwal, 1971). Yield potential also increased (Austin et al., 1980; L. T. Evans, 1980; Perry and Reeves, 1980; Kulshrestha and Jain, 1982). However, this yield potential increased markedly with the incorporation of the Norin 10 gene in wheat. These short, fertilizer-responsive wheats had been grown in Japan long before scientific plant breeding and were successfully used in Italy 60 years ago (Athwal, 1971). However, it was only when Vogel at Pullman and then Borlang at CIMMYT used this material that the so-called semidwarf wheats made such an impact on world wheat yields in the 1960s and 1970s. These wheats are resistant to lodging and have many tillers and grains per spikelet. Reduced stature and resistance to lodging are characteristics sought in the “all-crops ideotype;” increased number of grains per spikelet in the semidwarf wheats defines the expression of increased yield. On the other hand, the freetillering and relatively broad, lax leaves of these highly successful semidwarf varieties are a challenge to the all-crops ideotype. Many workers believe that only a small number of tillers is needed to give both maximum yields and sufficient plasticity to permit adaptation to the environment (MacKey, 1966; Hurd, 1969; Bingham, 1972; Jones and Kirby, 1977). This view was extended by Donald (1968a,b), who described a wheat ideotype for high grain yields with a short, strong stem, few small, erect leaves, a large ear in relation to the total dry matter (i.e., a high harvest index), an erect ear, awns, and a single culm. (In view of the authorship of that article, this wheat ideotype conforms to the common ideotype for all annual seed crops described in this article). Atsmon and Jacobs (1977) have produced uniculm wheat lines of medium height, high harvest index, and resistance to lodging; they appreciably outyielded the standard cultivar of the region. Further evidence for the potential of controlled tillering was presented by Islam and Sedgley (1981), who examined the effects of manually detillering wheat plants in the field to give biculms. The performance of these was compared with normal-tillered control plots of the same variety. The detillered plants outyielded the controls by 14 and 22%, respectively, in 1978 and 1979. B. BARLEY

Barley (Hordeurn distichurn) is a cereal with a growth form and physiology similar to wheat, so that most considerations of canopy structure, tillering, and leafiness are applicable to both species. Cultivated barleys vary in height from brachytic forms (40 cm), widely grown in the Middle East, to tall forms (1.5 m) (Reid and Weibe, 1968). They tiller freely at low density, especially the tworowed types, but there is a mutant form with a single stem per plant (uniculm). There are considerable differences in leaf size, with extremes of leaf shortness in

CONVERGENT EVOLUTION OF ANNUAL SEED CROPS

123

brachytic kinds and of leaf narrowness in the mutant form governed by a single gene. Two-rowed barleys generally have narrower leaves than six-rowed forms. During this century there have been two principal trends relating to plant form and productivity, as illustrated among varieties released, in the United Kingdom. There has been a progressive reduction in height from about 1 m (cv. Spratt, pre-1900) to semidwarfs of about 70 cm. This has been accompanied by an increase in harvest index from about 0.4 to 0.5 (Cannell, 1968; Hayes, 1970; Donald and Hamblin, 1976). This increase has not been consciously sought but is the result of a substantially constant biological yield concomitant with advances achieved by selection for grain yield, early flowering, and reduced plant height (Hamblin and Rosielle, 1983). Interest in plant form as a contributory feature to future yield relates to height (further reduction seems probable), reduced leafiness, and less tillering. Jones and Kirby (1977) believe that although tillering is invaluable for adaptation to the environment, it can serve this role adequately with only a small number of tillers. Donald (1977), using his wheat ideotype as his model, has produced semidwarf, uniculm barleys which, when sown at about double the standard seed rate, outyield the leading local cultivars by 15-20%. However, these lines were not evaluated for grain quality. The initial attempts to produce radically new cereal plants (Atsmon and Jacobs, 1977; Donald, 1979) are sufficiently promising to warrant further effort. As well as the increases in yield that are evidently attainable through dwarfing and the elimination of tillering, further substantial increases may be possible through the development of nonleafy lines with short, narrow, erect leaves (Hamblin and Donald, 1974). The retention of awns seems desirable (Frey, 1971). The use of biological yield and harvest index as a means of interpreting behavior during breeding programs for yield has been strongly advocated (Donald and Hamblin, 1976).

Until 1900, the typical cultivated rice was a tall, strongly competitive plant which had emerged by natural selection within man’s crops because of its capacity to suppress weeds and more dwarf kinds of rice (Jennings, 1964; Athwal, 1971). It had long, broad, drooping leaves and thick culms, was strongly photoperiodic, and was subject to serious lodging, particularly if fertilizer was applied. The first development of a more productive rice of communal habit was in Japan early this century, when cultivars of Oryzu juponicu were bred with erect habit, reduced height, short, stiff straw, and fewer tillers. These varieties did not lodge with heavy applications of nitrogen. This was followed by 0. indicu varieties of similar noncompetitive habit, first in Taiwan with the release

124

C. M.DONALD AND J. HAMBLIN

of the cultivar Takhung Native 1 (TN1) in 1956, and then in 1966 at the International Rice Research Institute with the release of IR8, a variety that transformed rice yields over a great part of Southeast Asia. These two 0.indica cultivars, TNl and IR8, each derived their semidwarf habit and erect leaf growth from the variety Dee-geo-woo-gen, a mutant from an old Chinese variety, Woogen (Athwal, 1971). Additional features of IR8 in relation to the common seed crop ideotype were its nonphotoperiodicity,permitting use over a much greater geographic area, and an increased harvest index. Six older, tall, competitive varieties had a mean harvest index of 0.36, whereas the dwarf, erect, short-leaved varieties had an index of 0.53 (Chandler, 1969). Poor tiller survival because of intense mutual shading and the cessation of growth after flowering were believed to contribute to the low harvest index of the tall, leafy varieties. The other features of the common ideotype and its culture that may offer opportunity in rice production are nontillering (Japanese cultivars already show duction in tiller number) add the use of high plant densities. These features are of course Wed. As long as most of the world’s rice is transplanted by hand at enormous human effort from seed bed to paddy field at low plant densities (about 20 plants/m2), heavy tillering is essential. But in situations where rice is broadcast or aerially sown there may be potential gains in yield from less freely t i l l e d or even uniculm rices of higher harvest index sown at heavier seeding rates. D. MAIZE

The wild progenitor of maize was probably relatively dwarf, with several tillers each having a terminal inflorescence carrying both male and female flowers and several small ears in leaf a i l s (Mangelsdorf, 1965). The terminal inflorescence broke easily, assisting seed dispersal. With the exception of the United States corn belt dent types, the principal Commercial types of maize were fully developed by the American Indians; little genetic advance was made until the development of commercial hybrid corn in the 1930s (Mangelsdorf, 1965; Galiiat, 1965). During domestication strong artificial selection by man for large ears occurred, but natural selection of fecund plants probably occurred in fertile, man-made environments (Wilkes, 1977). The trend to a single stem and large ear was a consequenceof man’s preference for large, easily hand-harvested cobs and easy cultivation between rows and of natural selection for tall competitiveplants with many offspring. Tall competitive plants were regarded favorably,’ but a direct consequence was low optimal plant stands [e.g., 26,0001ha was considered a high density in ‘An Iowan would boast, “I’m from Iowa where the tall corn grows!”

CONVERGENT EVOLUTION OF ANNUAL SEED CROPS

125

Iowa in 1924 (Stringfield, 1964)l. However, during the 1950s a growing appreciation of the interaction between genotype, density, and fertility occurred (Stringfield, 1964), particularly when it was found that dwarf plants suffered much less sterility (5%) than normal plants (62%) at high densities [105,000 plantslha (Sowell, 1960)]. The importance of leaf distribution was also realized. With leaves more vertically disposed above the cob at high densities, light penetrates deeper into the canopy and yields are higher (Pendleton et al., 1968; Winter and Ohlrogge, 1973; Vidovic, 1974; Pepper et ul., 1977). Horizontal leaves were better at low densities, whereas at intermediate densities or in widely spaced rows leaf angle was not important. If no response to leaf angle is found, this probably results from sampling too narrow a density range, from leaves that are not stiff enough along their entire length to maintain a constant leaf angle or from the range of leaf angles that are too small to allow differentiation (Mock and Pearce, 1975). Nonetheless, responses to high leaf angles occur only at leaf area indices rarely obtained in commercial crops. Mock and Pearce (1975) presented features for a maize ideotype that included (1) stiff and vertical leaves above the ear and horizontal leaves below; (2) maximum photosynthetic efficiency; (3) efficient conversion of photosynthates to grain; (4) short interval between pollen shed and silk emergence; (5) ear shoot prolificity; (6) small tassel size; (7) photoperiod insensitivity; (8) cold tolerance for germinating seeds and seedlings (in areas where soils are cold and wet at planting); (9) a grain-filling period as long as is practical; and (10) slow leaf senescence. Features (4) and (6) relate specifically to maize and (8) relates to summer crops. All other features (assuming that ear shoot prolificity and small tassels are components improving harvest index) are common to every annual seed crop. Temperature maize production now uses many of these ideas, and similar objectives are currently being applied to tropical maize. Tropical cultivars are often tall (up to 3.5 m) and leafy, an competitive evolutionary response. They lodge easily and have low harvest indices (less than 0.35). Selection at CIMMYT for reduced height and leafhess within the cultivar Tuxpeno (CIMMYT, 1979) has reduced height by 8 cm/cycle so that plants now are only 60% of their original height; at the same time yield increased 190 kg/ha or 3% per cycle (cf. comments on Gardner’s mass selection program, Section IV,A,5). This increase is associated with reduced lodging, higher harvest index [0.35-0.48 during 15 cycles; cf. comments of Hamblin and Rosielle (1983) on the height-harvest index relationship], and increased crowding tolerance (optimum planting density of 45,000 plantslha at cycle 12 and of 60,000at cycle 15). Flowering was 13 days earlier and there were three leaves less below the ear. Thus similar plant type-density relationships occur in both temperate and tropical regions. Questions for future investigation are, What will be the equilibrium situation between these features and yield in maize? Should we examine the potential of

126

C.

M. DONALD AND J. HAMBLlN

maize crops sown at 160,OOO plants/ha (25-cm square planted) producing leaf area indices similar to other cereal crops but with improved canopy-light relationships, a low incidence of barrenness, and a high harvest index? E. SORGHUM

Only circumstantial evidence is available regarding the early history of sorghum (Sorghum bicolor). Doggett (1965) proposed that domestication first occurred in the Abyssinia-Sudan region about 5000 years ago. However, Harlan (1971) considers that sorghum had a more diffuse sub-Saharan origin. From there it spread to other parts of Africa, to India before lo00 B.c., and to China about A.D. 1300. The wild relatives of sorghum, widely distributed over the African continent, characteristically have large, pyramidal, loose inflorescences with spreading branches. Although mainly annual, some are perennial with short rhizomes; the racemes articulate at maturity, assisting natural spread; and they have small grains (de Wet and Huckabay, 1968; de Wet and Schechter, 1977). Cultivated grain sorghums have heads of varying degrees of compactness, from loose to extremely dense with tough rachises and persistent spikelets, features ascribable to selection by man and to natural selection, respectively. Competition for light within sown crops gave tall types a powerful advantage, so that village crops in Africa may be as tall as 3.5 m (Goldsworthy, 1970). Grain sorghum in the United States prior to 1928 was commonly 140-180 cm tall (Quinby and Martin, 1954). These cultivars were annual or weakly perennial, although without rhizomes. They were capable of regrowth from the crown to produce a second crop, permitting ratooning. Because of the wide geographic distribution of cultivated sorghums and the free hybridization between genotypes, many distinctive races can now be found (de Wet and Huckabay, 1968; de Wet and Harlan, 1971). Breeding programs with sorghum have had several clear-cut objectives. Through the use of dwarfing genes, striking reductions in height have been achieved with associated increases in grain yields. By 1953, 98% of the grain sorghum cultivars in the United States were of dwarf stature and harvested by combine (Quinby and Martin, 1954). Reductions from 1.5 to 1.2 m in singledwarf material and to 0.75 m in double-dwarf lines were typical of the reductions in height (Queensland Department of Agriculture, 1970). In Africa the use of dwarfing genes occurred later, partly because of the value of the tall stems for building and fodder in village life. Reductions in height are, however, now occurring in that continent (Goldsworthy, 1970). Another objective has been the incorporation of one or more genes for insensitivity to photoperiod, giving earlier flowering and permitting the progressive extension of sorghum to cooler areas of shorter season. At least three additive genes are involved (Quinby and Karper, 1945; Quinby and Martin, 1954). Since

CONVERGENT EVOLUTION OF ANNUAL SEED CROPS

127

1954, the great advance has been the discovery of cytoplasmic sterility, permitting commercial use of hybrid sorghums with yields about one-third greater than those of pure lines. The heterotic manifestations are higher metabolic efficiency, increased height, earlier flowering and longer grain-filling period, greater vegetative yield, and increased grain size and yield (Quinby, 1963). With reduced height, increased vigor, and higher soil fertility, there has been a growing need to manage the crop so as to regulate panicle number per square meter, taking account of the seeding rate, estimated seedling establishment, and the probable tillering behavior (Ross and Eastin, 1972). The row spacing adopted is commonly as close as will permit cultivation (75 cm, or double rows 30 cm apart, at 100 cm); dry-land populations of 50,000-80,000 and of 250-300,OOO plants/ha under imgation have been adopted in the United States. The natural evolution, under cultivation, to very tall competitive plants has thus been followed by a controlled move toward communal plants, that is, toward dwarf stature and much-reduced tillering. Some sorghum cultivars are described as “single-stemmed,” although they tiller at low densities. The opportunities for further progress toward communal plants and higher grain yields seem to lie in further increases in plant density; the development of lines of strictly uniculm habit, shorter, narrower leaves, more erect leaf disposition, and markedly narrower rows without interrow cultivation (Clegg, 1972). F. COMMON OR AMERICAN BEAN

The earliest known cultivation of the common bean (Phaseolus vulgaris) was at least 7000 years ago (Kaplan and McNeish, 1960; Kaplan el a l . , 1973). Beans probably evolved over a wide area (Harlan, 1971; A. M. Evans, 1980); they were a valuable component of the American Indian diet, and their use extended over much of central and north America to about 42”N and over western South America. P . vulgaris has been used both for green beans and as dry beans; it is with the latter use, as a seed-bearing, annual field crop, that we are concerned here. The wild progenitor is P . vulgaris f. aborigineus, the climbing thicket bean, a perennial form with strong branching and a tuberous root. The cultivated species is very variable in growth habit, ranging from indeterminate climbing types to determinate bush types with 3-6 nodes on the primary stem (A. M. Evans, 1980). The climber, grown on a trellis or in association with maize, was the earlier cultivated form, with a branching, indetenninate habit of growth. The dwarf or bush type, used for seed production in presentday mechanized agriculture, is known (from vegetative remains) to have been grown as a crop in indigenous Mexican agriculture at least 800 years ago. It is of determinate growth habit, the main stem and each branch having a terminal flower after 3-6 nodes. The responses to domestication in American Phaseolus beans are summarized

128

C. M.

DONALD AND J. HAMBLIN Table I

Rcspoase of Phase&

v u f g d is to -u

chuacteristic of plant as _____

Type of selectionb

Wild

DomaSticated

2 2 1

W d Y pmnnial Short-day plant Indeterminate Many nodes Scrambling habit Small seeds Hard-seeded Physiological dormancy Testa colors and patterns few Small leaves Pods dehiscent Stems thin

Annual Day neutral Determinate Few nodes More erect to erect habit Large seeds soft-seeded No dormancy Testa colors and patterns, many Large leaves Pads indehiscent Stems thick

?

2 1 and 2 2 2 1 2 2 2

“From A. M. Evpos (1980),Smartt (1969).aad Purseglove (1968).

’1, pmbably conscious selection by man; 2, natural selection in agricultural environment.

and classified in Table I. The mechanisms of some of the changes may be debated, and some changes may have multiple causes, but the grouping of most of them is self-evident. Only man himself could have selected the dwarf determinate form; as Smartt (1969) remarks, “In nature or mixed cultivation the dwarf determinate mutant would have been effectively lethal . . . man has preserved and propagated a key mutant.” We thus see that the American Indians deliberately selected and developed for cropping a shorter, less competitive plant; a selection of the unfit. This step has been repeated in wheat and rice by modem workers loo0 years later. The dramatic increase in seed size, although undoubtedly leading to a reduced number of propagules, must have also been attained through deliberate selection. Some seed colors may have had natural selective advantage (fungistatic properties of pigments, less predation by birds), but the choice of particular colors by man has been the all-powerful factor in the local evolution of color patterns. Various consequences result from man’s conscious selection, particularly effects on growth form related to selection for reduced height. However, many important characteristics of modem field beans result from natural selection within the climatic or cultural environment of man’s crops. The most notable of these is the annual habit, an ability to complete the life cycle before killing frosts prevent the production of viable seeds. The perennial habit suffices in subtropical areas, but did not permit survival as the cultivation of the bean extended northward.The responses in time of maturity, photoperiodism, and ready germination

CONVERGENT EVOLUTION OF ANNUAL SEED CROPS

129

were all responses to the climatic or man-made environment, and large leaves gave competitive advantage over neighbors for light. The loss of pod dehiscence enabled survival of the harvested seed to be sown the following year (A. M. Evans, 1980). There is no doubt that the evolution of the bean under domestication in the Americas was in many ways more advanced, by several centuries, than the evolution of rice as a crop in Asia or of wheat as a crop in Europe. What developments offer further increases in seed yields in the common bean? It was suggested by Adams (1973) that major reduction in branching is desirable, so that each plant has a main stem and a few short lateral branches with many pods at each nude. Smaller leaves are also indicated as a means of securing deeper light penetration into the canopy. To be effective, these changes must be accompanied by increased density of the stand and strong pursuit of improved harvest index of communal plants growing in a strongly competitive crop situation. G. FIELDBEAN Viciufubu includes the field bean and the broad bean; the former, Viciufubu var. minor, is here considered. It is an erect annual with a main stem and, depending on plant density, one to several lateral stems. Each stem is indeterminate in growth, with 5-10 basal vegetative nodes and about 10 nodes with axillary inflorescences followed by a continuing production of vegetative nodes (Poulsen, 1977; Chapman and Peat, 1978). Most of the seed is produced by the main stem, with one or two pods per inflorescence and four to six seeds per pod. There is competition both among the developing pods and between the pods and the further vegetative growth (Chapman and Peat, 1978), a situation closely comparable to the tall sorghum genotypes discussed earlier. The weaknesses in this plant structure are evident, namely, unnecessary height and vegetative growth associated with the indeterminate production of sterile nodes above the pods. Various useful genes, principally simple recessives, are available, including those for dwarf stature and for a terminal inflorescence. Crossing has shown (Chapman and Peat, 1978) that there are excellent prospects for developing field beans of reduced, erect stature with upright pods borne terminally and leaf sizes reduced to no more than two leaflets; the problem of massive amounts of vegetative tissue passing through the harvesting machinery would thus be alleviated. The yields of these semidwarf determinate types so far are less than those of current tall cultivars because of the inadequate yield of seed per node; more seeds per pod are sought. Two points may be made: first, the reported tendency of determinate forms to produce more branches (Chapman and Peat, 1978), may cancel the gains achieved through reduction of the vegetative tissues of the main stem. Nonbranching determinate forms seem highly desirable. Second, any test-

130

C. M.DONALD AND J. HAMBLIN

ing of such types should be undertaken within high density communities with continued emphasis on harvest index as a guide to efficiency within the biomass of the crop. The efficiency or yield of isolated plants is irrelevant. A disadvantage of reduced branching and increased density may be seed requirements, as the large seed weight of this species will mean that seed costs will be appreciable.

H. SOYBEAN The wild ancestor of the cultivated soybean (Glycine man) is believed to be Glycine soja, indigenous to China, the Soviet Union, Korea, Japan, and Taiwan. G . l l u u ~and G . soju have few barriers to hybridization and on this and other grounds are regarded as conspecific (Hymowitz and Newell, 1980). Both species are annuals, but although the wild G . soju is a slender twiner, characteristic of hedges and roadsides, the cultivated soybean is a bushy shrub. The domesticated plant differs also in having reduced dehiscence of the pods and larger seeds of higher oil content. The soybean was probably domesticated in the eastern half of north China in the eleventh century B.c., spreading to Southeast Asia in the early centuries A.D. It was not known to European agriculture until the early eighth century nor to North American agriculture until the 1850s (Hymovitz and Newell, 1977, 1980). . Three growth habits are present in soya: determinate, semideterminate, and indeterminate; genetic control is by two genes. There are marked differences in the source-sink relationships between these types (Shibles, 1980). Narrow rows and higher plant populations often produce yield increases (Costa et ul., 1980). This may result from improved light relationships (Shaw and Weber, 1967) or improved water-use efficiency (Peters and Johnson, 1960; Timmons et al., 1967). Narrow leaf types give better light penetration, but this was not associated with increased yields (Hicks et al., 1969), although they had high water-use efficiencies (Hiebsch et ul., 1976). The potential for manipulating the soybean plant to develop communal plants appears excellent. However, as they will be poor competitors in mixtures, and as competitive ability is related to branching, height, and late maturity (Mumaw and Weber, 1957; Hinson and Hanson, 1962; Schutz and Brim, 1967), care must be taken to ensure their retention in segregating populations. Also, they must be yield tested in pure culture at high densities if their full yield potential is to be realized. I.

PEAS

Davies (1977a,b) has reviewed the dramatic developments in the pea (Pisum surivum) crop. There has been a marked reduction in stature from a height of 1-2

CONVERGENT EVOLUTION OF ANNUAL SEED CROPS

131

m for garden peas to 0.3-0.6 m for field peas. Yet even with this considerable dwarfing, two major physical problems remain: the great bulk of vegetative material to be handled during harvest and the frequency of severe lodging (amounting almost to certainty) with loss of the canopy structure and further deterioration of the light profile. These dwarf pea crops, despite their reduced height, have much in common with the old, tall rice varieties (large, horizontally disposed leaves and poor physical stability). Two mutant genes now offer the prospect of improved canopy structure. The first reduces the leaflets to tendrils and the second reduces the leafy stipules to small bracts. With only the f m t of these genes the plant is known as “semi-leafless”; with both, it is “leafless.” Here then is a dramatic reduction in leafhess. Leafless, and particularly semileafless, crops promise to outyield standard varieties (Davies, 1977a,b; Hedley and Ambrose, 1981). The advantages for seed crops may be several. First, leaflessness permits a much deeper penetration of light into the crop and thereby a more effective mean illumination of the photosynthetic surfaces; second, the interlocking tendrils give such effective mutual support that lodging cannot occur; third, the reduction of vegetative parts contributes to a higher harvest index; and forth, leafless peas may use water more efficiently than leafy types. Perhaps the radical structure of the canopy of these peas may offer, for the f m t time, prospects of yields from legumes more closely comparable to those of cereals. A feature of the pea crop that warrants fuller examination is the extent of vegetative branching, which has already been reduced in some dwarf genotypes. Increased sowing rates of leafless, nonbranching plants probably would improve the crop productivity by these most unusual plants. It is of interest at this point to note the features that existing leafless pea plants have in common with the semidwarf rice varieties: reduced stature, reduced leafhess, better light profile, improved physical stability, improved synchrony of flowering, and, almost certainly, better harvest indices.

In pre-Columbian times, all the cultivated cottons of Central and South America were perennial shrubs confined to tropical regions. These perennial American cottons founded the crops of southern Europe, Africa, and India, but since the mid-eighteenth century, three annual forms have evolved within these crops; upland cotton (Gossypium hirsutum), and sea island and Egyptian cottons (barbadense) (Phillips, 1976). There was a progressive change under cultivation from xeric, wild species to cultigens adapted to more fertile soils and more abundant water (Stebbins, 1974). A major selective force was the extension of cropping into temperate regions, where the frost-free season was progressively shorter. Differential seed production, in favor of plants adapted to the climatic,

132

C. M. DONALD AND J. HAMBLIN

soil fertility, and water regimes of new environments, ensured the natural selection of mesic, early-flowering annuals. Cotton culture, as exemplified by that in the United States, has been of branched, annual shrubs, typically in rows about 1 m apart with about 5-15 cm between plants. Where irrigation is practiced, these rows run centrally along flattopped “hills,” between irrigation furrows spaced at 1 m. In the mid-l960s, however, a major change in cotton culture was foreshadowed through the study of “narrow-row cotton.” In the f m t paper on narrow-row cotton, Ray and Hudspeth (1966) stated that the primary objective was to determine whether yields could be increased substantially through high popylations with large amounts of fertilizer and irrigation water. They found (Brashears et al., 1%8)that population increase was effective in raising yield only if it was achieved by the closer spacing of the rows rather than by increasing plant number within the row. When the population was increased to about 250,000/ha and the mean row width decreased to 50 cm (i.e., with 2 rows 40 cm apart on each 1-m hill), the following changes ensued: reduced plant stature, nonbranching, frost avoidance through earlier maturity (8-10 days), more simultaneous ripening of the bolls (more uniform cotton quality), and higher yield. Similar results have been reported elsewhere (for a review see Low and McMahon, 1973). The study of narrow-row cotton culture led cotton workers to ask themselves two questions: Can a more suitable genotype be developed for use in narrow rows at high population density? and, Can machinery be developed to harvest narrow rows, pferably in a “once-over” operation? The outcome has been a trend toward dwarf, determinate cotton varieties, with bolls borne on short fruiting stalks so that they lie close to the stalk. These varieties are also described as “stom proof.” They can, as was hoped, be harvested at a single stroke. These cotton cultivars and the system under which they are grown are parallel to the common ideotype to be discussed in the next section in many aspects. Bhardwaj et af. (1971), working in India, reported a negative correlation between yield of seed cotton and both plant height and leaf area. They emphasized the need for more dwarf cultivars with fewer branches and less leaf area. Constable (1977) states that in Australia there is also a need for varieties bred specifically for narrow-row culture with reduced leafhess. There are clear opportunities to improve the light profde of the crop through the use of Okra types of cotton, which have deeply cleft leaves of much reduced area, but the field evidence in favor of such foliage is as yet inconclusive (Andries et al.. 1969, 1970; Constable, 1977; Pegelow et al., 1977). Certainly, Okra leaf is unlikely to offer advantages in 1-m rows, but it may well prove of significant value at high density in narrower rows. There has been no clear statement regarding the influence of narrow-row culture on the ratio of lint or

CONVERGENT EVOLUTION OF ANNUAL SEED CROPS

133

fruit weight to biological yields (harvest index), but there are good prospects for progress. The trend in cotton toward short, less branched, less leafy plants has occurred by a different path than that in other crop species. In most annual seed crops, the trend toward such plants has arisen through breeding for increased yield or through theoretical considerations of improved plant form, with the consideration of the value of increased plant density or reduced row spacing for such communal plants then following. In cotton, the development of shorter, less branched, less leafy plants did not arise in that way. The first move was an agronomic step toward narrower rows and denser populations, as a means of improving exploitation of the environment; it resulted in phenotypic changes to smaller, earlier maturing, less leafy plants of the standard cultivars. The second phase was the consideration of the need for suitable genotypes for this new cultural environment, cultivars that were genetically smaller, less branched, and less leafy, rather than to allow phenotypic suppression at high densities. The outcome will be the same as in other crops; only the sequence has differed. A specific limitation in cotton to the plant type or the adoption of narrower rows and high densities lies in agronomic practice. Although tests have shown that cropping with 12-cm rows and plants spaced at 8 cm in the row (1 million plants/hectare) have given very high yields (Kirk et ul., 1969), such spacings probably will not prove practicable under irrigation. This suggests that future cultivars for natural rainfall conditions and for irrigation may be appreciably different in plant form, especially in height and degree of branching, depending on the row spacing and the density at which they are sown.

K. SUNFLOWER Wild sunflower (Heliunrhus unnuus) is distributed from southern Canada to northern Mexico, with its greatest abundance in the southwestern United States. It is a single-stemmed, tall annual, commonly much branched with many small heads (2-5 cm in diameter) and with ovate or cordate leaves (10-20 cm long) (Purseglove, 1968; Heiser, 1978). No perennial ancestor has been recognized though there are many perennial species of Heliunthus native to the region. Archeological evidence indicates that the sunflower may have been cultivated in Arizona and Mexico as early as 3000 B.c., possibly earlier than maize in that region. It was grown for its edible seed and used principally as flour, but possibly also for oil (Putt, 1978). The types cultivated by the Indians were freely branched with many small heads, but other types had a main stem and one principal head. There is some evidence that prior to European settlement, some Indian tribes may already have had monocephalic, unbranched, very tall sunflowers with massive heads (Heiser, 1978; Fick, 1978).

134

C.M.DONALD AND J . HAMBLIN

The development of modem sunflowers began with its introduction to Russia; by 1900, nearly half a million hectares were planted there. The Russian cultivars of the nineteenth century became progressively taller (up to 4 m) (Purseglove, 1968), almost certainly as a consequence of natural selection, and many village cultivars developed. Subsequent breeding was aimed at reduced height, earliness, and high oil content. The crop returned to North America with direct commercial seed importation from Russia of very tall (2-4 m), late cultivars, notably Mammoth Russian, in the late nineteenth century. These tall cultivars were at first used mostly for silage, but in 1936 and 1950 Canada and the United States, respectively, began breeding for high seed yield and oil production, using lines carried in by Memmonite immigrants and some newer lines from the Soviet Union. Some of these were earlier, dwarf (1-1.5 m) cultivars with small seed (Putt,1978). Considerable increases in yield followed the development of hybrid sunflower in the late 1940s and especially the discovery and use of cytoplasmic male sterility and genetic fertility restoration in the 1970s. Modem cultivars are dwarf (0.8-1.2 m), unbranched plants with a large head, seemingly desirable attributes in an annual seed plant. What major further changes in plant form might lead to worthwhile increases in grain yield by sunflower? The most evident weakness is the extremely large, undivided leaves, up to 40 cm long and 35 cm wide, with an almost horizontal disposition (within 8-10"; Hiroi and Monsi, 1966), the leaf pattern of a shade plant rather than of a crop grown in open fields. A very marked reduction in leaf size (perhaps even to 20% of the leaf size in modem cultivars), together with the development of a semierect leaf disposition, would give greatly improved light penetration into the crop, increased crop growth, and, it may reasonably be expected, greater yield. Many consequences would follow. The heads would be smaller and the stems would be lighter and shorter. Yield per plant would be low and the plant population needed to exploit the environment and give full yields would be considerably greater than those presently in use; narrow rows would be indicated. Interrow cultivation might be neither practicable nor necessary. The crop would, sadly, be reduced in beauty but increased in harvest index.

VI. A BASIC IDEOTYPE FOR ALL ANNUAL SEED CROPS Within the generd field of agronomy several developments are occuning which, to maximize grain yield, require integration through plant breeding. The development and use of suitable herbicides in many crops, especially those traditionally sown in wide rows to allow cultivation for weed control, has permit-

CONVERGENT EVOLUTION OF ANNUAL SEED CROPS

135

ted higher densities, better plant arrangements, and higher yields (e.g., cotton, maize, sorghum, soybeans; see earlier discussion). With increasing crop density, optimizing canopy structure to maximize light fixation will be a potential avenue to increased crop yields. The theory behind canopy optimization is based on physical principles (Monsi and Saeki, 1953; Davidson and Philip, 1956; Wilson, 1960; Donald, 1961, 1962; Monteith, 1965a,b; de Wit, 1965; and many others). In a simplified model, three aspects of crop canopy structure affect light penetration; the leaf angle to the incident light and the vertical, and horizontal, distributions of the leaves. The deeper that light penetrates into the canopy, the greater is the photosynthetic capacity of the crop (Wilson, 1960; Blackman, 1961); penetration is increased if leaves are evenly distributed both vertically and horizontally and have a high leaf angle (assuming a high angle for the sun) (Wilson, 1960). Differences in cereal canopy structure have been related to yield differences [see Tanaka et al., 1964, 1966; Jennings, 1964; Jennings and Beachell, 1965; Beachell and Jennings, 1965; and Matsushima et al., 1964 (for rice); Hamblin, 1971; Hamblin and Donald, 1974; and Tanner et al., 1966 (for wheat, barley, and oats); Pendleton el al., 1968; Pepper et al., 1977; and Williams et al., 1968 (for maize)]. However, in many cases the lines used were not isogenic and results may be related to other factors such as reduced disease incidence, improved water use (Trenbath and Angus, 1975), or even slightly improved carbohydrate supply at critical times of development (Fischer, 1981). Blackman (1961) suggested that narrow, dissected leaves would be better than round or cordate leaves in dicotyledonous crops. The use of the Okra leaf type in cotton would appear to confirm the potential of this approach (Andries et al., 1969; Constable, 1977). The most extreme case of altered canopy structure is that of peas, in which the leaves have been dispensed with entirely, markedly altering the pattern of light distribution down the profile; yields have been increased (Hedley and Ambrose, 1981). Despite the criticisms of Trenbath and Angus (1975) concerning the comparisons of nonisogenic lines, it is probable that improved canopy morphology will lead to improved yields in many grain crops. This should occur more rapidly in short C , crops that are grown at high density than in tall C, crops grown at lower densities (Evans and Wardlaw, 1976) and in environments where the solar incidence in the growing season is high (Trenbath and Angus, 1975). However, the rate of change will also depend on other factors of crop production affecting yield-density relationships. When crops are grown at high levels of fertility, yields (at least of dry matter if not of grain) are frequently increased but crops tend to lodge. Lodging resistance, primarily through reduced height, is a breeding objective of many programs directed at high yield potential. Lodging resistance may also be increased if a plant has thick strong stems. These may allow a greater accumulation

136

C. M. DONALD AND J. HAMBLIN

of stored photosynthate during the vegetative phase, which if it can be retranslocated to the grain at a later date will increase the harvest index. If biological yield is constant d u d height also will have a direct benefit in terms of an improved harvest index (Hamblin and Rosielle, 1983). Within a given cross, short plants tend to have short leaves (Chowdhry and Allan, 1966; Hamblin, 1971). On the average, short leaves will be more upright thanlong leaves because they have less bending momenL Therefore, short plants may also have an advantage in terms of canopy structure at high densities. In many agricultural situations the length of the growing season is clearly defined. This may be because of drought, frost, the rotational needs of the following crop, or other factors. Within that defined season, there will be an optimum relationship between the vegetative and reproductive phases of crop growth, in terms of both phenology and dry matter production. This problem has recently been discussed by Fischer (1979, 1981) for dry-land Mediterranean situations. He concludes that, within a given environment, there is an optimum level of dry matter production at anthesis for maximum grain yield. If biological yield at anthesis is above that optimum, then there is insufficient water to maximize grain yield; if biological yield is below the optimum, then there is insufficient sink for maximum grain yield (Fischer, 1979, 1981). In many situations, early flowering allows a longer period of grain filling and higher yields (Thorne, 1966). Early flowering may also put grain development into a more favorable season (Fischer, 1981); however, it may reduce the biological yield at flowering to suboptimal levels. This can be countered by selecting types with vigorous early growth, by growing crops at high levels of nutrition, and by using higher seeding rates (Fischer, 1981). Uniculm cereals would have an advantage here; seeding at high rates would give rapid development of leaf area without the presence of tillers. In many situations the development of photoperiod insensitivity allows wide adaptation for a variety. A major problem for agronomists aiming at optimum levels of dry matter production at flowering is to manipulate the cropping strategy to maximize the probability of achieving that optimum. This is particularly difficult in branching and indeterminate crops in which there is little control over the amount of vegetative dry matter likely to be developed. The tendency to determin:ncy, which will allow some control of the relationship between vegetative and reproductive growth, is apparent in several species (beans, cotton, soya). There is a similar tendency toward reduced branching (tillering) in many species (cotton, cereals). The logical development is to nonbranching (uniculm cereals) so that it is possible to manipulate the relationship between vegetative and reproductive growth by adjusting density. This option is already available in the uniculm species, that is, in maize and sunflowers. Workers on these crops, in contrast to people working on other grain crops do not consider the uniculm habit unusual. In species such as sorghum and rice, selection for the strictly annual habit will probably increase yield potential slightly as no resources would be diverted to

CONVERGENT EVOLUTION OF ANNUAL SEED CROPS

137

perennating organs. Many of the characteristics considered in this section will automatically lead to a high harvest index; these include reduced height, earlier flowering, and nontillering. To date, however, the critical experiments on the use of harvest index per se as a selection criterion either have not been carried out or have produced inconclusive results (Hamblin and Rosielle, 1983). As a long-term objective, however, selection for harvest index alone must eventually lead to diminishing returns if biological yield remains constant. There is widespread acceptance of the value of reduced stature, reduced tiller number, and early flowering in many circumstances, but there is little interest in the uniculm (or nonbranching) habit; shorter, narrower, and more erect leaves, higher harvest index, increased plant populations, and narrower rows than the present 18- to 20-cm spacing. Yet these features, it is proposed, provide notable opportunities to increase yields in all environments. It was proposed that there is much to gain in the breeding of crop plants by designing ideotypes, “biological model which is expected to perform or behave in a predictable manner within a defined environment and . . . to yield a greater quantity or quality of grain, oil or other useful product when developed as a cultivar” (Donald, 1968a, p. 389). It is here proposed that the ideotypes of all annual crops grown for their seed will have major features in common, even to the extent that a basic ideotype can be conceived for cereal crops, cotton, peas, beans, soybeans, linseed, sunflower, or any other annual seed crop. One may observe that breeding toward many of these features is already in progress in many annual seed species, and further that there are trends toward like, though often independently conceived, agronomic practices. There may be much to gain by recognizing and systematizing those trends in plant breeding and agronomy. Most of the features of this common ideotype arise directly from the proposed need for communal plants sown at high density. Various other useful features and practices for annual seed crops which can be postulated are set out in Table 11. It will be seen that despite some conflict in the plant features needed to meet particular criteria of crop performance, a clear picture emerges. The principal characteristics of the ideotype proposed for all annual seed crops and their culture are 1. Strictly annual habit 2. Erect growth form 3. Dwarf stature 4. Strong stems 5. Unbranches or nontillered habit 6. Reduced foliage (smaller, shorter, narrower, or fewer leaves) 7. Erect leaf disposition 8. Determinate habit 9. High harvest index

C. M. DONALD AND J. HAMBLIN

138

10. Nonphotoperiodic for most but not all situations 11. Early flowering for most but not all situations 12. High population density 13. Narrow rows or square planted Table II The Features of a Common Ideotype for All Seed-Producing Annual Crops, Together with Associated Cultural Practices Feature of crop Pure culture sown at high density

Features of ideotype

Good plant performance among like neighbors sown at high density, hence communal plants needed; plant yield in isolation or in competition with other genotypes of no relevance Strictly annual habit Determinate growth; plant death at seed ripeness; loss of residual features of perenniality (i.e., of vegetative branching, tillering, or vegetative storage organs) Crop must not lodge or collapse Plants of sound physical structure; short stature, strong or flexible stems, nonbranching, nontillering, nonleafy Effective form and disposition of foli- Deep light penetration within the leafy canopy; small, age for light utilization narrow or divided, erect leaves High seed yield sought High biological yield, attainable through high sowing rate, rapid emergence, rapid attainment of optimum LAI, high net assimilation rate High harvest index, involving annual habit, no excessive use of resources on plant framework, short stature, light stems, nonbranching, nonleafy Large sink for photosynthates, many seeds per unit of biological yield, long interval flowering to maturity, no sterility at high plant density Absence of those features associated with strong comMinimal competition between plants petitive ability (i.e., absence of tallness, large or horizontally disposed Leaves, branching, or widely ramifying root system) Plant density and plant arrangement to High plant density to compensate for lack of branching and lack of leafiness; close approach to uniform spacbe appropriate to the communal plant form ing through use of narrow rows Effective response to high nutrient Limited increase in competition between plants as fertillevels ity is raised; absence of plant responses giving increased competitive ability, especially minimal increase in height, leafhess, or branching As appropriate to the climatic region but commonly Wide climatic adaptation including nonphotoperiodicity; earliness of flowering to avoid early or late frosts, cold soil or cold irrigation water early in the season, drought, or wet or wintry conditions at harvest; wide temperature tolerance

CONVERGENT EVOLUTION OF ANNUAL SEED CROPS

139

ACKNOWLEDGMENTS We would like to thank Drs. W. J. Collins, R. A. Fischer, R. Knight, A. J. Rathjen, A. A. Rosielle, and W. R. Stem, and Mr. N. J. Halse, for comments and criticisms of the manuscript. Dr. Hamblin was supported by a Wheat Industry Research Council Grant and this is gratefully acknowledged.

REFERENCES Adair, C. R., and Jones, J. W. 1946. J. Am. SOC. Agron.38, 708-716. Adams, M. W. 1973. Semin. Ser. No. 2E, pp. 266-278. CIAT, Cali, Columbia. Akihama, T. 1968. Ikushugaku Zusshi (Jpn. J . Breed) 18, 213-216. Allard, R. W., and Adams, J. 1969. Am. Nut. 103, 621-645. Allard, R. W., and Jain, S . K. 1962. Evolurion 16, 90-191. Andries, J. A,, Jones, J. E., Sloane, L. W., and Marshall, J. G. 1969. Crop Sci. 9, 705-710. Andries, J. A., Jones, J. E., Sloane, L. W., and Marshall, J. G. 1970. Crop Sci. 10, 403-407. Athwal, D. S. 1971. Q. Rev. Biol. 46, 1-34. Atsmon, D., and Jacobs, E. 1977. Crop Sci. 17, 31-35. Austin, R. B., Bingham, J., Blackwell, R. D., Evans, L. T., Ford, M. A., Morgan, C. L., and Taylor, M. 1980. J . Agric. Sci. 94, 675-689. Beachell, H. M., and Jennings, P. E. 1965. In “The Mineral Nutrition of the Rice Plant,” pp. 29-75. John Hopkins Press, Baltimore, Maryland. Bell, G. D. H. 1963. Barley Genet. 1, 285-306. Bell, G. D. H. 1965. In “Essays on Crop Plant Evolution” (J. B. Hutchinson, ed.), pp. 70-102. Cambridge, Univ. Press, London. Bhardwaj, S. W., Nath, V., and Mehra, R. B. (1971). Indian J. Agric. Sci. 41, 519-523. Bingham, J. 1972. Eucarpia Congr. Eur. Assoc. Res. Plant Breed. 6, 15-29. Blackman, G. E. 1961. Amos Mem. Lecture E. Malling Res. Srn. 5th. Blijenburg, J. G., and Sneep, J. 1975. Euphyrica 24, 305-315. Brashears, A. D., Kirk, I. W., and Hudspeth, E. B. 1968. Texas Agric. Exp. Srn. Misc. Publ. No. 872, 1-7. Bremner, P. M. J. 1969. J. Agric. Sci. 72, 273-280. Cannell, R. Q. 1968. J. Univ. Newcasrle-upon-Tyne Agric. SOC. 22, 3. Chandler, R. F., Jr. 1969. I n “Physiological Aspects of Crop Yield” (I. D. Eastin, F. A. Hoskins, C. Y. Sullivan, and C. M. H. Van Bavel, eds.), pp. 265-285. Amer. Soc. Agron., Madison, Wisconsin. Chang, T. 1976. Euphyrica 25, 425-441. Chapman, G. P., and Peat, W. E. 1978. Outlook Agric. 9, 267-272. Chowdhry, A. R., and Allan, R. E. 1966. Crop Sci. 6, 28-31. Christian, C. S., and Grey, S. G. 1941. CSIR J. 14, 59-68. CIMMYT 1979. CIMMYTRev. Maize Sect. Spec. Res. Proj. Mexico, pp. 35-37. Clegg, M. D. 1972. In “Sorghum in the Seventies” (G. P. Rao and L. R. House, eds.), pp. 279-301. IBH Publ. Co., New DeIhi. Clements, F. E., Weaver, J. E., and Hanson, H. C. 1929. Carnegie Inst. Washington Publ. No. 398. Constable, G. A. 1977. Aust. J. Exp. Agric. Anim. H u b . 17, 135-142. Costa, J. A., Oplinger, E. S., and Pendleton, J. W. 1980. Agron. J. 72, 153-156. Darlington, C. D. 1969. In “The Domestication of Plants and Animals’’ (P. J. Ucko and G. W. Dimbleby, eds.), pp. 67-72. Duckworth, London.

C. M. DONALD AND J. HAMBLIN

140

Darwin, C. 1868. “The Variation of Animals and Plants under Domestication.” Murray, London. Davidson, J. L., and Philip, J. R. 1956. I n “Climatology and Mimlimatology.” UNESCO, Paris. Davies, D. R. 1977a. Appl. Biol. 2, 87-127. Davies, D. R. 1977b. Sci. Prog. (Oxjord)64, 201-214. de Wet, J. M. J., and Harlan, J. R. 1971. Ewlwion 25, 128-135. de Wet, J. M. J., and Huckabay, J. P. 1%7. Evolution 21, 787-802. de Wet, J. M. J., and Schechter, Y. 1977. I n “Crop Resources” (D. J. Seigler), pp. 179-191. Academic Press, New Yo&. de Wit, C. T. I. 1%5. Centre for Agricultural Publications and Documentation. Agric. Res. Rep. 663.

Doggett, H. 1%5. In “Essays on Crop Plant Evaluation” (J. B. Hutchinson, ed.), pp. 50-69. Cambridge Univ. Press, London. . EX^. BWl. 15,283-313. Donald, C. M. 1961. S y m ~ SOC. Donald, C. M. 1962. J. Aust. Inst. Agric. Sci. 28, 171-178. Donald, C. M. 1%3. Adv. Agron. 15, 1-118. Donald, C. M. 1%8a. Euphyrica 17, 385-403. Donald, C. M. 1968b. Proc. Int. Wheat Genet. Symp. 3rd, pp. 359-369. Donald, C. M. 1979. J. Agric. Sci. 93, 261-269. Donald, C. M. 1981. I n “Wheat Science-Today and Tomomw” (L.T. Evans and W. J. Peacock, eds.), pp. 361-405. Cambridge Univ. Press, London. Donald, C. M., and Hamblin, J. 1976. Adv. Agron. 28, 361-405. Engledow, F. L. 1925. J. Agric. Sci. 15, 125-146. Evans, A. M. 1973. Semin. Ser. No. 2E, pp. 279-286. CIAT, Cali Columbia. Evans, A. M. 1980.In“‘Advances in Legume Science” (R. J. Summefield and A. H. Bunting), pp. 337-347. Stationery Office, London. Evans, L. E., and Bhatt,G. M. 1977: Can.J . Planr Sci. 57,45-47. Evans, L. T. 1980. J. Aust. Inst. Agric. Sci. 46, 3-13. Evans, L. T., and Dunstone, R. L. 1970. AW. J . Biol. Sci. 23,725-741. Evans, L.T., and Wardlaw, I. F. 1976. Adv. Agron. 28, 301-359. Fick, G. N. 1978. I n “Sunflower Science andTechnology” (J. F. Carter, ed.), pp. 279-315. Amer. Soc. Agron., Madison. Wisconsin. Fick, G. N., and Swders, C. M. 1975. Crop Sci. 67, 743-745. Fischer, R. A. 1978. Proc. Int. Wheat Genet. Symp. 5th, pp. 767-777. Fischer, R. A. 1979. J. Aust. Inst. Agric. Sci. 45, 83-94. Fiiher, R. A. 1981. P h Soil 58, 249-278. FRY. K. J. 1971. ASA Spec. Publ. No. 20. 15-58. Galinat, W. C. 1%5. Econ. Bot. 19, 350-357. Gardner, C. 0. 1961. CropSci. 1, 241-215. Gardner,C. 0. 1%8. Proc. Int. Congr. Genet. 12th, pp. 57-77. Gardner, C. 0. 1%9. I n “Induced Mutations in Plants,” pp. 469-477. Atomic Energy Agency, Vienna. Goldsworthy, P. R. 1970. J. Agric. Sci. 75, 109-122. W u s , J. E., Thomas, R. L.,and B a m d , J. 1976. Crop Sci. 16,673-677. Hamblii, J. 1971. Ph.D. Thesis, Univ. ofA&kzide, Australia. Hamblin, J. 1975. Euphytica 24,435-445. Hamblin, J., and Donald, C. M. 1974. Euphytica 23, 535-542. Hamblii, J., and Morton, J. R. 1977. Euphyrica 26, 75-83. Hamblii, J., and Rosielle, A. A. 1983. The contribution of genetics to plant breeding. Raeger (in press). Hamblin, J., and Rowell, J. G. 1975. Euphyrica 24, 221-228.

CONVERGENT EVOLUTION OF ANNUAL SEED CROPS

141

Hamblin, J., Knight, R., and Atkinson, M. J. 1978. Euphyrica 27, 497-505. Harlan, J. R. 1971. Science (Washington, D.C.) 174,46-474. Harlan, J. R., deWet, J. M. J., and Price, E. G. 1973. Evolution 27, 311-325. Harlan, M.V., and Martini, M.L. 1938. J . Agric. Res. (Washington, D.C.) 57, 189-199. Hayes, J. D. 1970. Rep. Welsh P h r Breed. Srn. (Aberysrwyrh, Wales). Hedley, C. L., and Ambrose, M. J. 1981. Adv. Agron. 34, 225-277. Heiser, C. B. 1978. I n “Sunflower Science and Technology” (J. F. Carter, ed.),pp. 31-53. h e r . Soc. Agron., Madison, Wisconsin. Hicks, D. R., Pendleton, J. W., Bernard, R. L., and Johnson, T. J. 1969. Agron. J . 61, 290-293. Hiebsch, C. K., Kanemasu, E. T.,and Nickell, C. D. 1976. Can. J. Plant Sci. 56, 455-458. Hinson, K.,and Hanson, D. 1962. Crop Sci. 2, 117-123. Hiroi, T., and M o d , M. 1966. J. Fac. Sci. Univ. Tokyo 9 Sect. 3, 241-285. Hwd, E. A. 1969. Euphytica 18, 217-226. Hutchinson, J. B. 1965. I n “Essays on Crop Plant Evolution” (J. B. Hutchinson, ed.), pp. 166-181. Cambridge Univ. Press, London. Hymowitz, T., and Newell, C. A. 1977. I n “Crop Resources”(D. S.Siegler, ed.), p. 209. Academic Press, New York. Hymowitz, T., and Newell, C. A. 1980. I n “Advances in Legume Science” (R. J. Summerfieldand A. H. Bunting, eds.), pp. 251-264. HM Stationery Office, London. Islam, T.M. T., and Sedgley, R. H. 1981. Euphyrica 30, 277-282. Jennings, P. R. 1964. Crop Sci. 4, 13-15. Jennings, P. R., and Aquino, R. C. 1968. Evolution 22, 529-542. Jennings, P. R., and Beachell, H. M. 1965. I n “Mineral Nutrition of the Rice Plant,” pp. 449-457. John Hopkins Press, Baltimore, Maryland. Jennings, P. R., and de Jesus,+J., Jr. 1968. Evolution 22, 119-124. Jennings, P. R., and Herrera, R . M. 1968. Evolurion 22, 332-336. Jensen, N. F., and Federer, W. T. 1964. Crop Sci. 4, 641-645. Johnston, P. R. 1972. W.D. Thesis, Univ. of Adelaide, Australia. Jones, H. G., and Kirby, E. J. M. 1977. J . Agric. Sci. 88, 391-397. Kaplan, L., and McNeish, R. S. 1960. Bor. Mus. Led. Harv. Univ. 19, 33. Kaplan, L.,Lynch, T. F., and Smith, C. E. 1973. Science (Washington,D.C.) 179, 76. Kern, J. J., and Atkins, R. E. 1970. Agron. J . 62, 183-186. Khalifa, M. A., and Qualset, C. 0. 1974. Crop Sci. 14, 795-799. Khalifa, M. A., and Qualset, C. 0. 1975. Crop Sci. 15, 640-644. Khan, M. A., and Tsunoda, S. 1970. Ikushuguku Zusshi (Jpn. J . Breed.) 20, 133-140. Kirk, I. W., Brasheam, A. D., and Hudspeth, E. B. 1969. Texas Agric. Exp. Srn. Misc. Publ. No. 937.

Klages, K. H. W. 1936. J . Am. Soc. Agron. 28, 935-940. Kranz, A. R. 1966. Biol. Zentralbl. 85, 681-734. Kulshrestha, V. P., and Jain, H. K. 1982. Z. Pflanzenzuecht. 89, 19-30. Laude, H. H., and Swanson, A. F. 1942. J. Am. SOC.Agron. 3, 270-274. Lonnquist, J. H., Cota, A., and Gardner, C. 0. 1966. Crop Sci. 6, 330-332. Low, A., and McMahon, J. P. 1973. Corron Grow. Rev. 50, 130-149. MacKey; J. 1966. Yugoslav Symp. Res. Whear Srh, pp. 37-48. Mangelsdorf, P. C. 1965. I n “Essays on Crop Plant Evolution” (J. B. Hutchinson, ed.), pp. 23-49, Cambridge Univ. Press, London. Matsushima, S., Tanaka, T.,and Hoshino, H. 1964. Nippon Sakumorsu Gakkui Kiji (Proc. Crop Sci. SOC. Jpn.) 33, 44-48. McGinnis, R. G . , and Shebeski, L. H. 1968. Proc. Inr. Whear Genet. Symp. 3rd, pp. 109-114. Mock, J. J., and Pearce, R. B. 1975. Euphyrica 24, 613-623.

142

C. M. DONALD AND J. HAMBLIN

Monsi, M., and Saeki, T. 1953. Jpn. J. Bor. 14, 22-52. Monteith, J. L. 1965a. Field Crop Abstr. 18, 213-219. Monteith, J. L. 1%5b. Ann. Bot. (London) 29, 17-37. Montgomery, E. G. 1912. Bull. Nebraska Agric. Exp. Stn. 127, 3-22. Mom-Val, C. A., and Miller, P. A. 1975. Crop Sci. 15, 479-482. Mumaw, C. R., and Weber, C. R. 1957. Agron. J. 49, 154-160. Nicholson, A. J. 1962. I n “The Evolution of Living Organisms” (G. W. Leeper, ed.), pp. 62-73. Melbourne Univ. Press, Australia. OBrien, L. 1979. Ausr. J . Agric. Res. 30, 587-595. Pal, B. P., Upadhyaya, Y. M., Khan, H. R., and Ramanujam, S. 1960. Indian J . Gener. Plant Breed. 20, 102-112. Passioura, J. B. 1972. Aust. J. Agric. Res. 23, 745-752. Pegelow, E. J., Buxton, D. R., Briggs, R. E., Murramoto, H., and Gensler, W. G. 1977. Crop Sci. 17, 1-4. Pendleton, J. W., and Seif, R. D. 1962. Crop Sci. 2, 154-156. Pendleton, J. W., Smith, G. E., Winter, S. R., and Johnston, R. J. 1968. Agron. J. 60,422-424. Pepper, G. E., Pearce, R. B., and Mock, J. J. 1977. Crop Sci. 17, 883-886. Percival, J. 1921. “The Wheat Plant.” Duckworth, London. Perry, M. W., and Reeves, J. T. 1980. “Farmnote.” Western Australian Dept. Agric., Avondale Project 1979. Peters, D. B., and Johnson, L. C. 1960. Agron. J. 52, 687-699. Fhillips, L. L. 1976. In “The Evolution of Crop Plants” (N. W. Simmonds, ed.), pp. 198-200. Longmans, London. Philipschenko, J. 1972. “Variabilitat und Variation.” Berlin. Poulson, M. H. 1977. J . Agric. Sci. 89, 643-654. Puckridge, D. W., and Donald, C. M. 1%7. Aust. J. Agric. Res. 18, 193-211. Purseglove, J. W. 1%8. “Tropical Crops,” Vol. 1. Longmans, London. Putt, E. D. 1978. In “Sunflower Science and Technology” (J. F. Carter, ed.),pp. 3-29. Amer. Soc. Agron., Madison, Wisconsin. Queensland Department of Agriculture 1970. Advisory Leaflet No. 1082. Quinby, J. R. 1963. CropSci. 3, 283-291. Quinby, J. R., and Karper, R. E. 1945. J. Am. SOC. Agron. 37, 916-936. Quinby, J. R., and Martin, J. H. 1954. A h . Agron. 6, 305-359. Quisenberry, J. E., Roark, B., Bilbro, I. D., and Ray, L. L. 1978. Crop Sci. 18,799-801. Ray, L. L., and Hudspeth, E. B. 1966. S w t h Plains Res. Exr. Cenr. Curr. Res. Rep. No. 66-5, 1-13. Reid, D. A,, and Wiebe, G. A. 1968. In “Barley,” pp. 61-84. USDA, Agric. Handbook No. 338. Rosielle, A. A., and Frey, K. J. 1975. Crop Sci. 15, 544-547. Ross, W. W., and Eastin, J. D. 1972. Field Crop Absrr. 25, 169-174. Sakai, K. 1955. Cold Spring Harbor Symp. Quanr. Biol. 20, 137-157. Schutz, W. M., and Brim, C. A. 1967. Crop Sci. 7, 371-376. Schwanitz, F. 1966. “The Origin of Cultivated Plants.” Harvard Univ. Press, Cambridge, Massachusetts. Shaw, R. H., and Weber, C. R . , 1967. Agron. J . 59, 155-159. Shibles, R. 1980. In “Advances in Legume Science” (R. J. Summerfield and A. H. Bunting), pp. 279-286. HM Stationery Office. London. Smartt, J. 1%9. In “The Domestication and Exploitation of Plants and Animals” (P. J. Ucko and G. W. Dimbleby, eds.), pp. 451-462. Duckworth, London. Smith, 0. D., Kleese, R. A., and Stuthman, D. D. 1970. Crop Sci. 10, 381-384. Soetono, and Donald, C. M. 1980. Aust. J. Agric. Res. 31, 455-470.

CONVERGENT EVOLUTION OF ANNUAL SEED CROPS

143

Sowell, W. F. 1960. Diss. Abstr. 20, 2476. Stebbins, G. L. 1974. “Flowering Plants: Evolution above the Species Level.” Harvard Univ. Press, Cambridge, Massachusetts. Stringfield, G. H. 1964. Adv. Agron. 16, 101. Suneson, C. A. 1949. Agron. J. 41,459-461. Suneson, C. A., and Wiebe, G. A. 1942. J . Am. Soc. Agron. 34, 1052-1056. Tanaka, A., Kawana, S . A., Garcia, C. V., Parao, F. T., and Ramirz, E. 1964. IRRI Tech. Bull. No. 7.

Tanaka, A., Kawana, S. A., and Yamaguchi, J. 1966. IRRI Res. Bull. No. 3. Tanner, J. W.,Gardener, C. J., Stoskopf, N. C., and Reinbergs, E. 1966. Cun. J . Plant Sci. 46, 690.

Thome, G . N. 1966. In “The Growth of Cereals and Grasses” (F. L. Mlthorpe and J. D. Ivins, eds.), pp. 88-120. Butterworth, London. Timmons, D. R., Holt, R. F., and Thompson, R. L. 1%7. Agron. J . 59, 262-265. Trenbath, B. R., and Angus, J. F. 1975. FieM Crop Absrr. 28, 231-244. Turesson, G. 1922. Hereditus, 3, 100-113. Vidovic, J. 1974. B i d . Plant. (Prague) 16, 174-183. Walker, J. T. 1969. Biol. Rev. 44, 207-242. Wiebe, G. A., Petr, F. C., and Stevens, W. 1963. In “Statistical Genetics and Plant Breeding,” pp. 546-557. NAS, NRC Publ. 982. Wilcox, J. R., and Schapaugh, W. T. 1980. Crop Sci. 20, 809-811. Wilkes, G. 1977. In “Crop Resources“ ( 0 . S . Seigler, ed.), p p . 211-223. Academic Press, New York. Williams, W. A., Loomis, R. S.,Duncan, W. G . , Dovrat, A,, and F. Nunez, A. 1968. Crop Sci. 8, 303-308.

Wilson, J. W. 1960. Proc. IN. Grussl. Congr. 8th, pp. 275-279. Winter, S . R., and Ohlrogge, A. J. 1973. Agron. J . 65, 395-397.

This Page Intentionally Left Blank

ADVANCES IN AGRONOMY, VOL. 36

CURRENT STATUS AND FUTURE PROSPECTS FOR BREEDING HYBRID RICE AND WHEAT S. S. Virmanil and Ian B. Edwards2 1 International Rice

Research Institute, Manila, Philippines ‘Pioneer Hi-Bred International, Inc., Glyndon, Minnesota

I.

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

Introduction

11. Heterosis in

111. IV.

V.

VI. VII.

VIII.

IX. X. XI.

B. Heterosis for Other Plant Characters. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ............... C. Combining Ability . . . . . . . . . . . . . . . . . . . . . Advantages of Hybrids over Conventionally Bred Cytoplasmic-Genetic Male Sterility Systems in Rice and Wheat . . . . A. Early Research ........................................ B. Major Sources of Cytoplasmic Male Sterility. . . . . . . . . . . . . . . . . . . . . . . . . . . C. Additional Sources of Cytoplasmic Male Sterilit D. Techniques for Cytoplasmic Differentiation. . . . E. Cytoplasmic Effects on Other Plant Characters . . . . . . . . . . . . . . . . . . . . . . . . . F. Cytoplasmic Effects on Disease Resistance .................. Fertility Restoration. . . . . . . . . . A. Sources of Restorer Genes ................................. B. Inheritance of Restoration . . . . . . . . . . . . . . . . . . . . . . . . . C. Environmental Effects on Male Fertility Restoration. . . D. Influence of Female Genetic Background on Fertility R Use of Chemical Pollen Suppressants in Hybrid Production.. Factors Affecting Cross-Fertilization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ................. A. Flowering Behavior . . . . . . . . . . . . . B. Floral Structure.. . . . . . . . . . . . . . . . . . . . . . . . . C. Effect of Pollinator Distance . . . ......................... D. Effect of Plant Height and Other Morphological Traits. . . . .............. .............. Seed Production ........................................ A. Multiplication of Cytoplasmic Male-Sterile and Maintainer ines . . . . . . . . . . ...... B. Hybrid Seed Production.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . C. Disease Problems Associated with Seed Production . . . . . . . ..... D. Seed Quality in Hybrids and Their Inbred Lines . . . . . . . . . . .............. QualityofHybrids ....................................... Economic Considerations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .............. Problems ............................................................. A. Rice ............................................................ B. Wheat ..........................................................

.

.

145

146 147 147 151 154 155 157 157 159 163 164 166 168 169 169 I 74 177 179 180 183 183 185 188 190 191 191 193 194 195 196 198 200 200 201

Copyright 0 by Academic F’resr. Inc. All rights of reproauction in any form reserved.

ISBN 0-12-000736-3

146 XII.

S. S. VIRMANI AND IAN

B. EDWARDS

Conclusion.. ......................................................... A. Current Outlook.. ................................................ B. Future Strategies.................................................. References ...........................................................

202 202 203 206

1. INTRODUCTION Rice and whe& together constitute the world’s two most important food crops. Estimates (U.S.D.A., 1982) of the total land area devoted to production of the two cereals indicate that approximately 380 million ha are evenly divided between wheat and rice cultivation. Wheat is the staple food of many developed countries and constitutes the leading cereal in terms of total world production. Because of the importance of rice in many developing countries, especially in Asia, considerable effort has been expended during the past 2 decades to develop varieties capable of responding to improved management and to reduce production hazards through the incorporation of genes resistant to major biological, chemical, and physical stresses. The successful development of hybrid maize in the 1930s provided an important impetus for breeders of other crops, including self-pollinatingcereals such as wheat, rice, barley, and often cross-pollinating sorghum, to utilize the principles of hybrid production. The basis for such genetic manipulation is the phenomenon of hybrid vigor, the tendency for the offspring of crossed varieties to have greater productivity than the parental varieties. Unlike maize, the floral biology of rice and wheat ensures that both crops are almost 100% self-pollinating. Consequently, selection for a more open flowering habit, with improved anther extrusion in the male parent and stigma receptivity in the female parent, was crucial to successful development of inbreds. Cytoplasmic male sterility (CMS) was visualized as an essential genetic tool to develop F, hybrids in self-pollinating crops. Kihara (1951) was the first to report the Occurrence of cytoplasmic male sterility in wheat. Later, Wilson and Ross (1962) established the existence of usable male sterility in wheat from the interaction of the common wheat nucleus with Triticum timopheevi cytoplasm. Schmidt et al. (1962) and Wilson and Ross (1962) found that they could restore fertility in the “Bison” cytoplasmic male-sterile line by crossing it with a T . timopheevi bread-wheat derivative. This prompted extensive research into hybrid wheat production in the United States and in other countries. Consequently, some commercial hybrids were developed and marketed in 1975, but production has been limited. In rice, the role of the cytoplasm in causing male sterility was fist reported by Sampath and Mohanty (1954) and Weeraratne (1954). The first cytoplasmic

HYBRID RICE AND WHEAT

147

male-sterile line in cultivated rice was developed by Shinjyo and Omura (1966a). Additional cytoplasmic male-sterile lines were developed in the early 1970s at the International Rice Research Institute (WU) (Athwal and Virmani, 1972), in the United States (Carnahan et al., 1972), and in China (Yuan, 1972). The Chinese were the first to use cytoplasmic male sterility to develop commercial F, rice hybrids in 1973 (Hunan Agricultural Bureau and Revolutionary Committee, 1977; Lin and Yuan, 1980). About 6 million ha are currently planted to hybrid rices in China, and hybrids have 20-30% higher yields than the best semidwarf commercial varieties. Developments in China have encouraged rice and wheat breeders elsewhere to expore the value of hybrid breeding for further increases in the yield potentials of these crops. This article reviews the current status and future prospects for breeding hybrid rice and wheat. An attempt has also been made to analyze existing breeding methodologies and to suggest some future strategies for hybrid improvement.

II. HETEROSIS IN RICE AND WHEAT Heterosis was f i s t reported in wheat when Freeman (1919) found that F, plants were generally taller than the tall parent. In rice, Jones (1926) observed that some F, hybrids had more culms and higher yields than their parents. Subsequently, other workers have reported the Occurrence of this phenomenon in various agronomic traits of rice and wheat such as yield, grain weight, number of grains per panicle or spike, number of panicles or spikes per plant, plant height, number of days to flowering, and general plant vigor. A critical prerequisite for the successful production of hybrid varieties is that sufficient hybrid vigor (heterosis) be available through specific parental combinations, so that yields of hybrids would significantly exceed those obtained from the best conventionally bred varities available; this difference is known as standard heterosis. The literature on heterosis in wheat has been reviewed by Briggle (1963), Johnson and Schmidt (1968), and Zeven (1972), and that on rice has been reviewed by Chang et al. (1973), Davis and Rutger (1976), and Virmani et al. (1981). A. HETEROSIS FOR YIELDAND YIELDCOMPONENTS

Reports in the literature have provided ample evidence of significant positive mid- and high-parent heterosis for yield, ranging from 1.9 to 368.9% in rice (Virmani et al., 1981) and from 0 to 100% in wheat (Briggle, 1963). However, a common problem in many of the reports in rice and in earlier reports in wheat was the limited scope and application of many of the studies. Generally, only a

148

S. S. VIRMANI AND IAN B. EDWARDS

small number of crosses were evaluaw, parental selection was not necessarily designed to maximhe heterosis;small populations were space planted either in the field or in greenhouses; noncommercial and unproductive varieties were frequently used (thereby eliminating the opportunity to evaluate standard heterosis); and the effects on height, maturity, and yield components were measured more often than heterosis for grain yield. Despite these limitations, the levels of heterosis have been high in certain cross combinations. However, Murayama et al. (1974)provided evidence that heterosis in rice is not influenced by plant spacing and soil fertility. In wheat, Jost and Glatki-Jost (1976)noted that excep tional hybrids can produce more tillers per unit area than inbreds, regardless of the seeding rate. Suggestions of exploiting heterosis commercially by developing F, rice hybrids have been made from time to time (Stansel and Craigmiles, 1966;Shinjyo and Omura, 1966a,b; Yuan, 1966, 1972;Craigmiles er al., 1968;Huang, 1970; Watanabe, 1971; Athwal and Virmani, 1972; Carnahan et al., 1972; Swaminathan et al., 1972;Baldi, 1976). However, difficulties in hybrid seed production discouraged most of the researchers from continuing their efforts, the notable exceptions being Chinese scientists (Yuan, 1966, 1972). In wheat, siflicant hybrid advantages have been measured in some instances, while other studies have reported no hybrid advantage (Kronstad and Foote, 1964;Larrea,1966;Brown etal.. 1966;Briggle ef al., 1967;Fonseca and Patterson, 1968; Livers and Heyne, 1968; Wells and Lay, 1970; Singh and Singh, 1971;Bitzer and Fu, 1972;Allan, 1973;Widner and Lebsock, 1973;Jost and Glatki-Jost, 1976; Yadav and Murty, 1976;Jost ef al., 1976b;Hughes and Bodden, 1978;Cregan and Busch, 1978;Mihaljev, 1980;Jost and Jost, 1980; Bailey et al., 1980;Wilson et al., 1980). Livers and Heyne (1968)reported on a comprehensive 4-year study to determine hybrid vigor by intercrossing 9 varieties of well-adapted winter wheats at Hayes, Kansas. The 36 hybrids collectively exceeded all varieties by 20,37,37,and 35% in the 4 years (1964-1967), respectively, with an average hybrid superiority of 32%. The best hybrid was consistently better than the best variety for the area. Similar results were obtained when 10 hybrids were compared with leading varieties at three locations (Livers and Heyne, 1966). When wheat hybrids were made using cytoplasmic male sterility-fertility restoration systems and field tested at optimum population rates, the results were less favorable than those obtained from hand-produced hybrids (Allan, 1973; Hayward, 1975; Johnson, 1977, 1978; Edwards er al., 1980). Allan (1973) found high-parent hetemis ranging from 23 to 113% among soft white winter wheat hybrids grown in the state of Washington, but the results were highly sitespecific; when averaged over 4 locations, they indicated that no hybrid had outyielded the high parent. Jost and Milohnic (1975)tested 5 hybrids in Yugoslavia and, in this small sample, only 1 hybrid showed high-parent heterosis.

HYBRID RICE AND WHEAT

149

Similarly, Hayward (1975) reported on a 1970-1971 study of 39 hard red winter wheat hybrids and 6 high-yielding check cultivars in Kansas; he found only 1 hybrid that equalled the yield of the highest check. However, Hayward did show significant improvements among later hybrids tested in 1973-1974; the mean yield of 4 hybrids over three locations was 19%greater than the mean of 8 check cultivars, and the best hybrid yielded 13.7%more than the top cultivar. Some possible reasons for the relatively poor performance of CMS-produced hybrids in the early studies are inadequate fertility restoration, adverse effects of T. rimopheevi cytoplasm, heavy selection pressure for restoration ability with less emphasis on agronomic performance during restorer development, and limited testing of different hybrid combinations. Most of the hard red winter wheat hybrids evaluated have shown more specific adaptability to certain regions and winter-hardiness zones than common check cultivars. Johnson (1977, 1978) reported on tests conducted at 11 sites in five states (Texas, Oklahoma, Kansas, Colorado, and Nebraska) during 1975-1979 and 1976-1977. Of the 15 hybrids tested in 1976, the leading hybrid produced 70 kg/ha less than the check cultivar ‘Centurk.’ In 1977,16 hybrids averaged 70 kg/ha less than Centurk, although the best hybrid exceeded Centurk by 130 kg/ha. Johnson concluded that the hybrids were not sufficiently superior to justify their use over the best available varieties. Published reports of spring wheat hybrid evaluation have been more limited compared with winter wheats. Edwards et al. (1980) reported on 1978 and 1979 tests with a series of spring wheat hybrids. Although the levels of high-parent heterosis ranged up to 35%, the top-yielding hybrids exhibited standard heterosis of only 10-14% above the leading check cultivar ‘Era.’In summary, hybrid wheats based on the cytoplasmic-genetic system have not expressed as much heterosis as the early handproduced hybrids, and the yield advantages to date have not been sufficiently great to justify widespread commercial production. Experiments on heterosis in rice conducted at Davis, California (Rutger and Shinjyo, 1980) indicated significant yield superiority of 11 of 153 rice hybrids over the best check variety. Standard heterosis ranged from 16 to 63% and averaged 41%. Hybrid corn seed producers in the U.S. maintained that this frequency (11 of 153 combinations) and degree of standard heterosis (41%) would make the prospects of hybrid rice exciting if sufficient hybrid seed could be produced (Rutger and Shinjyo, 1980). Studies conducted at the International Rice Research Institute in the Philippines during 1980-1981 have shown levels of as much as 73, 59, and 34% for mid-parent, high-parent, and standard heterosis, respectively (Virmani et al., 1982). Hand-crossed F, hybrids produced from elite breeding lines yielded up to 6.2 ton/ha compared with 5.0 ton/ha from the best check variety (‘IR42’)in the wet season, and 10.4 ton/ha compared with 7.9 ton/ha (IR54) under irrigation during the dry season. Of a total of 202 F, hybrids evaluated for yield during

150

S.

S. VIRMANI AND IAN B. EDWARDS

1980-1982, 63% showed positive high-parent heterosis (4-64.3%) and 50% showed positive standard heterosis (0.1-46.4%). Yields of the F, were found to be positively correlated with the parental mean. (r = 0.45**) and high-parent values (r = 0.33**). The parents had been selected for high per se yield performance, diverse genetic background, and resistance to diseases and insects incorporated though conventional breeding, Saini et al. (1974) also observed positive standard heterosis when the F, hybrids were derived from selected parents with improved plant type. However, the association between parental yield per se and F, hybrid performance may vary with the genetic background of the inbreds, and Khaleque et al. (1977) found no association between parental and hybrid yield performance. The most comprehensive commercial utilization of heterosis in rice has been that reported from the People’s Republic of China (Li, 1977; Lin, 1977; Lin and Yuan, 1980); more than 12 hybrids were officially released prior to 1980 (Shen, 1980). Yields under large-scale production have exceeded the best conventionally bred varieties by 20-30%. Results from replicated yield trials are given in Table I, and the data indicate that although the hybrids had fewer effective panicles per square meter, they had significantly more filled grains per panicle and larger seeds. The highest individual yield obtained from the F, hybrids was 12.8 tonslha, compared with 10.4 tons/ha from a conventionallybred variety (L. P. Yuan, personal communication). The major yield components in rice and wheat are number of panicles (or spikes) per square meter, spikelet number per panicle (or spike), spikelet fertility percentage, and 1000-grain weight. Significant positive mid-parent, high-parent, and/or standard heterosis have been observed for one or more of these components in a number of rice crosses (Pillai, 1961; Namboodiri, 1963; Rao, 1965; Dhulappanavar and Mensikai, 1967; Karunakaran, 1968; Carnahan et al., 1972; Chang et al., 1973; Mohanty and Mohapatra, 1973; Saini and Kumar, 1973; Sivasubramanian and Madhava Menon, 1973; Murayama et al., 1974; Saki er al., 1974; Parmar, 1974; Paramsivan, 1975; Davis and Rutger, 1976; Mallick ef al., 1978; Rutger and Shinjyo, 1980; Virmani et al., 1981, 1982). Virmani el al. (1981) observed negative heterosis for panicle number per square meter, but in combinations showing positive mid- and high-parent heterosis for yield this was overcompensated by positive heterosis in spikelets per panicle. Most crosses showing significant standard heterosis for yield have been found to show heterosis for more than one component (Saini et al., 1974; Mauya and Singh, 1978; Virmani et al., 1981, 1982). Results obtained in China and at the IRRI indicate that heterotic F, combinations usually show an increased sink size through increases in spikelets per panicle, spikelet fertility percentage, and 1000-grain weight. In wheat, Livers and Heyne (1968) pointed out that each yield component was important but that no single one was predominant in determining yield. Their

HYBRID RICE AND WHEAT

151

Table I Yield and Yield Components of Hybrid Rice Varieties in Regional Teats of 27 Sites in Hunan Province, china0 Hybrid combinations or check 1977 Wei You 6 Shan You 6 Nan You 6 Zhao You 6 Dong Ting Wan Xian (ck) 1979 Wei You 6 v20 x s Tan You 4 Dong Ting Wan Xian (ck)

Yield (tonsha)

Effective panicledm2

Filled grains per panicle

weight (g)

6.1 6.1 6.0 5.9 5.2

324 308 339 318 374

86.1 95.1 86.4 85.4 75.1

26.7 25.4 24.5 26.0 22.0

6.5 6.3 6.1 5.4

328 266 328 345

78.0 72.7 66.9 77.7

27.0 33.9 31.9 21.7

1OOO-grain

“Data from VimuCni ef d.(1981).

data showed that although top-yielding hybrids tend to have relatively high values in all three components, good performance is possible with a low value for any one component if the other two components have high values. Yieldcomponent compensation has been well documented in the literature (Donald, 1962; Bingham, 1967), and several workers have concluded that yield-component selection has limited value in breeding programs (Rasmusson and Camel, 1970; Fisher, 1975). However, kernel weight has been considered the most independent yield component because it is the last component developed, and its level of expression should not produce a compensating change in other components. In contrast, Sinha and Khanna (1975) hypothesized that heterosis in wheat will have commercial utility only when yield per spike increases, because tiller number per plant is strongly influenced by environment and can be manipulated by seeding rate. Briggle et al. (1967) also noted that heterosis for tiller number decreased as the population increased in b o a parents and hybrids. However, Jost and Glatki-Jost (1976) found that exceptional hybrids had the capacity to produce more tillers per unit area than inbreds irrespective of seeding rate, and Wilson et al. (1980) found the 17% high-parent heterosis in full dwarf X semidwarf hybrids to be primarily the result of an increase in spikes per unit area. B. HETEROSISFOR OFHER PLANT CHARACTERS

In both rice and wheat, a number of studies have shown that heterosis for plant height is highly cross specific (Pillai, 1961; Namboodiri, 1963; Briggle et al.,

152

S. S. VIRMANI AND IAN

B. EDWARDS

1964;Rao, 1965; Dhulappanavar and Mensikai, 1967; Livers and Heyne, 1968; Karunakaran, 1968; Amaya et al., 1972; Bitzer and Fu, 1972; Sivasubramanian and Menon, 1973; Ingold, 1974; Paramsivan, 1975; Khaleque et al.. 1977; Sreekumari et al., 1977; Mallick et al.. 1978; Wilson et al., 1980), and significant positive as well as negative heterosis has been reported. In rice, shorter height of F, hybrids may be attributed to the higher photoperiod sensitivity of some hybrids in comparison to their parents. In wheat, various studies have shown F, hybrids to exceed the tall-parent value (Ingold, 1974), exceed the midparent value (Amaya et al., 1972), approximate the mid-parent value (Wilson et al., 1980), and, in the case of “Olsen-dwarf” derivatives, approach the dwarfparent value (I. B. Edwards, personal observation). Because height is one expression of vigor that may lead to unfavorable grain/straw ratios and belowoptimum yields as a result of lodging, a number of hybrid programs are manipulating dwarfing genes in parents to obtain desirable height expression in the hybrids. A number of workers have observed the growth duration of rice hybrids to be shorter than the mid-parent value and, in some cases, shorter than that of the early parent (Dhulappanavar and Mensikai, 1967; Karunakaran, 1968; Chang et al., 1973; Bardhan Roy et al., 1975; Khaleque et al., 1977; Mallick et al., 1978). The dominance of Efgenes for the short basic vegetative phase (BVP) has been pointed out by Chang et al. (1969). However, in subtemperate-to-temperate China, most of the heterotic rice hybrids are later maturing than their parents (Lin and Yuan, 1980). This apparent discrepancy requires further investigation. Heading dates of wheat hybrids made on both normal and alien cytoplasms have tended to be earlier than the mid-parent value (Livers and Heyne, 1968; Amaya et al.. 1972; Bitzer and Fu, 1972; Wilson et al., 1980) and, in some combinations, to exceed the early parent value (Bitzer and Fu, 1972; Jost et al., 1976a; Jost and Hayward, 1980). I. B. Edwards (Table 11) found different sets of spring wheat hybrids, evaluated over a 5-year period, to consistently show dominance or overdominance for earliness. The latter is a desirable trait in the northern spring wheat region of the United States, because heat stress is frequently encountered during the early grain-filling period. A number of researchers have emphasized the need to maintain a complementary balance between “source” (photosynthate supply) and “sink” (potential grains) in cereals. Sinha and Khanna (1975) proposed that both source and sink capacity should increase in order to improve yield, and Virmani et al. (1981, 1982) observed this phenomenon in F, rice hybrids derived from semidwarf parents. Jennings (1967) found significant heterosis for vegetative growth to be negatively associated with yield in hybrids derived from tall parents. In contrast, Virmani et al. (1981, 1982) observed significant standard heterosis for both vegetative growth and grain yield in certain hybrid combinations. Increases in

HYBRID RICE AND WHEAT

153

Table II Average Hybrid Performance of Spring Wheat during a 5-Year Period“

Measurement Days to 50% heading Low parent Hybrid High parent Height (in.) Low parent Hybrid High parent Yield (kgha) Low parent Hybrid High parent

1978 (9)b

1979 (12)

1980 (13)

1981 (9)

1982 (32)

55.5 54.7 59.2

56.4 56.I 59.3

57.5 56.6 61.1

64.9 65.2 68.8

49.7 48.0 51.5

30.6 33.3 34.3

30.9 34.0 35.I

28.9 31.6 32.6

28.7 32.2 32.9

29.2 32.5 31.8

38.4 52.1 47.8

29.6 43.2 36.9

53.4 55.6 58.7

34.9 30.2 42.6

51.7 62.7 66.3

“Source: I. B. Edwards, Pioneer Hi-Bred International, Inc. bNumber of hybrids evaluated with their parents.

grain yield in certain hybrid combinations have been attributed to a more efficient distribution of dry matter in the plant, and the harvest index (ratio of grain weightkotal plant weight) has been examined by several workers (Sinha and Khanna, 1975). Benson (1978) found the high yields and heterosis in four spring wheat hybrids to result from both increased plant weight and harvest index. He attempted to use harvest index as a screening technique for yield in spring wheat hybrids. However, although a high correlation was shown between harvest index and yield in conventional-sized plots (2.44 X 1.22 m), the harvest index of small, single-row plots showed only a weak correlation with yield in conventional plots. In rice research, heterosis has been observed for such traits as cold tolerance (Sawada and Takahashi, 1977), salt tolerance (Akbar and Yabuno, 1975), photoperiod sensitivity, and rooting habit (Lin and Yuan, 1980). Chinese F, hybrids showed heterosis for root penetration rate, depth and width of the rhizosphere, number of adventitious roots per plant, and number of root fibrils (Anonymous, 1977; Lin and Yuan, 1980). Preliminary observations made at IRRI have also indicated that some hybrids are superior to their parents at comparable growth stages with regard to total root dry weight and root number, length, diameter, and pulling force (O’Toole and Soemartono, 1981). Superiority of F, rice hybrids in such physiological traits as photosynthetic area, chlorophyll content per unit area, photosynthetic efficiency, and mitochondrial activity has been reported in China (Hunan Agricultural College Depart-

154

S. S. VIRMANI AND IAN B. EDWARDS

ment of Chemistry, 1977; Lin and Yuan, 1980) and elsewhere (McDonald et al., 1971, 1974). In wheat, Sage and Hobson (1973) observed increased mitochondrial activity above the high-parent value in several mixtures, and found these to be significantly correlated with the percentage yield heterosis of fully restored hybrids grown at lower seed densities. Improvements in both plant type and physiological efficiency would appear to be a logical consequence of hybrid research in both rice and wheat. C . COMBINING ABILITY

The diallel analysis has been the major mating design used to estimate heterosis and the relative amounts of general combining ability (GCA) and specific combining ability (SCA) in rice and wheat. Most wheat studies have revealed that GCA is usually of greater relative importance for grain yield than is SCA (Kronstad and Foote, 1964; Brown et al., 1966; Gyawali et al., 1968; Walton, 1971; Bitzer and Fu, 1972; Widner and Lebsock, 1973). All workers reported significant GCA effects for grain yield, but significant SCA effects occurred only when the experiments were space planted (Kronstad and Foote, 1964; Gyawali er al., 1968; Yadav and Murty, 1976). The absence of SCA effects in competitive growth conditions suggests that nonadditive genetic variance may not be well expressed in wheat under these circumstances (Cregan and Busch, 1978). Widner and Lebsock (1973) evaluated a 10-parent diallel of genetically diverse durum wheat lines and found highly significant GCA effects for grain yield, tillers per unit area, kernels per spike, kernel weight, seedling vigor, maturity, height, and lodging. Specific combining activity effects among F, values were significant for kernel weight, seedling weight, and seedling vigor, suggesting that maximum grain production may be attainable under a system that can exploit both additive and nonadditive genetic effects. The largest levels of heterosis and the highest yielding hybrids involved genetically diverse parents, and other workers have concluded that in hybrid wheat the genetic diversity of the parents is as important as their mean performance (Nettevich, 1968; Yadav and Murty, 1976). The results of combining ability studies in rice have tended to be more variable than those in wheat. The predominant role of additive effects was established for all yield components except panicle number, which was affected by a certain level of nonallelic interaction (Chang et al., 1973; Li, 1975). Several workers have found high GCA effects in the parents to be associated with maximum SCA effects and heterosis for yield in the resulting hybrids (Ranganathan et al., 1973; Parmar, 1974; Maurya and Singh, 1977; Singh, 1977; Khaleque et al., 1977; Rahman et af.,,1981). In contrast, a number of inheritance studies have suggested dominant gene action for yield and/or yield components (Wu, 1968a,b; Chang, 1971; Sivasubramanian and Madava Menon, 1973; Shaalan el al.. 1975;

HYBRID RICE AND WHEAT

155

Singh and Nanda, 1976; Singh et af., 1979, 1980; Rahman et af., 1981). Chang (1980) found 1-2 pairs of dominant genes to affect the expression of heterosis for panicle and grain characteristics. Finally, some workers have suggested that their results showed little relationship between combining ability effects and the manifestation of heterosis in the corresponding hybrids (Mohanty and Mohapatra, 1973; Parmar, 1974; Singh and Nanda, 1976; Maurya and Singh, 1977; Rao et al., 1980; Haque et al., 1981; Rahman et al., 1981). Clearly there is a need for hybrid rice programs to investigate this subject further. The question of inbreeding depression has received comparatively little attention in rice and wheat although this is of major significance in assessing the merits of hybrid versus conventional breeding. Cregan and Busch (1978) studied the F, ,F,-F, bulks, and F5 lines from an eight-parent spring wheat diallel at two locations. The F, yields showed significant GCA and SCA mean squares. The latter was attributed to additive X additive epistasis and, although it was present in the F, progeny, it was less apparent in later generations. The significant F, heterosis and SCA for yield, coupled with significant inbreeding depression (0.23% yield reduction per 1% decrease in heterozygosity), indicated the possible desirability of F, hybrids to maximize yields. However, no F, hybrid significantly outyielded the best F5 line tested, and it was unclear whether yields would be maximized by pure line or F,-hybrid development. The inbreeding depression in these crosses between genetically related parents, although significant, was substantially smaller (one-half to one-third) than that reported in maize. This was attributed to the significantly less dominant genetic variance. Yadav and Murty (1976) were able to show varying levels of inbreeding depression in their eight-parent spring wheat diallel study. A high level of inbreeding depression was associated with high heterotic effects in diverse crosses. It is evident that hybrid programs should establish separate heterotic pools for male and female inbred development, that the genetic relationships between these pools should be minimized, and that further studies of the relationship between heterosis and inbreeding depression should be conducted.

Ill. ADVANTAGES OF HYBRIDS OVER CONVENTIONALLY BRED VARIETIES Hybrid advantages are not simply a function of heterosis. Three factors affect the end result: (1) breeding-method efficiency (a rate-of-progress factor), (2) the negative or positive effects of the cytoplasmic male sterility-fertility restoration system used to produce the hybrid, and (3) the inherent heterosis. Although the levels of heterosis in rice and wheat are comparable to those obtained in maize and sorghum, comparatively few studies have reported eco-

156

S. S. VIRMANI AND IAN

B. EDWARDS

nomically significant yield advantages over the best conventional varieties. The most promising results on hybrid rice have come from China (Lin and Yuan, 1980), where F, hybrids outyielded conventionally bred varieties by 20-30% under varying levels of on-farm management. Preliminary results from IRRI indicated that leading hybrids had high-parent and standard heterosis of 0-64 and 0-4796, respectively. In wheat, few studies have reported economically significant yield advantages of F, hybrids over the best conventional varieties. However, much of the research on hybrid wheat has been directed at perfecting the genetic system. Only since the early 1970s have the leading hybrid programs devoted more research effort toward the agronomic improvement of male inbreds (restorer lies). The identification of varieties potentially useful as female inbreds, their rapid incorporation into a male-sterile conversion program, and the maintenance of pure seed have provided a management challenge to hybrid breeders. It is against this background that one must assess the advantages of hybrids over conventionally bred varieties. A strong variety breeding program is fundamental for the production of female inbreds; to this extent hybrids and varieties are both complementary and competitive. When separate, large, and genetically diverse “pools” of male and female inbreds are available to a hybrid breeding program, it is reasonable to assume that consistently higher levels of heterosis will be obtained. The hybrid program may have considerable advantage over conventional programs that frequently suffer from inbreeding situations in which several of the top parents used in crosses have varying degrees of genetic relationship. At this point, the rate of progress factor in hybrid production becomes significant [i.e., the generations of selections (usually F,-F,) in conventional breeding are bypassed and the testing phase is immediate]. The hybrid,breeding approach can expedite the incorporation of dominant genes for resistance to major diseases and insects. For example, IR26 is a rice restorer line containing dominant genes for resistance to the brown planthopper and bacterial leaf blight; it has conferred resistance to a number of hybrids (Shen, 1980). If resistance is conditioned by recessive genes, these would have to be incorporated into both parents. In wheat, hybrids offer an advantage for trait complementation of certain quality factors. For example, a mixing time in dough development that is either too long or too short is considered an undesirable trait. Results indicate that mixing time in the hybrid is intermediate between that of the two parents, and this would produce a desirable result in such a cross. More vigorous vegetative growth, taller height, stronger root systems, and higher photoperiod sensitivity are some of the traits already observed in F, rice hybrids compared with their parents. These traits may aid in suppressing weed competition and enable hybrids to adjust to varying water and nutrient regimes. Yap and Chang (1976) repoM that hybrids performed better under dryland than under wetland conditions.

HYBRID RICE AND WHEAT

157

No yield-enhancing effects have been attributed to T. timopheevi or other cytoplasms used in hybrid wheat production. However, the significant nuclear-cytoplasmic interactions that have been obtained in certain instances (Gomaa, 1973) should not preclude such a possibility. So far, the use of T. timopheevi has produced no apparent adverse effects on agronomic traits, although unpublished data obtained by Hayward (1975) did suggest that heterosis was slightly lower in T. timopheevi compared with normal cytoplasm hybrids. In summary, environmental adaptation (i.e., the response of hybrids to heat and moisture stress, various insects, and diseases) combined with such factors as photoperiod sensitivity (rice), winter survival (winter wheats), and maturity will all contribute to relative hybrid advantage. These factors should be considered in conjunction with yield data. More extensive hybrid evaluation is now being conducted by a number of rice and wheat programs, and the results are encouraging.

IV. CYTOPLASMIC-GENETIC MALE STERILITY SYSTEMS IN RICE AND WHEAT The development of F, hybrid varieties of rice and wheat, both self-pollinating crops, must involve the use of an effective male sterility system. Among the available male sterility systems (genetic, cytoplasmic, cytoplasmic-genetic, and chemically induced), the cytoplasmic-genetic sterility system has been found the most effective and practical. Almost all of the 6 million ha of cultivated F, rice hybrids in China are developed from cytosterile and restorer lines. In wheat, all experimental as well as commercial hybrids that have been developed involve use of the cytoplasmic-genetic male sterility system. A. EARLYRFSEARCH

1 . Rice

The role of cytoplasm in causing male sterility in rice was first reported in 1954 (Weeraratne, 1954; Sampath and Mohanty, 1954). Katsuo and Mizushima (1958) observed completely male-sterile plants in the progeny of the first backcross Oryza sativa f. spontanealo. sativa cv. Fujisaka 5*. In the reciprocal cross, however, no male-sterile plant was observed, indicating the role of cytoplasmic factor@)and nuclear gene(s) interactions in inducing male sterility. Kitamura (1962a,b) observed slightly lower seed fertility in an indica/japonica hybrid, Tadukan/Norin 8, than in the parental varieties. Backcrossing with male parent Norin 8 increased spikelet sterility in the test progenies. Spikelet sterility

158

S. S. VIRMANI

AND IAN B. EDWARDS

was attributed to nondehiscence of anthers, because both male and female gametes were nomal. Shinjyo and Omura (1966a)developed the first cytoplasmic male-sterile line in cultivated rice by substituting nuclear genes of a japonica variety, Taichung 65, into the cytoplasm of indica variety Chinsurah Boro 11 (Shinjyo, 1970). Watanabe et’uf. (1968) observed male sterility in the progeny of the indica-japonica cross (Lead/Fujisaka 5), but those of the reciprocal cross were fertile. However, no male-sterile line was developed. Eiickson (1969)and Carnahan er uf. (1972)developed cytoplasmic male-sterile lines from crosses of an indica variety, Birco (PI279120),with Californian japonica rice varieties Calrose, Caloro, and Colusa; the F, plants were almost completely sterile whereas the reciprocal crosses produced about 50% seed set. The three Californian varieties, when used as recurrent paternal parents, always gave higher sterility in the Birco cytoplasm than in their own. The sterility increased with succeeding backcrossing of Californian japonica varieties into Birco cytoplasm, and the third backcross generation plants became completely male sterile (Camahan et ul., 1972). Watanabe (1971) also reported development of cytoplasmic-genetic male-sterile lines by means of indica-japonica crosses. A cytosterile line possessing. 0. gluberrima cytoplasma in the genetic background of variety Colusa was also developed in California (Camahan et u f . , 1972). Athwal and Virmani (1972)developed a cytoplasmic male-sterile line at the IRRI by substituting nuclear genes of indica rice variety Pankhari 203 into the cytoplasm of a semidwarf indica variety, Taichung Native 1. The first cytoplasmic male-sterile line used to develop commercial F, rice hybrids was developed in China in 1973 from a sterile plant (wild-aborted) occurring naturally in a wild rice population (Oryzu sutivu f. spontunea or 0. perennis) on Hainan Island in 1970 (Hunan Provincial Rice Research Institute, 1977;Yuan, 1977). Subsequently, cytoplasmic male-sterile lines have been developed from various accessions of 0. sutivu f. spontunea, indica variety Gambiaca (from Africa), and the Chinese variety 0-Shan-Tao-Bai (Lin and Yuan, 1980). Rutger and Shinjyo (1980)studied the distribution of male-sterile cytoplasms in various geographical forms of 0. perennis. In Asian and American strains, the frequencies of male-sterile cytoplasm were about 64 and 48,respectively. No malesterile cytoplasm was found in the African and Oceanean strains. 2 . Wheat

The first report of cytoplasmically induced male sterility (CMS) in wheat was that of Kihara (1951),who obtained cytosterile plants by substitution backcrossing of the common wheat genome into Aegilops cuudutu cytoplasm. Plants with A. cuudufu cytoplasm also showed partial female sterility and pistilloidy. Fukasawa (1953)obtained CMS plants from successive backcrosses of Aegifo-

HYBRID RICE AND WHEAT

159

tricum X Triticum durum". The Aegilotricum species had been synthesized from a cross of Aegilops ovutu with T. durum. Male-sterile T. durum plants with A. o v m cytoplasm had reduced plant height and delayed maturity compared with normal durum plants, and these effects were consistent through subsequent generations of backcrossing. In the reciprocal cross, A. ovutu plants with T. durum cytoplasm showed male and female fertility comparable to that of normal A. ovutu plants, but the T . durum cytoplasm delayed maturity and reduced plant height. Other studies by Japanese researchers established the presence of CMS plants in the intergeneric crosses, and the sterility persisted through backcross generations (Fukasawa, 1957, 1958, 1959; Kihara, 1958; Kihara and Tsunewaki, 1961).

B. MAJORSOURCES OF CYTOPLASMIC MALESTERILITY

I . Rice From the foregoing review of the literature, 19 sources of cytoplasmic male sterility in rice can be identified (Table III). Five of these [i.e., wild rice (designated as wild aborted or WA type), 0. sutivu f. spontuneu, Chinsurah Boro II (BT type), Gambiaca (Gam type), and 0-Shan-Tao-Bail are being used. More than 100 cytoplasmic male-sterile lines in indica and japonica backgrounds derived from these sources are currently available in China. The male-sterile lines from China are classified into three basic groups according to genetic properties and relation between restorer and maintainer lines (Lin and Yuan, 1980). Group I . The WA cytosteriles are typical of this group, but Gam type and some male-sterile lines derived from 0.sutivu f. spontuneu also belong here. The function of the male sterility gene is sporophytic; pollen grains abort at the uninucleate stage. Maintainer lines are found in both indica and japonica rices. Group ZZ. This group consists of BT-type male-sterile lines developed by Shinjyo and Omura (1966a). The function of the male sterility gene is gametophytic; pollen grains abort between the binucleate and the trinucleate stages. The restoration spectrum of Group I1 is wider than that of Group I; it is easier to sterilize japonica varieties than indica varieties to this type of male sterility. Group ZZZ. The cytoplasmic-genetic male sterility mechanism of this group is derived from some 0. sativa f. spontuneu lines. The Hong-Lien is typical of this group. Pollen grains abort at the binucleate stage and the relation between restorers and maintainers in this group is in contrast to that of Group I. For example, the maintainers of Group I, such as Zhen Shan 97 and Er-Jiu-Ai 4, become restorers of this group, and the restorers of Group I, such as Tai-Yin 1, are good maintainers of this group. Among the various sources of cytosterility, cytoplasmic male-sterile lines

S. S. VIRMANI AND IAN B. EDWARDS

160

Table IU Cytoplnsmie sourceS Identifkd to Induce Male Sterility in Rice

Cytoplasm source

PTB16

Nuclear source ?

Male-sterile lines developed (number)

-

Oryza sativa f. spontanea

Fujisaka 5

-

Oryzaf-

Fujisaka 5

-

Fujisaka 5 Several indica and japonica rim Norin 8 Taichung 65

Lead B h (PI279120)

w u 10 Several japonica rice varieties in China Fujisaka 5 calrose, calm

Orya glaberrima

Colusa IR36

Taichung (Native) 1

Pankhari 203

Akebono Wild rice with aborted pollen (0.sativa f. spontanea or 0.perennis) or WA

0.glaberrima

0.@pogon (KR 7) Gambiaca

0-Shan-Tao-Bai

Several

1

Several 2 or 3

1 1 -

Reference Weeraratne (1954);Sampath and Mohanty (1954) Katsuo and Mizushima (1958) Katsuo and Mizushima (1958) Heu and Chae (1 970) Lin and Yuan (1980) Kitamura (1962a) Shinjyo and Omura (1966a,b) Lin and Yuan (1980) L. P. Yuan (personal communication) Watanabe et al. (1%8) Erickson (1%9); Carnahan er al. (1972) Carnahan er al. (1972) s. s. Virmalli (unpublished) Athwal and V i (1972) Yabuno (1977) Yuan (1972);Lin and Yuan (1980)

Er-Jiu-Nan 1,

Several

Zhen Shan 97, V20, V41, and several other indica and j a p onica rices Taichung 65 Gang-Yi-Ya Ai Zhao, Yat-AiZhao Toride 1 and several other indica and japonica rices

Several

Cheng and Huang (1979) Lin and Yuan (1980)

Several

Lm and Yuan (1980)

HYBRID RICE AND WHEAT

161

Table III Continued

Cytoplasm source IARI 10061 IARI 10560 Jeerege Samba 0. perennis (Wl080) 0 . perennis (W1092)

Nuclear source

Male-sterile lines developed (number)

IARI 11445 IARI 11445

Reference

-

Parmar et al. (1981)

-

Shinjyo et al. (1981) Shinjyo and Motomura

IR24 Taichung 65 Taichung 65

(1981)

derived from the WA cytosterility system have been found to be the most stable in China and at the IRRI for their complete or nearly complete pollen sterility (Lin and Yuan, 1980; Virmani et al., 1981). According to L. P. Yuan (personal communication), the probability of developing stable male-sterile lines is higher from relatively wider crosses where the female parent is a primitive line and the male parent is an advanced line. The closer the relation between the two parents, the harder it is to obtain a stable male-sterile line, and vice versa. In the IRFU hybrid rice breeding program, 11 cytosterile lines representing 5 cytoplasmic sources (i.e., Gambiaca, Birco, 0 . sativa f. spontanea of Group I, Taichung Native 1, and BT) are available. Only 7 of these lines [Zhen Shan 97A, V20A, Er-Jiu Nan lA, and V41A (all WA type), Yar Ai Zhao A (Gam type), Pankhari 203A (TN type), and Wu 10A (BT type)] are relatively stable for pollen sterility. The lines MS519A and MS577A possess stainable pollen as do fertile plants, but these pollen grains do not germinate or affect fertilization. All these lines are highly susceptible to major diseases and insects in the tropics, and they cannot be used to develop commercial F, hybrids. Pankhari 203A is tall and photoperiod sensitive, and Wu 10A is a japonica type. At the IRRI, the cytosterility system(s) of some of these lines is being transferred into the genetic background of improved breeding lines and varieties that possess disease and insect resistance.

2 . Wheat Early Japanese research on cytoplasms, coupled with the discovery by Wilson and Ross (1962) that T. timopheevi cytoplasm induces male sterility, led to the

162

S. S. VIRMANI

AND IAN B. EDWARDS

establishment of major research programs in Japan, the United States, and Bulgaria to determine the cytoplasmic variation in species of the genera Triticum and Aegilops. Cytoplasmic male sterility in wheat was reviewed by Maan (1973a) and by Sage (1976). Over 15 different cytoplasms have been recognized, several of which induce male sterility in common wheat and could form a basis for alternative systems of hybrid production (Maan,1975; Mukai and Tsunewaki, 1980). Nearly all hybrid wheat breeding research continues to be based on the T. timopheevi system, and its widespread use has been largely the result of its apparently neutral effect on agronomic and quality characters. Most other cytoplasms from Triticum and Aegilops have deleterious effects on various traits (Maan, 1973a; Sage, 1976). Of the altemative cytoplasms available, most show no advantage over that of T. timopheevi, and time and resources limit change. However, the potential for genetic vulnerability to a major disease is always present when a single cytoplasm is used; the southern corn leaf blight epidemic in the United States in 1970 (caused by Helminthosporium maydis, race T) is a good example. Ghiasi and Lucken (1982a) compared the reactions of A. speltoides and T. timopheevi cytoplasms to various restorer gene combinations and examined a number of agronomic and quality traits. They concluded that A. speltoides cytoplasm can be used interchangeably with that of T. timopheevi in hybrid wheat breeding, providing an alternative that can broaden genetic variability. On the basis of nucleocytoplasmicinteractions (Maanand Lucken, 1971, 1972) and cytogenetic evidence (Kimber and Athwal, 1972; Kimber, 1973; Shands and Kimber, 1973), it has been suggested that A. speltoides may have contributed the G genome and cytoplasm to T. timopheevi. However, the restorer line R5 (T. zhukovskyi/3* ‘Justin’) is an effective restorer for T. timpheevi cytoplasm but not for A. speltoides cytoplasm, and this interaction provides the genetic basis for differentiation of the two cytoplasms. Gomaa and Lucken (1973) compared the breeding behavior of the restorers R5 and BR4704 in T. timopheevi and T . boeoticum cytoplasms. Fertility restoration (RB genes effective in T. boeoticum cytoplasm were also effective in T. timopheevi cytoplasm, but not necessarily vice versa. The reduction in vigor noted when the genomes of common wheat are substituted into T. boeoticum cytoplasm (Hori and Tsunewaki, 1967; Maan and Lucken, 1967, 1972; Gomaa, 1973) has curtailed the use of T. boeoticum as an alternative cytoplasmic source for hybrid breeding. It should be recognized from the previous statements that nucleocytoplasmic interactions are often significant; in T. timopheevi cytoplasm, small differences between male-sterile lines and their maintainers have also been noted for a number of traits (Jost et al., 1976b). Several researchers have also observed reductions in germination and seedling vigor with progressive backcrossing in certain wheat genomes into T. timopheevi cytoplasm.

HYBRID RICE AND WHEAT

163

C. ADDITIONAL SOURCES OF CYTOPLASMIC MALESTERILITY

1 . Rice

Although a number of sources for cytoplasmic male sterility in rice have been identified, more than 90% of the area planted to hybrid rice in China is occupied by hybrids derived from WA cytosterile lines. This situation makes hybrid rice in China potentially vulnerable to disease or insect epidemics. Work is in progress in China (Lin and Yuan, 1980)and at the IRRI (Virmani etal., 1981)to diversify usable sources of cytoplasmic male sterility for hybrid rice development. Rice species (i.e., Oryza gluberrimu, 0. fatuu, Asian forms of 0. perennis, and 0 . rufipogon) and varieties (i.e., PTB16, Tadukan, Lead 35, Akeboro, IARI 10061, IARI 10560,and Jeerege Samba) that are known to induce cytoplasmic male sterility may result in cytosterile lines that possess different cytosterility systems. The use of protoplast fusion techniques should expedite the development of new cytosterile lines (E. C. Cocking, personal communication).

2 . Wheat Additional cytosterility systems that supply male-sterile plants with good vigor and female fertility have been produced with the cytoplasms of Zhukovskyi (2n = 42; AAA’A’GG), Triticum araraticum (2n = 28; AAGG), and T. dicoccoides var. nudiglumis (2n = 28;AAGG) (Maan, 1975;Maan and Lucken, 1971). The fertility restoration systems of the previously mentioned cytosterility systems are also under complex genetic control and cause difficulties in breeding agronomically suitable restorer lines. Franekowiak et al. (1976) presented a proposal for hybrid wheat using Aegilops squarrosa cytoplasm that seeks to avoid the breeding of restorer lines. The D genome of common wheat contains genetic factor(s) €or restoration of fertility of T. aestivum with A. squarrosa cytoplasm. The nucleus of T. aestivum was substituted into A. squurrosa cytoplasm and the seed was treated with a mutagenic agent (ethyl methanesulfonate, EMS)to inactivate the critical gene(s) that causes fertility. Ten male-sterility mutants from an M, population of 45,000 plants expressed sterility in F, or F, generation, which indicates control by a single recessive gene. Crosses with four spring wheats produced completely fertile F, progeny. However, the major weakness of the system was that no malesterile genes that function specifically in the A. squarrosa cytoplasm were found. The development of fertile T. aestivum B lines with homozygous recessive genes for the maintenance of the A lines was not completed. Mukai and Tsunewaki (1979)proposed a similar system using the cytoplasms of Aegilops kotschyi and A. variabilis. When 12 common wheat genotypes were

164

S. S. VlRMANI AND IAN B. EDWARDS

substituted into these cytoplasms, 3 were found to be male sterile. Crosses with Chinese Spring produced a fertile F, hybrid, and restoration was attributed primarily to a single dominant gene (Rfvl).Trificurnspelfa var. duhumelianum carries a gene on chromosome 1B which interacts with A. kofschyi and A . vuriabilis cytoplasms to give male-sterile plants. Cultivars that carry the 1B/lR rye translocation and thereby lack the short, satellited arm of chromosome 1B display the same male-sterile interaction. Normal wheat cultivars carry gene(@ that overcome this sterility and therefore constitute male or restorer parents for hybrids. Comparisons with T. fimpheevi cytoplasm for 12 agronomiccharacters showed that A. bfschyi cytoplasm influenced only dry matter (reduced to 12%). The A. vuriabilis cytoplasm reduced plant height 496, ear number 18%, and dry matter 26%. Several hybrid programs in the United States and elsewhere are currently evaluating this system.

D. TECHNIQUES FOR CYTOPLASMIC DIFPERENTIA~ON

The techniques available for detecting cytoplasmic variation in a crop species are

1. Substitution backcrossing 2. Use of cytoplasm-differentiating genes 3. Interaction of restorer (Rfigenes with the cytoplasm 4. Study of pollen abortion patterns 5 . Restriction endonuclease fragment analyses of organelle DNAs. Work on these lines in rice has been limited. Chinese scientists have used techniques 2, 3, and 4 and reported that the WA cytosterility system is different from the BT system because the restorer gene(s) for the former have sporophytic action and those for the latter have gametophytic action (Y. Y. Dong, unpublished). Shinjyo (1969, 1975) and Kinoshita et al. (1980) have also reported gametophytic action of the restorer gene for BT cytosterilelines. By studying the pollen abortion pattern of different CMS lines, Xu (1982) found that pollen of WA cytosterile lines abort at the uninucleate stage and those in BT cytosteriles abort at the binucleate and trinucleate stages (Q. L. Jiang, unpublished). Chaudhary ef al. (1981) also established differences between WA, TNl, and BT male-sterile lines maintained at IRRIon the basis of their pollen abortion pattern. Chaudhary ef al. further suggested that the pollen abortion stage in a CMS line depended on the distance of relation of its cytoplasmic and nuclear donor parents. It appears that cytosterile lines with pollen abortion at the uninucleate stage are more stable for complete pollen sterility than lines with pollen abortion at the binucleate or trinucleate stage. Cytoplasmic variation in the Triticinae was reviewed by Maan (1973b, 1975)

HYBRID RICE AND WHEAT

165

and Sage (1976). The literature reveals that all genomically distinct species of Triticwn and related genera examined differ cytoplasmically. The sterility- fertility interactions between genomes from Triticum spp. and cytoplasms from Aegilops spp. indicate that male fertility-restoring genes derived from one cytoplasm donor species may restore fertility to male-sterile wheats that have cytoplasms of one or more of the other related species. Maan (1973b) drew attention to the fact that a number of factors will influence the expression and detection of cytoplasmic effects: (1) stability of the cytoplasm, (2) stability of the nuclear genome, (3) choice of cytoplasm and genome donor, (4) persistence of nuclear genes from the cytoplasm donor, and ( 5 ) genotype-environment interactions. In substitution backcrossing of the genome of one species into the alien cytoplasm of another, sufficient backcrosses are required to eliminate all nuclear genes derived from the cytoplasm donor species and produce a new nucleocytoplasmic combination. If this results in relatively stable male sterility, the cytoplasms of the two species involved in the cross may be considered distinct. Maan (1973b) reported that genomes of common wheat and durum wheat generally have similar interactions resulting in male sterility and abnormalities of plant growth with the cytoplasms of most of the related species. When they differed, the durum genomes were more sensitive to certain alien cytoplasms than the common wheat genomes. Sasakuma and Maan (1978) introduced T. durum genomes into the cytoplasms of 6 species of Triticum, 14 species of Aegilops, and 1 species each of Secale and Haynuldia. O f the 22 alloplasmic lines, 14 were completely male sterile, 4 were partially fertile, and the rest, which had the cytoplasms of T. dicoccoides, A . kotschyi, A. variabilis, or H . villosa, had normal fertility. When new nucleocytoplasmic combinations are made using T. aestivwn or T. durum genomes, differences between cytoplasms in traits other than male sterility often occur. These differences indicate that distinctions between cytoplasms cannot be based on male sterility alone. This subject is dealt with jn Section IV,E. Cytoplasm-differentiating nuclear genes include male fertility-restoring genes, male fertility-inhibiting genes, and genes affecting plant vigor in various ways. The relationships among these different genes are not clearly understood. However, research during the past decade has led several hybrid breeding programs to apply the terms “hard-to-restore” and “easy-to-restore” to genotypes in which male fertility-inhibiting genes are present or absent, respectively. The cytoplasms of T. timopheevi and A. speltoides cannot be differentiated by the presence of T. aestivum genomes, because male sterility is the only deleterious effect in both. However, substitution of the genome of T. timopheevi into A . speltoides cytoplasm produces male-sterile plants with reduced vigor (Maan and Lucken, 1972). Therefore, nuclear gene differences between the genomes must control the behavior of T. timopheevi and A. speltoides cyto-

166

S . S . VIRMANI AND IAN B. EDWARDS

plasms with either T. aestivum or T. tinwpheevi genomes. Such genes were termed cytoplasm-differentiating genes. In addition, T . tinwpheevi and A. speltoides cytoplasms differ in their reaction with restorer R5 (Maan, 1973a). The latter restores T. rimopkevi but not A. speltoides cytoplasm. This type of information on the components of interacting male sterility-male fertility restoration systems is important in hybrid wheat breeding. Restriction endonuclease fragment analysis of organelle DNAs, which is effective in demonstrating the heterogeneity of mitochondrial (MT) DNA among normal, fertile (Levings and Pring, 1977) and male-sterile cytoplasms in corn (Pring and Levings, 1978; Conde et al., 1979; Pring el al., 1980) and sorghum (Pring et al., 1982). Li and Liu (1983) found the heterogeneity of chloroplast (CT) DNA in CMS and maintaining lines in wheat, corn, and rape and suggested that changes in CT DNA may be involved in CMS.These techniques have not yet been used in rice and wheat for differentiating and identifying cytoplasm sources. E. CYTOPLASMIC Emcrs ON OTHER PLANTCHARACTERS

The effects of sterility-inducing cytoplasm on morphological traits have been reported for tobacco (Clayton, 1950; Chaplin and Ford, 1965), maize (Grogan and Sarvella, 1964; Grogan et al., 1971), and sorghum (Lenz and Atkins, 1981). Such effects in rice hybrids developed from WA cytosteriles V41A and Zhen Shan 97A and four fertility restorer lines (IR24, IR30, Xin-ni-ai-he, and Shiu Lian gu) have been reported in China (Lu et al., 1981). A comparison of F, hybrids A X R and B X R indicated that ‘Sterile’ cytoplasm had negative effects on number of spikelets per panicle, number of filled grains per panicle, 1OOOgrain weight, and yield per plant, although it had a positive effect on number of tillers per hill. Observations at the IRRI c o n f m that such effects are present but are cross specific; therefore it should be possible to eliminate the negative effects of cytoplasmic male sterility through the selection of appropriate restorer lines. Gomaa (1973) measured quantitatively and qualitatively the effects of T. boeoticum and T . timopheevi cytoplasms on five spring wheat cultivars. The male steriles in T. tinwpheevi cytoplasm were generally similar to their normal counterparts for the characters measured. With T . boeoticum cytoplasm a combined analysis revealed significant cultivar cytoplasm interactions for seedling vigor, vigor at heading stage, and spike length. C. F. Hayward (1975, unpublished) compared crosses of three restorer (male) lines with two male-sterile (T. tinwpheevi) lines and their normal cytoplasmic (B line) counterparts. Hybrids made on normal cytoplasm had yields 7.1% higher than their T . timopheevi counterparts, although the magnitude differed considerably between crosses (-3.5-22.0%). Mean heterosis levels averaged 19.9% in the normal cytoplasm and 12.8% in the T. tinwpheevi cytoplasm.

Table IV

Interactions between the Genomes of T. dunun, T. aesh'vum, and T. tinropheevi and the cytoplasms of Species of Aegilops, Tritieum, Secale, and Haynoldioa

Cytoplasm donor

Aegilops species A. speltoides A. bicornis A. longissimae A. sharonensisc A. mutica A. comosa A. heldreichii A. uniaristata A. caudatac A. umbellulata A. squarrosa A. cyIindrica A. ventricosa A. crassa A. ovata A. triaristata A. biuncialis A. columnaris A. juvanalis A. varzhbilis A. kotschyii A. triuncialisc Triticum species T. monoemcum T. boeoticum T. dicoccoides T. dicmcum T. dunun T. aestivum T. macho T. dicoccoides var. nudigIumis T. timopheevi T. araraticum T. zhukovskyi Other species Secale cereale Haynaldia villosa

Chromosome number (2n) and genome symbol

Nucleocytoplasmic interactionsb

T. durum

T. aestivum

T. timopheevi

14, SS 14, SbSb 14, SISI 14, SISI 14, M'Mt 14, MM 14, MM 14, MUMU 14, CC 14, C U C U 14, DD 28, CCDD 28, DDM'JM'J 28, DIDLMM 28, CUCuMM 28, CUCUMM 28, W M M 28, CuCuMM 42, DDMMCUCU 28, CUCUSISL 28, CuC'JSIS1 28, CUCUCC 14, 14, 28, 28, 28, 42, 42, 28,

AA AA AABB AABB AABB AABBDD AABBDD AAGG

28, AAGG 28, AAGG 42, AAAAGG 14, RR 14, HH

"After Maan (1975) and Sasakuma and Maan (1978). bF,male fertile; PF, partially fertile; S,male sterile; FS,female sterile; N, normal vigor; NN, near n o d vigor; BN, below normal vigor; w.weak (markedly reduced vigor); Z, zygote elimination (nonviable seed); B, bushy (stunted); E, early maturity; L, delayed maturity. =Some evidence of intraspecific cytoplasmic variability has been obtained by using two 01 more accessions.

168

S. S. VIRMANI AND IAN B. EDWARDS

The effects of various nucleocytoplasmiccombinations involving the genomes of common and durum wheat on a range of phenotypic characters were reviewed by Maan (1973a,b, 1975) and Sage (1976). Sasakuma and Maan (1978) reported on an extensive study in which the genomes of T. durum (selection 56-1) were substituted into 22 cytoplasms, and the effects on pollen fertility, seed set, heading date, and plant vigor were measured. Table IV represents the combined data of Maan (1975), Sasakuma and Maan (1978), and some data supplied by S. S. Maan (unpublished). It should be pointed out that the effects recorded are those that were most easily observable. In the future, more precise measurements may reveal different degrees of increased or reduced fertility and plant vigor. As has already been reported for rice, different accessions within certain species show nucleocytoplasmic interactions that differ sufficiently to suggest that cytoplasmic differences are present; these are indicated in Table IV (S.S. Maan, personal communication). The principal cytoplasmic effects resulting from various nucleocytoplasmic combinations included reduced vigor, delayed maturity, pistilloidy, and nongerminating grain. Mukai and Tsunewaki (1979) also showed that the degree of phenotypic deviation for a number of traits varied depending on the cultivar or genome used. The literature on gene interaction indicates that most genes do not act in isolation from other genes. Therefore, the alien cytoplasms and nuclear genes controlling cytoplasmic effects from alien sources may alter the phenotypic expression of various agronomic triats. Although no yield-enhancing effects from nucleocytoplasmicinteractions have been reported, the possibility of such an Occurrence should not be disregarded in the future as increasing numbers of substitutions are made by hybrid breeders.

F. CYTOPLASMIC E m s ON DISEASE RESISTANCE Although work to date has not revealed any relationship between CMS and disease susceptibility in rice (Y. Y.Dong, unpublished data), the desirability of hybrid rice and wheat breeders using more than one source of cytoplasmic male sterility as a safeguard against potential disease epidemics is widely recognized. Washington and Maan (1974) tested alloplasmic lines of T. aestivum (cultivars Chris and Selkirk) and T. durum with three physiologic races of wheat leaf rust (Pucciniu recondiru) at the seedling and adult plant stages. Although euplasmic and alloplasmic Selkirk lines were resistant to all races at both stages, differences were found in the cultivar Chris. Both forms were seedling susceptible,but adult plants of euplasmic Chris were resistant, whereas certain alloplasmic lines were susceptible or moderately susceptible to all races used. Other alloplasmic Chris lines were susceptible to one race but not to the other two. These results indicate that certain alien cytoplasms may alter the expansion of host nuclear genes for resistance to certain physiologic races of leaf rust. Furthermore, the host parasite

HYBRID RICE AND WHEAT

169

interaction was influenced by host cytoplasmic factors, host nuclear genes, and the rust fungus.

V. FERTILITY RESTORATION A. SOURCES OF RESTORERGENES

1. Rice

The practical use of cytoplasmic-genetic male sterility in developing hybrid varieties in grain crops is possible only when the effective restorer lines are identified and/or developed. In rice, effective restorer lines for the WA, Gam, and BT cytosterility systems have been identified among cultivated rice varieties and elite breeding lines (Shinjyo, 1969, 1972a,b, 1975; Lin and Yuan, 1980). Effective restorer lines for cytosteriles Pankhari 203A and MS577A have not been identified. Shinjyo (1975) found that 35% of 153 rice varieties that originated from outside Japan were effective restorers for the BT cytosterile line. Restoration ability was related to grain type (Matsuo, 1952). Effective restorer varieties were mainly distributed in the tropics where indica rice was exclusively grown. The Fist set of effective restorer lines used in China in commercial F, hybrids involving the WA cytosterility system was identified in 1973 (Hunan Provincial Rice Research Institute, 1977; Lin and Yuan, 1980). Since then, a number of these lines have been selected and/or developed for various cytosterility systems. The frequency of restorer lines, higher among rice varieties originating in lower latitudes, was about 20% of 75 varieties from southwestern and southern China but only 7.5% of 438 varieties from the Yangtze Valley (Hunan Academy of Agricultural Sciences, unpublished). The frequency of restorer lines was even less among varieties from nothern China, eastern Europe, Japan, and Korea; the highest frequency (35%) was found among 197 varieties originating in the lower latitudes of south and southeast Asia. The restoring ability of rice varieties has been found to be somewhat related to their origin. The frequency of restorer lines is higher and restoration ability is stronger among varieties closely related to the WA cytosterility source; it is higher among indicas because indica varieties originalted earlier than japonica varieties and are closely related to wild rice. The frequency of restorer lines among japonica varieties is negligible (Shinjyo, 1975; Lin and Yuan, 1980); consequently, japonica F, rice hybrids in China have been bred by transferring restorer gene(s) into the male parents from indica rice. Among indica rices, restorer lines have been more frequent among late-matur-

170

S. S. VIRMANI AND IAN B. EDWARDS

Table V Restorer Lines Used Widely in Commercial F1Rice Hybrids in China Line

Restorer of CMS system

Origin

IR24 IR26 IR30 IR661 IR665 Gu 154 Gu 223 Gui 630 Tai Yin 1 Yinni Ai He Ke Zhen 145 Zhao Hui 3533 c55 c57 Beijing 300 5350 5154 Hong Zhao Nou

WA, Gambiaca WA, Gambiaca WA, Gambiaca WA, Gambiaca WA, Gambiaca WA, Gambiaca WA, Gambiaca WA, Gambiaca WA, Gambiaca WA, Gambiaca WA, Gambiaca WA, Gambiaca Chinsurah Born I1 (BT) Chinsurah Boro I1 (BT) Chinsurah Boro I1 (BT) Chinsurah Boro I1 (BT) Chinsurah Born I1 (BT) Chinsurah Boro I1 (BT) Nan guang zhan

IRRI IRRI IRRI IRRI IRRI Cuba/IRRI CubdRRI CubamRRI ThailanMRRI Indonesia China China China China China China China China China

ing than among early-maturing varieties, perhaps because late-maturing indicas are primitive, relatively closer to wild rice. The restorer lines used widely in China are listed in Table V. The IR24, IR26, IR661, and IR665 restorers of the most promising, most widely cultivated indica hybrids in China originated at the IRRI. Additional restorers are being selected from IRRI elite breeding lines that possess multiple disease and insect resistances to be used in developing hybrid rices for the tropics and subtropics. 2 . Wheat The development of fertility restorer lines (R lines) which restore complete male fertility to F, hybrids presented a major challenge to hybrid wheat breeders and has constituted a key determinant of the pace of hybrid wheat development. The male fertility-restoring genes or gene combinations and cytoplasm used in hybrid wheat represents a substantial introgression of germ plasm from related species. Krupnov (1971) pointed out that all species capable of donating a cytoplasm that causes male sterility in common wheat are also sources of the corresponding Rf genes. This is to be expected if it is the absence of specific

HYBRID RICE AND WHEAT

171

genes, caused by their elimination during backcrossing, that causes pollen development to break down in male-sterile nucleocytoplasmic combinations. Wilson and Ross (1962) made the necessary breakthrough for hybrid wheat breeding when they successfully transferred Rf genes from T. timopheevi to common wheat by substitution backcrossing. Schmidt et ul. (1962) found malesterile and male-fertile plants in ‘Nebraska 542437,’ a T. timopheevi derivative (1279A9-111-4) crossed to ‘Nebred.’ Crosses between the male-sterile and male-fertile plants were fertile, demonstrating that a suitable restorer for T. timopheevi cytoplasm had been found. Most Rf genes are derived from T. timopheevi, but a number also occur in common wheat (Zeven, 1968; Lucken, 1973). Johnson and Schmidt (1968) pointed out that the discovery of Rf genes in a number of hexaploid and tetraploid varieties that provide at least partial restoration of CMS (T. timopheevi) lines suggests the possibility that at least one of the genes in the Nebraska restorer may have come from common wheat. Evidence from both the Kansas and Nebraska program suggested that a gene in ‘Cheyenne’ winter wheat is the same as one from T. timopheevi. When certain wheat cultivars are crossed into a CMS line, partial fertility varying from only a few seeds at the base of the spike to nearly full seed set may be observed in F, or BC, (F,) plants. In the latter case this would indicate that fertility is expressed only when the genes are homozygous. Zeven (1967) found that only ‘Redman’ spring wheat and a line from T. mcha (No. 2) caused partial fertility in A. cuudutu cytoplasm; in contrast, he was able to identify 30 common wheats and the T. macha (No. 2) line as having Rf genes for T. timopheevi cytoplasm. Joppa and McNeal (1969) identified 3 common wheats (PI 167841, PI 277013, and PI 277016) with genes for pollen fertility restoration to CMS (T. timopheevi) Selkirk. Oehler and Ingold (1966) reported that the hexaploid European cultivar ‘Primepi’ restored fertility to CMS (T. timopheevi) lines, and Milohnic and Jost (1974) reported Primepi as showing the highest level of restoration among the group of common wheats they tested. Lists of common wheats that carry Rf genes have been provided by several researchers (Zeven, 1967; Porter and Merkle, 1967; Johnson and Schmidt, 1968; Jost and Milohnic, 1976a; Ghiasi and Lucken, 1982b), and it is now a comparatively common Occurrence for hybrid wheat breeders to encounter Rf genes in common wheat varieties during substitution backcrossing into T. timopheevi cytoplasm. The reasons for this occurrence of Rf genes in conventional cultivars is not known. It can be hypothesized that partial Rf genes are retained because of pleiotropic effects that positively influence the phenotype, because they are linked with other desirable genes, or because their retention is a chance occurrence. Hughes and Bodden (1977) suggested that the high frequency of Rfgenes in wheats developed at the Plant Breeding Institute (Cambridge, England) may have been the result of linkage with genes for resistance to powdery mildew

Table VI

Origin of Some Sources of Male-Fertility-RestoringGenes Used in Hybrid Wheat Breeding Program

c 4

E3

Cytoplasm restored

Designation

-

Aegilops ovata P168

ABD-1

-

Aegilops caudata ABD- 13 R8 R9

Triticum timopheevi

R10 R1-Lee R2-Sonora 64 (= CIMMYT restorer)

Reference and/or other information

Derivation T. dicoccoides var. kotschyanum T. aestivum var. erythrospermum, where chromosome 1D is replaced by A. caudata chromosome C-sat-2 (1C) T. dicoccoides var. spontaneo-nigramlA.squarrosa typica No. 2 T. compactum T. dicoccum VemaYA. squarrosa strangulata A. speltoideslChmese Spring//4*Chris A. speltoideslChinese Spring/l2*SelkirkChinese Spring A. caudata/9*Chris//R8, F8 T. timopheevil2*Hussar-Hard. Fed//Comet Hussar-HardRed/Nebr (= Nebr 542437) T. iimopheevil2*Marquid/Sonora64

Fukasawa (1955) Kihara (1963a) Rf gene; Tahir and Tsunewaki (1971) Tahir (1969) Kihara (1963b)

Kihara and Tsunewaki (1964) Maan (1973a) Maan (1973a)

Maan (1973a) Schmidt et al. (1962)

Rf genes; Talaat (1969) CIMMYT (unpublished); Rf genes; Bahl and Maan (1973)

R3 (Wilson or Kansas restorer) R4 or RD RC RK

R5

R6 or BR4704 R7

TX73C9610-1

-

W 4

VS73-786 149-2-3 R11 R12, RE1 R13, RE3

T. timopheevi/3*Marquis T. T. T. T. T.

timopheevi-A. squarrosa/3*Dirk timopheevi-A. squarrosa/3*Canthatch rimopheevi-A. squarrosa/3*Kam zhukovskyi/3*Justin boeoticum-A. squarrosalT. durumNChinese spring T. boeoticum/2*T. durum/l2*Selkirk T. spelra var. duhumelianum T. aestivwn cv. Rimepi T. timopheevi/6/n*msBison/4*Selkirk/5/~s Bison/2*Selkirk/4/T. timopheevi/Justin//Marquis/3/Justin Complex cross-Rf source: Primepi, Maris Beacon, and R3 T. arararicum/3*Selkirk, F7 T. dicoccoides var. nudiglumis No. 1/6* T. durum T. dicoccoides var. nudiglumis No. 4/T. durum

Wilson and Ross (1962); Rf genes; Livers (1964) University of Manitoba, Canada

Rf genes; Yen et al. (1969)

Maan and Lucken (1967, 1970, 1972) Maan and Lucken (1970, 1972) Maan and Lucken (1970) Kihara and Tsunewaki (1966) Oehler and Ingold (1966) Gilmore et al. (1978)

Jost (1980) Maan (1973a) Maan (1973a); Sasakuma and Maan (1978) Maan (1973a); Sasakuma and Maan (1978)

174

S . S . VIRMAM AND IAN

B. EDWARDS

(Erysiphe graminis f. sp.) or other genes affecting productivity. In contrast, Ghiasi and Lucken (1982b) evaluated 40 F, lines derived from a single F, plant which was heterozygous for a gene(s) for partial fertility restoration. Testcrossing revealed that 27 lines had the Rf gene but 13 lacked it. Two years of evaluation for agronomic and disease traits failed to explain the retention of the partial Rf gene in the conventional variety breeding program for reasons other than chance. Sources of Rf genes widely used in hybrid breeding programs have been cited by several researchers (Maan and Lucken, 1972; Zeven, 1972; Maan, 1973a; Sage, 1976; Sasakuma and Maan, 1978; Jost, 1980), and a number of these sources are listed in Table VI. Two factors should be mentioned: first, the genome of the cytoplasm donor is clearly not the only source of Rf genes; and second, a number of the restoration sources are capable of restoring more than the cytoplasm indicated. For example, Maan and Lucken (1970) obtained highly fertile F, hybrids of normal vigor by crossing R5 and R6 into A lines with the cytoplasm of either T. timopheevi or the amphidiploid T. boeoticum-A. squarrow. Maan (1973a) transferred Rf genes to T . aestivum and T. durum genomes from several accessions of T. uraruticum (R1 1) and T. dicoccoides var. nudiglomis (R12, R13). These T. durum restorer lines restored fertility to CMS lines of durum with cytoplasms from several species of Triticum and Aegilops.

B. INHEIUTANCE OF RESTORATION Results of experiments designed to study the inheritance of fertility restorer genes using conventional genetic analysis are often difficult to interpret. The variable penetrance and expressivity of certain Rf genes, the presence or absence of fertility inhibitor genes, genetic background, and environment all affect genetic ratios in segregating populations. The information available on the two crops is reviewed in the following sections. I . Rice

Shinjyo (1969) identified a fertility-restoring single dominant gene, Rf, in the variety Chinsurah Boro I1 for the BT cytosterile line; its effect was gametophytic in the male sterility-inducing cytoplasm (MS-Boro). Kitamura (1962a) reported that high fertility in the F, hybrids of cytosterile Ta 820 (Kitamura, 1962b) was conditioned by a recessive gene combined with modifiers or polygenes. Shinjyo (1973, however, contended that the fertility-restoring gene of the variety Tadukan used by Kitamura (1962a) behaved similarly to the restorer gene from Chinsurah Boro II. Kitamura’s work had indicated that cytosterile Ta 820 pos-

HYBRID RICE AND WHEAT

175

sessed functional male sterility, because there was no abortion of male and female gametes but the anthers were nondehiscent. These two authors may have been working with two different systems of cytosterility. Shinjyo (1975) also studied the genetics of fertility restoration in the male sterility-inducing cytoplasm of the rice variety Lead identified by Watanabe et al. (1968). Two restorers, Fukuyama (a japonica variety) and a strain of MS-Boro Rf, Taichung 65 (Shinjyo et al., 1974), possessed the Rfx and the Rf gene, respectively, and had gametophytic effects in the Lead rice cytoplasm. The Rfx gene gave weaker restoration than did the Rfgene in the cytoplasm of Chinsurah Boro 11. Neither the Rf nor the Rfx genes restored the fertility of 0. sativa f. spontanea cytoplasm identified by Katsuo and Mizushima (1958). Shinjyo (1975) could not find any restorer source for Chinsurah Boro I1 cytoplasm, which had both effective and weak restoring genes. It was therefore concluded that the weak (Rfx) and effective (RB genes were probably allelic. The genetics of restoration in cytosterile lines developed by Katsuo and Mizushima (1958), Erickson (1%9), and Athwal and Virmani (1972) has not been reported. Therefore, the fertility restoration genes identified by Shinjyo (1969, 1975) cannot be related to fertility restoration in the cytosterile lines developed by the other workers. Inheritance of fertility restoration for cytosterile lines developed and used in China, although determined, is not well documented in the literature available outside China. Information gathered by S. S. Virmani during his various visits to China, and the results obtained by Gao (1981), indicate that restoration in the WA cytosterile line is controlled by one or two pairs of dominant genes, and the effect of the restoration genes is sporophytic. In some crosses, however, fertility restoration in WA cytosterile lines has also been found to be quantitative in inheritance (Anonymous, 1976).

2 . Wheat Sage (1976) reviewed the literature on fertility restoration and pointed out that in many studies the initial genetic explanation was proved by subsequent work to be an oversimplification. For example, restoration of T. tirnopheevicytoplasm by Primepi has been reported to be caused by a single dominant gene (Oehler and Ingold, 1966; Goujon and Ingold, 1967), two incompletely dominant genes with a major and a minor effect, respectively (Miller, 1970; Schmidt et al., 1971; Miller et al., 1974), two incompletely dominant genes with the epistatic action of a single recessive gene (Nettevich and Naumov, 1970), and one major and one minor dominant gene which act in a complementary manner together with one modifier gene and one inhibitor gene (Shebeski, 1971). Using monosomic analysis, it was shown that two Rf genes on chromosomes 1B and 5D controlled male fertility restoration in Primepi (Bahl, 1971; Bahl and Maan, 1973).

S. S. VIRMANI AND

176

IAN B. EDWARDS

Table W Chromoeomes Shown by Mom#lomie Analysis to InniKnee Male Sterility Restoration

Restorer line

Chromosome location of Rf genes

Reference

R1-Lee W-Sonora 64 R3

lA, 5A, 7D lA, 6B, 7D lA, 7D

Talaat (1969); Bahl and Maan (1973) Bahl (1971); Bahl and Maan (1973) Talaat (1969); Bahl and Maan (1973); Robertson and Curtis

R4-RD (Dirk)

lA, 7D

R4-RC (Canthatch) R4-RK (Karn)

6B, 6D lA, 6B lA, 7B, 7D 1B lB, 5D 1C (Sat-2) lB, 4B, 7D 6B

Yen et al. (1969); Talaat (1969); Bahl and Maan (1973) Yen et al. (1969) Yen et al. (1969) Bahl (1971); Bahl and Maan (1973) Tahir and Tsunewaki (1%9) Bahl (1971); Bahl and Maan (1973) Tahir and Tsunewaki (1971) Zeven (1970) Gilmore et al. (1977)

lA, 6B

Bravo (1982)

R5 T. spelta var. duhamelianwn Primepi P168 Minister Tam W-1oQ (rye translocation) R113

Monosomic analysis has been the principal technique used to reinforce the results obtained by conventional genetic analysis of restoration, and the results of analyses on several restorer lines are summarized in Table VII.Most monosomic analyses have been conducted on restorer lines that were the initial introgression lines or were early sources of restorer genes for T. timpheevi cytoplasm. However, Bravo (1982) used R113, a restorer line from the North Dakota State University hybrid program that had proved superior to other R lines in its ability to confer complete male fertility to F, hybrids. Monosomic analysis indicated that R113 had restorer genes on chromosomes 1A and 6B. Restorer genes of lower penetrance or modifier genes were located on chromosomes lB, 4B, and 2D, and inhibitors of restoration were located on chromosomes 5A, 6A, and 5B. The penetrance of these genes appeared to be influenced by environment, (i.e., they were detected in only 1 of 2 years). Because no major new restorer gene was detected, it was suggested that the high restoration capacity of R113 may result from the balanced, cumulative effects of genes on several chromosomes. The hybrid breeder is therefore selecting restorer lines from populations which are segregating for several genes, both restorer genes per se and modifying genes that may control the level of restoration needed for productive F, hybrids.

HYBRID RICE AND

177

WHEAT

c. ENVIRONMENTAL EFFECTS ON MALE FERTILITY RESTORATION The effects of environmental factors on male fertility restoration in hybrid rice would appear to be genotype-specific. Several of the commercial hybrid combinations involving restorer lines IR24, IR26, IR661, Gui 154, and others are widely adapted in China, implying that their restoration ability is stable in various enviornments. Lin and Yuan (1980), however, reported that the rate of seed set in hybrid rice, particularly under unfavorable climatic conditions, was only about 80%, less than that of the best conventional varieties in China. Combinations with better seed set than leading conventional varieties, even under unfavorable conditions, have been developed, confirming that the stability of fertility restoration in hybrid rice is genotype specific. IR54, another restorer line identified at the IRRI, has given normal seed set in the F, combination Zhen Shan 97A/IR54, both at the IRRI and in Fukien Province, China (Ren Cui Yang, personal communication). Results of a joint study between IRRI and China, aimed at identifying effective restorer lines among elite breeding materials developed at IRRI and other national programs, indicated that about 15 and 24% of these lines were effective restorers in China and at the IRRI, respectively, but only 6% were effective restorers at both sites (Table VIII). The restoration ability of the remaining restorer lines was site specific; the frequency of restorer genotypes was higher in a tropical than in a subtropical environment. In wheat, the effects of various environmental parameters such as temperature, photoperiod, and moisture stress on general growth and development are well

Table VIII Restoration Ability of 218 Elite Breeding Lines Tested in China and at the IRRI (1981)a Test in China ~

Test at the IRRI

Restorer

Partial restorer

Restorer Partial restorer Partial restorer; maintainer Partial maintainer Maintainer

13 17

37 66

-

-

-3 33

No data

Partial maintainer

Maintainer

Total

-

2 4 4 -

3 25 7 8 15 -5

2

53 111 7 10 23 14 -

113

63

9

218

3 -

4

“Figures given ar? number of lines. About 90% of the breeding lines were developed at the IRRI; the rest were developed in other national programs.

178

S. S. VIRMAM

AND IAN B. EDWARDS

documented. These factors may also influence the penetrance and expressivity of

Rf genes, and in some cases the interactions of cytoplasmic male-sterile lines with restoring genes in different environments have been highly unstable. This has been verified in the International Wheat Restorer Germplasm Screening Nursery (IWRGSN), where common sets of hybrids grown in several countries showed an unusually high level of instability and environmental sensitivity associated with fertility restoration (Jost, 1979, 1980, 1981). The genetic background of both male-sterile and restorer parents influenced seed set, but it was not possible to identify the environmental components that caused the wide variation between sites. Although several R lines in the IWRGSN have produced hybrids with relatively stable restoration and seed set, Lucken (1982) expressed doubt as to whether complete and normal fertility in F, hybrids has been achieved under all environmental conditions. In some instances, seed set may be complete, but the anthers may be slightly malformed at the tip of the spike. Based on field observations of restoration from Mexico to the northern United States, Wilson (1968a) classified environments as shallow sterile (Mexico), sterile (KansadOklahoma), and deep sterile (northern United StatesICanada). He suggested that restorers adequate in the shallow-sterile environment may be inadequate in the deep-sterile environment. Environmental factors that influence seed set and/or affect the stability of fertility restoration are listed in Table IX.

Table IX Environmental Factors That Af€ect the Stability of Fertility Restoration Environmental effects that may depress seed set Indirect effects Pot experiments Greenhouse environment Planting date Location

Short growing season Limited plant development Climatic effects Long photoperiod Long day and Low temperatures High temperatures Dry conditions High temperature and dry conditions

Reference

Miri et al. (1970) Rajki and Rajki (1966); McCuistion (1968); Schmidt er al. (1971) Kihara (1970) Monteagudo et al. (1967); Lucken and Maan (1967); Wilson (1968b); Schmidt er al. (1971); Jost (1979, 1980, 1981) Wilson (1968a) Wilson (1968a) Nanda and Chinoy (1945); Welsh and Klatt (1971); Johnson and Patterson (1973) Fukasawa (1953); Meletti (1961) Welsh and Klatt (1971); Johnson and Patterson (1973) Bingham (1966) Mihaljev (1972, 1976a,b)

HYBRID RICE AND WHEAT

179

It is apparent that temperature effects are complex and several interactions occur. Although higher temperatures prevail in the southern United States, flowering occurs earlier in the hard red winter wheats (April) as compared with the hard red spring wheats (July) in the northern United States. Heat stress can occur in the spring wheat region in July; consequently, spring wheats may encounter environmental stresses such as long photoperiods, cold temperatures (in some seasons), heat stress during pollen formation and anthesis (in some seasons), and dry conditions (where they prevail), which limit plant development. This has provided a challenge in restorer-line breeding and has necessitated the use of long-term R-line testers to monitor environmental effects each year when new testcross hybrids are evaluated in the field. D. INFLUENCEOF FEMALE GENETICBACKGROUND ON FERTILITY RESTORATION

Although no published data on this subject are available for rice, the widespread occurrence of intervarietal hybrid sterility attributable to gametic development (GD) nuclear genes randomly distributed among rice varieties (Oka, 1953, 1954, 1963) suggests that the genetic background of the female parent could influence the pollen and spikelet fertility of F, hybrids. In wheat, variation in the ease of restoration (EOR) of a number of genotypes in T. tirnopheevi cytoplasm has been reported by several workers (Lucken and Maan, 1967; Wilson, 1968a; Mihaljev, 1976a,b; Trupp, 1976; Jost, 1979, 1980, 1981). Wilson (1968a) interpreted this phenomenon as being the result of variations in sterility gene number and/or effectiveness among CMS lines, or of the presence of fertility genes that act in a complementary or additive fashion with restorer genes. Excess sterility genes could act as inhibitors of pollen fertility restoration in the F, generation. Trupp (1976) evaluated a range of soft red winter wheats for ease of restoration and found wide variability in the latter, a small portion of which could be accounted for by the presence of minor genes for restoration in the maintenance line. By using partially effective R-line testers as female parents, Trupp established that EOR is a genetically controlled trait for which prediction could be accomplished with a satisfactory level of accuracy. Jost (1980) reported that in the Second IWRGSN (1979), MS-Butte had the highest EOR (107.5% seed set) and MS-Tobari sib had the lowest (64.0% seed set). In the 1980 nurseries, the highest and lowest EOR levels were recorded in MS-Maris Hobbit (139.3%) and MS-Abe (47.1%) (Jost, 1981). It may be concluded that hybrid programs should place high emphasis on evaluating EOR among potential female inbreds in addition to searching for more effective Rf gene combinations. There is increasing evidence that stable restoration and consistent seed set under different environmental conditions results from

180

S. S. VIRMAM AND IAN B. EDWARDS

a balance between major Rf genes and modifer (fertility enhancing) genes that improve EOR.

VI. USE OF CHEMICAL POLLEN SUPPRESSANTS IN HYBRID PRODUCTION An alternative system for producing hybrids in rice and wheat is the use of chemical pollen suppressants (CPS). Male sterility induced by CPS is relatively convenient to use because there is no need to maintain it. Any variety can be sterilized and used as the female parent of a hybrid, and there is no need to identify a restorer line for making a commercial hybrid. Because there is no genetic segregation for male sterility in the F2 population, it is possible for farmers to plant heterotic F2 populations. An ideal chemical pollen suppressant should

1. Selectively induce only pollen sterility without affecting female fertility 2. Be systemic or sufficiently persistent to sterilize both early and late tillers 3. Have a reasonably broad “window” or target period of application to overcome the effects of adverse weather conditions and varible crop growth and permit treatment of large hectarages 4. Have minimum side effects on plant growth 5. Not induce only functional male sterility with viable pollen and nondehiscent anthers.

The proximity of the stamens and pistil within rice and wheat florets and the coordinated sequence of morphological development limit the ways to affect male and female fertility differentially. However, there exists an inherent differential sensitivity between male and female flowers, as is evidenced when environmental stresses such as high temperatures, low temperatures (frost, in wheat), and drought cause male but not female sterility. These observations and others, that several different nuclear and cytoplasmic genes can cause male sterility, and that there is both gametophytic and sporophytic control of male sterility in cereals, suggest several possible mechanisms to suppress pollen development without affecting the female (Lucken, 1982). A number of researchers have evaluated the use of CPS in wheat (Chopra et al., 1960; Porter and Weise, 1961; Rowell and Miller, 1971, 1974; Bennett and Hughes, 1972; St. Pierre and Trudel, 1972; Trupp, 1972; Borghi et al., 1973; Hughes et al., 1974; Jan et al., 1974; Johnson and Brown, 1976; Mihaljev, 1976a; Miller and Lucken, 1977; Dotlacil and Apltaverova, 1978; Jan and Rowell, 1981). The first studies (Chopra et al., 1960; Porter and Weise, 1961)

HYBRID RICE AND WHEAT

181

reported on the use of maleic hydrazide (250-1000 ppm); both of these groups obtained complete pollen sterility, but considerable female sterility and plant damage was also encountered. Foliar application of 2-chloroethyl phosphoric acid (ethaphon or ethrel) induced male sterility in wheat without significantly affecting female fertility (Law and Stoskopf, 1973; Hughes et al., 1974; Rowell and Miller, 1974). However, restricted spike emergence (phytotoxicity) and the need for precision in the time of application (narrow target period) have limited the commercial utilization of this chemical. Fairey and Stoskopf (1975) have since reported that soil application of granular ethephone overcame the phytotoxic effects associated with folial application, although high rates were required and average sterility was less than 100%. Ethephon has also been evaluated in rice (Perez et al., 1973;Cheng and Huang, 1978; Parmar et al., 1979a). Perez et al. (1973) applied ethephon (lo00 ppm) in three applications at 2-day intervals at early boot stage. Pollen sterility was 67% effective, but female sterility and phytotoxic effects were also encountered. Parmar et al. (1979a) used considerably higher dosage rates (a single application of 6000-8000 ppm at the boot stage or split applications of 4000-6000 ppm 1 week before and at boot stage). Pollen sterility was 94% effective but extremely low seed set resulted, and it was not clear whether insufficient pollen load or female sterility was the major cause of this low seed set. Cheng and Huang (1978) also reprted female sterility, and current research on rice suggests that ethephon is largely ineffective because application rates sufficiently high to induce pollen sterility also cause female sterility. The compound RH53 1 [sodium 1-(p-chloropheny1)-1,Zdihydro4,6-dimethyl-Zoxonicotinate; Rohm and Haas Co.] has been evaluated as a CPS in wheat and rice (Perez et al., 1973; Jan et al., 1974). RH531 gave a mean pollen sterility of 99% when applied at 100 ppm at the prebooting stage in rice (Perez et al., 1973), but it also caused complete ovule sterility. Jan et al. (1974) found that treatment with RH531 several days prior to meiosis at a rate of 2.0 kg/ha active ingredient (ai) gave maximum reduction in fertility with the spring wheats ‘Anza’ and ‘Yecora 70.’ Anza showed increased spike density and thickened floral tissues and was more sensitive to RH53 1 than was Yecora. Floret opening was not conducive to cross-pollination, and very low seed set resulted. Clearly, the effective use and versatility of CPS will depend on genotype-chemical, environment-chemical, and genotype-environment-chemical interactions. Where desirable heterotic combinations are identified, maximum seed production may be expected to result from the specific tailoring of application rates and timing for the cultivar and environment. Jan and Rowell (1981) compared high and low application rates of ethephon, RH532, and RH2956 on wheat tillers at various stages of development in Anza and Yecora 70. Treatment with RH2956

182

S. S. VIRMANI AND IAN B. EDWARDS

at the high application rate induced uniform and maximum male sterility in early and late tillers of both Anza and Yecora 70. Treatment with RH522 (high rate) was effective for Anza but not for Yecora 70. Ethephon proved the least effective, affecting only the late tillers treated at or before meiosis. Miller and Lucken (1977) evaluated four compounds (RH531, RH532, RH2956, and RH4667) for northern spring wheats. Several chemical companies in the United States are currently evaluating CPS compounds in wheat, and one company has commenced limited CPS production on several soft red winter and hard red winter hybrids. Chemical pollen suppressant rice hybrids have been released in China. Two chemicals, zinc methyl arsenate (CH,As,03Zn) and sodium methyl arsenate (CH,As,Na), have proved effective when sprayed 5-9 days before flowering at a concentration of 0.02% (Anonymous, 1978a); the sodium salt is superior. Three hybrids (Gang-hue-da-tian, Gua hua 2, and Bei-hua gu 67) were released for commercial production and have performed well. The costs of preparing these CPS compounds are also low (Shen, 1980). Other Chinese institutionshave evaluated CPS compounds. The Hunan Academy of Agricultural Sciences (unpublished) has reported finding zinc and sodium methly arsenate (200 ppm), calcium sulfamate, and fluoroacetamide (15OO-2OOO ppm) to be effective male sterilants. However, the general consensus apparently indicates a preference for CMS systems over CPS systems in hybrid production. The advantages of using a CPS method for hybrid seed production may be summarized. (1) Breeding procedures are simplified by eliminating the need for cytoplasms and Rf genes; (2) costly and difficult operations of male-sterile increase are avoided; (3) genotypes with poor anther extrusion can still be used as female parents; (4) the time lag in converting promising new genotypes into CMS (female) inbreds is avoided; and (5) evaluation of lines for general and specific combining ability is simpler. The principal disadvantages of the CPS method are; (1) chemical dosage levels sufficient to ensure male sterility often induce some female sterility and reduced seed set as compared with CMS hybrid production; (2) extended rainfall and prolonged winds can prevent field application of the chemical at the optimum time; (3) genotype-environment-chemical interactions must be evaluated; and (4) male fertility or selfing in hybrid fields can result in seed that does not conform to seed law specifications for a hybrid. Clearly, both systems have their advantages and disadvantages, but they can complement one another in advancing the development of hybrid rice and wheat. Chemical pollen suppressants can aid CMS programs in evaluating the combining ability of new lines prior to male-sterile conversion. Conversely, CMS lines can be used to evaluate the performance of a CPS. By comparing seed sets on equivalent (unsprayed) CMS lines and normal lines treated with a CPS, a measure of the effect of the chemical on female fertility can be determined.

HYBRID RICE AND WHEAT

183

VII. FACTORS AFFECTING CROSS-FERTILIZATION Floral structure, anthesis, and anther dehiscence patterns in rice (Van Breda de Haan, 1913; Hector, 1913; Rodrigo, 1925) and wheat (Percival, 1921; Leighty and Sando, 1924) make these crops strictly self pollinating. The extent of natural outcrossing in cultivated varieties varies from 0 to 6.8%in rice (see Sahadevan and Namboodiri, 1963) and from 0 to 4%in wheat (Heyne and Smith, 1967). In wild rice forms of 0. perennis Moench, 16.5-100% outcrossing has been observed (Sakai and Narise, 1959; Oka and Morishima, 1967). Male-sterile plants of cultivated rice have shown outcrossing of 0-44% (Stansel and Craigmiles, 1966; Athwal and Virmani, 1972; Carnahan et al., 1972) and 20-92% (Trees, 1975). Seed set as much as 75%on male-sterile wheats is a common occurrence. Variability in extent of natural outcrossing in these crops can be attributed to variations in flowering behavior, floral characteristics of varieties or species, and variations in environmental factors. An analysis of the factors affecting cross fertilization in the two groups should be useful. A. FLOWERING BEHAVIOR

Early descriptions of the flowering process were provided by Rodrigo (1925) in rice, and by Percival (1921) and Leighty and Sando (1924) in wheat. Rice researchers have addressed the subject of flowering behavior on the basis of bloom initiation, blooming duration of the panicle, and the angle and duration of floret opening, and wheat researchers have examined factors affecting flower opening, the duration of stigma receptivity, and pollen viability. Flowers of rice and wheat reach their peak opening between 9:00 and 11:30 AM on a sunny day. In contrast to rice, a second flush of flower opening is common around 3:OO to 5:OO PM in wheat, although Parmar et al. (1979~) identified two rice cultivars in India (IAFU 6193-B and IARI 7216) which also showed a second flowering flush between 5:30 and 6:OO PM. Work at the IRRI has also shown species differences in the time at which peak flowering takes place; approximately 60% of the spikelets of 0. glaberrimu flowered at 9:OOAM, whereas less than 5% of those in 0. sutivu reached anthesis at this time (WU, 1978). Flowers of both crops have two lodicules located at the base of the ovary, and their primary function is to open the floret at anthesis. As pollination approaches, the lodicules increase in turgidity, pushing apart the palea and lemma (Kadam, 1933), and consequently they acquire a significance in hybrid breeding because of their role in aiding anther extrusion and pollen interception. Studies on rice by Parmar et al. (1979~)suggested that the inner floral organs, such as the filaments and stigma, exert slight pressure on the linear joints of the palea and lemma as they become turgid whereas the lodicules exert leverage pressure at the

184

S. S. VIRMANI AND IAN B. EDWARDS

basal joint of the lemma and palea. Simultaneously the anthers protrude and thereby help increase the angle of opening, which varies among varieties from 25" in long, slender spikelets to 35" in short, coarse spikelets. In wheat, the lodicules are dependent on adequate moisture and cool temperatures for proper functioning. McNeal and Ziegler (1975) measured lodicules taken from spring wheat florets at daily intervals after heading and found an increase in size until the sixth day after complete emergence from the boot. Lodicules from both emasculated and male-sterile spikes were significantlyheavier, wider, and thicker than those from the spikes of fertile cultivars. Significant varietal differences in lodicule size were also noted. Varietal differences in the duration of flower opening from 28 to 93 min have been reported in rice (Virmani and Athwal, 1973; Parmar et al., 1979~);floret opening in CMS rice plants ranged from 105 to 280 min (IRRI,1983). Parmar et al. (1979~)also noted a longer duration of flower opening in late-maturing varieties (50-70 min) compared with early-maturing varieties (28-35 min). Delay or failure in pollination was found to prolong flowering (Grist, 1953) and, consequently CMS plants have a longer duration of floret opening than fertile plants (Saran et al.. 1971). A mean temperature of 2530°C and a relative humidity of 7 5 4 0 % was found to provide optimum conditions for anthesis in male-sterile, maintainer, and restorer lines, with the flowering period lasting 6-7 days (Hunan Academy of Agricultural Sciences, personal communication). Higher or lower temperatures affected synchronization;both low (18-26°C) and high (27-35°C) temperatures affected a number of agronomic and floral characters that influence outcrossing. High temperatures delayed anthesis and slightly increased the period of floret opening (IRRI, 1983). Genetic studies of traits affecting flowering behavior in rice have been limited, Estimates of heritability and the genetic coefficient of variation for floweropening duration were found to be low (Virmani and Athwal, 1973). Nagao (1951) reported nonclosure of glumes after anthesis to be a simple, recessive trait, whereas others have shown that temperature, light, and humidity markedly influence blooming behavior (Chu et al., 1970; IRRI, 1983). In wheat, flowering behavior has been examined from the perspective of the duration of stigma receptivity rather than from that of floret opening per se. Prolonged stigma receptivity is considered essential, because exact synchronization of anthesis in the male with flowering in the female is often difficult to accomplish in seed production blocks. Stigma receptivity has generally been measured by seed set on the male sterile, and most studies suggest that environmental effects are of greater significance than genetic effects (Bardier, 1960; Rajki and Rajki, 1966; Johnson and Schmidt, 1968; Zeven, 1968; Khan et al., 1973). Under field conditions, fluctuations in temperature and relative humidity each day, and from day to day, generally encompass both optimum and adverse conditions. The period of stigma receptivity is reduced as a direct result of these

HYBRID RICE AND WHEAT

185

fluctuations, and studies have shown a range from as few as 2 days (Khan et af., 1973) to as many as 13 days (Rajki and Rajki, 1966) in wheat, and from 2 to 6 days in rice (Virmani and Tan, 1982; Hunan Academy of Agricultural Sciences, personal communication). Stigma receptivity has also been shown to decrease as the elapsed period flowering and actual pollination increases (Imrie, 1966). Johnson et af. (1967) suggested that parental lines should flower on the same day to maximize seed set and reduce floral disease, but others have suggested that the pollinator should flower 1-4 days later than the male sterile. A 2-year study of seed set at two North Dakota locations by Miller and Lucken (1976) suggested that seed set is maximized by planting the maintainer a projected 44°C growingdegree days after the male sterile. This difference indicated that maximum anthesis in the maintainer occurred approximately 4 days ahead of optimum floret opening in the male sterile. Because seed set constitutes an indirect measure of stigma receptivity, the question of pollen viability cannot be excluded. In growth-chamber studies, Watkins and Curtis (1967) found that wheat pollen viability decreased as the temperature increased, for all relative humidities tested (20, 50, and 80% RH), and also as the relative humidity decreased, for all temperatures tested (65, 75, 85, and 95°F). Field experiments in Colorado c o n f i i e d the growth-chamber findings, and similar results were reported by Zeven (1968) in Holland and by Mihaljev (1976a,b) in Yugoslavia.

B. FLORALSTRUCTURE

Stigma size, style length, stigma exsertion, stigma receptivity, anther size, filament size, and pollen number are important floral characters that influence outcrossing in rice and wheat. Significant varietal differences in these traits have been observed in cultivated and wild rices (Copeland, 1924; Sampath, 1962; Oka and Morishima, 1967; Virmani and Athwal, 1973; Lyakhovkin and Singil’Din, 1975; Parmar et al., 1979b; Virmani ef al., 1980a; IRRI, 1983) and in wheat (Cahn, 1925; Kherde et al., 1967; Atashi-Rang, 1970; De Vries, 1974a;Jost and Milohnic, 1976a,b). The range of variation for these traits and the source of desirable floral traits as reported in the literature are given in Tables X and XI. Jost and Milohnic (1976b) tested 20 Rf sources in wheat and found the highest pollen count in the cultivar Primepi. There was a highly significant positive correlation between restoration ability and number of pollen grains per anther in the 3 years of testing and for the three female testers used. They suggested that anther length and numberofpollen grains per anther might be used to identify plants in segregating progenies that possess Rf genes, thereby reducing the number of testcrosses made each year. However, Jost and Glatki-Jost examined parents and progeny from the cross Primepi X R1 (Nebraska) at two sites. The F7

S. S. VIRMANI AND IAN B. EDWARDS

186

Table X Range of Variation and Varietal Sources P d p Maximum Value of Floral Traits Muenring Outcrossing in Rice as Reported in the Literature Trait Stigma length (mm) Stigma breadth (mm) Style length (mm)

Extent of stigma exsertion (%) Anther length (mm) Anther breadth (mm) Filament length (mm) Pollen nurnberhther

Varietal source possessing maximum value

Range of variation Cultivated, 0.2-2.6 Wild, 0.3-5.0 Cultivated, 0.2-1.0 Wild, 0.4-1.3 Cultivated, 0.6-3.2 Wild, I .2-2.3 Cultivated, 0.2-87.8 Wild, 0-100 Cultivated, 0.9-3.7 Wild, 1.6-5.4 Cultivated, 0.2-0.9 Wild, 0.2-1.0 Cultivated, 0-23 Wild, 0-14 Cultivated, 463-3833

IARI6637, IAR17332, IARI10979A" Genetic stock 6209-36 Not reporteda Genetic stock 6209-3 IARI7332, IARI10754, IARI10871, IARI10979A, IARIl 1205" Oryza sativa f. spontaneaa BPI76-1 (n.s.)= Genetic stock 6209-36 IARI5819, lAR15823a Oryza australiensis" IARI5819, IARI5823a Oryza sufivu f. spontanea" Not known Not known IR13526-41- 1-2'

"Parmar et al. (1979b). bIRRl (1963). c V h a n i and Athwal(I973). "Hunan Academy of Agricultural Sciences (personal communication). (unpublished).

lines and parents had significantly shorter anthers at Szeged than at Zagreb, but the mean number of pollen grains per anther for all lines was essentially the same. Substantial line-site interactions were obtained, and the authors questioned the validity of using anther dimensions as a basis for selecting genotypes containing Rf genes from segregating progenies. Inheritance studies (Virmani and Athwal, 1974) for floral traits such as anther length, stigma length, and stigma exsertion in rice indicated that these traits were governed by polygenes. Huang and Huang (1978) reported that stigma exsertion was dominant, partially dominant, or recessive depending on the cross. They also found a negative correlation between stigma exsertion and pollen fertility. Both additive and nonadditive effects were important in the inheritance of floral traits (Virmani and Athwal, 1974). The prevalence of duplicate type epistasis was considered a barrier to fixing these traits at higher levels of manifestation through conventional breeding. Therefore, recurrent selection in biparental progenies was proposed by Virmani and Athwal(l974) to improve these characters,

187

HYBRID RICE AND WHEAT Table XI

Range of Variation in Anther Size and Number of Pollen Grains in Wheat as Reported in the Literature Number of cultivars tested

Range of variation

Reference

Anther length (mm) 30 5 9 20 45 26

3.45-5.09 2.99-3.84 3.22-4.29 3.29-5.07 3.19-4.41 3.67-5.25

Atashi-Rang (1970) De Vries (1974a,b) Fisher (1977) Jost and Milohnic (1976b) Kherde er al. (1967) Milohnic and Jost (1970)

Anther width (mm) 20 45

0.29-0.97 0.57-1.07

Jost and Milohnic (1976b) Kherde er al. (1967)

Number of pollen grainslanther 22 4 5 4 8 20 1

581-2153 856-1380 1746-2856 2236-3022 2867-3867 2188-3983 8100

Beri and Anand (1971) Cahn (1925)

De Vries (1974a) D’Souza (1 970) Joppa et al. (1968) Jost and Milohnic (1976b) Yeung and Larter (1972)

because they may have to be transferred from wild rice in which an undesirable linkage relationship exists among the floral traits and agronomic characteristics. Attempts to transfer the long stigma (5 mm) trait of the genetic stock designated as 6209-3 (IRRI,1983) into the genetic background of selected maintainer lines, from which it would be transferred to cytoplasmic male-sterile lines in rice, are being made at the IFUU and the Sichuan Academy of Agricultural Sciences, China (IRRI, 1983). Atashi-Rang and Lucken (1978) found significant differences in both GCA and SCA effects of parents for anther length, anther extrusion, and glume tenacity in wheat. The GCA-SCA variance component ratio was approximately 5:l for anther length, 1:1 for anther extrusion, and 8:l for glume tenacity. Narrow-sense heritabilities on a per spike basis were 0.61 for anther length, 0.19 for anther extrusion, and 0.54 for glume tenacity. The authors suggested that selection for altered anther length and glume tenacity should be effective, and that selection in stress environments could provide a means of improving and stabilizing anther extrusion data.

188

S. S. VIRMAIW AND IAN B. EDWARDS

Ghiasi and Lucken (1982b) reported that visual mass selection for anther extrusion in F3 and F5 bulks of spring wheat R/B, R/R, and A/R crosses effectively increased the mean anther extrusion of successive generations. This increase was 19% per cycle in the F3 bulk mean. Selection for anther extrusion did not adversely affect other agronomic and quality traits. Komaki and Tsunewaki (1981) found that anther length was correlated with flowering data among parent lines as well as the F, progeny of 13 cross combinations. Anther length and flowering date fitted a curvilinear regression wherein reduction in anther length in both early- and late-flowering cultivars was attributed to poorer environmental conditions for floral development. This phenomenon was not found in the F2, and it was concluded that the genes controlling flowering date differ from those controlling anther length. The heritability estimate for anther length was 0.65. The pollen load in the air at a given time is a function of the amount of pollen produced per anther, the amount of anther extrusion, and the number of anthers per unit area. Joppa ef al. (1968) investigatedthe relative pollen-shedding ability of 11 hard red spring wheats and 3 durum varieties. Percentage of anther extrusion had the largest direct effect on pollen shedding. The expected pollen load in the air per 10 mm2 was compared with the actual number obtained on slides situated within each plot. The calculated and observed numbers of pollen grains were of the same order of magnitude, with a correlation coefficient (r) of 0.91. Their results indicated that the relative pollen-shedding capacity of a variety can be predicted from a knowledge of the number of pollen grains per anther, percentage of anther extrusion, and fertile florets per plot. These studies indicate that anther length and extrusion score are the simplest measurements of pollen-shedding capacity for R-line breeding. After the confirmation of restoration ability based on testcross fertility, the yield evaluation of new R lines will have dual significance. The economics of hybrid production will be improved, and high-yielding lines will have more fertile florets per unit area and an increased pollen load. C. E

m

OF

POLLINATOR DISTANCE

Provided that an adequate pollen load is present in the field, and that there is correct synchronization of flowering between male and female parents, seed set on male-sterile plants is also influenced by the distances that pollen can travel and remain viable. De Vries (1971) reviewed the environmental factors influencing pollen viability in wheat. The factors investigated by rice and wheat researchers have included both pollinator distance and wind direction. Rodrigo (1925) observed that in cultivated rice varieties pollen grains are carried by the wind, in the same plane as the flower itself, as far as 1.5-2.0 m

HYBRID RICE AND WHEAT

189

from the source depending on wind velocity. Virmani er al. (1980b) found that seed set from cross-pollination was higher on plants located 30 cm from the pollen source than on those located 30-110 cm from the source. However, differences were less pronounced when the CMS line was located downwind from the pollinator, and it was concluded that satisfactory seed set could be obtained in hybrid seed production plots by growing five rows of the CMS line (1-1.25 m wide) alternately with one row of the restorer (pollinator) line. Similar malelfemale ratios are used in China (Lin and Yuan, 1980). Pollen dispersal trials conducted in China (Hunan Academy of Agricultural Sciences, personal communication) have shown that in hybrid seed production plots with isolation distances of 10, 20, 30, and 40 m, contamination resulting from foreign pollen was 5.2, 1.0, 0.2, and O I , respectively. Consequently, a minimum isolation distance of 40 m is set for hybrid seed production in China. In wheat, a number of studies on the effects of pollinator distance and wind direction on seed set have been conducted (Kihara and Tsunewaki, 1964; Holland and Roberts, 1966; Porter er d.,1966; Bitzer and Patterson, 1967; Popov and Gotsov, 1968; Rajki and Rajki, 1968; Keydel, 1969; Tsunewaki, 1969; Ezrokhin and Nettevich, 1970; Anand and Beri, 1971; Stoskopf and Rai, 1972; De Vries, 1974b; Miller and Lucken, 1976). However, a major limitation of several of these studies is that they were conducted in very small plots where the basic pollen load was so small, and the seed set was so low, that they provide little guidance for commercial field production of hybrid seed. In Kansas, Holland and Roberts (1966) showed that percentage seed set decreased from 57.6 to 15.3% as distance of the female increased from 0.3 to 29.0 m from the pollen source. The importance of wind direction on percentage seed set was demonstrated by De Vries (1974b) in Holland and by Bitzer and Patterson (1967) in Indiana, where wind direction affected seed set in their experiments as much as range in pollinator distance (1.5-7.6 m). Evidence to date indicates that the isolation requirements for hybrid wheat seed production (and particularly malesterile increase) are greater than those indicated for rice. Jensen (1968) showed that although 90% of the wheat pollen shed remained within 6 m of its source, pollen could travel as far as 60 m. Researchers at Pioneer Hi-Bred International, Inc. (Kansas) conducted pollen slide studies and found that viable pollen could be obtained as far as lo00 m from a very large pollen source (C. F. Hayward, personal communication). In another Kansas study, Arp (1967) measured onethird more pollen at 12.2 m from the pollen source than at 1.5 m, which again confirms the potential buoyancy of air-borne wheat pollen. It appears that the size of the pollinator strip and its direction relative to the female will have a considerable effect on seed set, effects of pollinator distance, and applicability of the data to commercial seed production situations. The buoyancy of the pollen grain and the distances that viable pollen can travel will

S . S . VIRMANI AND IAN B. EDWARDS

190

influence the isolation requirements for male-sterile increase and hybrid production blocks.

D. EFFEC~ OF

PLANT HEIGHTAND OTHERMORPHOLOGICAL TRAITS

Because wind velocity will affect the rate at which pollen descends following dispersal from an extruded anther, a number of wheat researchers have suggested that there would be a positive effect on seed set if the male-sterile parent were shorter than the pollen donor (Lelley, 1966; Lein, 1967; Zeven, 1969; De Vries, 1972; Fisher, 1977). De Vries (1972) showed that the pollen concentration 20 cm above the spike level was considerably less than at the spike level, which in turn was less than that measured 20 cm below the spike level. Lelley (1966) showed that wheat pollen under calm conditions falls from a height of 1 m at a rate of 60 cm/sec. Fisher (1977) found a significant decrease in seed set when a tall CMS line was matched with a semidwarf pollinator. These studies suggest the need for a taller pollinator. However, it is possible that insufficient attention was paid to the pollen-shedding ability of short-statured wheats in some of these studies. Several workers have reported lower anther extrusion among semidwarf lines compared with conventional lines, and this factor could be of greater significance than height effects per se. Semidwarf restorer lines with good anther extrusion are-known to produce excellent seed set (more than 75%) on taller females (I. B. Edwards, personal observation). Chinese researchers have used rice restorer lines that are 10-20 cm taller than the CMS line. Use of a “recessive tall” gene in rice (Rutger and Camahan, 1981) should also enable the use of even taller R lines with semidwarf CMS lines to develop semidwarf hybrids. Other traits that might hinder cross-pollination in rice include long and broad flag leaves and incomplete panicle exsertion. Consequently, selection of CMS and restorer lines with short, narrow flag leaves and good panicle exsertion would be advantageous. Parental lines possessing high tillering capacity and/or larger numbers of spikelets per panicle would also enhance cross-pollination potential. Although Sections VI1,A-D suggest that much has been learned about the external and internal floral features that control access to wind-borne pollen, we have yet to establish a floral trait in the female that is as important for crossfertilization as is anther extrusion in the pollinator. That some females set more seed through cross-pollination than others becomes evident during male-sterile increases. The reasons for these differences are not clearly understood at present, but there is significant economic significance to improving our understanding of those floral traits in the female that enhance cross-fertilization and in learning how to manipulate them in breeding programs.

HYBRID RICE AND WHEAT

191

VIII. SEED PRODUCTION Hybrid seed production using the cytoplasmic-genetic male sterility system involves three steps: (1) multiplication of CMS (A) lines; (2) multiplication of maintainer (B) and restorer (R) lines; and (3) production of hybrid seed (A X R). Multiplication of B and R lines is done in the same manner as with conventional varieties; however, multiplication of A line and production of hybrid seed require different methods. A. MULTIPLICATION OF CYTOPLASMIC MALESTWLE AND MAINTAINER LINES

Procedures for CMS and maintainer line multiplication in rice have been described by Chinese researchers, and the methods are highly labor intensive (Ye, 1980). Four rows of A line are alternated with two rows of B line (2:l ratio), and to improve synchronization of flowering a split seeding of the maintainer is carried out at 4 and 8 days after seeding the male-sterile line. Both the A and the B lines are transplanted on the same day, and plants from the early and late B-line planting are alternated in the rows. Cross-pollination is improved by planting the rows perpendicular to the prevailing wind direction, and flag leaves of both the A and the B lines are clipped at the boot stage to aid pollen movement. This procedure requires 25 man-days/ha and has been found to increase seed yields by 42.9% compared with unclipped checks (Lin and Yuan, 1980). One or two applications of 20 ppm gibberellic acid are applied to the A and B lines after clipping to improve panicle exsertion. Pollination is improved on calm days by manual techniques designed to agitate the pollen parents. Rope pulling (Fig. 1) and shaking of the pollen parent with a pole (Fig. 2) require 5 and 15 man-days/ha, respectively. Isolation in time (21 days) or distance (100 m) from other rice crops prevents contamination by foreign pollen. Using these techniques, the average seed yields per production hectare in China are 400-900 kg of A line and 1000-1500 kg of B line. The two techniques commonly used to evaluate the most economical production scheme in wheat have involved varying the ratio of male sterile to pollinator and varying the basic drill strip width. Seed-increase methods were reviewed by Zeven (1974) and by Miller and Lucken (1976). MS/pollinator ratios (Wilson, 1968a; Schmidt et al., 1971; Rogers and Lucken, 1973; De Vries, 1974b; Miller et al., 1974; Miller and Lucken, 1976) evaluated have been 1:2, 1:1, 2:1, and 3:l. Several conclusions can be drawn from these studies. (I) The 2:l MS/pollinator ratio has usually proved the most effective when yields are expressed on a production unit basis; (2) increasing drill strip widths has resulted in decreased

192

S. S. VIRMANI AND IAN B. EDWARDS

FIG. 1. Supplementary pollination using rope-pulling method as practiced in China.

FIG. 2. Supplementarypollination using pole as practiced in China.

HYBRID RICE A N D WHEAT

193

seed set toward the center of the strip, the level of reduction being affected by genotype and environment; and (3) any environmental conditions or production practices that enhance normal wheat yields will positively affect male-sterile production yields. The effect of the alien cytoplasm on CMS wheat lines usually has been delay of seedling emergence and flowering compared with the maintainer line. Miller and Lucken (1976) evaluated the growing-degree day technique for timing planting dates between the male-sterile and maintainer lines and found seed set to be maximized by planting the maintainer a projected 44°C growing-degree days after the male-sterile line. B. HYBRIDSEEDPRODUCTION

Similar techniques are used in hybrid seed production. The selected R line serves as the pollen donor parent for the male-sterile A line, and seed harvested from the A line constitutes the commercial hybrid seed. Seed harvested from the R line is the result of self-fertilization and may be used again in hybrid production, or the surplus can be sold as commercial grain. The effects of strip width and parent ratio on pollen load and distribution are basic considerations in hybrid production. Increasing the ratio of female to pollinator decreases pollen production per unit area and also increases the distance over which pollen must travel. A change to a narrower drill strip width without altering the ratio alters only the distance pollen must travel and may result in a more homogeneous pollen distribution. Hybrid rice seed production in China involves six to eight rows of A line and one to two rows of R line. These parental lines are seeded and transplanted on different dates, depending on their growth duration, to synchronize their flowering. The R line is seeded on three or four different dates, but seedlings of different ages are transplanted the same day in the field to increase the duration of pollen availability at flowering. The other techniques of seed production, such as direction of row planting, flag leaf cutting, application of gibberellin, supplementary pollination, and isolation distance, are the same as described in the preceding section. Using these techniques, seed yields of 0.45-1.5 tonslha have been obtained in China (Lin and Yuan, 1980). In one instance, seed set on a CMS line in a hybrid seed production plot was 74% (Hunan Provincial Rice Research Institute, 1977). Average seed yield in hybrid seed production plots has increased steadily up to 2.5 tons/ha as seed producers have become more experienced. In hybrid seed production plots at the IRRI, 18-34% natural outcrossing on three cytosterile lines has been observed, resulting in a hybrid seed yield of 0.66-1.68 tons/ha (S. S. Virmani, unpublished).

194

S. S. VIRMANI AND IAN B. EDWARDS

Miller and Lucken (1976)used the restorer line R5 and four females in hybrid wheat production blocks at 1 :1, 2:1, and 3:1 MS/pollinator ratios in nine environments. Mean hybrid seed yields were 1.37,1.00,and 0.84 tons/ha, respectively, but the yields were 0.69,0.67,and 0.63 tons/ha when expressed on a production hectare basis. Narrowing the basic drill strip width from 3.1 to 1.5 m at the 1:l and 2:l ratios did not increase seed yields. Miller and Lucken concluded that adequate hybrid seed production could be accomplished with the existing drill sections (2.1-3.7 m) used on North Dakota farms. Experience gained from hybrid seed production in a number of environments has demonstrated the importance of using several sites to spread the risks and ensure a more consistent seed supply. It has now been established that high seed set can be obtained, given the right parental combinations and favorable conditions. Lucken (1982)cited data from hard red winter wheat hybrid production (Dwight Glenn, DeKalb Agresearch, Inc.) in which female (hybrid) seed yields ranged from 90 to 101% in three irrigation environments. A 3:l ratio was used, with the female strip width ranging from 21.8 to 18.9 m. As an alternative to the drill strip method of hybrid production, consideration has been given to blend production (Caroline, 1977;Mann, 1977). This method involves seeding a mixture of male-sterile and restorer seed, and Carolina (1 977) tested the effects of varying the blend proportions (70,80,and 90%) of two CMS lines (Minn 11-54-30and DeKalb 3047) in hybrid production blocks with three restorer lines (R101,R106,and R5). The percentage of hybrid seed was 81, 74, and 63% when entire blocks of 90,80, and 70% female blend were harvested. Hybrid seeds yields were significantly reduced at the 90% blend compared with the 70 and 80% blends because of inadequate pollen load. Mann (1977)investigated the possibility of using different seed sizes in the pollinator and female parents and mechanically separating the female (hybrid) seed from the R-line component in a mixture or blend. Separation was achieved with a Carter Disk Separator (Carter-Day Co., Minneapolis), provided the kernel size ranges within each component did not overlap. In practice, such a seed size differential would severely restrict the number of parental combinations that could be used. The extent of natural outcrossing on cytosterile lines in rice and wheat can be further increased by selection and/or breeding of CMS, maintainer, and restorer lines that possess desirable floral as well as agronomic characters influencing outcrossing. A recurrent selection program involving genetic male sterility (Athwal and Borlaug, 1967) should help reconstitute parental lines that possess an improved level of outcrossing. C. DISEASE PROBLEMS ASSOCIATED WITH SEED PRODUCTION

Ergot (Claviceps purpureu) is a potential problem in the two phases of hybrid wheat production that utilize a male-sterile parent, and infection can be particu-

HYBRID RICE AND WHEAT

195

larly severe in the northern spring wheat region of the United States. Most cultivars have morphological resistance, an exclusion mechanism resulting from the short time the florets are open during flowering. This form of resistance is undesirable in hybrid production where open flowering is essential in female parents, and a physiological or biochemical resistance is necessary. Schmidt et al. (197 1) attributed the increase in ergot infection to increasing divergence of blooming of the male-sterile and pollinator lines. Although others have observed this phenomenon, genotype differences are present, and Schmidt (1976) found partial resistance to ergot in the spring wheats ‘Chris’ and ‘Cajeme 71,’ intermediate susceptibility in ‘Waldron,’ ‘Bonanza,’ and ‘Kenya Farmer,’ and high susceptibility in ‘Nadadores 63. ’ Another disease, more common in hybrid wheat, is loose smut (Ustilago tritici), which is also partly the consequence of a more open flowering habit in the parents. Systemic chemicals (oxathiins) are available for loose smut control, and seed treatment is essential in some regions. Observations on early-generation breeding lines by I. B. Edwards (unpublished) would suggest that restorer lines are significantly more susceptible to loose smut than are conventional B lines. Whether this is a function of genetic background, nucleocytoplasmicinteraction with T. timopheevi, open-flowering habit, or a combination of these factors has not been clearly established. No specific disease problems have been reported in hybrid rice seed production plots in China. However, under tropical conditions where high levels of disease inoculum are present, the potential for disease problems remains high. D. SEEDQUALITY IN HYBRIDSAND THEIR INBRED LINES

A number of lines of common wheat, when converted into CMS (T. timupheevi) lines, have been found to produce shriveled kernels (Johnson el al., 1967; Rai et al., 1970; Schmidt et al., 1971). These shriveled kernels were extremely prone to preharvest sprouting (Rai and Stoskopf, 1974; Jonsson, 1976; Rai, 1979) and resulted in a serious loss of seed viability under unfavorable environmental conditions. The shriveling appears to result from a nucleocytoplasmic interaction and is associated with higher a-amylase activities (Rai, 1979). The degree of kernel shriveling was influenced by the genetic background of the B lines, the level of seed set, and environmental conditions. Unpublished data by I. B. Edwards would support the suggestion that a nucleocytoplasmic interaction causes shriveling. Certain spring wheat R lines in T. timopheevi cytoplasm have shown significant kernel shriveling and test-weight reduction in the absence of environmental stress, but other R lines have produced high test weights in the same experiment. In those affected R lines where reciprocal crosses were made into normal (T. aestivum) cytoplasm, a significant im-

196

S . S . VIRMANI AND IAN B. EDWARDS

provement in kernel pluinpness resulted. Comparativeinformation on seed quality of hybrid rice is not presently available.

IX. QUALITY OF HYBRIDS Because most rice is consumed as whole grain, there is conwm that the codring quality of commercial F, hybrids (F, grains) would be impaired because of genetic segregation for chemical characteristics of the grain. Results from China, however, indicate that there is no apparent adverse effect on the cooking quality of the F, hybrids. The table quality and swelling ratio of hybrids were intermediate between those of their parents (Lin and Yuan, 1980). The protein content in hybrid rice grains was reported as 9-1 1% (Lin and Yuan, 1980) and 10-12% (Anonymous, 1978b), which is 1-4% higher than that of conventionally bred variaties. Mohamed Sayed (1975) found that protein bodies in an intervarietal rice hybrid (‘Cherumodan’/ ‘ADT 27’) were uniformly distributed from the periphery to the interior of the grain. Protein was more concentrated in the peripheral layers of the parents, and milling and polishing caused a greater protein reduction in the parents than in the hybrid. Dzuba and Kolesnikov (1976) found additional protein complexes in F, hybrids that were not present in the parents. Heterotic effects for amylose content and alkali spreading value (Singh et al., 1977) and for protein content (Chao, 1972; Singh et al., 1977) have been reported. In general, these traits showed a nonsignificant association among themselves as well as with yield, indicating that simultaneous improvement of yield, protein, and cooking quality should be possible in F, hybrids (Singh et al., 1977). Five high-yielding rice hybrids and their parents were evaluated for physicochemical and cooked-rice characteristics at the IRRI in the Philippines. Hybrids were generally intermediate in characteristics and occasionally approached or exceeded the high parent in certain quality traits. Therefore, by complementation of quality traits in the parents, it should be possible to produce F, hybrids of desirable grain quality. The basic definition of quality in wheat will vary with the market class. Discussion of quality in this article will be confined to those traits that affect the milling and bread-making properties of hybrids in the hard red winter and spring wheat classes. Two factors of importance in maintaining quality in hybrid combination are a knowledge of the heritabilities and mode of gene action among quality traits and a knowledge of the effects of T. timophemi and other cytoplasmic sources on bread-making quality. Larrea (1966) evaluated parents and their F, and F, progeny from an eightparent diallel for nine quality traits in spring wheat. Parental performance pre-

HYBRID RICE AND WHEAT

197

dicted F, performance for milling and most baking characteristics, and only in a few cases did the hybrid exceed the best parent. The high narrow-sense heritabilities (close to 1) for test weight, loaf volume, mixogram pattern, and protein revealed the close relationship between the F, and mid-parent values for these traits. The heritability for flour yield was also high (0.72). Most GCA variances were significant, indicating the importance of additive gene action. Only crumb color had a significant SCA variance, and this trait had a low heritability. Larrea suggested that the advantage of additive gene action for quality traits was the ability to select parents and predict hybrid quality. However, the parents should be of good or at least of fair quality to be useful, and this could restrict the use of certain diverse germplasms. In contrast, Shebeski (1966) pointed out that parents of high quality may not necessarily confer these traits to their hybrids. ‘Pembina,’ a good quality spring wheat, had excellent GCA for all quality traits measured, while ‘Canthatch’ (another high quality wheat) proved a very poor combiner among the genetically diverse group of wheats examined. Many of the initial reports on the effects of T. timopheevi cytoplasm on the bread-making quality of hybrids were published in 1966 (Gilles and Sibbitt, 1966; Larrea, 1966; Rooney et al., 1966; Shebeski, 1966; Wilson and Villegas, 1966). Wilson and Villegas (1966) reported that T. timopheevi had little or no adverse effect on dough-mixing and sedimentation properties, and the quality of F, hybrids was intermediate between the quality of the parents. Rooney et al. (1969) crossed the R line ‘BA 130’ to the male-sterile (T. timopheevi) and maintainer (T. aestivurn) lines of the three winter wheats, and found that cytoplasm had no effect on several quality traits. Only in the A lines does T. timopheevi appear to have an adverse effect on quality. Schmidt et al. (1971) found dough-mixing time, mixing tolerance, and loaf volume of CMS ‘Gage’ to be impaired compared with normal Gage. However, hybrids of CMS Gage and Bison, pollinated by NBR 3547, had high protein, intermediate mixing properties, and good loaf volume. Thus, NBR 3457 may have contained genes (possibly linked with the Rf genes) that overcame the adverse effects of T. timopheevi cytoplasm on the quality of Gage. Aside from T. timopheevi studies, comparatively little information is available on the quality of fertile alloplasmic wheats with alien cytoplasms. Busch and Maan (1978) substituted the genomes of the hard red spring ( H R S ) wheats Chris and Selkirk into the cytoplasms of T. macha, T. dicoccoides, and A . squarrosa. Agronomic traits were not adversely affected, but the limited quality data indicated that loaf volume was the trait most adversely affected, followed by slightly reduced grain protein, compared with normal (T. aestivurn) cytoplasm. In contrast, Kofoid and Maan (1982) reported on a study of 14 alloplasmic lines, each with the nuclear genome of Selkirk and cytoplasm from a species of Triticum, Aegilops, or Haynuldia grown at three locations. Lines with a cytoplasm from a species of Aegilops had more protein and better loaf volume, and in general the

198

S. S. VIRMANI AND IAN B. EDWARDS

alloplasmic lines exceeded the euplasmic control for these traits. Sasaki et ul. (1979) found that four cytoplasms (T. fimopheevi, A . cuudutu, Aegilops umbellukztu, and Secule cereule) enhanced grain protein compared with the euplasmic control. However, they attributed these differences to the effects of the cytoplasm on yield-component characters. In summary, the need for using alternative cytoplasms in hybrid production has already been emphasized, and these preliminary quality studies may assume a greater significance in the future.

X. ECONOMIC CONSIDERATIONS The adoption of hybrid breeding technology in a sexually propagated crop plant implied that the farmer would be buying the F, seed every crop season, because he would not be able to use his harvest as seed to raise the next crop. Acceptance of hybrid seeds by farmers usually depends on the relative cost of hybrid seed compared with the economic gain obtained by the cultivation of the hybrid over the nonhybrid variety. The price of commercially produced hybrid seed will reflect research and development expenditures as well as the costs of production, processing, and marketing. Of these, the actual cost of seed production will be the single most important factor (Johnson and Schmidt, 1968). Key factors that will influence the cost of seed production are the percentage of seed set in the male-sterile increase and hybrid production fields, the ratio of female parent to pollinator, the seed multiplication ratio, and the proximity of the seed production operation to the areas of commercial seed use. Currently, China is the only country growing.hybrid rice. The following information was made available to S. S. Virmani during several visits: (1) the relative seed costs, in United States dollars, are $1.20 and $0.12/kg for hybrid and varietal seed, respectively. At a seeding rate of 20-25 kg/ha, the added cost of growing hybrid seed is $25.00 per ha; (2) hybrid yields average 1 ton/ha above varietal yields, and provide an added income of $120/ha (or $95 net); (3) the extra production costs for hybrid seed are $290/ha, and seed producers harvest about 0.75-1 ton/ha of hybrid seed and 1 ton/ha of restorer seed, compared with variety seed yields of 5 tons/ha. Thus, with the seed price differential of 1O:l (hybrid/variety), the hybrid seed producers have an extra income of approximately $130-430/ha. This advantage has prompted communes to specialize in hybrid seed production (L. P. Yuan, personal communication). The 20-33% yield advantage from hybrids has made hybrid rice production a profitable venture for both farmer and seedsman. The levels of standard heterosis needed to cover the added costs of hybrid seed at commercial yield levels ranging from 1 to 10 tons/ha are shown in Fig. 3. Under irrigated conditions, where yields of 5-6 tons/ha are attainable, standard

HYBRID RICE AND WHEAT 24

-0. 5

199

p I ton/ho

15

25

I

1

35

45

Increased cost hybrid seed/ha [over cost nonhybrid (US$)]

FIG.3. Percentage of standard heterosis necessary to pay the additional cost of hybrid seed at different commercial yield levels in rice. The price of paddy is assumed to be $2OO/ton (Virmani et al., 1981).

heterosis of 3.7-4.5% would cover added seed costs up to $45/ha. In rainfed situations, where yields vary from 1 to 3 tons/ha, standard heterosis of 8-22.5% would cover seed costs. Compared with rice, wheat has the problems of requiring higher seeding rates; also, its seed multiplication ratios are lower, and most of it is grown under dryland conditions where farm yields average slightly less than 1 ton/ha (35 bu/ac). Sage (1967) drew attention to the constraints of plant population and seed multiplication ratio in wheat compared with maize and sorghum. Lucken (1982) suggested that seed multiplication ratios for wheat, maize, and sunflower were 25,215, and 450, respectively, under comparable growing conditions in the Red River Valley of North Dakota and Minnesota. An important consequence of these constraints is the amount of lead time needed for initial increase of the parents, and the additional land area that would have to be set aside for hybrid seed production (approximately double) compared with conventional variety increases. Male-sterile increase, in particular, is a difficult and costly operation, necessitating intensive management and adequate isolation. Reitz (1967) suggested that a seed set below 50% would result in prohibitively high hybrid seed costs. Hayward (1975) suggested that seed set in the 50-70% range, with an MS/pollinator ratio of 2: 1-3: 1, would make hybrid wheat economically feasible. Mounting evidence suggests that with producible parents a

200

S. S. VIRMANI AND IAN B. EDWARDS

seed set in excess of 70% is attainable, and several seed companies in the United

States are using irrigation to improve the consistency of hybrid seed production. Johnson and Schmidt (1968) presented data showing the amounts of standard heterosis necessary to pay the additional costs of hybrid seed at different commercial yield levels and at different additional seed costs per hectare (comparable to the variables used in Fig. 3). In 1982, two seed companies in the United States marketed winter wheat hybrids at $l.lO/kg, compared with approximately %0.40/kg for variety seed. At a seeding rate of 65-70 kglha, an average yield of 0.95 tons/ha in the winter wheat region, and a grain market price of $150/ton, standard heterosis levels of 11-16% would cover the additional seed costs. This is closely comparable to the figures provided for dry-land rice. Important commercial production experience has been gained in the United States during the past decade, and some marked improvements in efficiency have been accomplished as a result of using more producible inbreds. Any production practice that will increase yields and reduce seed rates will improve the profitability of hybrid seed production. Several workers have suggested that seed rates of adapted hybrids could be reduced by as much as 50% without reducing yield, because of the heterotic expression of emergence, vigor, and tillering capacity in hybrids. Conflicting results have been obtained in a number of studies, which do not provide adequate justification for seed-rate reductions in hybrid wheat environments.

XI. PROBLEMS

Although hybrid rice has been developed and cultivated in China, several problems remain. First, the commercial hybrids have a longer growth duration (125-140 days) than do varieties and thus cannot be grown during the first cropping season (March-April to June-July) which is short because of the low temperature prevailing before March-April. Second, inadequate heterosis and lack of suitable restorers in japonica rice has prevented extensive use of hybrids in the japonica rice belt north of the Yellow River. Third, difficulties remain in hybrid seed set, and production is highly labor intensive. Last, the lack of multiple disease and insect resistance in the current hybrids has prevented their spread to the subtropical areas. However, researchers in China and at the IRRI report progress in shortening the maturation period, transferring Rf genes from indica to japonica varieties, improving pest and disease resistance, and selecting for traits that will improve hybrid seed set. Outside China, there are a number of factors that will affect the spread of

HYBRID RICE AND WHEAT

20 1

hybrid rice. Many scientists are not fully convinced that heterosis sufficient to pay for the extra seed cost exists, and the development of economically feasible seed production methods has yet to be resolved. The extent to which the laborintensive and time-consuming techniques of hybrid seed production developed in China can be adopted elsewhere is uncertain. In the absence of comprehensive analytical data on hybrid rice seed production and cultivation in China, many countries are skeptical about the economics of hybrid rice. Finally, many riceproducing countries do not have an efficient organizational setup to produce, certify, and market hybrid seed.

B. WHEAT

A major constraint in hybrid wheat development relates to the high seeding rate and low multiplication ratio discussed in the preceding section. These increase the labor and costs of hybrid testing and limit the number of hybrids evaluated in a single season. In maize, sorghum, and sunflower a single pollination produces enough seed to conduct yield trials at several locations, whereas a corresponding hand pollination in wheat may produce only 10-30 seeds. In consequence, inadequate (in general) combining ability studies have been conducted in hybrid wheat programs to realize the potential of the material on hand. Considerable research effort has been spent on developing an adequate and stable pollen fertility restoration system, and less selection pressure has been imposed for the agronomic performance of male parents. This factor, coupled with delays in converting desirable B lines into male steriles, has resulted in hybrids with comparatively low levels of standard heterosis because yield gains have been made through conventional breeding. A further problem is the producibility of the female parent. Most hybrid programs to date have tended to incorporate agronomically desirable varieties or lines into male steriles. Comparatively few crosses have been made with the objective of improving female producibility, and there is a growing realization that the objective of a true B-line program can differ considerably from those of a conventional variety breeding program for such traits as plant height, anther extrusion, and combining ability. Finally, a problem encountered especially in the spring wheat region is obtaining consistent seed production yields. However, considerable progress has been made in the hard red winter wheat areas. Hybrids have been marketed on a limited scale in the United States since 1974, but cropped hectarage has remained small and performance has not lived up to expectations. A number of institutions, both private and public, embarked upon hybrid research in the mid 1960s. The complexity of the system and the magnitude of the task were not adequately appreciated by some administrators, and continued funding became a problem. The small yield advantages of hybrids

202

S. S. VIRMANI AND IAN B. EDWARDS

were attributed by some to limited heterosis in wheat rather than to a lack of combining ability studies on an adequate number of parents.

XII. CONCLUSION A. CURRENT OUTLOOK

A review of rice breeding during the past decade would indicate that although conventional breeding procedures have succeeded in maintaining adequate levels of pest and disease resistance in cultivars, they have not shown significant advances beyond the yield plateau established by IRS. Yield results from numerous wheat research programs have been somewhat varied; certain countries and regional programs have reported genetic gains and others a more static situation. Both crops would undoubtedly benefit from the adoption of more efficient breeding procedures, and the exploitation of heterosis through hybrid breeding offers an important option. Significant heterosis for yield and other agronomic and physiological characteristics have been reported in the literature on rice and wheat. Rice hybrids are grown on approximately 6 million ha in China, and hybrid wheats are now being routinely produced, tested, and marketed by several hybrid wheat programs in the United States. Small quantities of hybrid wheat seed are also being marketed in several other countries, and the potential for hybrid rice production is being evaluated in India, Indonesia, the Philippines, and South Korea, and by two United States-based seed companies. Aside from the considerable breeding progress made during the past 20 years, rice and wheat researchers have played a major role in increasing our understanding of nucleocytoplasmic interactions. Several usabIe cytoplasmic male-sterile systems are available in diverse genetic backgrounds and may be used to sterilize maintainer varieties (B lines) that possess desirable agronomic characteristics, insect and disease resistance, and adaptability to local conditions. The frequency of rice maintainer lines is adequate among elite breeding lines developed by national and international programs, and both these and restorer lines have been found to possess multiple disease and insect resistance. Although a number of wheat B lines contain genes for partial fertility restoration, most are maintainers that can be easily converted into totally male-sterile lines through nuclear substitution into an alien cytoplasm. Because of the complexity of the restoration system, the frequency of R lines that provide complete fertility restoration with desirable agronomic and disease resistance traits is proportionately lower. Standard heterosis of as much as 30% for rice hybrids and 20% for wheat hybrids, has been obtained.

HYBRID RICE AND WHEAT

203

Hybrid seed production techniques have been developed for both crops, and rice seed yields of 0.45-1.5 tons/ha have been obtained in China. Hybrid wheat seed yields of 0.6-2.0 tons/ha have been obtained in the United States. Genetic variability in floral characters influencing outcrossing have been obtained in both crops, and breeders should be able to further improve the outcrossing potential of maintainer and restorer lines. Results obtained in China and at the IRRIindicate that the problems found in breeding and producing hybrid rice in China, and those foreseen elsewhere, are technically surmountable. Progress made in the United States since 1969 in developing a stable male sterility-fertility restoration system, in increasing the levels of standard heterosis, and in stabilizing and improving hybrid yields provides a sound basis for expanding the use of this technology. The adoption of F, hybrids in rice and wheat will ultimately depend on (1) the magnitude of the yield advantage obtained; (2) the cost/benefit ratio of using hybrid versus pureline seed; and (3) the efficiency of seed production, certification, and distribution agencies available in the country.

B. FUTURESTRATEGIES

Hybrid rice and wheat breeding should provide farmers with an opportunity to improve productivity, particularly in potential high-yield areas and where conventional breeding has apparently reached a yield plateau. The early vegetative vigor and stronger root system of heterotic rice hybrids should make them adapted to rainfed areas, and there is limited evidence to suggest that certain wheat hybrids may also have improved dry-land adaptability. Sorghum hybrids in India have shown greater yield advantages under drought conditions than under adequate moisture conditions. Some spring wheat hybrids appear to show a similar response, but both rice and wheat hybrids require further evaluation under stress conditions to confirm whether these advantages exist. Heterosis for protein content has not been clearly established in rice and wheat, but the identification of high-protein hybrids could further enhance the prospects for both crops. Aside from China, countries that may have prospects for hybrid rice in the near f u m e include India, Indonesia, the Philippines, and South Korea. In the United States, the prospects for hybrid rice are limited at present by the high costs of seed production and the unacceptable quality of heterotic hybrids introduced from China. There are 32 million ha of wheat grown annually in the United States, and Reitz (1967) estimated that hybrid wheat could find a place on 12 million ha. This represents a formidable challenge to the seed industry to manage a large hectarage of hybrid production fields and provides a strong incentive for increasing the scope of hybrid development. Seed companies and producers will have to face the realities of environmental effects upon seed set

204

S. S. VIRMANI AND IAN

B. EDWARDS

US. wheat crop is produced in low rainfall areas, irrigation may prove a prequisite for increasing seed set and stabilizing seed production. The approach of the breeder to stabilizing production will be to continue identifying floral features that improve pollen load and female receptivity, to establish variation for these traits, and to incorporate favorable genes into new inbred lines. The stability and adaptability of available A, B, and R rice lines should be studied in different environments for which hybrids are to be developed. To protect F, hybrids against potential genetic vulnerability to disease and insect epidemics which may be linked to the use of a certain CMS system, studies of the effects of existing cytoplasms on resistance or susceptibility are essential, and alternative sources of CMS should be sought. In rice, a suitable alternative to the W A cytosterility system needs to be discovered; in wheat, Aegilops speltoides appears to offer the best alternative to T. timophemi, and adequate Rf genes are available. A sterility-inducing factor(s) in the cytoplasms of available CMS lines should be identified by restriction endonuclease fragment analyses of organelle DNAs. This will not only help in differentiating the available CMS systems but could provide a means of screening for prospective donors for CMS and fertility restoration. Interspecific and intraspecific hybridization and/or tissue culture, protoplast culture, and fusion techniques should be examined to identify new potential sources of cytoplasmic male sterility. As an alternative to the cytoplasmic-genetic system, the search for effective chemical pollen suppressants for rice and wheat should be continued. Whether the CPS system of hybrid production will be as effective as the CMS system in the long run remains to be seen. However, the use of a CPS system in facilitating combining ability studies will be an important new ingredient in hybrid testing. There is now improved understanding of the importance of a balance between genes for pollen fertility restoration in the male parent and fertility-enhancing genes that affect the ease of restoration in the female. It may be possible to identify CMS females that exhibit partial fertility in a favorable (shallow-sterile) environment. Such lines in hybrid combination may interact with restorer genes to provide a highly fertile hybrid that is stable over a range of climatic conditions. In wheat, monosomic analyses provide a definitive means of identifying restorer genes but are time consuming and expensive. The development of a key for the identification of restorer genes (or gene groups), possibly based on a combination of differential lines and several cytoplasms, would be worthy of investigation; it would provide hybrid breeders with a more simple means of managing and deploying Rf genes in their programs. Another breeding strategy available to the hybrid wheat breeder is using the semidwarfing genes Rh? I and Rht 2 in different parents to ensure short stature. and, because much of the

HYBRID RICE AND WHEAT

205

Wilson et al. (1980) reported on the successful development of full dwarf x semidwarf hybrid wheats. They suggested that in developing semidwarf lines, some of the best genetic combinations for vigor will probably result in too much height. The use of Olsen dwarf derivatives with partial dominance for dwarfism offers an additional means of improving lodging resistance in hybrids targeted for high-potential or irrigation environments. If a dominant gene for semidwarfing is discovered in rice, tall varieties possessing wider genetic diversity can be used as parents to develop hybrids with high-yield potential (Athwal and Virmani, 1972). Improved techniques of hybrid seed production should be developed to economize seed production cost. Critical economic analyses of seed production and rice and wheat hybrid cultivation are required to determine under what situations and with what cost/benefit ratio hybrid breeding technology can be adopted in different countries. Because of the considerable differences that exist between wheat and rice, a comparative assessment of the prospects for developing hybrids is neither simple nor entirely valid. However, a few factors merit consideration: 1. Because rice is first sown in seedbeds and subsequently transplanted into production fields, it has been possible to match parents with widely different maturities. In contrast, because of environmental constraints and seed production methods, hybrid wheat breeders have had to work within clearly defined maturity groups. This can limit genetic variability and may be partially responsible for the lower (15-20% versus 20-30%) levels of standard heterosis reported in wheat versus rice. 2. Stable CMS systems in rice are derived from both intraspecific (BT and Gam type) and interspecific (WA) crosses. In wheat, CMS systems are developed by substituting the genomes of desirable genotypes into an alien cytoplasm and backcrossing to the nuclear donor parent. Although the effects of the alien cytoplasm on heterosis in wheat have not been widely studied, it seems that the effects (positive or negative) will be cross specific. 3. Fertility restoration appears to be less complexly inherited in rice than in wheat. Therefore, although Rf genes (and their associated linkage blocks) have been obtained from a number of common wheats, related genera, and species, their transfer into desirable agronomic types has been more time consuming. 4. The extent of natural cross fertilization in hybrid rice production blocks is lower at the present time than in hybrid wheat production blocks (35-45% versus 50-80%), and a considerable amount of hand labor has been used to produce hybrid rice seed. This could limit hybrid rice production to those countries where labor is abundant and inexpensive. 5 . Seeding rates of rice hybrids are lower (20-25 kg/ha) than those of wheat (60-70 kg/ha), and seed multiplication rates in hybrid production blocks are higher (approximately twice). This could offer certain advantages to the rice farmer and seed producer.

206

S . S . VIRMANI AND IAN B. EDWARDS

6. Comparative costs of hybrid versus pure-line seed suggest a 6-10 1 ratio for rice in China versus a 3 . 5 1 ratio for wheat in the United States. These price ratios will undoubtedly fluctuate but could influence farmer preference.

In addition to the factors influencing farmer preference for hybrid versus pureline seed, the capability of countries to organize production, certification, and distribution of hybrid seed will have an important bearing on industrial expansion. Progress during the next decade will determine how much hybrids in these two major cereal crops can help to increase world food production.

REFERENCES Akbar, M., and Yabuno, T. 1975. Ikushugaku Zusshi (Jpn. J . Breed.) 25, 215-220. Allan, R. E. 1973. Froc. Int. Wheat Genet. Symp. 4th. pp. 311-317. Amaya, A. A., Busch, R. H.,and Lebsock, K. L. 1972. Crop Sci. 12, 478-481. Anand, S. C., and Ben, S. M. 1971. Indian J . Gener. 31, 132-135. Anonymous. 1976. Zhon Guo Nongye Kexue (Chinese Agric. Sci.) 1, 51-65. Anonymous. 1977. Int. Rice Res. Newsl. 2, 27-28. Anonymous. 1978a. Chih Wu Hsuch Fa0 (Acta Bor. Sin.) 20, 305-313. Anonymous. 1978b. Agric. Ind. Life 40, 10. Arp, A. L. 1967. M.S. Thesis, Kansas State Univ., Manhattan. Atashi-Rang, G. 1970. Ph.D. Thesis, North Dakota State Univ., Fargo. Atashi-Rang, G., and Lucken, K. A. 1978. Crop Sci. 18, 267-272. Athwal, D. S . , and Borlaug, N. E. 1967. Indian J . Genet. 27, 136-142. Athwal, D. S., and Virmani, S. S. 1972. In “Rice Breeding,” pp. 615-620. Int. Rice Res. Inst., Los B&m, Philippines. Bahl, P. N. 1971. Ph.D. Thesis, North Dakota State Univ., Fargo. Bahl, P. N., and Maan, S. S. 1973. Crop Sci. 13, 317-320. Bailey, T. B., Jr., Qualset, C. O., and Cox, D. F. 1980. Crop Sci. 20, 339-342. Baldi, G. 1976. Ann. Accaa‘. Agric. Torino 119, 15-23. Badhan Roy, S. K., Mitra, A. K., and Mukherji, D. K. 1975. Sci. Culr. 41, 118-119. Bardier, N. G. 1960. Plant Breed. Abstr. 32, No. 4599. Bennett, M. D., and Hughes, W. G. 1972. Nature (London) 240,566-568. Benson, R. M. 1978. M.S. Thesis, North Dakota State Univ., Fargo. Ben, S. M., and Anand, S. C. 1971. Euphytica 20, 327-332. Bingham, J. 1966. Ann. Appl. B i d . 57, 365-377. Bingham, J. 1967. J . Agric. Sci. 68, 411-422. Bitzer, M. I., and Fu, S. H. 1972. Crop Sci. 12, 35-37. Bitzer, M. J., and Patterson, F. L. 1967. Crop Sci. 7, 482-484. Borghi, B., Bonali, F., and Boggini, G. 1973. Proc. Int. Wheat Genet. Symp. 4th. pp. 337-343. Bravo, J. M. 1982. M.S. Thesis, North Dakota State Univ., Fargo. Briggle, L. W.. 1963. Crop Sci. 3, 407-412. Briggle, L. W., Dawn, R. J., and Stevens, H. 1964. Crop Sci. 4, 220-223. Briggle, L. W., Cox, E. L., and Hayes, R. M. 1967. Crop Sci. 7,465-470. Brown, C. M., Weibel, R. D., and Seif, R. D. 1966. Crop Sci. 6, 382-383. Busch, R. H., and Maan, S. S. 1978. Crop Sci. 18, 864-866. Cahn, E. 1925. J . Am. SOC.Agron. 17, 591-595.

HYBRID RICE AND WHEAT

207

Carnahan, H. L., Erickson, J. R., Tseng, S. T., and Rutger, J. N. 1972. In “Rice Breeding,” pp. 603-607. Int. Rice Res. Inst., Los Baiios, Philippines. Caroline, J. J. 1977. M.S. Thesis, North Dakota State Univ., Fargo. Chang, T. M. 1971. Nung Yeh Yen Chih (J. Tuiwun Agric. Res.) 20, 9-26. Chang, T. M. 1980. Chung-huu Nung Yeh Yen (J. Agric. Res. of China) 29, 107-114. Chang, T. T., Li, C. C., and Vergara, B. S. 1%9. Euphyticu 18, 79-91. Chang, T. T., Li, C. C., and Tagumpay, 0. 1973. Bor. Bull. Acud. Sin. 14, 83-93. Chao, C. N. 1972. T’ui-wan Nung Yeh Shuung Yueh K u n (Tuiwun Agric.) 8,60-65. Chaplin, J. F., and Ford, Z. T. 1%5. Crop Sci. 5, 436-438. Chaudhary, R. C., Virmani, S. S., and Khush, G. S. 1981. Oryzu 18, 149-142. Cheng, Y. G., and Huang, Z. S. 1978. Chung-huu Nung Yeh Yen (J. Agric. Res. China) 27, 267-290. Cheng, Y. K., and Huang, C. S. 1979. Chung-huu Nung Hsueh Hui Puo (J. Agric. Assoc. China) 106, 11-22. Chopra, V. L., Jain, S. K.,and Swaminathan, M. S. 1960. Indian J . Genet. 20, 188-199. Chu, Y. E., Tsai, K. S., and Li, H. W. 1970. Bof. Bull. Acud. Sin. 11, 105-112. Clayton, E. E. 1950. J . Hered. 41, 171-175. Conde, M. F., Pring, D. R., and Levings, C. S., III 1979. J. Hered. 70, 2-4. Copeland, E. B. 1924. “Rice.” MacMillan, London. Craigmiles, J. P., Stansel, J. W., and Flinchum, W. T. 1968. Crop Sci. 8, 720-722. Cregan, P. B., and Busch, R. H. 1978. Crop Sci. 18, 247-251. Davis, M. D., and Rutger, J. N. 1976. Euphyricu 25, 587-595. De Vries, A. P. 1971. Euphyricu 20, 152-170. De Vries, A. P. 1972. Euphyricu 21, 185-203. De Vries, A. P. 1974a. Euphyricu 23, 11-l9. De Vries, A. P. 1974b. Euphyticu 23,601-622. Dhulappanavar, C. V., and Mensikai, S. W. 1967. Mysore J . Agric. Sci. 1, 117-122. Donald, C. M. 1962. J. Ausr. Inst. Agric. Sci. 28, 171-178. Dotlacil, L., and Apltaverova, M. 1978. Euphyricu 27, 353-360. D’Souza, L. 1970. 2. Pflun zen Zuechr. 63, 246-269. Dzuba, V. A., and Kolesnikov, G. P. 1976. Risu 19,24-26 (Byul. Nauch. Tekh. Inf. No. VIVII). Edwards, I. B., Thompson, W. G., and Pingree, D. W. 1980. Annu. Wheat Newsl. 26, 36-37. Erickson, J. R. 1969. Agron. Absir., p. 6. Ezrokhin, L. M., and Nettevich, E. D. 1970. Plunr Breed. Absrr. 42, 7320. Fairey, D. T., and Stoskopf, N. C. 1975. Crop Sci. 15, 29-32. Fisher, J. J. 1977. M.S. Thesis, North Dakota State Univ., Fargo. Fisher, R. A. 1975. Proc. Int. Winrer Wheur Conf. 2nd. pp. 178-1%. Fonseca, S., and Patterson, F. L. 1968. Crop Sci. 8, 85-88. Franckowiak, J. D., Maan, S. S., and Williams, N. D. 1976. Crop Sci. 16, 725-728. Freeman, G.F. 1919. Generics 4, 1-9. Fukasawa, H. 1953. Cyrologiu 18, 167-175. Fukasawa, H. 1955. Cytologiu 20, 211-217. Fukasawa, H. 1957. Cyrologiu 22, 30-39. Fukasawa, H. 1958. Wheurlnf. S e n . 7, 21. Fukasawa, H. 1959. Jpn. J . Bor. 17, 55-91. Gao, M. W. 1981. Acra Genet. Sin. 8, 66-74. Ghiasi, H . , and Lucken, K. A. 1982a. Crop Sci. 22, 527-531. Ghiasi, H., and Lucken, K. A. 1982b. Euphyricu 31, 253-259. Gilles, K. A., and Sibbitt, L. D. 1966. Cereal Sci. Today 11, 510-516. Gilmore, E. C., Porter, K. B., and Tuleen, N. A. 1977. Annu. Wheat Newsl. 23, 112-113.

208

S. S. VIRMANI AND

IAN B. EDWARDS

G i ,E. C., Porter, K. B., and Tuleen. N. A. 1978. Annu. Wheut Newsl. 24, 130-131. GMnaa, A. S. 1973. W.D.Thesis, North Dakota State Univ.. Fargo. Gomaa, A. S., and Lucken, K. A. 1973. Proc. Int. Wheut Genet. Symp. 4th, pp. 345-350. Goujon, C., and Ingold, M. 1%7. C.R. Hebd. Seunces Acud. Sci. Ser. D 264, 334-336. Grist, D. H. 1953. “Rice.” Longmans Green, London. Grogan, C. O., and Sarvella, P. 1964. Crop Sci. 4,567-570. Grogan, C. O., Francis, C. A., and Sarvella, P. A. 1971. Crop Sci. 11, 295-297. Gyawali, K. K., Qualset, C. O., and Yamazaki,W. T. lW8.Crop Sci. 8,322-324. Haque, M. M., Faridi, M.N.I., Razzaque, C. A., and Newaz, M. A. 1981 .Indim J. Agric. Sci. 51, 711-714. Hayward, C. F. 1975. Proc. Int. Winter Wheut Conj. 2nd. pp. 84-104. Hector, S. P. 1913. Mem. Dep. Agric. Indiu Bot. Ser. 6, 1-10. Heu, M. H., and Chae, Y. A. 1970. Res. Rep. Rurul Dev. (Crop) Koreu 13, 9-13. Heyne, E. G., and Smith,G.S. 1967. I n “Wheat and Wheat Improvement” (K. S. Quisenberry and L. P. Reitz, eds.), pp. 269-306. Amer. Soc. Agron., Madison, Wisconsin. Holland, R. F., and Roberts, T. H. 1966. Publ. DeKalb Agric. Assoc. Hori, T., and Tsunewaki, K. 1967. Jpn. J. Breed. 19, 19-24. Huang, C. H. 1970. JCRR Plant Ind. Ser. 29, 1-5. H u g , C. S., and Huang, H. H. 1978. Publ. Agric. Res. Inst. Taipei, pp. 21-255. Hughes, W. G., and Bodden, J. J. 1977. Theor. Appl. Genet. 50, 129-135. Hughes, W. G., and Bodden, J. J. 1978. Theor. Appl, Genet. 53, 219-228. Hughes, W. G., Bennett, M. D., Bodden, J. J., and Galanopoulou, S. 1974. Ann. Appl. Biol. 76, 243-252. Hunan Agricultural Bureau and Revolutionary Committee of Yan-Pai-du Commune. 1977. Actu Bot. Sin. 19, 47-52. Hunan Agricultural College, Department of Chemistry. 1977. Acta Bot. Sin. 19, 226-236. Hunan Provincial Rice Research Institue. 1977. utongguo Nongye Kexue (ChineseAgric. Sci.) 1, 21-26. Imrie, B. C. 1966. Aust. J. Exp. Agric. Anim. Husb. 6, 175-178. Ingold, M. 1974. Extruit Rech. Agron. Suisse 13, 235-243. International Rice Research Institute. 1978. “Annual Report 1977.” Int. Rice Res. Inst., Los Baf~os, Philippines. International Rice Research Institute. 1983. “Annual Report 1981.” Int. Rice Res. Inst., Los BafIos, wilippines. Jan, C. C., and Rowell, P. L. 1981. Euphytica 30, 501-504. Jan, C. C., Qualet, C. O., and Vogt, H. E. 1974. Euphytica 23, 78-85. Jennings, P. R. 1967. Int. Rice Comm. Newsl. 16, 24-26. Jensen, N. F. 1968. Annu. Wheat Newsl. 15, 26-28. Johnson, J. W., and Patterson, F. L. 1973. Crop Sci. 13, 92-95. Johnson, R. R., and Brown, C. M. 1976. Crop Sci. 16, 384-387. Johnson, V. A. 1977. Farm Seed Conf. Am. Seed Trade Assoc. 23rd. pp. 27-34. Johnson, V. A. 1978. Cereal Foods World 23, 84-86. Johnson, V. A., and Schmidt, I. W. 1968. Adv. Agron. 20, 199-223. Johnson, V. A., Schmidt, J. W., and Mattern, P. J. 1967. Q w l . Plant. Muter. Veg. 14, 193-21 1 . Jones, J. W. 1926. J. Am. SOC.Agron. 18,423-428. Jonsson, J. 0. 1976. Cereal Res. Commun. 4, 115-119. Joppa, L. R., and McNeal, F. H. 1969. Crop Sci. 9, 839-840. Joppa, L. R., McNeal, F. H., and Berg, M. A. 1968. Crop Sci. 8, 487-490. Jost, M. 1979. Int. Wheut Restorer Germphm Screening Nur. 2nd. pp. 1-64.

HYBRID RICE AND WHEAT

209

Jost, M. 1980. Inr. Wheat Restorer Germplasm Screening Nur. 3rd, pp. 1-67. Jost, M. 1981. Inr. Wheat Restorer Germplasm Screening Nur. 4th, pp. 1-41. Jost, M., and Glatki-Jost, M. 1976. I n “Hybrid Wheat Research in Yugoslavia,” pp. 155-163. Univ. of Zagreb, Zagreb, Yugoslavia. Jost, M., and Hayward, C. F. 1980. Theor. Appl. Genet. 57, 177-180. Jost, M., and Jost, M. 1980. Proc. Inr. Wheat Conf. 3rd. pp. 122-136. Jost, M., and Milohnic, J. 1975. Proc. Int. Winter Wheat Conf. 2nd, pp. 153-162. Jost, M., and Milohnic, J. 1976a. I n “Hybrid Wheat Research in Yugoslavia,” pp. 73-77. Univ. of Zagreb, Zagreb, Yugoslavia. Jost, M., and Milohnic, J. 1976b. I n “Hybrid Wheat Research in Yugoslavia,” pp. 93-98. Univ. of Zagreb, Zagreb, Yugoslavia. Jost, M., Glatki-Jost, M., and Milohnic, J. 1976a. I n “Hybrid Wheat Research in Yugoslavia,” pp. 113-121. University of Zagreb, Zagreb, Yugoslavia. Jost, M., Glatki-Jost, M., Hrust, V., and Milohnic, J. 1976b. In “Hybrid Wheat Research in Yugoslavia,” pp. 39-57. Univ. of Zagreb, Zagreb, Yugoslavia. Kadam, B. S. 1933. IndiunJ. Agric. Sci. 3, 577-588. Karunakaran, K. 1968. Agric. Res. J . Kerufu 6 , 9-14. Katsuo, K., and Mizushima, U. 1958. Ikushugaku Zasshi (Jpn. J . Breed.) 8, 1-5. Keydel, F. 1969. Vorg. F . Pflonzenzuchter. 12, 123-142. Khaleque, M. A., Jorder, 0. T.,and Eunus, A. M. 1977. Bangladesh J. Agric. Sci. 4, 137-145. Khan, M. N. A., Heyne, E. G., and Arp, A. L. 1973. Crop Sci. 13, 223-226. Kherde, M. K., Atkins, I. M., Merkle, 0. G., and Porter, K. B. 1967. Crop Sci. 7, 389-394. Kihara, H. 1951. Cyrologiu 16, 117-193. Kihara, H. 1958. Proc. Inr. Congr. Genet. 106th 1, 142-171. Kihara, H. 1963a. Seiken Jiho 15, 13-23. Kihara, H. 1963b. Proc. Inr. Congr. Genet. 11th 1, 231. Kihara, H. 1970. Seiken Jiho 22, 107-1 11. Kihara, H., and Tsunewaki, K. 1961. Seiken Jiho 12, 1-10. Kihara, H., and Tsunewaki, K. 1964. Seiken Jiho 16, 1-14. Kihara, H., and Tsunewaki, K. 1966. Seiken Jiho 18, 55-63. Kimber, G. 1973. Proc. Inr. Wheat Gener. Symp. 4rh, pp. 81-85. Kimber, G., and Athwal, R. S. 1972. Proc. Narl. Acud. Sci. USA 69, 912-915. Kinoshita, T., Mori, K., and Takahashi, M. 1980. J . Fuc. Agric. Hokkuido Univ. 60, 23-41. Kitamura, E. 1962a. Ikushuguki Zusshi (Jpn. J . Breed.) 12, 81-84. Kitamura, E. 1962b. lkushugaku Zusshi (Jpn. J . Breed.) 12, 166-168. Kofoid, K. D., and Maan, S. S. 1982. Crop Sci. 22, 725-729. Komaki, M. K.,and Tsunewaki, K. 1981 . Euphyricu 30, 45-53. Kronstad, W. E., and Foote, W. H. 1964. Crop Sci. 4, 616-619. Krupnov, V. A. 1971. Genetika (Moscow)7, 159-174. Lama, P. E. 1966. M.S. Thesis, North Dakota State Univ., Fargo. Law, J., and Stoskopf, N. C. 1973. Can. J . Plant Sci. 53, 765-766. Leighty, C. E., and Sando, W.J. 1924. J . Agric. Res. (Wushingfon, D.C.) 27, 231-244. k i n , A. 1%7. Vorrr. Pfinzen. Zucht. 11, 97-105. Lelley, J. 1966. Zuechrer 36, 314-317. Lenz, M. C., and Atkins, R. E. 1981. Crop Sci. 21, 946-950. Levings, C. S., 111, and €‘ring, D. R. 1977. J. Hered. 68, 350-354. Li, C. C. 1975. Chung-huu Nung Hsueh Hui Puo (J. Agric. Assoc. China) 92, 41-56. Li, J., and Liu, Y-N. 1983. Theor. Appl. Gener. 64, 231-238. Li, P. H. 1977. Acru Bot. Sin. 19, 7-10.

210

S . S. VIRMANI AND IAN B. EDWARDS

Lin, S. C. 1977. Int. Rice Res. Newsl. 2, 27-28. Lin, S. C., and Yuan, L. P. 1980. In “Innovative Approaches to Rice Breeding,” pp. 35-51. Int. Rice Res. Inst., Los Bairos, Philippines. Livers, R. W., and Heyne, E. G. 1966. Agron. Absrr. 58, 11. Livers, R. W., and Heyne, E. G. 1968. Proc. Inr. Wheat Genet. Symp. 3rd. pp. 431-436. Lu, H. R., Yang, Rf. C., and Liu, K. M. 1981. Fujien Agric. Coll. Sci. Res. Bull. No. 79062. Lucken, K. A. 1973. Proc. Int. Wheat Genet. Symp. 4th. pp. 361-366. Lucken, K. A. 1982. Am. Soc. Agron. Publ. Wheat Yields (in press). Lucken, K. A., and Maan,S . S. 1967. Agron. Abstr. 14. Lyakhovkin, A. G., and Singil’Din, G. A. 1975. Selek. Semenovodmo 6, 28-29. Maan, S. S . 1973a. Proc. Symp. Gener. Breed. Durum Wheat Bari, Italy, pp. 117-138. Maan, S . S . 1973b. Proc. Inr. Wheat Gener. Symp. 4th. pp. 367-373. Maan, S. S. 1975. In “Prairie: A Multiple View” (M. K. Wali, ed.),pp. 255-281. UNv. of North Dakota Press, Grand Forks. Maan, S. S., and Lucken, K. A. 1967. Wheat Inf. Ser. Kyoto Univ. 23, 6-9. Maan, S. S., and Lucken, K. A. 1970. Euphyrica 19, 498-508. Maan, S. S . , and Lucken, K. A. 1971. J. Hered. 62, 149-152. Maan, S. S., and Lucken, K. A. 1972. Crop Sci. 12, 360-364. Mallick, E. H., Ghosh, H. N., and Bairagi, P. 1978. Indian J . Agric. Sci. 48, 384-386. Mann, H. A. 1977. M.S. Thesis, North Dakota State Univ., Fargo. Matsuo, T. 1952. Bull. Narl. Insr. Agric. Sci. (Japan), Ser. D . No. 3. 1-111. (In Japanese). Maurya, D. M., and Singh, D. P. 1977. Indian J . Agric. Sci. 47,65-70. Maurya, D. M., and Singh, D. P. 1978. Indian J . Genet. Plant Breed. 38, 71-76. McCuisition, W. L. 1968. Ph.D. Thesis, Oklahomn Stare Univ., Stillwater. McDonuld, D. J . , Gilmore, E. C.,and Sransel, J . W . 1971. Agron. Abstr., pp. 11-12. McDonald, D. J.. Stanse.1, J. W., and Gilmore, C. 1974. Proc. Gen. Congr. SABRA0 2nd. pp. 1067- 1373. McNeal, F. H., and Ziegler, D. M. 1975. Cereal Res. Commun. 3, 199-203. Meletti, P. 1961. Planr Breed. Absrr. 33, No. 2%1. Mihaljev, I. 1972. Conremp. Agric. 20 (Suppl.), 1-45. Mihaljev, I. 1976a. In “Hybrid Wheat Research in Yugoslavia,” pp. 59-66. Univ. of Zagreb, Zagreb, Yugoslavia. Mihaljev, I. 1976b. In “Hybrid Wheat Research in Yugoslavia,” pp. 67-72. Univ. of Zagreb, Zagreb, Yugoslavia. Mihaljev, I. 1980. Proc. Int. Wheat Con$ 3rd. pp. 102-109. Miller, J . F. 1970. M.S. Thesis, Univ. of Nebraska, Lincoln. Miller, J. F., and Lucken, K. A. 1976. Crop Sci. 16, 217-221. Miller, J. F., and Lucken, K. A. 1977. Euphyrica 26, 103-112. Miller, J. F., Rogers, K. J., and Lucken, K. A. 1974. Crop Sci. 14, 702-705. Milohnic, J., and Jost, M. 1970. Acra Agron. Acad. Sci. Hungaricae 19, 17-23. Milohnic, J., and Jost, M. 1974. Proc. Congr. Eucarpia 7th Miri, R. K., Amawate, J. S., and Jain, fi. K. 1970. Indian J . Genet. 30, 383-394. Mohamed Sayed, A. A. 1975. Agric. Res. J . Kerala 12, 120-123. Mohanty, H. K., and Mohapatra, K. C. 1973. Indian J . Genet. 33, 264-270. Monteagudo, A., Sanchez-Monge, E., and Lacadena, J. R. 1967. Bol. Inst. Nac. Invest. Agron. (Spain) 27, No. 56, 79-81. Mukai, Y., andTsunewaki, K. 1979. Theor. Appl. Genet. 54, 153-160. Mukai, Y., and Tsunewaki, K. 1980. In “Genetic Diversity of the Cytoplasm in Triricum and Aegilops” (K. Tsunewaki, ed.), pp. 237-250. Jpn. Soc. Remotion of Sci., Tokyo.

HYBRID RICE AND WHEAT

21 1

Murayama, S., Omura, T., and Miyazata, K. 1974. Ikurhugaku Zusshi (Jpn. J . Breed.) 24, 287-290. Nagao, S. 1951. Adv. Genet. 4, 181-212. Namboodiri, K. N. 1963. Rice News Teller 11, 92-96. Nanda, K. K., and Chinoy, J. J. 1945. Curr. Sci. 14, 241. Nettevich, E. D. 1968. Euphyrica 17, 54-62. Nettevich, E. D., and Naumov, A. A. 1970. Plant Breed. Abstr. 43, No. 4865. Oehler, E., and Ingold, M. 1966. Wheat Inf. Sew.Kyoto Univ. 22, 1-13. Oka, H. I. 1953. lkushugaku Zarshi (Jpn. J . Breed.) 3, 31-39. Oka, H. I. 1954. Ikushugaku Zusshi (Jpn. J . Breed.) 3, 1-6. Oka, H. I. 1963. In “Genetics and Cytogenetics,” pp. 158-174. Elsevier, New York. Oka, H. I., and Morishima, H. 1967. Evolution 21, 249-258. O’Toole, J. C., and Soemartono. 1981. Euphytica 30, 283-290. Paramsivan, K. S. 1975. Madras Agric. J . 62, 456-457. Parmar, K. S. 1974. Ph.D. Thesis, Sardar Patel Univ. Vallabh Uidyanayar, Gujarat, India. Parmar, K. S., Siddiq, E. A., and Swaminathan, M. S. 1979a. Indian J . Genet. 39, 529-541. Parmar, K. S., Siddiq, E. A,, and Swaminathan, M. S. 1979b. Indian J. Genet. 39, 551-559. Parmar, K. S., Siddiq, E. A., and Swaminathan, M. S. 1979c. Indian J . Genet. 39, 542-550. Parmar, K. S., Siddiq, E. A,, and Swaminathan, M. S. 1981. Z. PJlanzenzuechr. 86, 1-10. Percival, J. 1921. “The Wheat Plant.” Duckworth, London. Perez, A. T., Chang, T. T., Beachell, M. H., Vergara, B. S., and Marciano, A. 1973. SAERAO Newsl. 5, 133-139. Pillai, M. S. 1961. Rice News Teller 9, 15-17. Popov, P., and Gotsov, K. 1%8. Genet. Plant Breed. (Sofia) 1, 19-28. Porter, K. B., and Merkle, 0. G. 1967. Agron. Absrr, p. 16. Porter, K. B., and Weise, A. F. 1961. Crop Sci. 1, 381-382. Porter, K. B., Lahr, K. A., and Merkle, 0. G. 1966. Res. Texas 13,6. Pring, D. R., and Levings, C. S., III 1978. Generics 89, 121-136. Pring, D. R., Conde, M. F., and Levings, C. S., 111 1980. Crop Sci. 20, 159-162. Pring, D. R., Conde, M. F., and Schertz, J. F. 1982. Crop Sci. 22, 414-421. Rahman, M., Patwary, A. K.,and Miah, A. J. 1981. Indian J . Agric. Sci. 51, 543-546. Rai, R. K. 1979. Proc. Int. Wheat Genet. Symp. 5th. pp. 291-305. Rai, R. K.,and Stoskopf, N. C. 1974. Z. Pflanzenzuechr. 71, 307-318. Rai, R. K., Stoskopf, N. C., and Reinbergs, E. 1970. Can. J . Plant Sci. 50, 485-491. Rajki, E., and Rajki, S. 1966. Novenyrermeles (Crop Prod.) 15, 5-20. Rajki, E.,and Rajki, S. 1968. Euphytica 17 (Suppl. I),43-48. Ranganathan, T.B., Madhava Menon, P., and Sree Rangasamy, S. R. 1973. Madras Agri. J . 60, 1134- 1138. Rao, G. M. 1965. Andhra Agric. J . 12, 1-12. Rao, A. V., Krishna, T. S., and Prasad, A. S. R. 1980. Indian J . Agric. Sci. 50, 193-197. Rasmusson, D. C., and Camel, R. Q. 1970. Crop Sci. 10, 51-54. Reitz, L. P. 1967. In “Wheat and Wheat Improvement” (K.S. Quisenbemy and L. P. Reitz, eds.), pp. 1-18. Amer. Soc. Agron., Madison, Wisconsin. Robertson, L. D., and Curtis, B. C. 1967. Crop Sci. 7, 493-495. Rodrigo, P. A. 1925. Philipp. Agric. 14, 155-171. Rogers, K. J., and Lucken, K. A 1973. N.D. Farm Res. Rooney, L. W., Merkle, 0. G., Porter, K. B., and Lahr,K. A. 1966. In “Hybrid Wheat Research in Texas,” pp. 15-19. Texas A&M Univ., College Station. Rooney, L. W., Gustafson, C. B., and Porter, K. B. 1969. Crop Sci. 9,431-435.

212

S. S. VIRMANI AND IAN B. EDWARDS

Rowell, P. L., and Miller, D. G. 1971. Crop Sci. 11,629-631. Rowell, P. L., and Miller, D. G. 1974. Crop Sci. 14, 31-34. Rutger, J. N., and Carnahan, H.L. 1981. Crop Sci. 21, 373-376. Rutger, J. N., and Shinjyo, C. 1980. I n “Innovative Approaches to Rice Breeding,” pp. 53-66. Int. Rice Res. Inst., Los Bahos,.Philippines. Sage, G. C. M. 1967. Ann. Appl. Biol. 59, 319-321. Sage, G. C. M. 1976. Adv. Agron. 28, 265-298. Sage, G. C. M., and Hobson, G. E. 1973. Euphyrica 22, 61-69. Sahadevan, P. C., and N m W , K. M. N. 1963. Proc. Indiun Acud. Sci. Sect. B 58, 176-185. Saini, S. S., and Kumar, I. 1973. Indian J . Gener. 33, 197-200. Saini, S. S., Kumar, I., and Gagneja. M. R. 1974. Euphytica 23,219-224. Sakai, K. I., and Narise, T. 1959. Annu. Rep. Natl. Inst. Genet. 10, 65. Sampath, S. 1%2. Oryza 1, 1-29. Sampath, S., and Mohanty, H.K. 1954. Curr. Sci. 23, 182-183. Saran, S., Sahu, S. P., and h a m , M. A. 1971. Indian J. Agric. Sci. 41, 515-519. Sasaki, M., Yasummuru, Y., and Nekata, N. 1979. Proc. Int. Wheut Genet. Symp. 5rh, pp. 293-298. Sasakuma, T.,and Maan, S. S. 1978. Can. J . Genet. Cyrol. 20, 389-398. Sawada, S., and Takahashi, M. 1977. J . Fuc. Agric. Hokkaido Univ. 58,497-509. Schmidt, H.J. 1976. M.S. Thesis, North Dakota State Univ., Fargo. Schmidt, J. W., Johnson, V. A., and Maan, S. S. 1962. Nebr. Exp. Stn. Q. No. 9. Schmidt, J. W., Johnson, V. A., Moms, M. R., and Mattem, P. J. 1971. Seiken Jiho 22, 113-1 18. Shaalan, M. I., Gad, A. M., Ali, A. E., El Fawal, M. A., and Elkhishen, A. A. 1975. Egypr. J . Gener. Cytol. 4, 474. Shands, R., and Kimber, G. 1973. Proc. Inr. Wheat Genet. Symp. 4rh, pp. 101-108. Shebeski, L. H. 1966. Can. J . Genet. Cyrol. 8, 375-386. Shebeski, L. H. 1971. II$ Bull. Near Easr Cereal Improvement Prod. Proj. 8, pp. 1-7. Shen, J. H.1980. I n “Rice Improvement in China and Other Asian Countries,” pp. 9-36. Int. Rice Res. Inst., Los Bahos, Philippines. Shinjyo, C. 1969. Jpn. J. Genet. 44, 149-156. Shinjyo, C. 1970. Sci. Bull. Coll. Agric. Univ. Ryukus 17, 261-272. Shinjyo, C. 1972a. Jpn. J. Genet. 47, 237-243. Shijyo, C. 1972b. Ikushugaku Znsshi (Jpn. J . Breed.) 22, 329-333. Shinjyo, C. 1975. Sci. Bull. Coll. Agric. Univ. Ryukus 22, 1-57. Shinjyo, C., and Motomura, K. 1981. lkushugaku Zasshi (Jpn. J . Breed.) 31 (Suppl. 1). 240-241. Shinjyo, C . , and Omura,T. 1966a. Ikushuguku Zasshi (Jpn. J . Breed.) 16 (Suppl. l), 179-180. Shinjyo, C., and Omura,T. 1966b. Sci. Bull. Coll. Agric. Univ. Ryukus 22, 1-57. Shinjyo, C., Nishime, R., and Watanabe, Y. 1974. Ikushuguku Zusshi (Jpn. J . Breed.) 24 (Suppl. l), 130-131. Shinjyo, C., I s m e , Y.,and Tanaki, M. 1981. Ikushuguku Zusshi (Jpn. J . Breed.) 31 (Suppl. l), 238-239. Singh,D. P., and Nanda, J. S. 1976. Indian J . Genet. 36, 10-15. Singh, K. B., and Singh, J. K. 1971. Euphyticu 20, 586-590. Singh, N. B., Singh, H. G., and Singh, P. 1977. Indian J. Genet. 37, 347-352. Singh, R. P., Singh, R. R., Singh, S. P., and Singh, R. V. 1979. Oryza 16, 115-118. S i g h , R. P., Singh, R. R., Singh, S. P., and Singh, R. V. 1980. Orym 17, 24-27. Singh,R. S. 1977. Indian J. Genet. 37, 384-387. Sinha, S. K.,and Khanna, R. 1975. Agron. J. 27, 123-174. Sivasubramanian, S., and Madava Menon, P. 1973. Madras Agric. J . 60, 1097-1102. Sivasubramanian, S., and Menon, M. 1973. Mudrus Agric. J . 60, 1139-1144.

HYBRID RICE AND WHEAT

213

Sreekumari, A. J., Namboodki, K. M. N.,and George, M. K. 1977. Agric. Res. J . Kerala 15, 13-16. Stansel, J. W., and Craigmiles, J. P. 1966. Rice J. 69, 14-15, 46. Stoskopf, N. C., and Rai, R. K. 1972. Can. J . Plnnt Sci. 52, 387-393. St. Pierre, C. A., and Trudel, M. J. 1972. Agron. Abstr., p. 28. Swaminathan, M. S., Siddiq, E. A., and Sharma, S. D. 1972. In “Rice Breeding,” pp. 609-613. Int. Rice Res. Inst., Los Baiios, Philippines. Tahir, C. M. 1969. Whear I f . Sew. Kyoto Univ. 29, 15-17. Tahir, C. M., and Tsunewaki, K. 1969. Jpn. J. Genet. 44, 1-9. TAU, C. M., and Tsunewaki, K. 1971. Can. J . Genet. Cytol. 13, 14-19. Trees, S . C. 1975. Diss. Abstr. Inr. B 35, 3387-3388. T ~ p p C. , R. 1972. Agron. Abstr., p. 29. Trupp, C. R. 1976. Crop Sci. 16,453-456. Tsunewaki, K. 1969. Seiken J i b 21, 1-5. U.S. Department of Agriculture. 1982. Foreign Agriculture Circular No. FG 32-82. Van Breda &-Haan, JYI913. Bull. Econ. Indo-China 16, 16-28. Virmani, S. S., and Athwal, D. S . 1973. Crop Sci. 13, 66-67. Virmani, S . S., and Athwal, D. S. 1974. Crop Sci. 14, 350-353. V i a n i , S. S., and Tan, Z. H. 1982. Int. Rice Res. Newsl. 7 , 3. Vinnani, S. S., Khush, G. S., and Yang, R. C. 1980a. Inr. Rice Res. Newsl. 5, 4-5. Virmani, S. S., Khush, G. S., Bacalangco, E. H., and Yang, R. C. 1980b. Int. Rice Res. Newsl. 5, 5-6. Vinnani, S. S., Chaudhary, R. C., and Khush, G. S. 1981. Oryza 18, 67-84. Virmani, S. S., Aquino, R. C., and Khush, G. S. 1982. Theor. Appl. Genet. 63, 373-380. Walton, P.D. 1971. Crop Sci. 11, 422-424. Washington, W. J., and Maan, S. S. 1974. Crop Sci. 14, 903-905. Watanabe, Y. 1971. Oryza 8,9-16. Watanabe, Y.,Sakaguchi, S., and Kudo, M. 1968. Ikushugaku Zasshi (Jpn. J . Breed.) 18 (Suppl. 2), 77-78. Watkins, R. E., and Curtis, B. C. 1967. Crop Sci. Abstr., p. 17. Weeraratne, H. 1954. Trop. Agric. (Colombo) 110, 93-97. Wells, D. G., and Lay, C. L. 1970. Crop Sci. 10, 220-223. Welsh, J. R., and Klatt, A. R. 1971. Crop Sci. 11, 864-865. Widner, J. N., and Lebsock, K. L. 1973. Crop Sci. 13, 164-167. Wilson, J. A. 1968a. Euphytica (Suppl. l), 13-33. Wilson, J. A. 1968b. Proc. Inr. Wheat Genet. Symp. 3rd, pp. 423-430. Wilson, J. A., and Ross, W. M. 1%2. Whear lnf. S e n . 14, 29-30. Wilson, J. A., and Villegas, E. 1966. Cereal Sci. Today 11, 306-309. Wilson, J. A., Wilson, J. A., and Schmidt, H. J. 1980. Proc. In?. Whear Conf. 3rd, pp. 210-214. Wu, H.P. 1968a. Bor. Bull. Acad. Sin. 9, 1-9. Wu, H. P. 1968b. Bot. Bull. Acad. Sin. 9, 121-138. Xu, S. 1982. Sci. Agric. Sin. 2, 9-14. Yabuno, T. 1977. Euphytica 26, 451-463. Yadav, S. P.,and Murty, B. R. 1976. Indian J. Genet. 36, 184-196. Yap, T. C., and Chang, S. C. 1976. SABRA0 J . 8, 35-40. Ye, D. H. 1980. “Seed Production Techniques for Indica Type Three Line Hybrid Rice.” Agric. Publ. Soc., Beijing, China. Yen, F. S., Evans, L. E., and Larter, E. N. 1969. Can. J . Gener. Cyrol. 11, 531-546. Yeung, K. C., and Larter, E. N. 1972. Can. J . Plant Sci. 52, 569-574. Yuan, L. P. 1966. Sci. Bull. 4, 32-34.

S . S . VIRMANI AND IAN B. EDWARDS

214

Yuan,L. P. 1972. Proc. Workshop Genet. 2nd. Hainan Guangdong. Yuan,L. P. 1977. ulongguo Nongye Kexue (Chinese Agric. Sci.) 1, 27-31. Zeven, Zeven, Zeven, Zeven, Zeven,

A. C. A. C. A. C. A. C. A. C.

1967. Euphytica 16, 183-189. 1%8. Annu. WheatNewsl. 15,52. 1%9. Wheat Inf. Sew. 29, 8-9. 1970. Annu. Wheat Newsl. 17,56. 1972. Euphytica 21, 468-488.

ADVANCES IN AGRONOMY, VOL. 36

THERMODYNAMICS AND POTASSIUM EXCHANGE IN SOILS AND CLAY MINERALS Keith W. T. Goulding Soils and Plant Nutrition Department, Rothamsted Experimental Station Harpenden, Hertfordshire, United Kingdom

I. Introduction ... ....................................................... 11. The Thermodynamics of Ion-Exchange Equilibria. . . . . . . . . . . . . . . . . . . . . . . . . . . . A. Basic Equations . .. . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . .. ... . . . . . . . . . . . . B. Standard States.. . . . . . . . . . . . . . . . .. . . . . . . .. . . . . . .. . .. ... . . . . . . . . . . . C. Ionic Strength and Hysteresis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . D. Standard Free Energies, Enthalpies, and Entropies . . . . . . . . . . . . . . . . . . . . . . E. Adsorbed-Ion Activity Coefficients.. . . . . . . . . . . . . . . . , . . . . . . . , . F. Excess Functions . . . ...................... G. Incomplete Exchange ge............................. H. Ternary Exchange. .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. ... . . . . . . . . . . . . 111. Calorimetry in Ion-Exchange Studies.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . .. ... . . . .. . . . . . A. History and Techniques . . . . . . . . . . . . .. . . . . . . . . B. Standard, Integral, and Differential Enthalpies of Exchange.. . . . . . . . . . . . . . IV. Thermodynamics Applied to Potassium Exchange in Soils and Clay Minerals. . . . . A. Background.. . .. . . . . . . . . . . . . . . . . . .. . . . . .. . . . . . . . . .. . .. . . . . . . . . . .. B. Comparing Potassium with Other Cations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . C. Comparing Clay Minerals . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . D. Comparing Soils .................................................. E. Potassium Selectivity and Fixation . . . . . . . . . . . . . . . . . . . . . . . . . . . . F. Potassium Potentials. . . ..................................... V. Exchange Equilibrium and the xchange . ... .. . . . . . . . . . . VI. Summary and Conclusions . . . . ....................... VII. Appendix: List of Symbols.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References ...........................................................

.

215 217 217 221 223 223

227 227 228 228 230 233 233 236 241 243 248 253 256 258 259 260

1. INTRODUCTION In 1972, reporting on the second working session of the ninth Colloquium of the International Potash Institute (IPI) entitled “Ion Exchange System of the Soil,” Walsh said, “The use of thermodynamic functions . . . is in many cases 215

Cojyight 0 by Academic Ress, Inc. All rights of repoduaion in m y form reserved. ISBN 0-126007363

216

KEITH W.T.GOULDING

remote, as they tend to integrate a variable quantity over a range of K saturations to give a kind of ‘average’ value which may be intellectually satisfying but not always useful . . . As a final comment, it might be advanced that perhaps at least some of the work now in progress becomes too theoretical and academic, beiig far removed from what actually happens to the soil as a specific entity in the field.” These comments will no doubt find sympathy with many readers. However, in his concluding remarks at the same colloquium, Schroeder (1972) said, “In the field of potassium in the exchange system it is my opinion that the thermodynamic approach has helped to overcome the stagnation of the past two decades.” More recently, Cooke and Gething (1978) said, at the eleventh Congress of the P I , “Walsh (1972) gave a pessimistic assessment of practical progress but we have in fact advanced, if only in that empirical methods have been largely abandoned.” This article attempts to show that, as well as stimulating research on potassium exchange as suggested by Schroeder (1972), the use of thermodynamics has greatly increased our understanding of the exchange complex and potassium furation and release and therefore has had positive practical results. Results of the last 10 years are emphasized; research before 1972 has been well reviewed by Schuffelen (1972), van Blade1 (1972), and Talibudeen (1972). However, earlier papers considered to be of particular importance are discussed where relevant. When referring to the thermodynamics of potassium exchange, workers invariably mean exchange equilibria and thus equilibrium thermodynamics. The attainment of equilibrium in the laboratory is entirely possible, but in the field the exchange of potassium between soil and solution is a “dynamic equilibrium,” if it is an equilibrium at all (see Hoagland and Martin, 1933; Cooke and Gething, 1978; Cooke, 1979). Some have tried to accommodate this probem by applying nonequilibrium thermodynamics to ion exchange in soils (Ravina and de Bock, 1974), but although results from the application of such methods agreed with those obtained from equilibrium thermodynamics, the method has never been put to use. The history of the development of our understanding of ion exchange, from its discovery in agriculture by Way (1850) to the present, has been recounted by Thomas (1977). The history of the development of the mathematical equations and theoretical concepts used has been covered by Sposito (1981a,b). Only the essential elements of these approaches that specifically relate to potassium will be considered. The basic elements of the thermodynamic treatment of ion exchange will be examined and then the application of these to potassium exchange will be discussed. Thermodynamics is about equations. These will be kept to a minimum and as simple as possible, but they cannot be avoided because an understanding of them is necessary for a full appreciation of the thermodynamics of potassium exchange.

THERMODYNAMICS AND POTASSIUM EXCHANGE

217

II. THE THERMODYNAMICS OF ION-EXCHANGE EQUILlBRlA A. BASIC~ U A T I O N S

The 2:1 layer aluminosilicate minerals in soils have a permanent negative charge caused by isomorphous substitution in the lattice of AP for Si4+ and of Mg2+ for A P + . They also have a pH-dependent charge, which is negative at high pH because of the dissociation of protons from surface hydroxyl groups and positive at low pH because of the adsorption of protons by such groups. The pHdependent charge is therefore positive below and negative above the isoelectric point of the solid, and it is superimposed on, and invariably less than, the permanent negative charge. The excess negative charge is neutralized by adsorbed cations which exchange with other cations in solution when spatially accessible. The 1:l layer silicate minerals and amorphous oxides of Fe, Al, and Si usually have a pH-dependent charge that contributes to the exchange capacity of the inorganic fraction of the soil. Soil organic matter also has a pH-dependent charge, arising from the association and dissociation of protons linked with amino, phenolic, and carboxylic groups. Although it is usually only a small percentage of the soil by weight, organic matter has a negative charge in the range 2000-4000 peq/g; thus, it can make an important contribution to the cation-exchange capacity (CEC) of the soil. More complete descriptions of the nature of ion exchange and the sources of charge were given by Schuffelen (1972) and Talibudeen (1981). The exchange reaction between adsorbed cation A of valency u and solution cation B of valency v is described by the general equation +

vA-(soil),

For the Ca2 + K +

+

+ uBY+

uB-(soil),

+ vAu+

(1)

exchange, the equation becomes Ca-(soil)z

+ 2 K+

2 K-(soil)

+ CaZ+

(2)

The double arrow (=)implies that the reaction is reversible (Section 11,G). For a reversible reaction such as Eq. (1) at equilibrium, the thermodynamic equilibrium constant K is given by

K =

+I" [B-(s~il),]~[A~ [A-(soil),]"[BV 1"

(3)

+

or again, for Ca2 + K +

+

exchange,

K=

[K-(soil)12[Ca2 ] [Ca-(soil),][K l2 +

+

where the square brackets refer to activities.

(4)

KEITH W.T. GOULDING

218

Many workers have suggested empirical relationships similar to Eq. (3) in an attempt to define an equilibrium constant, because such a constant would be valuable to soil science for predicting the state of the equilibrium at different concentrations. Some of the better known exaniples are those of Ken (1928), Vanselow (1932), and Gapon (1933); all were well reviewed by Bolt (1967) and by Sposito (1981a). In a series of papers, Sposito (1977), Sposito and Mattigod (1979), Oster and Sposito (1980), and El-Prince and Sposito (1981) have shown that these empirical “constants” can be derived from thermodynamic principles. However, in practical tests none of them have been found to be truly constant over the whole of the exchange process, although some, such as Gapon’s constant, KG, have proved to be very useful in practice. They are thus better described as equilibrium or selectiviry coeficients. The true thermodynamic equilibrium constant is exactly what is required. Unfortunately, it cannot be obtained directly because, although the activities of ions in solution can be measured, those of adsorbed cations cannot. Nevertheless, the latter activities can be approximated by relating them to experimentally measurable quantities, and as Sposito (1981a,b) shows, all, of the empirical constants can be derived from Eq. (3) by choosing suitable expressions for the activities. The two most important forms of the selectivity coefficient as regards the thermodynamics of K+ exchange are those of Vanselow (1932) and Gaines and Thomas (1953). Vanselow (1932) approximated adsorbed-ion activities by mole fractions, N, and wrote for the reaction in Eq. (1)

K, =

Nf; [Au+Iv NS [Bv+IU

The Vanselow selectivity coefficient, K,, equals K only if the mixture is ideal, that is, if activities = mole fractions (Guggenheim, 1967). Ca2+-Mg2+ exchange and K -Rb exchange most nearly approximate this. When the mixture is not ideal, activities must be related to mole fractions by activity coefficients, J and thus, in the Vanselow convention, +

+

K =

NvB[Au N r A [Bv

+

+

1” 1”

where f A

= -[A1 NA

(7)

Gaines and Thomas (1953) also defined an adsorbed-ion activity coefficient, g, but using equivalent fractions, E. Thus

THERMODYNAMICS AND POTASSIUM EXCHANGE

219

(The question of the use of different conventions to define adsorbed-ion activity coefficients is discussed in full in Section II,E). Gaines and Thomas’ selectivity coefficient, K,, is thus

K, =

Eij[A” +Iv PA[BV+lU

(9)

and, by the Gaines and Thomas convention,

The Gaines and Thomas thermodynamic treatment of ion exchange, from which Eqs. (8)-(10) are taken, stimulated much research into ion exchange in both “pure” clays and soils and other exchange materials such as resins and zeolites (see Section IV,A). Therefore, when explaining the derivation of equations used, many workers have referred to the “Gaines and Thomas Method” or the “Gaines and Thomas Treatment” (e.g., Hutcheon, 1966; Deist and Talibudeen, 1967b; Laudelout et al., 1968a; Talibudeen, 1981). However, others have often referred to Argersinger et al. (1950) or to the “Argersinger Thermodynamic Approach” when making reference to the source of thermodynamicequations used (e.g., Jensen, 1973a). Argersinger et al. (1950) and Hogfeldt and co-workers (Ekedahl et al., 1950; Hogfeldt et al., 1950) first derived (independently) a set of general thermodynamic equations for ion exchange. They were based on the Vanselow (1932) convention of mole fractions. Gaines and Thomas (1953), although referring to Argersinger et al. (1950) and Ekedahl et al. (1950), made their own, thermodynamically more rigorous derivation of a set of equations based on equivalent fractions. For homovalent exchange (exchange between ions of the same valency), mole and equivalent fractions are equal and so the two approaches give the same results. For heterovalent exchange (exchange between ions of unequal valency), mole and equivalent fractions are not equal and thus neither are most of the thermodynamic parameters derived by the two methods. Any paper reporting thermodynamic data must therefore be carefully examined for the convention used, and of course direct comparison between data derived from the two conventions may not be possible (but see Section II,E). As stated previously, K, or K, do not equal K unless the mixture is ideal. However, K can be calculated from K, or K, by integrating over the whole exchange [i.e., EB = 0 to 1, as shown by Gaines and Thomas (1953)]. For an exchange as in &. (l), this gives a complex equation following Gaines and Thomas’ convention

220

KEITH W.T.GOULDING

The last term of the equation represents the change in water activity (in effect the change in water content of the soil) in going from an A-(soil) to a B-(soil). This term has been found in practice to be negligible (Gaines and Thomas, 1955; Laudelout and Thomas, 1965). The third term on the right-hand side of Eq. (11) is made zero by the choice of suitable standard states (Section II,B), or by assuming that g L = gff, which is not generally true. So there remains the simplified form of Eq. (1 1) most often used In K = (u-v)

+

L

In K, dE,

In the Vanselow convention, this becomes In K

=

I:

InK,dE,

We now have basic equations for obtaining a selectivity coefficient, Eq.( 5 ) or (9), and a thermodynamic equilibrium constant, Eq. (12) or (13), from experimental exchange equilibrium data. Other parameters are estimated as outlined in Section II,D,E, and F. Exchange isotherms are often presented in therodynamic analyses of exchange data. These are plots of the equivalent fraction of an adsorbed cation against that of the same cation in solution (Fig. 1). Their application is discussed in Section IV,A. Also, sometimes an ‘‘uncorrected selectivity coefficient” is used, called Khor KL (e.g., van Blade1 and Laudelout, 1967). This is given, again following the Gaines and Thomas convention, by

Equivaknt fraction of K+ in solution

.,

FIG.1. The exchange isotherm. A graph of the equivalent fraction of an adsorbed cation versus its equivalent fraction in solution. This example: K+-CaZ+ exchange on soil showing hysteresis. C a + K,0,K -+ Ca. After Deist and Talibudeen (1967a).

THERMODYNAMICS AND POTASSIUM EXCHANGE

22 1

where mA and mg are molarities. It thus represents a selectivity coefficient uncorrected for activities in solution. B. STANDARD STATES

To understand what these are and why they are important one must look at the definition of ion activity and of equilibrium itself. The condition for chemical equilibrium in any system is that the chemical potentials (p)of each component of the system are equal throughout the system. Thus in a cation-exchange reaction such as that given in Eiq. (l), vp[A-(soil),]

+ up[B"+] = up[B-(soil),] + vp[AU+1

(15)

But p represents an intrinsic chemical property that cannot be identified with a universal scale (such as temperature), nor accorded a reference value of zero in the absence of the substance to which it refers. It is thus necessary to adopt a conventional reference or standard state for the substance at which p is zero (Sposito, 1981b; Talibudeen, 1981). The chemical potential in its standard state is written as po,and it can be shown (see Sposito, 1981b, Chapter 2) that p = po + R T l n a

(16)

where R is the gas constant, T is the absolute temperature, and a is the activity. Thus the activity of an ion is a measure of the deviation of the chemical potential of that ion from its value in the standard state, and the activity of an ion in its standard state is 1. Therefore, before thermodynamic quantities for exchange equilibria can be calculated, standard states must be defined for each phase; their choice affects greatly the value of such quantities and their physical interpretation. The various standard states adopted for exchanger and solution phases were discussed in full by Sposito (1981b). A list of the more important ones, and the practical results of their use, is given in Table I. The only ones commonly used are those suggested by Gaines and Thomas (1953), with a slight modification for practical reasons. In practice, the standard state for adsorbed cations is taken as being a homoionic exchanger in equilibrium with a solution of the saturating cation at constant ionic strength. The experimental results can be obtained at several ionic strengths and extrapolated back to zero, the standard state specified by Gaines and Thomas, as suggested by van Blade1 and Laudelout (1967) (but see Section 11,C). However, it appears that the values of activities in exchange reactions on soils and clays depend very little on concentration (Jensen, 1973a; Jensen and Babcock, 1973); this is a fortunate result, as such an extrapolation is rarely made in practice.

Table I

Some of the Standard States Used in Calculating the Thermodynamic Parameters of Cation-Exchange Equilibria ~~

~~

Standard states Adsorbed phase

Solution phase

Implications

Reference

Activity = mole fraction when the Activity = molarity as concentration Can calculatef,Kv, etc., but all depend on Argersinger er al. (1950) latter = 1 +o ionic strength Homoionic exchanger in equilibrium Activity = molarity as concentration AG' expresses relative affiiity of exchanger Gaines and Thomas (1953) with an infinitely dilute solution +0 for cations of the ion Activity = mole fraction when the Activity = molarity as concentration AGO expresses relative affiinity of exchanger Babcock (1963) latter = 0.5. Components nor in +0 for cations when mole fraction = 0.5 equilibrium

THERMODYNAMICS AND POTASSIUM EXCHANGE

223

C. IONICSTRENGTH AND HYSTERESIS

van Blade1 and Laudelout (1967) found hysteresis of exchange isotherms during heterovalent exchange reactions involving the selectively adsorbed NH, ion (almost identical in size and hydration to K +). Hysteresis means that forward and reverse exchange isotherms are not the same, as in Fig. 1. They also found a large variation in the uncorrected selectivity coefficient, Kf, with ionic strength I and suggested that both were caused by clay aggregation at finite ionic strength. They reasoned that such aggregation would not occur at the standard state ionic strength of zero. Therefore, to avoid the problem of hysteresis and the need to calculate activity coefficients of ions in solution (y), they plotted log Kf against (2J)f (finding this empirically to be a linear relationship) and extrapolated to (2Z)l = 0 where, by definition, y = 1 and thus Kf = K,. This supported earlier theoretical work by Laudelout and Thomas (1965), who had derived an equation predicting a linear relationship between In K, and solution concentration at any one cation ratio. However, Laudelout et al. (1972) found a maximum change in In K , of only 9% in going from 0.01 M to 0.2 M ,showing that much of the variation in Kf is corrected for by calculating activity coefficients in solution. In addition, although isotherms for heterovalent exchange do often exhibit hysteresis, selectivity for the ion of higher valency, as shown by the exchange isotherm or Kf, increases continuously as ionic strength decreases. Thus, as the ionic strength approaches zero, isotherms become rectangular (i.e., become increasingly close to the x and y axes of the graph) and Kf tends to infinity (Barrer and Klinowski, 1974). Thus the log Kf versus ( U ) d relationship cannot have a finite linear slope over a large range of I, and any extrapblation to (2J)f = 0 which gives a finite value of log Kf is incorrect. It would seem much more sensible in experimental work, therefore, to calculate y values and use K, at a known ionic strength to determine ion selectivity. It is also worth noting that Barrer and Klinowski (1974) presented a method for calculating exchange isotherms (and therefore K, values) at any solution concentration when an isotherm has been experimentally measured. Thus with modem computing methods little effort is required to measure cation selectivity in a soil over a whole range of soil solution concentrations. +

D. STANDARD FREEENERGIES,ENTHALPIES,AND ENTROPIES

Many publications have examined cation selectivity during the exchange process by using selectivity coefficients and have drawn important conclusions from them (e.g., with respect to potassium, see Bolt et al., 1963; van Schouwenberg and Schuffelen, 1963; Marques, 1968). Often, however, the overall selectivity or preference of the soil for one of a pair of cations is required, perhaps for

KEITH W. T. GOLJLDING

224

comparison with other cation pairs (Section IV,A) or of soils (Section IV,B). This could be achieved through the thermodynamicequilibrium constant, which integrates selectivity over the whole exchange process, although it is usually expressed by the free energy function. The standard Gibbs free energy of exchange, AGO, is calculated from the experimentally determined thermodynamic equilibrium constant, K, using the relationship AGO = -RT In K

(17)

It is the difference in free energy between the two homoionic forms of the soil or clay at the chosen standard state. It has been stated that AGO defines the difference in the strength of binding between the soil and the two cations (Drake, 1964; Deist and Talibudeen, 1%7a), but this is incorrect. The free energy term is the sum of ion binding strength, expressed by the standard enthalpy of exchange, AH",and the degree of order of the system, expressed by the standard entropy of exchange, Af'. The relationship between these three standard functions is given by the familiar Gibbs equation

AG"

=

AH" - TAP

(18)

As well as being directly measurable by calorimetry (Section III), enthalpies can be calculated from measurementsof the thermodynamicequilibriumconstant at two temperatures, T, and T2,using the Van? Hoff equation ln(K2/K,) =

-AH(11T2 - l/Tl) R

The standard entropy of exchange is then calculated from AGO and AH'values using Eq. (18). The physical interpretation of the three parameters is discussed fully in Section IV. E. ADSORBED-ION ACTIVITY COEFFICIENTS

Absorbed-ion activity coefficients are central to the development of a set of thermodynamic equations describing cation exchange (Section 11,A). Although they cannot be measured experimentally, they can be calculated from the measured selectivity coefficient (for derivations, see Gaines and Thomas, 1953; Sposito, 1981b). For the general exchange reaction described in Eq.(l), and for Coefficients (g) defined by equivalent fractions according to the Gaines and Thomas conventions, vln g,

=

E,[ln K, - (u-v)] - I S l n K , dE,

THERMODYNAMICS AND POTASSIUM EXCHANGE

225

and

For Ca2+ + K + exchange, these become In g,

= EK(ln K,- 1) -

In K, dEK

and 2 In g, = (l-EK) (1-ln K,)

+

I

1

In K , dE,

(23)

EK

These equations have been used in the majority of papers where adsorbed-ion activity coefficients have been calculated (e.g., Hutcheon, 1966; Deist and Talibudeen, 1967a,b; Goulding and Talibudeen, 1980). The equations forf values, derived according to Vanselow ’s convention using mole fractions instead of equivalent fractions, are of necessity slightly different (see Sposito, 198lb) and have been used only by Jensen (1973a). Activity coefficients, by definition, correct the equivalent or mole fraction terms for departure from ideality (Section 11,A). They thus reflect the change in the status, or fugacity, of the ion held at exchange sites, and thus the heterogeneity in the exchange process, as is shown experimentally in Section IV,B,C, and D. Adsorbed-ion activity coefficients used in soil and clay studies have almost always been calculated according to Gaines and Thomas’ (1953) procedure, but Sposito and Mattigod (1979) and Sposito (1981a,b) have questioned this. They state that the Gaines and Thomas-type adsorbed-ion activity coefficients are not true thermodynamic activity coefficients because they are defined by equivalent fractions [Eq.@))I rather than by mole fractions as in Vanselow’s convention [Eq.(7)]. This problem has been discussed in detail by Goulding (1983). Briefly, although the absolute values of the two types of coefficients are not the same (except at Ei= 1, where gi=& = 1 by definition), plots of g j versus Ei are very similar to those of f i versus Ei, as shown in Fig. 2, and result in similar conclusions as to cation behavior during an exchange reaction. Also, as will be shown later (Section IV,B,2), the girelate to heterogeneity as shown by calorimetrically measured enthalpies of exchange and thus have a sensible physical interpretation. Sposito and Mattigod (1979) gave expressions relating gi and fi, and also K , and K,. In this article, the symbolfwill be used for adsorbed-ion activity coefficients, and all values referred to have been calculated according to the Gaines and Thomas convention.

KEITH W . T . GOULDING

226

.4-

O u)

4.0-B

L)

c

0

0.2

0.4

0.6

0.8

1.0

Fractional K saturation

FIG.2. Adsorbed ion activity coefficients, calculated according to Vanselow’s v) and Gaines and Thomas’ (g) conventions, as a function of fractional K + saturation for (A) Ca2+ + K+ exchange on Hanvell series soil, U.K. (Deist and Talibudeen, 1967a); (B) A13+ + K+ exchange on Palm Garden Soil, Tea Research Institute, Sri Lanka (Talibudeen, 1972). From Goulding (1983).

F. EXCESSFIJNCITONS

Excess functions form the “ultimate” calculation from exchange equilibrium data and have rarely been used in soil and clay studies. They account for the properties of the exchange complex in terms of the activity coefficients of both adsorbed ions and were first introduced in studies of ion exchange in zeolites (Barrer et al., 1963). As was found for adsorbed-ion activity coefficients (Section II,E), they describe exchange heterogeneity qualitatively in soils and clays (Goulding, 1980). Excess free energies (AGE), enthalpies (AHE), and entropies (AYE) are calculated at chosen cation saturations (EB)from the following equations:

UVACE= vE,RT In fA

+ u(l -E,)RT

ASE = [AHE - ACE]/T

In fB

(24)

(26)

A H E values can also be calculated from the temperature coefficient off, and fB (Talibudeen, 1971). Excess functions were used to describe NH,+-Sr2+ exchange on mont-

227

THERMODYNAMICS AND POTASSIUM EXCHANGE

morillonite by Laudelout et al. (1968b) and K -Ca2 on soils by Talibudeen (1971, 1972). +

+

and K -AP +

+

exchange

G . INCOMPLETE EXCHANGE AND MIXED EXCHANGERS

Incomplete exchange implies that in an exchange such as that described in Eq. (l), entering cations (B) cannot replace all the adsorbed cations (A). The reaction is thus not completely reversible. There are three cases in which this can occur: 1. A time-dependent hysteresis occurs between forward and reverse isotherms (i.e., a maximum in the equivalent fraction of the entering cation appears to have been reached, but it increases with time) 2. A definite maximum content of B is reached which is less than complete exchange and independent of temperature 3. A definite maximum is reached which varies with temperature. The first case cannot be analyzed by equilibrium thermodynamics, but Barrer et al. (1973) present a method for treating the second and third which will not be examined in detail because it has not been used in soil or clay exchange work. It involves separating exchange sites into those that can be occupied by A or B ions and those that can only be occupied by A ions. Selectivity coefficients and thermodynamic equilibrium constants are obtained for the two sets of sites separately. The ion-exchange complexes of soils are always mixed exchanger systems. As Sposito (1981b) says, thermodynamic systems in soil may often be treated as if they were homogeneous for the analysis of experimental data (and almost always have been in ion-exchange work). But soils are truly polyfunctional ion exchangers and really should be treated as such. Sposito shows how this can be achieved, based on work of Barrer and Klinowski (1979), again by splitting exchange sites into classes, considering each separately, and then obtaining a weighted geometric mean of the thermodynamic functions at the end. Such a treatment is very complex and has not yet been used in practice, although Munns (1976) separated K + adsorbed on volcanic ash soils into tightly and loosely bound fractions by a similar procedure. However, modem computing methods make the treatment of such mixed exchanger systems, and incomplete exchange, perfectly feasible. Thus, although not yet of great importance in relation to K + exchange, these methods may well prove more useful in the future. H. TERNARY EXCHANGE

Cation-exchange experiments in the laboratory can be restricted to binary (two-cation) exchange. In the field, however, ion exchange is rarely binary, although in many soils the real situation can be well approximated by considering

228

KEITH W.T.GOULDING

only the dominant cations (e.g., K+-Ca2+ in calcareous soils, K+-Na+ or Ca2+-Na + in saline soils, and K -A13 in acid soils). As a move toward a more realistic approximation of field conditions, attempts have been made to develop a thermodynamic treatment of ternary (three-cation) exchange. El-Prince and Babcock (1975) were the first to try this, basing their equations on a model developed by Wilson (1964) for calculating activity coefficients for three-component systems from mole fractions. It was thought then that all the constants in the model could be calculated from binary exchange data. El-Prince and Babcock (1975) calculated isotherms for Na+ -Rb+ -Cs exchange on Chambers montmorillonite and for Na -K -Cs exchange on attapulgite. These isotherms suggested that the qualitative selectivity rules that applied to binary exchange also applied to ternary exchange, in that selectivity followed the lyotropic series (Section IV,B,l). Wiedenfeld and Hossner (1978) used the same equations for Ca2+-MgZ+-Na+ exchange in saline soils, and plotted threedimensional exchange isotherms. They found that the results were “in agreement with recognized properties of the cations,” in that Ca2+ and Mg2+ were selectively adsorbed. In neither of these reports were experimental data provided to test the model, however. El-Prince et al. (1980) tested this “subregular model” of Wilson (1964) against data for NH, -Ba2 -La3 exchange on a Nevada montmorillonite and found calculated results in “reasonably good agreement with experimental data.” The model has been questioned by Chu and Sposito (1981). They calculated a set of general thermodynamic equations for ternary exchange and showed with them that the subregular model was not solely dependent on binary exchange data. One of the model constants required data from ternary exchange for its calculation, although its value was often insignificant by comparison with other terms. This perhaps explains the good agreement between calculated and experimental results found by El-Prince et al. (1980). Unfortunately, Chu and Sposito (1981) did not have enough experimental data for ternary exchange to test their set of equations. +

+

+

+

+

+

+

+

+

111. CALORIMETRY IN ION-EXCHANGE STUDIES A. HISTORY AND TECHNIQUES

The enthalpy change of a chemical reaction expresses the gain or loss of heat during the reaction. The reaction may be exothermic, in which case the change of enthalpy is negative and heat is lost to the surroundings. Alternatively it may be endothermic, in which case the enthalpy change is positive and heat is gained from the surroundings. Very few reactions have an enthalpy change of zero. The

THERMODYNAMICS AND POTASSIUM EXCHANGE

229

enthalpy change is the result of the breakage and formation of chemical bonds, and the enthalpy of a reaction is the sum of all such events in the reactants and products, including solvent molecules (i.e., hydration enthalpies). A negative enthalpy change implies stronger bonds within the products and a positive enthalpy change implies stronger bonds in the reactants. The enthalpy of ion exchange is therefore a direct measurement of binding strength and is an important quantity. Coleman (1952) was the first to use calorimetry to measure standard enthalpies of ion exchange (W). He obtained enthalpies of H -Na+ and H -K exchange on a “Volclay” bentonite and an ion-exchange resin by direct measurement, and also from enthalpies of neutralization of the H+-clay or resin by subtracting the enthalpy of neutralization of the corresponding acid and alkali. Results for these two methods agreed to within 0.5 W/mol. Enthalpies of Na+-K+, Ba2+-Ca2+, and Ca2+-K+ exchange were also measured and compared with values calculated from exchange isotherm data at two temperatures using Fq. (19) in Section I1,D. Agreement was again good, always within *lo% or k0.2 kJ/mol. Coleman’s values were somewhat uncertain because he made no attempt to assess the extent of exchange after the reaction, assuming that adding a large excess of the replacing ion would give complete exchange. [That this error was small can be inferred from the results of Maes and Cremers (1977) who obtained exchange levels of 93-94% in Ca2 -Na+ exchange using this “excess” method.] Cruickshank and Meares (1957) overcame this problem by measuring the enthalpy of the forward and reverse reactions of A-B exchange to a common equilibrium point. By ensuring that the total molality at equilibrium was the same for both experiments, and taking the algebraic difference between the enthalpies with correction for dilution, etc. (see Section III,B), they obtained the standard enthalpy of A-B exchange. Meanwhile, Calvet and Pratt (1956) designed a microcalorimeter for measuring enthalpies of various physicochemical and biological reactions. Martin and Laudelout (1963) compared three methods for determining AZf” using this type of calorimeter: +

+

+

+

1. Coleman’s (1952) method of measuring the enthalpy of neutralization of an H+-clay with an alkali, X - O H , which, after subtracting the enthalpy of neutralization, gives A Z f O for H -X exchange 2. For exchange not involving H + , the enthalpy of complete exchange is measured at several ionic strengths, I , and extrapolated to I = 0 where the measured enthalpy = AZf” by definition 3. Cruickshank and Meares’ (1957) method of measuring enthalpies of A + B and B + A exchange to a common equilibrium point and subtracting the values. +

230

KEITH W.T.GOULDING

The results from the three methods were in reasonable agreement; those between the second and third methods differed by as much as 1.2 kJ/eq, and those between the f m t method and the other two differed by as much as 3 kJ/eq.Many workers have since used the third method. Laudelout et al. (1968a) measured enthalpies of exchange for various ion pairs on Camp Berteau montmorillonite and obtained results in good agreement (10% lower) with those calculated from exchange isotherm data, as did Gast et al. (1969). Calorimetry has also been used to obtain a more detailed picture of enthalpy changes during an exchange reaction. In fact, the f m t important application of calorimetry to ion exchange was made by Barrer et al. (1963), who showed how enthalpy changes during ion-exchange reactions on zeolites, involving Li , Na+ , Cs+ , K + , Rb+ , and Ca2+, reflected different types of exchange sites. The integral and dzyerential enthalpies obtained by Barrer et al. (explained in Section III,B) were laboriously obtained by making many individual experiments during several days. The third method listed was used and the enthalpies were measured at as many as a dozen different equilibrium points. Pipetting and injecting devices have been developed which enable enthalpies at different cation saturations to be measured in one continuous experimental run. Harter and Kilcullen (1976) designed a pipetting device for adding and mixing exact amounts of solution to a Calvet microcalorimeter which makes multiple reactions possible. They tested the device on reactions between clay and organic materials. Unfortunately, in making the additions and mixes, the equipment generated amounts of frictional heat comparable with heats measured in ion-exchange experiments. Talibudeen et al. (1977) and Minter and Talibudeen (1982) developed an automated injection system for an LKB microcalorimeter and used it to measure enthalpies of K+-Ca2+ exchange on soils and clays. The technique enables a complete exchange reaction, including as many as 20 points on the isotherm, to be measured in 2-3 days; it is sensitive enough to measure heat changes as small as 0.1 ml. The technique is outlined in Section III,B and its applications are described in Section IV,A-E. +

B. STANDARD, INTEGRAL,AND DIFFERENTIAL ENTHALPIES OF EXCHANGE

The calorimetrically measured enthalpy of an ion-exchange reaction cannot be equatedper se with the standard enthalpy of exchange, M ,as obtained directly from exchange isotherm measurements at two or more temperatures. The measured enthalpy change represents the sum of all the enthalpy changes from (1) the cation exchange reaction, involving the exchange of cations and the hydration of cations and surface; (2) the solvation of the solid; (3) the dilution of the salt solutions when mixed, and (4) the mechanical injection and mixing processes in the calorimeter.

THERMODYNAMICS AND POTASSIUM EXCHANGE

23 1

The design of modem microcalorimeters, with two reaction cells connected electrically in opposition, enables these enthalpy changes (2)-(4) to be compensated experimentally (Maes et al., 1976; Talibudeen et al., 1977). We are thus left with component (l), the enthalpy of exchange, which corresponds to the experimental conditions used and not the standard state. However, the correction required to convert the experimental enthalpy to the standard enthalpy is merely the enthalpy of dilution of the two salts from their final state to infinite dilution (often referred to as the “difference in the apparent molar heat contents of the two salts”; see Laudelout et al., 1968a; Talibudeen et al., 1977). These values are available from tables, but at the concentrationsused are always within experimental error (Cruickshank and Meares, 1957; Barrer et al., 1963; Talibudeen et al., 1977). That this is a reasonable assumption is confirmed by the excellent agreement between enthalpies measured calorimetrically and those calculated from exchange isotherms (see Laudelout et al., 1968a; Maes et al., 1976). The standard enthalpy of exchange expresses the difference in binding strength between one homonionic form of an exchanger and another. Although this gross comparison is very useful, a more detailed picture of how binding strength varies at different cation ratios would be even more useful (see Section I). For K + exchange in particular, the first 5-1096 K + saturation is the part of the exchange process most important to crops, because the Kf saturation of a soil seldom exceeds 10% even after many years of treatment with K fertilizers. The measurement of enthalpies by calorimetry makes possible a detailed analysis of such regions as well as the determination of the enthalpy of the whole reaction, as shown by Talibudeen et al. (1977), Goulding and Talibudeen (1979, 1980), and Talibudeen and Goulding (1983a,b). The technique for Ca2+ --* K + exchange involves adding a small amount of KCl solution to a suspension of a known amount of the Ca2+ form of a soil or clay. The enthalpy of the resulting exchange of some of the Ca2+ ions for K + ions is measured and the procedure is repeated. The extent of exchange at each step is measured in a separate exchange isotherm experiment, and the cumulative, or integral, enthalpy change (AHx) can then be plotted against the cumulative K + saturation (x), as in Fig. 3. A detailed description of the method was given by Talibudeen et al. (1977). The reverse reaction, K + --* Ca2+, can also be followed to check variability and reversibility (as is also shown in Fig. 3). The value of AHx at x = 1 is equal to the standard enthalpy of exchange within experimental error as explained earlier. The mxversus x curve, in almost every case, takes the form of a series of linear segments separated by sharp changes in slope, a feature also noted for some ion-exchange reactions in zeolites by Barrer et al. (1963). This characteristic is very significant, suggesting sharply defined groups of homogeneous exchange sites within a heterogeneous structure. To ensure that this observation is correct, a model consisting of a series of linear segments is fitted by the least

KEITH W.T.GOULDWG

232

X

FIG. 3. The integral enthalpy of exchange (AH,)as a function of fractional K+ saturation (x) for K -Ca* + exchange on Upton montmofionite and Fithian ate.Some data points are omitted for clarity. D, Ca + K 0, K -P Ca. After Godding and Talibudeen (1980). +

squares approximation to the AHx versus x plot and compared with the best smooth curve that can be fitted (Goulding and Talibudeen, 1980). In every case the stepped straight lines give the best fit, and so there is a clear indication of heterogeneity in the exchange process. This heterogeneity is more clearly seen when the slope of the AHx versus x curve, the differential enthalpy of exchange [d(AH,)ldr], is plotted against x , as in Fig. 4.Such an analysis has been used to give new information on clay mineralogy (Section IV,C). To complete the set of differential functions, differential free energies [d(AGx)ldr]can be calculated from selectivity coefficients as shown by Clearfield and Kullberg (1974) with

Montmorillonite

0

0.2

0.6

0.4

0.8

1.0

X

FIG. 4. The differential enthalpy of exchange [d(M,)/dx)] as a function of fractional K+ saturation (x) for K+-Ca2+ exchange on Upton montmorilloniteand Fithian illite. After Goulding and Talibudeen (1980).

THERMODYNAMICS AND POTASSIUM EXCHANGE

233

Differential entropies can then be calculated using a modified form of E q . (18):

and entropy changes during an exchange reaction can thus be analyzed.

IV. THERMODYNAMICS APPLIED TO POTASSIUM EXCHANGE IN SOILS AND CLAY MINERALS A. BACKGROUND

Ion exchange was first studied systematically by Thompson (1850) and Way (1850, 1852), who examined the adsorption of ammonia by soil and the cations subsequently released. Such an interest in the subject from the aspect of plant nutrition has continued, but in the early-to-middle twentieth century, naturally occurring and synthetic zeolites, and then synthetic (organic) resinous exchangers, became the main field of study. Interest in clay minerals as cation exchangers was renewed in the 1940s and 1950s, however, when large quantities of radioactive wastes began to be produced. The fixing of such nuclides as 137Cs ,Y j r 2 ,'Wo2 ,and 64Zn2 in clay mineral deposits underground was seen as a cheap and easy way of disposing of them. Therefore much research effort was put into understanding and predicting their ion-exchange properties, and the Gaines and Thomas (1953) method (Section II,A) was a direct result of this. Thomas and co-workers in the United States (Gaines and Thomas, 1953, 1955;Faucher and Thomas, 1954; Merriam and Thomas, 1956) were responsible for most of the early applications to clay studies. Laudelout set up a research group in Belgium, again concerned primarily with clays (e.g., Martin and Laudelout, 1963; Laudelout and Thomas, 1965; Cremers and Laudelout, 1966; van Blade1 and Laudelout, 1967; Laudelout e l al., 1968a,b), and Gast and coworkers in the United States (Gast, 1968, 1969, 1972; Gast et al., 1969) concentrated on alkali metal cation selectivity. None of these groups was primarily interested in potassium or even in soils. Hutcheon (1966) first applied the Gaines and Thomas method specifically to potassium exchange, using montmorillonite as a relatively simple exchanger. Tailbudeen and co-workers in Great Britain (Deist and Talibudeen, 1967a,b; Coulter and Talibudeen, 1968; Talibudeen, 1972; Goulding and Talibudeen, 1979, 1980) used Gaines and Thomas' equations to study potassium exchange in soils and clays, and Jensen in Denmark also +

+

+

+

KEITH W.T.GOULDING

234

investigated K exchange using thermodynamic techniques (Jensen, 1972, 1973a,b, 1975; Jensen and Babcock, 1973) but based on equations developed by Argersinger et al. (1950) (see Section &A). The following parameters are derived from a thermodynamic analysis of cation exchange, presented with their physical interpretation. +

1. The exchange isotherm (Fig. 1) relates the equivalent fraction of the adsorbed cation with its equivalent fraction in solution. It can be used to indicate selectivity in an exchange process under certain conditions (see later) or to calculate selectivity coefficients. Exchange isotherms were classified by Sposito (1981b) into four common types, depending on their behavior at low values of the ordinate and abscissa (Fig. 5): (a) S type, indicative of an exchangeable ion whose relative affinity for the exchanger is not large; (b)L type, indicative of an ion with a high relative affinity for an exchanger; (c) H type, an extreme case of an L type; and (d)C type, a linear isotherm indicative of nonpreference. Isotherms for K+-Ca2+ exchange have been found to be S type (Hutcheon, 1966), L type (Jensen, 1973a), and H type (Deist and Talibudeen, 1967a), depending on temperature, concentration, and the exchanger. Isotherms can vary greatly with ionic strength (see Sposito, 1981b), hence the need for caution when interpreting them. However, an isotherm at one concentration can be used to calculate isotherms at any other concentration for the same cation pair, temperature, and exchanger (Section 11,C). 2. The (corrected) selectivity coefficient (K,) expresses the selectivity of an exchanger for a pair of cations at a certain cation ratio. It is less ambiguous than the exchange isotherm because it is virtually independent of ionic strength (Barrer and Klinowski, 1974; Sposito, 1981b). A plot of K,, or as is more commonly used, In K,, against fractional saturation gives a quantitative indication of selectivity changes during an exchange reaction.

FIG.5. The four classes of exchange isotherm. Sposito (1981b).

THERMODYNAMICS AND POTASSIUM EXCHANGE

235

3. The Gibbs free energy of exchange (AGO) expresses the overall selectivity of an exchanger at constant temperature and pressure, and independently of ionic strength. It has been called the driving force of a reaction. For A 4 B exchange, a negative AGO implies that B is the preferred or selected cation, and vice versa. 4. The enthalpy of exchange (A@) indicates the relative binding strength of the two cations and forms one part of the driving force (AGO) of a chemical reaction according to AGO = AHO - T A P

(29)

5 . The entropy of exchange (AS”)expresses the difference in degree of order of all components of the exchange between the two homoionic forms of the exchanger. It is the second part of the driving force of a reaction according to Eq. (29). Entropies have been used to assess the relative importance of solid and solution phase changes in exchange reactions (e.g., Hutcheon, 1966; Deist and Talibudeen, 1967b). 6. Adsorbed-ion activity coefficients u> reflect the fugacity of an adsorbed ion. Fugacity is the degree of freedom an ion has to leave the adsorbed state, relative to a standard state of maximum freedom of unity. Plots off versus fractional saturation therefore show how this “freedom” alters during the exchange, thus indicating exchange heterogeneity. 7. Excess thermodynamic functions also indicate’departurefrom ideality and thus heterogeneity but allow for the behavior of both cations (Talibudeen, 1971). 8. Integral and differential thermodynamic functions show how free energy, enthalpy, and entropy vary during an exchange reaction. Differential free energies are calculated directly fi’om selectivity coefficients (Section II1,B) and so give no more information than do the latter. However, directly measured differential enthalpies give a clear picture of exchange heterogeneity and provide another means of investigating surface chemistry. Differential entropies, although complex, can interpreted in terms of the structural order of all the components of the exchange system. In presenting the physical interpretation of the latter three sets of parameters, “exchange heterogeneity” has been mentioned (see also Sections II,E,F and 111,B). Heterogeneity of exchange is caused by one or more of the following: a heterogeneous distribution of ions on the exchanger; a heterogeneous distribution of exchange sites in terms of their position and energy; differences in the properties of the two cations (e.g., size, polarizability, and hydration); in soils, a heterogeneous clay mineralogy and a complex mixture of organic and inorganic exchangers. The application of thermodynamic reasoning to cation exchange in soils and clays, and of the Gaines and Thomas method in particular, is not without problems. The method assumes a constant exchange capacity and a negligible adsorp-

236

KEITH W.T. GOULDING

tion of anions from solution. The latter is usually true for cation exchange in soils and clays in dilute solutions. However, the CEC has been found to change during an exchange experiment, particularly for K+-Ca2+ exchange where some K+ may be fixed. Faucher and Thomas (1954) estimated that a change in CEC from 123 to 135 p q / g during Cs+-K+ exchange affected the isotherm by about 4% and was thus negligible. Hutcheon (1966) found a similar change in CEC during K+-Ca2+ exchange and therefore ignored it. However, Deist and Talibudeen (1967b) showed more satisfactorily that the decrease in CEC during Ca2+ --f K + exchange in their experiments on soils resulted from Ca2+ ions trapped inside the exchanger and not from an irreversible fixation of K + . They therefore suggested that if exchange sites were homogeneously distributed and some sites were physically restricted, the thermodynamic treatment was not invalidated. This view was supported experimentally when Goulding and Talibudeen (1980) obtained identical plots of In K, versus K + saturation (and thus identical AGO values) and identical AHO values for Ca2 --f K and K + Ca2 exchange on a montmorillonite, a kaolinite, and a vermiculite clay. Deist and Talibudeen (1967b) also discussed the problem of extrapolating exchange isotherms, and thus of calculating K, values, at very small saturations of the preferred ion where selectivity is high. They thought that a linear extrapolation was reasonable but could not estimate the errors involved. Sposito and Mattigod (1979) and Sposito (1981a,b) questioned the thermodynamic meaning, and thus the application, of adsorbed-ion activity coefficients and selectivity coefficients as calculated by the Gaines and Thomas method. This problem was discussed by Goulding (1983); the argument is summarized in Section II,E. Studies of the thermodynamics of potassium exchange have concentrated on either the comparison of the exchange properties of K + with those of other cations or the comparison of a series of soils or clays using potassium exchange with another cation as the reference point. The research will therefore be reviewed under these main headings, with special sections discussing potassium selectivity and furation, and potassium potentials. +

+

+

+

B. COMPARING POTASSIUM WITH OTHERCATIONS

The reviews by van Blade1 (1967, 1972) cover the early research in this area. The main aspects are the following: 1. Comparison of cation selectivities 2. Estimation of the relative effects of ionic polarizability and hydration 3. Comparison of the relative contributions of enthalpy and entropy to free energy 4. Use of exchange parameters for A-B exchange and A-C exchange to predict those for B-C exchange (Hess’s triangle rule)

237

THERMODYNAMICS AND POTASSIUM EXCHANGE

5 . Use of selectivity coefficients, adsorbed-ion activity coefficients, and ex-

cess thermodynamic functions to elucidate the behavior of the exchange complex in detail at various compositions of the exchange sites. The topic is best covered under the headings homovalent and heterovalent exchange. 1. ffomovalent Exchange Martin and Laudelout (1963) comprehensively examined the exchange of NH, with Na , K , Li ,Rb , and Cs on Camp Berteau montmorillonite. Selectivity, as expressed by In KL and AGO, was a function of ion polarizability (z /r, where z is the ion charge and r is the radius of the anhydrous cation). The order of selectivity for the monovalent alkali metal cations was that of the Hofmeister or lyotropic series Cs > Rb > NH, = K > H,O > Na > Li+ [the ionic radii of these anhydrous cations, in nanometers, are as follows: Cs+, 0.167; R b + , 0.147; NH4+, 0.143; K + , 0.133; Na + , 0.097; and L i+ , 0.068 (Weast, 1971); those of the hydrated cations are Cs+, 0.228; Rb+ ,0.228; K + , 0.232; Na+ , 0.276; and Li+ , 0.340 (Cotton and Wilkinson, 1972)l. The K + + M2+ exchange on a montmorillonite gave similar results for divalent cations (van Bladel, 1967): Ba2+ > Sr2+ > Ca2+ > Mg2+ [where the anhydrous ionic radii are Ba*+, 0.134; Sr2+, 0.112; Ca2+, 0.099; and Mg2+, 0.066 (Weast, 1971)l. It is interesting to note that the reverse order was found for Na+ -+ M2+ exchange on vermiculite by Wild and Keay (1964) because of the different characteristics of the Na+ ion and vermiculite. Deist and Talibudeen (1967a) obtained AGO and adsorbed-ion activity coefficient values for K -Na+ and K -Rb exchange on soils, which again gave the order of preference Rb+ > K+ > Na+ . The widely contrasting soils differed little in their selectivities. The mean AGO value for Na+ 3 K + exchange was -4.08 & 0.29 kJ/eq; for Rb+ -+ K + exchange it was -2.11 2 0.36 kJ/eq. By contrast the same soils exhibited AGO values for Ca2+ -+ K+ exchange ranging from -4.40 to - 14.30 kJ/eq. Adsorbed-ion activity coefficients showed that for K + -+ Na+ exchange (two ions of the same valency but different in size and thus in selectivity), fugacity for each ion increased smoothly with increasing saturation (Fig. 6). For K+-Rb+ exchange [two monovalent ions of very similar size and selectivity and thus with almost ideal behavior (Section &A)], fugacity changed very little from the standard state value of 1. Therefore, for homovalent exchange potassium selectivity is a function of its polarizability (i.e., its size only). Gast and co-workers (Gast, 1968, 1969, 1972; Gast et al., 1969) found the same selectivity series again for alkali metal cation exchange on Wyoming bentonite and Chambers montmorillonite; it was not changed by pH. They found +

+

+

+

+

+

+

+

+

+

+

+

+

+

+

+

+

KEITH W.T.GOULDING

238

1, or 0.4

-

0.2

3C

FIG. 6. Adsorbed ion activity coefficientsfK (0)and fNa (m) as a function of hctional K + saturation x for K+-Na+ exchange on soils. Deist and Talibudeen (1967a).

that selectivity was determined primarily by the enthalpy term, counterbalanced by a smaller entropy term, but that entropy changes became more important as the surface charge density of the clay increased. They concluded from these observations that selectivity was governed chiefly by electrostaticforces between hydrated cations and the surface, and assumed that all changes in entropy resulted from changes in ion hydration. Work on Na -Cs exchange on clays (Maes and Cremers, 1978) has supported the former conclusion, but not the latter. +

+

2 . Heterovalent Exchange More research effort has been expended on heterovalent exchange than on homovaknt exchange because the former is generally of greater agricultural and industrial importance. Results can be separated into mono-divalent, monotrivalent, and di-trivalent exchange. a. Mono-Divalent Exchange. From the early work, particularly of Hutcheon (1966), Deist and Talibudeen (1967a,b), and Laudelout et al. (1968a,b), and from the more recent work of Goulding and Talibudeen (1980, 1984a,b) and Talibudeen and Goulding (1983a,b), several conclusions can be drawn. 1. It is unwise to use the exchange isotherm to assign selectivity because it varies greatly with ionic strength and may show hysteresis. Deist and Talibudeen (1967b) suggested that the ionic strength effect resulted mainly from entropic forces. 2. Overall selectivity is best expressed by K or AGO, and changes in selectivity during the exchange process are best expressed by a graph of In K,, or the differential free energy, versus K+ saturation. In heterovalent exchange, AGO

239

THERMODYNAMICS AND POTASSIUM EXCHANGE

depends on ion size and valency, that is, the coulombic or charge factor. Thus AGO for Ca2 ---* Na2 exchange in soils is usually positive (Poonia and Talibudeen, 1977) and for Ca2+ ---* K + exchange it is usually, although not always, negative (Deist and Talibudeen, 1967b). Thus Goulding and Talibudeen (1984a,b) found AGO for Ca2 + K exchange on 14 soils to vary greatly (from +2.2 to - 10.0 kJ/eq) depending on mineralogy, pH, fertilizer, cropping history, and organic matter content. 3. The relative binding strength between two cations and the surface is expressed by the enthalpy of exchange (Hutcheon, 1966). Laudelout et af. (1968a) for M2+ + NH4+, and Hutcheon (1966), Deist and Talibudeen (1967b), and Goulding and Talibudeen (1979, 1980, 1984a,b) for Ca2+ + K + exchange, found that AtP was negative, implying stronger binding for K + , NH, , and, by implication, other selectively adsorbed cations such as Rb and Cs+. In addition, Goulding and Talibudeen (1984a,b) found that although this greater binding strength for K + in soils may be slightly decreased by K + fertilizer treatment, it never reverses to make Ca2+ the more strongly bound, even after 140 years of heavy FWM or inorganic K + fertilizer applications. The reaction M2+ --* K + becomes increasingly exothermic (i.e., the binding strength for K + increases) as the polarizability of M2+ decreases (van Bladel, 1967). This is because the less easy a cation is to polarize, the less easily it displaces K , and the stronger, relatively, is the K -surface bond. 4. The entropy of exchange expresses the rearrangement of cations, surfaces, and solvent molecules during the exchange process. Hutcheon (1966) and Deist and Talibudeen (1967b) found AF for Ca2+ + K + exchange on soils to be negative. They attempted a qualitative discussion of entropy changes in the solid and solution phases, and concluded that those in the solution phase must dominate, as they are large and negative for the K (solution) to Ca2 (solution) exchange. However, Goulding and Talibudeen (1980) found that Ca2+ + K + exchange on clay minerals was accompanied by a negative AF value for the expanded 2:l minerals vermiculite, illite, and montmorillonite, but by a positive AF value for the collapsed 2:l mineral muscovite mica. They therefore suggested that rearrangements within the solid phase contribute most to entropy changes. (For a full discussion, see Section IV,C.) 5 . Adsorbed-ion activity coefficients and excess functions for Ca2 + K exchange showed a marked difference in the behavior of Ca2+ and K + ions. Although the fugacity of Ca2+ decreased smoothly as K + saturation increased, that of K + varied greatly, the graph of fK versus K saturation showing maxima, minima, and inflections (Talibudeen, 1972) (see Fig. 2A). This was taken as reflecting the different distributions of Ca2+ and K + ions in the Gouy and Stem layers (Deist and Talibudeen, 1967a). The interpretation with respect to the surface is given in Section IV,D,l. It is interesting to note that the behavior of the selectively adsorbed K ion, as +

+

+

+

+

+

+

+

+

+

+

+

+

+

240

KEITH W.T.GOULDING

shown byf,, was influenced by both enthalpic and entropic forces, whereas that of the nonselectively adsorbed Caz+ ion depended only on entropic forces (Goulding and Talibudeen, 1980). This suggests the primacy of strength of binding in determining K+ selectivity and fixation (but see Section IV,E,l where the relative importances of enthalpy and entropy are discussed in full).

b. Mono-Trivalent Exchange. The only work on the thermodynamics of mono-trivalent exchange has been on K -AP exchange by Coulter (1969), Sin& and Talibudeen (1971), and Sivasubramaniam and Talibudeen (1972), which is summarized by Talibudeen (1972, 1981). Aluminium dominates the exchange complex in strongly leached acidic tropical soils. However, its ionexchange behavior is complicated by the existence of polymeric A l - O H forms at pH values above 4. For example, the precipitation of A l - O H polymers as “islands” in the interlayer space of 2:1 minerals prevents their collapse on K adsorption and thus prevents K+ fixation (Rich, 1972). Experiments at low pH values have clarified the K+-A13+ exchange process. Although exchange isotherms suggested A13 preference in all soils and clays (vermiculite, illite, and montmorillonite) examined, In K, and AGO values indicated K + preference in seven out of nine soils and all three clays (Talibudeen, 1972). It is important to note that if what Coulter (1969) called “difficultly exchangeable K ’ is ignored when calculating K, values, results sometimes suggest A13+ preference. However, this K + should be included in the calculations because it indicates strong K+ preference over the first 20-30% of A13 -K exchange because of specific (ion-size) effects. Illite, montmorillonite, and soils dominated by these minerals exhibited greatest preference for K + and less for vermiculite and chlorite. The presence of organic matter also decreased the preference of a soil for K over AP ,because organic matter chelates polyvalent, but not monovalent, cations (Talibudeen, 1981). The more strongly leached the soils (and therefore the less 2:l clay minerals present), the lower was the preference for K + . Indeed, when devoid of such minerals, soils preferred A13 , presumably because in other minerals there is no effect of ion size and valency alone controls selectivity. As in K -Ca2 exchange, fK versus K + saturation curves showed maxima and inflections, andf,, versus K + saturation curves changed smoothly (see Fig. 2B). This again shows the contrasting behavior of selectively adsorbed (K+) and nonselectively adsorbed (A13+) cations. +

+

+

+

+

+

+

+

+

+

+

c. Di-Trivalent Exchange. Only Ca2 -A13 exchange has been investigated (Coulter and Talibudeen, 1968). For vermiculite, illite, montmorillonite, and two acidic (pH 4.8) soils, exchange isotherms and In K, values showed strong A13+ preference which decreased with surface charge density for the clays, illustrating the importance of coulombic (charge or valency) effects in the exchange of such strongly hydrated ions. +

+

THERMODYNAMICS AND POTASSIUM EXCHANGE

24 1

C. COMPARING CLAYMINERALS

Potassium exchange with other cations has been used to examine and compare clay minerals (aluminosilicates), but only recently have thermodynamic methods been used for a systematic comparison of a suite of minerals. Homovalent exchange shows little difference between exchangers for any one cation pair, as was shown in Section IV,B, 1. Heterovalent exchange is therefore used, the main reference cation pair involved being K+-Ca2+ with, to a lesser extent, K+-A13+. Potassium selectivity and fixation in clays have also been of great interest, but these are considered separately in Section IV,E. Hutcheon (1966), examining K + -Ca2+ exchange on Chambers montmorillonite, related entropy changes to changes in lattice spacing and cation and surface hydration, and enthalpy changes to cation binding strength. He also related variation in the adsorbed-ion activity coefficients with changing K saturation to changes in lattice spacing. His aim was to segregate solid and solution effects, and, the behavior of ions in solution being fairly well defined, to learn more about those which occurred in the solid phase. He concluded that the overall exchange reaction was governed by a balance of interlayer cation hydration forces and attraction forces between the cations and the surface. A similar conclusion was reached by Gast (1969, 1972) and Gast et af. (1969). Although Hutcheon’s (1966) work was extremely thorough, it could be argued that he arrived at no new conclusions regarding clay properties, particularly for montmorillonite which had been extensively examined by other methods (e.g., Norrish, 1954). However, he was the first to apply thermodynamic methods specifically to K + exchange and to attempt a physical interpretation of the resulting data. He thus opened the way for a comparison of the more important clay minerals, which is of much greater interest than studies of a single clay or soil. Goulding and Talibudeen (1980) examined five aluminosilicate minerals chosen as representatives of groups of aluminosilicates commonly occurring in soils. They were a muscovite mica from Norway, Fithian illite, Montana vermiculite and Upton (Wyoming) montmorillonite from the United States, and a kaolinite from England. Free energies of exchange indicated that all of the minerals selectively adsorbed K + , selectivity decreasing in the order mica > vermiculite = illite > kaolinite > montmorillonite. Excepting kaolinite, this is approximately the same order of selectivity as that suggested by Talibudeen (1971) based on K -Ca2 exchange in soils and mineralogical data for those soils and identical to that found by Assa (1976a,b), who examined K+-Ca2+ exchange on a similar suite of minerals using Gapon’s constant. Because kaolinite is usually considered to have a very low preference for K if not actually preferring Ca2 ,the kaolinite examined by Goulding and Talibudeen (1980) had an anomalously high K + preference. A reason for this became apparent when +

+

+

+

+

242

KEITH W. T. GOULDING

-

*

0

0.2

O

0.6

0.4 X

1

0.8

0

FIG. 7. Differential enthalpy of exchange, [d(AHx)/)ldx)],as a function of fractional K+ saturation, x, for a Montana vermiculite and an English kaolinite. After Gouldmg and Talibudeen (1980).

the enthalpy values, and particularly the differential enthalpy curves, were examined. From the AH” values, the order of decreasing binding strength for K + was mica > illite = kaolinite > vermiculite > montmorillonite. Again, the kaolinite occupies an anomalously high position. Its differential enthalpy values were the same, within experimental error, as those of the Montana vermiculite (Fig. 7). Only the disribution of enthalpy values was different; there were about 160 p q / g of strong K binding sites in the vermiculite and only 19 peq/g of these sites in the kaolinite. This suggests, therefore, that the presence of about 2% by weight of a vermiculitic impurity [i.e., weathered micaceous interleaves of the type shown by Lee et al. (1975) using scanning electron microscopy] is responsible for the exchange characteristics of the kaolinite (see also Lim et al., 1980). The same effect can be seen in some data of Bansal (1982) on K+-Ni2+ exchange in a kaolinite from Bath, South Carolina (United States). The Lw” and AGO values indicated a stronger binding and selectivity for Ni2+. However, plots of K,,fK, AGE, and M Eversus Ni2+ saturation all suggested the existence of a few sites (about 20 w q / g of a total CEC of 107 peq/g) which had high selectivity for K . The differential enthalpy curves of Goulding and Talibudeen (1980) also suggested the presence of a small amount of mica in the montmorillonite, modifying its exchange properties. Subsequent research on kaolinites and montmorillonitesfrom different sources (Talibudeen and Goulding, 1983a,b) has shown that virtually no so-called montmorillonite is completely free of micaceous impurities (the < 0.2 pm fraction of an Upton montmorillonite was the only “pure” sample found), and that all of the cation-exchange properties of kaolinites can be explained by 2:l mineral impurities. Such detailed quantitative mineralogical analyses, which have hitherto been impossible using X-ray diffraction techniques, thus open up new possibilities in analysis. +

+

THERMODYNAMICS AND POTASSIUM EXCHANGE

243

Integral and differential entropy values of the five minerals studied by Goulding and Talibudeen (1980) supported to some extent Hutcheon’s (1966) view of the importance of lattice expansion and contraction, but not the dominance of solution forces over solid effects. X-ray diffraction evidence by Plancon et al. (1979) had suggested that montmorillonite surfaces rearrange and reorder before collapsing when the mineral is subjected to wetting and drylng cycles following K -Ca2 exchange. Based on this and their own data, Goulding and Talibudeen (1980) suggested three physical mechanisms that contribute to entropy changes during Ca2 +-K exchange in clays: (1) in the solid phase, replacing Ca2+ by K+ realigns the aluminosilicate layers in the 001 direction such that the hexagonal holes in adjacent sheets can accommodate K + ions [as suggested long ago by Jackson (1963)], resulting in a negative entropy change; (2) in the solid phase again, replacing Ca2+ by K + increases the randomness of distribution of exchangeable cations, resulting in a positive entropy change; (3) in the solution phase, replacing K + by Ca2+ increases the structural order of water molecules, decreasing the entropy of the system. The fact that all the expanded 2: 1 minerals examined exhibited negative entropy changes during Ca2+ + K + exchange, whereas muscovite mica, a collapsed 2: 1 mineral, exhibited a positive entropy change, suggested that the solid-phase effects (1) and (2) dominate. Adsorbed-ion activity coefficients gave little new evidence on mineral characteristics, but agreed quantitatively with differential enthalpy values as to the disposition of site groups (see Section IV,D,I). +

+

+

+

D. COMPARING Sons

1. Temperate Soils Deist and Talibudeen (1967a,b) first used a thermodynamic treatment of cation exchange to compare soils, examining K+-Na+, K+-Rb+, and K -Ca2 exchange on 10 important British arable soils. The clay content of the soils ranged from 13 to 4396, the pH ranged from 5.4 to 7.1, and the CEC (measured by M NH,Ac leaching at pH 7) ranged from 124 to 307 Feq/g. Little differentiation between soils was apparent in the homoionic exchange reactions, and the results can be summarized by saying that all soils preferred K to Na+ , and Rb and K + , in accordance with the expected order of selectivity (see Section IV,B, 1). Adsorbed-ion activity coefficients showed no heterogeneity for these cation pairs, preference for one ion over another being equally distributed over all the exchange sites. For Ca2+ + K + exchange, the characteristics were very different. Some isotherms showed hysteresis, and that of a Harwell series soil exhibited selectivity reversal. Free energy changes showed that K+ was preferred to Ca2+ on +

+

+

+

KEITH W.T. GOULDING

244

all the soils, and a comparison of enthalpy and entropy changes demonstrated that this was always because of stronger binding of K+ (negative AH"),offset by an increase in order (negative AS"). The soils differed greatly in their AGO, W , and ASO values because of differences in their clay and silt mineralogies. Graphs of adsorbed-potassium activity coefficients (fK)versus K+ saturation exhibited maxima, minima, and inflections (Fig. 2A). Later (Talibudeen, 1971, 1972) these characteristics were qualitatively related to the clay mineralogy of the soils, the assumption being that each maximum or inflection reflected a changeover of Ca2 form to K form by one group of K -selective sites after another. It was thought that such graphs might improve the description of soil mineralogies. Excess functions (Section I1,F) were also used for this purpose, but being calculated from fK and fca values offered no additional information. In relation to practical agriculture, Talibudeen (1971) thought that the change in excess free energy with K + saturation expressed the reciprocal of the K + buffering capacity of a soil (seealso Section IV,F,l). Soils in which this function changed least as K + saturation approached zero were expected to release most K+ . This was not tested experimentally, but changes infK (from which A P is derived) as K+ saturation approached zero agreed qualitatively with K uptake by ryegrass in pots from the soils. This and later work (Talibudeen and Weir, 1972) also explained the unusual K+-Ca2+ exchange characteristics of the Harwell series soil mentioned earlier as the result of the presence of a zeolite, clinoptilolite, mainly in the coarse clay and fine silt (0.3-5 pm) fraction of the soil. Calorimetric measurements of enthalpies of K -Ca2 exchange have shown that the interpretation of changes infK versus K + saturation curves and excess functions in terms of mineralogical differences were correct (Goulding, 1980; Goulding and Talibudeen, 1980). The maxima in the former coincided with the major steps in differential enthalpy curves from K+-selective to nonselective sites, as shown in Fig. 8, However, differential enthalpies were a much better guide to heterogeneity and soil mineralogy than fK values because, being directly measured by calorimetry, they were much more precise. Similar enthalpy measurements on the separated particle size fractions of a soil (Goulding and Talibudeen, 1979) showed the dominance of the fine (<0.2 pm) and coarse (0.2-2 pm) clay fractions in determining exchange characteristics, with the silt fraction making a small contribution. This work also showed that because of irreversible changes in the clay surfaces brought about by the separation procedure the parts did not comprise the whole. Further work on K+-Ca2+ exchange in temperate soils (Goulding and Talibudeen, 1984a,b) has shown more clearly how the exchange properties of soils are determined predominantly by clay mineralogy and there is a complex interaction of this with pH, organic matter, and manurial history. A study of unfertilized and K+-fertilized plots of eight soils with widely differing characteristics (pH +

+

+

+

+

+

THERMODYNAMICS AND POTASSIUM EXCHANGE

245

X

FIG. 8. The adsorbed K+ activity coefficient,fK (A), and differential enthalpy of exchange, d(Alf,)ldr (B), versus fractional K + saturation, x, for Ca*+ + K + exchange on Montana vermiculite. After Goulding and Talibudeen (1980).

range, 4.8-7.5; clay content range, 8-35%; CEC range, 61-176 peq/g) showed the following: 1. A soil that contained 60-70% mica in its clay fraction had almost the same exchange characteristics as those of a muscovite mica examined earlier (Goulding and Talibudeen, 1980). 2. Residual K + , from long-term applications of K + fertilizers, decreased the K preference of soils, primarily by decreasing the binding strength of K relative to Ca2+. There were direct relationships between the decrease in K+ preference and (a) percentage K saturation of the soils and (b)the duration of K fertilizer treatment. Perhaps more significantly, the differential enthalpy-K+ saturation relationship showed that over the first 10% of exchange the decrease in K + binding strength was directly related to the increase in percentage K + saturation following K + treatment, and thus naturally to a combination of the amount of K + applied and t'iie duration of the treatment. 3. Organic matter residues, from long-term heavy applications of manure, decreased K + preference, chiefly by increasing disorder in the system (i.e.. decreasing AS"). However, they sometimes had a very complex effect on exchange characteristics, resulting, it was suggested, from the residues being present as a coating on mineral surfaces. 4. K+ preference decreased with increasing surface charge density (SCD), as predicted by double-layer theory. +

+

+

+

KEITH W.T. GOULDING

246

5 . Ca2 + K exchange was accompanied by a negative entropy change in soils dominated by expanding 2: 1 minerals, and a positive entropy change in a soil dominated by mica, supporting previous findings in clay mineral studies (see Section IV,C). +

+

2. Tropical Soils Most strongly leached tropical soils are acidic, and thus aluminium ions dominate the exchange complex. Studies of AP -K exchange on such soils from Malaysia and Sri Lanka were made by Singh and Talibudeen (1971) and Sivasubramaniam and Talibudeen (1972), and are summarized by Talibudeen (1972). Of the Malaysian soils, seven out of nine exhibited K + preference, as indicated by a negative AGO value, associated with a preponderance of montmorillonitic and illitic minerals. The A13 selectivity of the other two soils was because of chloritic and vermiculitic minerals dominating their clay mineralogy. Plots of selectivity coefficients versus K + saturation suggested three types of exchange sites for the two soils that exhibited AP preference: K+ -specific, nonspecific, and AP+-specific sites. For the other seven soils which preferred K + , no AP+-specific sites were apparent. Udo (1978) felt that the dominance of kaolinite in many strongly weathered West African soils was important, and therefore examined Ca2+ K + exchange on a kaolinitic soil clay from Nigera. He found a preference for K over Ca2+, which accorded with his own past work (Udo, 1976) using nonthermodynamic methods, and with the results of Jensen (1973a). The K preference , was again the result of stronger binding of K over Ca2 (negative W ) offset by an increase in order (negative ASo) accompanying K + adsorption. Jensen (1973a) attributed the strong binding of, and selectivity for, K to K -specific sites on kaolinite minerals. However, Goulding and Talibudeen (1980) attributed similar properties of an English kaolinite to a small amount of a vermiculitic impurity (<2% by weight; Section IV,C). Possibly small amounts of such minerals were present in the kaolinites examined by Jensen (1973a) and Udo (1978). +

+

+

+

-

+

+

+

+

+

+

3. Arid Soils

Little research has been done on the thermodynamicsof K exchange in soils of arid regions. Most studies have involved the use of empirical selectivity coefficients such as that of Gapon (1933). For instance, Raju and Raman (1978) found K to be preferred to Na and Ca2 according to the Gapon coefficient and exchange isotherm. They attributed K + preference to the large micaceous mineral content of soil. The only study of K+ exchange that includes the calculation of thermodynamic quantities is that by Singh et al. (1981). They examined K + +

+

+

+

247

THERMODYNAMICS AND POTASSIUM EXCHANGE

(Ca2 +Mg2 ) exchange in eight Indian semiarid soils. Their main aim was to find a constant selectivity coefficient for predictive purposes, but all those tested were unsatisfactory. They arbitrarily divided the exchange isotherm and the selectivity coefficient versus K saturation plot into three sections representifig sites with high, medium, and low K + selectivity. For each soil they then tried to relate the proportion of sites in each class to mineralogy, with some qualitative success. The values of AGO calculated showed that there was a preference for K + over (Ca2++Mg2+) in all the soils, but they were very inaccurate because of the large extrapolations at K saturations near 0 and 1. The degree of K preference as expressed by the selectivity coefficient and AGO was used to predict the K + fixation properties of the soils with a view to aiding the assessment of K fertilizer requirements. Soil in arid regions often suffers from salinity. Much research has therefore been done on Na+-Ca2+ exchange in saline soils, including thermodynamic studies (Poonia and Talibudeen, 1977; Talibudeen, 1981), but there has been little research on K+-Na+ exchange in these soils. However, Talibudeen (1972) extended the conclusions of Deist and Talibudeen (1967a), based on K+-Na+ exchange on some British temperate soils, to saline soils. He commented that the uniformity in K + preference in very contrasting soils, and the fact that this K + preference is distributed evenly over all exchange sites as shown by adsorbed-K+ activity coefficients (Section IV,B, l), suggested that K+ fixation would not be a problem in saline soils. He also suggested that if mica minerals are present, K + release to crops should be adequate, and thus there may be no urgent practical need for studies of K exchange in saline soils. +

+

+

+

+

+

+

4 . Volcanic Ash Soils

Volcanic ash soils also described as soils with variable sur&ace charge because of the preponderance of pH-dependent charge on the aluminosilicate minerals (such as imogolite and allophane) and oxides which dominate them. All those who have worked on such soils have used the exchange isotherm and/or empirical selectivity coefficients to assess K + exchange characteristics. Thus the research cannot be said to involve a thermodynamic analysis, but it is interesting to note that the two main findings accord well with results from thermodynamic data of K + exchange on other soil types: 1. Volcanic ash soils are generally selective for K over Na (Galindo and Bingham, 1977), but for Caz+ and Mg2+ over K + (Gessa, 1970; Munns, 1976; Galindo and Bingham, 1977; Perrott, 1981). This is because the aluminosilicate minerals in these soils are usually amorphous, precluding ion-size effects; usually, valency alone controls selectivity. Selectivity for the divalent ion in monodivalent exchange increases with pH (i.e., as the number of negative sites, and +

+

KEITH W.T.GOULDING

248

thus the surface charge density, increases). These results are in agreement with diffuse double-layer thmry (see Eriksson, 1952). 2. Cation preference depends strongly on mineralogy as well as on pH (Perrott, 1981). Perrott found specific adsorption of K + in K+-Mg2+ exchange by some low A13+:(A13++Si4+) ratio aluminosilicates, which he related to some crystallinity, with structural channels selective for K + . This section on K exchange thermodynamics in soils has separated soils on a climatic basis. However, the discussion shows that climate is important only in its effects on the clay mineralogy of soils through weathering. Mineralogy is clearly the dominant factor in determining K exchange properties, sometimes modified by pH and manurial history. +

+

E. POTASSIUM S E L E C ~ VAND ~ ~FIXATION Y Selectivity can be defined as the preferential adsorption of one cation over another. The terms selectivity andpreference are thus used synonymously. Fixation has been defined as the sorption of cations in a nonexchangeable or difficultly exchangeable state (Grim, 1953) (i.e., the conversion of exchangeable or water-soluble cations into a form that cannot be readily extracted with neutral salt solutions). Mengel and Kirkby (1980) considered K selectivity and fixation to be an essential aspect of K + availability in the soil. They took the traditional view that organic matter and kaolinite minerals have low K+ selectivity, and that 2 1 minerals have high K + selectivity and high K + binding strength. That this is not always true, however, was shown in Section IV,C and is elaborated in the following sections. +

I. Selectivity Not surprisingly, most selectivity studies have involved the use of a selectivity coefficient. For example, van Schouwenberg and Schuffelen (1963), Bolt et al. (1963), and Assa (1976a,b) inferred from changes in Gapon’s constant, KG, that there were three types of exchange sites in soils and soil clay minerals distinguished by their physical position in the lattice, namely, planar, edge, and interlayer (Fig. 9). The K+ selectivity of these sites decreased in the order interlayer >> edge > planar, interlayer sites having the ability to fix K + ions. This argument was used to explain the selective adsorption of K + by soils, the weathering of soil micas to illites producing highly K -selective wedge sites, shown in Fig. 9 (van Schouwenberg and Schuffelen, 1963). Lee and Rowel1 (1973, 1975) also thought that wedge sites were responsible for the selectivity of K+ over Ca2+ in a kaolinite and a bentonite that they examined, the wedges resulting from aggregation of clay lamellae. However, Assa (1976a,b) found K + +

THERMODYNAMICS AND POTASSIUM EXCHANGE

249

FIG. 9. A schematic 2:l clay mineral particle, showing Planar, Edge, and Interlattice exchange sites and a Wedge Zone. After Schuffelen (1972).

selectivity in biotite mica, muscovite mica, illite, and vermiculite but not in montmorillonite. The preference for K + was attributed to its low hydration number and high polarizability. Thermodynamic measurements have added some details. In particular, examining free energies, enthalpies, and entropies separately enables the forces that bring about selective adsorption to be delineated. From the results examined to date, Ca2+ + K + exchange is always accompanied by a negative enthalpy change (i.e., K + is bound more strongly than Ca2+). This K + selective tendency is sometimes enhanced by a positive entropy change (e.g., Ca2+ .--* K + exchange on muscovite mica; Goulding and Talibudeen, 1980), but far more often it is offset by a negative entropy change (e.g., Ca2+ + K + exchange on soils; see Hutcheon, 1966; Deist and Talibudeen, 1967b; Ca2+ --* K + exchange on illite, vermiculite, and montmorillonite; see Goulding and Talibudeen, 1980). Differential enthalpy measurements in particular clarify the causes of selectivity. For example, the high K + selectivity [d(AG,)l& = -20 kl/eq] of a small number of exchange sites in Fithian illite was solely the result of the ability of these sites to bind K + very strongly [d(AH,)l& = -20 kl/eq; Fig. 4). Again, the somewhat lower K selectivity of a few sites in Upton montmorillonite and an English kaolinite was attributed to the presence of very small amounts of mica and vermiculite impurities, respectively, on the basis of d(AH,)l& values (Section IV,C). The apportionment of degrees of K + selectivity to three types of exchange sites by van Schouwenberg and Schuffelen (1963) and Bolt et al. (1963) involved an arbitrary division of the smooth curve of a plot of Gapon’s constant against K + saturation into three parts (Fig. 10). Of the minerals examined by Goulding and Talibudeen (1980) that exhibited K’ selectivity, none had three sets of site groups as reflected in their differential enthalpy values; the illite had four (Fig. 4), the vermiculite had two (Fig. 7), and the mica had one. Also, the site groupings indicated by differential enthalpy values do not necessarily reflect planar, edge, and interlayer sites or other structural features, because surface charge density differences may be equally important (see Section IV,E,2). So, although the recent thermodynamic data do support different site groupings in clays and soils, they do not prove that planar, edge, and interlayer sites are always responsible for the different degrees of K + selectivity. Combining all of +

KEITH W.T.GOULDING

250

h

lnterlayer

L Planar

I

0

K’saturation

100 %

FIG. 10. Gapon’s constant, KG,for K+-Ca2+ exchange as a function of K + saturation, showing the apportionment of Planar, Edge, and Interlattice exchange sites.

these observations, the dominance of ion binding strength in determining K + selectivity in illite and vermiculite may well suggest that surface charge density is the most important factor for these minerals. The dominance of entropic effects (i.e., the physical arrangement of cations and surfaces) in determining K+ selectivity in mica suggests that here wedge sites may be the main factor. 2 . Fixation

The fixation of K+ by soils, and the resultant “loss” of nutrient K + , is agricuiturally very important. Most research on cation fixation in soils and clays has therefore concentrated on K+ and the equally important NH,+ ion. The distinction, in the literature, between selectivity (preference for one cation over another) and fixation is not always clear. Indeed, Talibudeen (1981) appears to equate selectivity and fixation. However, there is a difference. As shown in Section IV,E, 1, most soils and soil clay minerals selectively adsorb K over Ca2 ,but most of this K is still readily exchangeable. Fixed K , as defined earlier, is not readily exchangeable and only certain adsorption sites fix K + . There is, at the least, a distinction between very rapid and very slow rates of exchange. Mengel and Kirkby (1980) attributed K fixation to interlayer exchange sites (see Fig. 9). They wrote, “Potassium fixation takes place by means of K + adsorption to these K+-specific binding sites of the interlayer zone . . . the replacement of interlayer K+ by larger cation species (Ca2+, MgZ+)expands the lattice and wedge zones are formed. The reverse process occurs when these larger cation species are replaced from interlayer sites by K + or NH,+ . The contraction of the mineral is accompanied by a decrease in cation exchange capacity. This is the process by which K+-depleted 2 1 clay minerals fix K+ .” The picture is a little more complex than this, as the review of fixation by Sawhney (1972) shows. The factors that contribute to cation fixation in general can be compiled from his review and those of Talibudeen (1973, 1981): +

+

+

+

+

THERMODYNAMICS AND POTASSIUM EXCHANGE

25 1

1. The hydration energy of the cation. Cations with low hydration energies cause interlayer dehydration and layer collapse, and thus cations such as K+, NH,+ , Rb+ , and Cs+ are fixed, whereas those with larger hydration energies (e.g., Na+, Ca2+, and Mg2+) are not. Sawhney (1972) considered this the major factor. 2. Closely linked with (1) is the exact matching of anhydrous cation size to the structural hexagonal cavities in the tetrahedral layer. Cations that fit snugly facilitate layer collapse. 3. Total layer charge. The higher the charge, the stronger the binding. 4. The site of negative charge (i.e., whether isomorphous substitution is in the tetrahedral or octahedral layer). That in the tetrahedral layer is closer to the adsorbed cations and so may hold them more tightly (but see Weir, 1965). 5 . Wedge zones or frayed edges, at the edge of books of aluminosilicate sheets, fix cations through spatial matching. 6. Structural faults and cracks, and other changes in the lattice resulting from impurities, create a limited number of exchange sites which may also fix cations through spatial matching. 7. Orientation of the hydroxyl group in the octahedral layer. Trioctahedral micas, with OH- oriented normal to the basal plane, strongly repel K + and other cations; dioctahedral micas, with OH- oriented at an angle of 16" to the basal plane, repel K + and other cations much less strongly, allowing a shorter K-0 bond length. 8. Substitution of F- for OH- in the octahedral layer increases K+ retention in trioctahedral micas by removing the OH- repulsion effect. 9. Drying, or more correctly, alternate wetting and drying of the clay, facilitates the rearrangement and close matching of surfaces. Thermodynamic exchange data can partially clarify the mechanism. However, it must be remembered that when cations have been fixed they do not take part in the (rapid) cation-exchange process, although they may readily take part in a slower diffusion-controlled exchange. The realignment of aluminosilicate layers of expanded 2: 1 minerals in the 001 direction following K + sorption, which produces a negative entropy change as explained in Section IV,C, precedes layer collapse and K + fixation (Plancon et al., 1979). This suggests that items (1) and (2) may be of primary importance in determining K + fixation in these minerals. A positive entropy change for the replacement of CaZ+ by K + on muscovite mica, as found by Goulding and Talibudeen (1980), suggests that little or no such rearrangement occurs in this mineral and supports the view that wedge sites [in effect, items (5) and (6)] are important in causing K + fmation in micas. For applications to practical agriculture, values of the activity coefficient of adsorbed K U;;) as K saturation ( x ) approaches zero were used by Talibudeen +

+

KEITH W.T. GOULDING

252

(1971) to predict the amount of nonexchangeable K + released by soils. The smaller the value of fK at x 3 0 and the steeper the slope of fK versus x over this range, the less nonexchangeable K + is released. The corollary to this, of course, is that the lower is fK and the steeper the fK versus x curve as x 0, the more readily K + will be fixed. This idea has not been quantitatively tested, although a qualitative confmtion is found in Section IV,D, 1. Eberl (1980) has suggested a new approach to the problem of predicting selectivity and fmation using thermodynamicmethods. Cations are considered to be fixed when adsorbed at exchange sites in unexpanded, unhydrated, or “dry” interlayers. When adsorbed on hydrated or “wet” surfaces, cations are defined as exchangeable. Following equations developed by Eisenmann (1961, 1962)for cation-selective electrodes, AGO values for exchange reactions in the interlayer space of nonexpanded or dry 2: 1 minerals such as illite were calculated from free energies of cation hydration and electrostatic attraction between the cations and the surface (the latter was taken to be a function of radius and charge of the cation, and the ionic field strength of the clay). Values of AGO for exchange reactions on all other wet mineral surfaces were calculated from a formula given by Cruickshank and Meares (1957):

AGO = =T(+ - In y m ) , ,,

+,

-WT(+ - In ym),-,,,,,

(30)

where the m. and y are osmotic coefficients, mean ionic molalites, and mean ionic activity coefficients, respectively. The calculations were applied to the alkali metal cations Li+, Na+, K + , Rb+, Cs+, and also H+ by Eberl(l980). AGO was plotted against the equivalent anionic radius of a clay (r,) where

ra = ( - a b / 8 ~ C ) ~ / ~

(31)

and ub is the unit cell area of the clay and C is the layer charge in equivalents per half unit cell. Curves of AGO versus ra for wet and dry clays were plotted on the same graph at constant concentration. The cation with the lowest AGO at any ra was the most preferred by the clay. The order of selectivity was found to change with r, and thus with clay type. For a montmorillonite-typeclay, the order of selectivity was Cs+ > Rb+ > K + > Na+ > Li+ > H+ (cf. Section IV,B,l). Measured and predicted values for AGO for Cs+ + B + exchange on 15 montmorillonites were in fairly good agreement, with five exceptions attributed to charge heterogeneity. Cation fixation, by definition, occurred at r, values less than, or at an equivalent layer charge greater than, those values at which the wet and dry AGO versus ra curves intersected. Values of layer charge obtained for the fixation of K and Na+ in smectites, and thus the formation of K+ and Na+ illite, agreed with those obtained by other means (e.g., number of smectite layers in interstratified illite/smectite versus layer charge). Eberl (1980) suggested that the calculations partly explained the greater stability of illite over muscovite mica to weathering, because the former was shown to have a much stronger preference for K + over +

253

THERMODYNAMICS AND POTASSIUM EXCHANGE

H + than the latter. They also suggested a reason for cation segregation or demixing in clays: selectivity was a function of layer charge that is heterogeneous in the smectites.

F. POTASSIUM POTENTIALS The concept of a nutrient potential was suggested by Schofield (1955) as a measure of the work the plant must do to remove nutrients from the soil. The idea was f m t applied to potassium by Woodruff (1955a,b) who expressed the thermodynamic potential of K + in the soil as its partial molar free energy, referred to that of calcium:

AG,,,,

= RT I n ( a , l a ~ ~ )

(32)

The function a,la,!l,2 is called the activity ratio (AR) or intensity (I)of K + in the soil. Arnold (1962) reasoned that for soils containing both Ca2+ and Mg2+, these ions can be assumed to behave similarly, and thus the K + potential could be redefined with reference to both ions as (33) + Mg = RT In ( a K 1 a Z 2 + ~ g ) It is measured, as indicated by Eq.(33), as the difference between the chemical AGK,Ca

potentials or activities of K + and (Ca2+ +Mgz +) in a dilute solution in equilibrium with the soil. The K potential is in effect a measure of the free energy of available K referred to some reference cation(s). In the case just mentioned, Ca2+ and Mg2+ are the reference pair; these ions dominate the exchange complex of most temperate soils. In acidic soils, AP+ or (AP +Ca2+ +Mg2+) have been used as the reference cations (Tinker, 1964a,b; Singh and Talibudeen, 1969). Potassium potentials were reviewed by Beckett (1971) and Talibudeen (1981), and general nutrient potentials were derived by Talibudeen (1974). The concept of nutrient potentials has been applied to soil potassium chemistry in two ways: (1) the quantity/intensity @/Z) relationship, which relates the change in exchangeable K + with the activity ratio or, less commonly, with the K potential, has been used to characterize the variability of K availability in soils; and (2) the relationship between crop growth and nutrient potential has been used to identify critical potentials that are characteristic of a plant during a particular growth phase. These two practical applications will be considered briefly, particularly to elucidate the role of potentials in relation to the thermodynamic parameters defined in Section 11. +

+

+

+

+

1. Q l l Curves

The Q/I relationship is an empirical relationship linking the K + intensity, measured as the activity ratio (AR), or the K + potential with the quantity of

254

KEITH W. T. GOULDING

AK

FIG. 11. Typical Q/l relations for soil potassium, where Q / l is the change in exchangeable K+ , ILK, versus the activity ratio, ARK-c,. A, normal clay loam soil; B, heavy clay soil; C, peaty organic soil; D,soil with little or no fixed charge. After Beckett (1971).

exchangeable K+ , usually measured as the change in adsorbed K + (AK) (Fig. 11). The use of Q/f curves was reviewed by Addiscott and Talibudeen (1969) and Beckett (1971). In earlier work it was suggested that Q/f curves indicated the ability of a soil to buffer changes in K+ activity against depletion by crops, distinguished between K -selective and nonselective sites, and predicted K fixation and release properties (Beckett, 1971). However, some workers have the opinion that Q l f curves only give a qualitative estimate of K+-selective sites and a rather ambiguous estimate of buffer capacity (Addiscott and Talibudeen, 1969). In particular, Q/f curves do not unambiguously reflect changes in the K + exchange characteristicsof soils caused by different cropping and manurial treatments. Beckett and Nafady (1968) showed that the Q/Z relationship is not altered by short-term (20-30 years) addition or release of K+ , and, although Mathews and Beckett (1962) detected differences in soils of the Broadbalk Experiment at Rothamsted following over 100 years of K + manuring or extraction, Johnston and Addiscott (1971) found no such differences in these soils. By contrast, standard free energy and enthalpy parameters (Goulding, 1981) and, more particularly, differential enthalpy changes (Goulding and Talibudeen, 1984a,b) have been found to distinguish clearly between soils (Section IV,D,l), They also quantitatively distinguish K -selective and nonselective adsorption sites (Section IV,E,l). Thus, in these two areas Gaines and Thomas-type thermodynamic parameters show considerable advantages over Q / l curves. As regards buffer capacity, Addiscott and Talibudeen (1969) thought that this would be defined best by the slope of the K saturation versus differential free energy curve, and Talibudeen (1971) suggested that the reciprocal of the excess free energy versus K+ saturation curve would also be a good estimate. Neither of these have been tested, however, and Q/Z curves are still used extensively for this purpose (e.g., Goedert er al., 1975; Sparks and Liebhardt, 1981). Adopting a slightly different approach, Karamanos and Turner (1977) suggested that a freeenergy or a potential term that included both the intensity and quantity factors would describe K+ availability more completely. They therefore +

+

+

+

THERMODYNAMICS AND POTASSIUM EXCHANGE

255

used the Gaines and Thomas-type thermodynamic equilibrium constant. However, they did not apply it directly; for the reaction CaZf +2

(i.e., K +

--f

KX

CaX2 + 2 K +

(34)

Ca2+ exchange or K + release), they defined a free energy term,

AF, as that required to make the reaction shown in Eq. (34) proceed from its

equilibrium state [subscript e in Eq. (35)] to a new state [subscript n in Ji!q. (331; that is the free energy involved in releasing a certain amount of K + . The free energy term AF is thus defined as a ratio of two values of K,

AF

(35)

= RT ln(K,IK,,)

and should therefore more correctly be written as d(AF).As written here, if AF is positive, K + is the preferred ion (because the reaction is K + +. Ca2+ exchange). The choice of the “new” value is arbitrary and depends on the required value of the activity ratio (i.e., concentration of K + in the soil reckoned adequate). Karamanos and Turner chose a value such that the reaction in Eq. (34) always moved to the right (i.e., K + was released) and thus the new AR had to be greater than the lowest AR measured. The free energy term AF was tested on a series of 23 soils and assessed by comparing it with In AR, (where AR, is the activity ratio at which K is neither adsorbed nor desorbed). The value AF was taken to be a better parameter for defining a soil’s ability to release K than AR, or K potential, on the unusual basis of the correlations of these parameters with the content of mica and other 2:l minerals in the soils. The same idea was applied to results from a long-term field experiment, growing a pearl millet-wheat rotation, by Singh et al. (1982). However, they used the true standard free energy of K+ + Ca2+ exchange, AGO, calculated by the Gaines and Thomas (1953) method (Section II,D), and not AF. Here again, because the reaction examined is K + --* Ca2+ exchange, the more negative AGO is, the less K+ is preferred and the more easily it is released. The experimental plots fertilized with K + had more negative AGO values than the unfertilized plots, expressing improved K+-releasing power over the “nil” plots, as expected. However, AGO correlated less well with K + uptake by these field crops than A h . +

+

+

2 . Critical Potassium Potentials in Crops Potassium potentials were reviewed by Beckett (1972) and Talihudeen and Page (1978), and the whole concept of a potential governing nutrient uptake was questioned by Nye (1968). Research since Beckett’s review (1972) has involved attempts to obtain potentials characteristic of a single aspect of crop growth, particularly an exhaustion potential (e.g., Addiscott and Johnston, 1975). Page and Talibudeen (1982a,b) have attempted to define and measure a much wider

KEITH W.T. GOULDING

256

K+ potential RG.12. A diagrammatic representation of the effect of soil nutrient potential on crop performance. After Talibudeen (1974).

range of potentials than this. They reasoned that [extending Schofield‘s (1955) hypothesis] potentials are dependent on the crop and the environment but not on the soil. A schematic curve based on such a concept, given by Talibudeen (1974), is shown in Fig. 12; exhaustion, response, optimum, luxury uptake, and toxicity potentials are defined. Page and Talibudeen (1982a) attempted to measure these for perennial ryegrass (Loliumperenne) and creeping red fescue (Festucu rubru genuinu) on six contrasting soil types in pot experiments. Only exhaustion and optimum potentials were obtained, and these depended on soil type because, it was suspected, of the large crop/soil ratio necessitatedby the use of small pots in constant environment chambers. In a second pot experiment in which five crops [i.e., spring wheat (Triticum uestivum), maize (Zeu mays), peas (Pisum sutivum L.), field beans (Viciufubu), and sugar beets (Beta vulgaris)] were grown in a loamy sand soil, the same authors were able to measure exhaustion, optimum, and toxicity potentials (Page and Talibudeen, 1982b). Exhaustion potentials were similar for all crops except beans, for which they were much lower, implying a greater ability than the other crops to extract K + . Optimum potentials increased with decreasing K concentration and content of the seeds. It was suggested that this was related to differences in the K + requirements of the crops and/or to their ability to extract K + from the soil. Toxicity potentials were relatively high, and some were even positive. They were considered probably far greater than any occurring in a natural system. +

V. EXCHANGE EQUILIBRIUM AND THE KINETICS OF POTASSIUM EXCHANGE In a paper entitled, “Toward a connection between ionic equilibrium and ionic migration in clay gels,” Thomas (1965) asked whether a relationship existed

THERMODYNAMICS AND POTASSIUM EXCHANGE

251

between the activation enthalpy (AH$) of an ion-exchange reaction and the standard enthalpy of exchange for the solid phase (AW,), where

q=m-w

(36)

and is the standard enthalpy of exchange for the liquid (solution) phase. He suggested that the best way to establish such a connection was through selfdiffusion studies. Unfortunately, no further publication has appeared clarifying this idea or confirming the relationship. The paper was also somewhat puzzling in that a direct relationship between chemical equilibria and kinetics of completely reversible reactions (such as ion exchange is taken to be) was established long ago (e.g., see Glasstone et al., 1941; Laidler, 1965). Use of this was made by Keay and Wild (1961), who calculated the enthalpy and entropy of Na+Mg2+ exchange on a vermiculite from measurements of the kinetics of Na+ -+ Mg2+ and Mg2+ + Na+ exchange. The relationship was applied for the first time to K+-exchange studies by Sparks and Jardine (1981), who calculated standard thermodynamic parameters of K -Ca2 exchange from measurements of the kinetics of this reaction. The apparent adsorption rate coefficient for K + adsorption (k:) is given by +

+

lOg(1 - K,/K,)

=

kLt

(37)

and the apparent desorption rate coefficient for K + desorption (k> is given by log(K,lKo)

=

k&t

(38)

where K, is the amount of K + adsorbed on soil exchange sites at time t , K, the amount at equilibrium, and KOthe amount at zero time. For a completely reversible reaction, the thermodynamic equilibrium constant is given by K = k!JkA

(39)

(Glasstone et al., 1941). Thus the free energy of exchange is readily obtainable if kk and ki can be measured. The energy of activation for K + adsorption (E,) and desorption (Ed) is calculated from the Arrenhius equation d In ki - Ei

dT

RF

where i = a or d. Then, following a similar argument, the standard enthalpy of exchange is obtained from the difference between activation energies of the forward and reverse reactions

AiT

= E, - Ed

(41)

The standard entropy of exchange can then be obtained from the Gibbs equation. Sparks and Jardine (198 1) measured values of kL and kA at 3,25, and 40°C, and accordingly calculated AGO, AHO, and A,!?. They also calcuated AGO, AW,and

KEITH W.T.GOULDING

258

Table II

Standard Free Energies (AG3, Enthalpies (AlT), and Entropies (W,of Caz+ Exchange on Soils, Calculated from Equilibrium and Kineticsa

+

K+

AG

Lw"

A$

Method of calculation

(kcaUmo1)

(kcaVmol)

(caYmoyK)

Kinetics data Eyring's absolute reaction rate theory Equilibrium data (Deist and Talibudeen, 1967b)

- 1.274 -1.290

-1.690 -1.680

-1.4 -1.3

-1.05 to -3.42

-0.91 to -8.49

-2.89 to -22.10

'Data after Sparks and Jardiie (1981). Adapted from Soil Sci. SOC.Am.J . 45, 1097-1099. By permission of the Soil Science Society of America Journal.

AS' using Eyring's absolute reaction rate theory (Laidler, 1965). In this method free energies (AGS),enthalpies (AH$), and entropies (ASS) of activation for forward and reverse reactions are calculated, and the difference between them is again AGO, A W , and W .These data are shown in Table 11, and the agreement between the methods is excellent. Unfortunately the authors did not also calculate these quantities from exchange isotherm data and compare the results, which would have been the best experimental verification of the theory. However, values for Ca2+ + K + exchange on a silt loam soil at 25°C were of the same order of magnitude as those found by Deist and Talibudeen (1967b) for a range of soils at the same temperature (Table II). VI. SUMMARY AND CONCLUSIONS The essential elements of a thermodynamic analysis of cation-exchange equilibria have been presented, and their application to potassium exchange in soils and clays has been examined. It is true to say that useful applications have so far been limited to characterizing the exchange properties of soils and establishing the selectivity of various soils and clays for potassium. No thermodynamic parameter has yet been found to predict crop yield or response to K + fertilizer from soil K+ measurements; K+ potentials seem closest to being used in this way. Perhaps the most promising new application suggested in this article, although not directly related to agriculture, could have useful indirect benefits. The use of calorimetrically measured enthalpies of exchange to detect small amounts of impurities in formerly pure clay minerals may well aid the clay mineralogist. It may also aid agriculturalists, because a more detailed picture of soil mineralogy will help to explain and predict the reaction of a soil when K +

THERMODYNAMICS AND POTASSIUM EXCHANGE

259

fertilizer is added. The differential enthalpies and entropies used for this purpose certainly avoid Walsh’s complaint (see Section I) that thermodynamic functions integrate variable quantities giving a (misleading) average value. However, one of the most important areas of agricultural research at the present time is the modeling of soil-plant processes (Cooke, 1979). Cation exchange, and in particular the adsorption and release of nutrient K + , should be part of this. As Cooke and Gething (1978) said, “Probably the practical returns from the basic work resulting from applying thermodynamic principles to soil systems, begun 25 years ago, will come from the development of models that lead to more efficient fertiliser recommendations.” Empirical models have proved unsatisfactory, but a basic set of thermodynamic equations describing cation exchange is now available. Rigorous and theoretically sound, and being based on no assumptions, it is universally applicable. Although the equations were initially written to describe only simple systems (e.g., binary exchange an “pure” exchangers), equations are now being developed to describe more closely the complicated situation found in multi-cation-exchange reactions in soils, and they are thus relevant to agriculture. The equations may be initially disconcerting to the agronomist because of their complexity, but the availability of computers makes them accessible and usable. VII. APPENDIX: LIST OF SYMBOLS Activity of cation A Unit cell area of a clay Activity ratio Layer charge of a clay Differential free energy of exchange Differential enthalpy of exchange Differential entropy of exchange Equivalent fraction of cation A Energy of activation for K+ adsorption Energy of activation for K + desorption Activity coefficient of adsorbed cation A, calculated according to the Vanselow convention up to the end of Section II,E and according to the Gaines and Thomas convention thereafter Activity coefficient of adsorbed cation A, calculated according to the Gaines and Thomas convention up to the end of Section 1I.E Ionic strength “Intensity” of K + in the soil (Section IV,F) Thermodynamic equilibrium constant Gaines and Thomas selectivity coefficient Uncorrected Gaines and Thomas selectivity coefficient Gapon constant (or selectivity Coefficient) Vanselow selectivity coefficient

KEITH W.T. GOULDING Uncorrected Vanselow selectivity coefficient Amount of K+ adsorbed at time t (Section V) Amount of K+ adsorbed at zero time (Section V) Amount of K+ adsorbed at equilibrium (Section V) Apparent adsoption rate coefficient (Section V) Apparent desorption rate coefficient (Section V) Mean ionic molality [Eq. (30)] Molarity of cation A Mole fraction of cation A Number of moles of water sorbed by the exchanger [Eq.(1 l)] “quantity” of K+ in the soil (Section IV,F) Gas constant Anhydrous radius of an ion Equivalent anionic radius of a clay (Section IV,E,2) Absolute temperature

Time Valency of cation A Valency of cation B Fractional K saturation of exchange capacity Ion charge (Section IV,B,l) Activity coefficient of cation A in solution Free energy term defined by Karamanos and Turner (1977) (Section IV,F) Potassium potential with reference to calcium Excess free energy of exchange Standard Gibbs free energy of exchange Activation energy Excess enthalpy of exchange Standard enthalpy of exchange Integral enthalpy of exchange Activation enthalpy Standard enthalpy of exchange for liquid or solution phase Standard enthalpy of exchange for solid phase Change in adsorbed (exchangeable) K+ (Section IV,F) Excess entropy of exchange Standard entropy of exchange Activation entropy Chemical potential Chemical potential in standard or reference state Osmotic coefficient [Eq.(30)] +

ACKNOWLEDGMENT The critical assessment of this article by Dr.0.Talibudeen is gratefully acknowledged.

REFERENCES Addiscott, T. M., and Johnston, A. E. 1975. J . Agric. Sci. 84,513-524. Addiscott, T. M., and Talibudeen, 0. 1%9. Potush Rev. (Berm) Subj. 4, 45th Suite, pp. 1-24.

THERMODYNAMICS AND POTASSIUM EXCHANGE

26 1

Argersinger, W. J., Davidson, A. W., and Bonner, 0. D. 1950. Truns. Kuns. Acad. Sci. 53, 404-410. Arnold, P. W. 1962. Proc. Fert. SOC.72, 25-43. Assa, A. D. 1976a. Cuh. ORSTOM Ser. Pedol. 14, 219-226. Assa, A. D. 1976b. Cuh. ORSTOM Ser. Pedol. 14, 279-286. Babcock, K. L. (1963). Hilgurdiu 34, 417-542. Bansal, 0. P. 1982. J. Soil Sci. 33, 63-71. Barrer, R. M., and Klinowski, J. 1974. J. Chem. SOC.Furuduy Truns. I 70, 2080-2091. Barrer, R. M., and Klinowski, J. 1979. J. Chem. Soc. Furuday Truns. I 75, 247-251. Barrer, R. M., Rees, L. V. C.,and Ward, D. J. 1%3. Proc. R. Soc.London Ser. A 273, 180-197. Barrer, R. M., Klinowski, J., and Sherry, H. S. 1973. J . Chem. SOC. Furaduy Tram. 2 69, 1669- 1676. Beckett, P. H. T. 1971. Potash Rev. (Berne) Subj. 5 , 30th Suite, pp. 1-41. Beckett, P. H. T. 1972. Adv. Agron. 24, 379-412. Beckett, P. H. T., and Nafady, M. H. M. 1968. J. Soil Sci. 19, 216-236. Bolt, G. H. 1967. Nerh. J. Agric. Sci. 15, 81-103. Bolt, G. H., Sumner, M. E., and Kamphorst, A. 1963. Soil Sci. SOC. Am. Proc. 27, 294-299. Calvet, R., and Pratt, H. 1956. “Microcalorimetrie,” Masson, Paris. Chu, S.-Y., and Sposito, G. 1981. Soil Sci. SOC. Am. J . 45, 1084-1089. Clearfield, A., and Kullberg, L. H. 1974. J. Phys. Chem. 78, 1812-1817. Coleman, N. T. 1952. Soil Sci. 74, 115-125. Cooke, G. W. 1979. J. Soil Sci. 30, 187-213. Cooke, G. W., and Gething, P. A. 1978. In “Potassium ResearcLReview and Trends,” pp. 361-405. Int. Potash Inst., Beme. Cotton, F. A., and Wilkinson, G. 1972. “Advanced Inorganic Chemistry.” Wiley (Interscience), London. Coulter, B. S. 1969. J . Soil Sci. 20, 72-83. Coulter, B. S . , and Talibudeen, 0. 1968. J. Soil Sci. 19, 237-250. Cremers, A., and Laudelout, H. 1966. Soil Sci. SOC. Am. Proc. 30, 570-576. Cruickshank, E. H., and Meares, P. 1957. Trans. Furaduy SOC. 53, 1289-1298. Deist, J., and Talibudeen, 0. 1967a. J . Soil Sci. 18, 125-137. Deist, J., and Talibudeen, 0. 1967b. J. Soil Sci. 18, 138-148. Drake, M. 1964. In “Chemistry of the Soil” (F. E. Bear, ed.), pp. 395-444. Reinhold, New York. Eberl, D. D. 1980. Clays Clay Miner. 28, 161-172. Eisenmann, G. 1961. In “Symposium on Membrane Transport and Metabolism” (A. Kleinzeller and A. Kotyk, eds.), pp. 163-179. Academic Press, New York. Eisenmann, G. 1962. Biophys. J . 2 (Suppl.), 259-323. Ekedahl, E., Hogfeldt, E., and Sillen, L. G. 1950. Actu Chem. Scund. 4, 556-558. El-Prince, A., and Babcock, K. L. 1975. Soil Sci. 120, 332-338. El-Prince, A., and Sposito, G. 1981. Soil Sci. SOC. Am. J . 45, 277-282. El-Prince, A., Vanselow, A. P., and Sposito, G. 1980. Soil Sci. SOC.Am. J. 44, 964-969. Eriksson, E. 1952. Soil Sci. 74, 103-113. Faucher, J. A., and Thomas, H. C. 1954. J. Chem. Phys. 22, 258-261. Gaines, G. L., and Thomas, H. C. 1953. J. Chem. Phys. 21, 714-718. Gaines, G. L., and Thomas, H. C. 1955. J. Chem. Phys. 23, 2322-2326. Galindo, G. G., and Bingham, F. T. 1977. Soil Sci. SOC. Am. J . 41, 883-886. Gapon, Y. N. 1933. J. Gen. Chem. USSR (Engl. Transl. of Zh. Obshch. Khim.) 3, 144-160. Gast, R. G. 1968. Truns. Int. Congr. Soil Sci. 9th. 1, 587-596. Gast, R. G. 1969. SoilSci. SOC.Am. Proc. 33, 37-41. Gast, R. G. 1972. Soil Sci. SOC. Am. Proc. 36, 14-19. Gast, R. G., van Blade], R., and Deshpande, K. B. 1969. Soil Sci. SOC.Am. Proc. 33, 661-664.

262

KEITH W. T. GOULDING

Gessa, C. 1970. Agrochimica 14, 188-197. Glasstone, S.,Laidler, K. J., and Eyring, H.1941. “The Theory of Rate Processes.” McGraw-Hill, New York. Goedert, W. J., Corey, R. B., and Syers, J. K. 1975. Soil Sci. lu), 107-1 11. Goulding, K. W. T. 1980. Ph.D.Thesis, UNv. of London. Goulding, K. W. T. 1981. J. Sci. FoodAgric. 32, 667-670. Gouldmg, K. W. T. 1983. J. Soil Sci. 34, 69-74. Gouldmg, K. W. T., andTalibudeen, 0. 1979. J. SoilSci. 30, 291-302. Goulding, K. W. T., and Talibudeen, 0. 1980. J . Colloid Interface Sci. 78, 15-24. Gouldmg, K. W. T., and Talibudeen, 0. 1984a. J . Soil Sci. 35, in press. Gouldmg,K. W. T., and Talibudeen, 0. 1984b. J. Soil Sci. 35, in press. Grim,R. E. 1953. “Clay Mineralogy.” McGraw-Hill, New York. Guggenheim, E. A. 1%7. “Thermodynamics.” North-Holland, Amsterdam. Harter, R. D., and Kilcullen, B. M. 1976. Soil Sci. SOC. Am. J . 40,612-614. Hoagland, D. R., and Martin, J. C. 1933. Soil Sci. 36, 1-32. Hogfeldt, E., Ekedahl, E., and Sillen, L. G. 1950. Acra Chem. S c u d . 4, 828-829. Hutcheon, A. T. 1966. J. Soil Sci. 17, 339-355. Jackson, M. L. 1963. Ckzys Clay Miner. 11, 29-46. Jensen, H. E. 1972, Arsskr. K . Vet. Landbokjsk (Copenhagen) 1972, pp. 88-103. Jensen, H. E. 1973a. Agrochimicu 17, 181-190. Jensen, H. E. 1973b. Agrochimicu 17, 191-201. Jensen, H. E. 1975. Agrochimicu 19, 257-261. Jensen, H.E., and Babcock, K. L. 1973. Hilgardiu 41, 475-488. Johnston, A. E., and Addiscott, T. M. 1971. J. Agric. Sci. 76, 539-552. Karamanos, R. E., and Turner, R. C. 1977. Geodermu 17, 209-218. Keay, J., and Wild, A. 1961. Soil Sci. 92, 54-60. Kerr, H.W. 1928. J. Am. SOC. Agron. 20, 309-335. Laidler, K. J. 1965. “Chemical Kinetics,” 2nd ed. McGraw-Hill, New York. Laudelout, H., and Thomas, H. C. 1965. J. Phys. Chem. 69, 339-340. Laudelout, H., van Bladel, R., Bolt, G. H., and Page, A. L. 1%8a. Trans. Faruduy Soc. @4, 1477- 1488. Laudelout, H., van Bladel, R., Gilbert, M., and Cremers, A. 1%8b. Truns. Int. Congr. Soil Sci. 9th 1, 565-575. Laudelout, H., van Bladel, R., and Robeyns, J. 1972. Soil Sci. Soc.Am. Proc. 36, 30-34. Lee, R., and Rowell, D. L. 1973. N.Z. J. Sci. 16, 839-846. Lee, R.,and Rowell, D. L. 1975. J. Soil Sci. 26,418-425. Lee, S . Y.,Jackson, M. L., and Brown, J. L. 1975. Clays Clay Miner. 23, 125-129. Lim, C. H., Jackson, M. L., Koons, R. D., and Helmke, P. A. 1980. Clays Clay Miner. 28, 223-229. Maes, A., and Cremers, A. 1977. 1. Chem. SOC. Faruduy Trans. 1 , 73, 1807-1814. Maes, A., and Cremers, A. 1978. J. Chem. SOC.Faruday Trans. I , 74, 1234-1241. Maes, A., Peigneur, P., and Cremers, A. 1976. Proc. Inr. Clay Con$. Mexico City, 1975. pp. 3 19-329. Marques, I. M. 1968. Potash Rev. (Berne) Subj. 5 , Suite 29, 1-5. Martin, H., and Laudelout, H. 1963. J. Chim. Phys. 60, 1086-1099. Mathews, B. C., and Beckett, P. H. T. 1%2. J. Agric. Sci. 58, 59-64. Mengel, K.,and K d b y , E. A. 1980. A h . Agron. 33, 59-110. Memam, C. N., andmomas, H. C. 1956. J. Chem. Phys. 24,993-995. Minter, B. A., and Talibudeen, 0. 1982. Lab. Pruct. 31, 1094-1096. Munns, D. N. 1976. Soil Sci. SOC.Am. J. 40, 841-845.

THERMODYNAMICS AND POTASSIUM EXCHANGE

263

Norrish, K. 1954. Discuss. Faraday SOC. 18, 120-134. Nye, P. H. 1968. J. Soil Sci. 19, 205-215. Oster, J. D., and Sposito, G. 1980. Soil Sci. SOC. Am. J . 44, 258-260. Page, M. B., and Talibudeen, 0. 1982a. J. Soil Sci. 33, 329-347. Page, M. B., and Talibudeen, 0. 1982b. J . Soil Sci. 33, 771-778. Perrott, K. W. 1981. Geoderma 26, 311-322. Plancon, A., Besson, G., Gaultier, J. P., Mamy, J., and Tchoubar, C. 1979. Proc. Int. Clay Conf. 6th. pp. 45-54. Poonia, S. R., and Talibudeen, 0. 1977. J . Soil Sci. 28, 276-288. Raju, A. S., and Raman, K. V. 1978. Indian J. Agric. Chem. 11, 21-25. Ravina, I., and de Bock, J. 1974. Soil Sci. Soc.Am. Proc. 38, 45-49. Rich, C. I. 1972. In “Potassium in the Soil,” pp. 15-31. Int. Potash Inst., Berne. Sawhney, B. L. 1972. Clays Clay Miner. 20, 93-100. Schofield, R. K. 1955. Soils Fert. 18, 373-375. Schroeder, D. 1972. In “Potassium in the Soil,” pp. 203-204. Int. Potash Inst., Berne. Schuffelen, A. C. 1972. In “Potassium in the Soil,” pp. 75-88. Int. Potash Inst., Beme. Singh, D., Pal, R., and Poonia, S . R. 1981. Geoderma 25, 55-62. Singh, M., Singh, A. P., and Mittal, S . B. 1982. Plant Soil 65, 375-382. Singh, M. M.,and Talibudeen, 0. 1969. J. Rubber Res. Insr. Malay. 21, 240-249. Singh, M. M., and Talibudeen, 0. 1971. Proc. Int. Symp. Soil Ferr. Eval. New Delhi 1, 85-95. Sivasubramaniam, S., and Talibudeen, 0. 1972. J. Soil Sci. 23, 163-176. Sparks, D. L., and Jardine, P. M. 1981. Soil Sci. SOC.Am. J . 45, 1094-1099. Sparks, D. L., and Liebhardt, W . C. 1981. Soil Sci. SOC.Am. J . 45, 786-790. Sposito, G. 1977. Soil Sci. Soc. Am. J . 41, 1205-1206. Sposito, G. 1981a. In “Chemistry in the Soil Environment,” (M. Stelly, ed.-in-chief), pp. 13-29. Am. Soc. Agron., Madison, Wisconsin (Am. SOC. Agron. Spec. Publ. No. 40). Sposito, G. 1981b. “The Thermodynamics of Soil Solutions.” Oxford Univ. Press (Clarendon), Oxford. Sposito, G., and Mattigod, S . V. 1979. Clays Clay Miner. 27, 125-128. Talibudeen, 0. 1971. Proc. Int. Symp. Soil Fert. Eval. New Delhi 1, 97-103. Talibudeen, 0. 1972. In “Potassium in the Soil,” pp. 97-112. Int. Potash Inst., Berne. Talibudeen, 0. 1973. Rep. Prog. Appl. Chem. 58, 402-408. Talibudeen, 0. 1974. Soils Fen. 37, 41-45. Talibudeen, 0. 1981. In “The Chemistry of Soil Processes” (D. J. Greenland and M.H. B. Hayes, eds.), pp. 115-1 17. Wiley, Chichester. Talibudeen, O., and Goulding, K. W. T. 1983a. Clays Clay Miner. 31, 37-42. Talibudeen, O., and Goulding, K. W. T. 1983b. Clays Clay Miner. 31, 137-142. Talibudeen, O., and Page, M. B. 1978. F A 0 Soils Bull. 37, 88-99. Talibudeen, 0..and Weir, A. H. 1972. J. Soil Sci. 23,456-474. Talibudeen, O., Goulding, K. W.T., Edwards, B. S., and Minter, B. S . 1977. Lab. Pracr. 26, 952-955. Thomas, G. W. 1977. Soil Sci. Soc.Am. J. 41, 230-238. Thomas, H. C. 1965. In “Plant Nutrient Supply and Movement,” pp. 4-19. IAEA, Vienna. (IAEA Tech. Rep. Ser. No. 48). Thompson, H. S . 1850. J . R . Agric. SOC.Engl. 11, 68-74. Tinker, P. B. 1964a. J . Soil Sci. 15, 24-34. Tinker, P. B. 1964b. J. Soil Sci. 15, 35-41. Udo, E. J. 1976. J. WestAfr. Sci. Assoc. 21, 241-248. Udo, E . J. 1978. Soil Sci. Soc.Am. J . 42, 556-560. van Bladel, R. 1967. Potash Rev. (Berne). Subj. 4, 38th Suite, pp. 1-12.

264

KEITH W. T. GOULDING

van Bladel, R. 1972. I n “Potassium in the Soil,” pp. 89-%. Int. Potash Inst., Beme. van Bladel, R., and Laudelout, H. 1%7. Soil Sci. 104, 134-137. van Schouwenberg, J. Ch., and SchufTelen, A. C. 1%3. Neth. J. Agric. Sci. 11, 13-22. Vanselow, A. P. 1932. Soil Sci. 33,95-113. Walsh, T. 1972. I n “Potassium in the Soil,” pp. 141-144. Int. Potash Inst., Beme. Way, J. T. 1850. J. R. Agric. Soc. Engl. 11, 313-379. Way, J. T. 1952. J. R. Agric. Soc. Engl. 13, 123-143. Weast, R. C. 1971. “Handbook of Chemistry and Physics.” Chemical Rubber Co., Cleveland, Ohio. Weir, A. H. 1965. Clay Miner. 6, 17-22. Wiedenfeld, R. P., and Hossner, L. R. 1978. Soil Sci. Soc. Am. J . 42,709-712. Wild, A., and Keay, J. 1964. J . Soil Sci. 15, 135-144. Wilson, G. M. 1964. J. Am. Chem. Soc. 86,127-130. Woodruff, C. M. 1955a. Soil Sci. Soc. Am. Proc. 19, 36-40. Woodruff, C. M. 195%. Soil Sci. Soc. Am. Proc. 19, 167-171.

ADVANCES IN AGRONOMY, VOL. 36

HERBICIDE ANTIDOTES: DEVELOPMENT, CHEMISTRY, AND MODE OF ACTION Kriton K. Hatzios Department of Plant Pathology and Physiology Virginia Polytechnic Institute and State University Blacksburg, Virginia

I. Introduction ................................................... 11. Development of Herbicide Antidotes ............ ..................... A. History ............................. B. Search for Herbicide Antidotes ...................................... 111. Chemistry of Herbicide Antidotes. ...........

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

al, and Toxicological Properties C. Analytical Procedures.. ............................................ IV. Field Performance of Herbicide Antidotes ........ .................. A. Applications of Herbicide Antidotes .................................. B. Factors Affecting Field Performance of Herbicide Antidotes .............. C. Adverse Effects of Herbicide Antidotes ............. V. Mode of Action of Herbicide Antidotes.. .................................. A. Mode of Antidotal Action of 1$-Naphthalic Anhydride. ................. B. Mode of Antidotal Action of R-25788 ................................ C. Mode of Antidotal Action of CGA-43089 .... .................... VI. Degradation of Herbicide Antidotes in Plants ............................... VII. summary ............................................................ References ................. ........................... .....

265 270 270 273 292 292 294 296 296 2% 299 300 30 1 302 304 308 309 310

310

1. INTRODUCTION Organic herbicides represent the most effective weapon available to farmers worldwide in their war against weeds. A fundamental reason for the widespread use of these chemicals in modern agriculture is their ability to control selectively a wide spectrum of weeds in a variety of crops. Generally, however, the differential response of plant species to a given herbicide applied at normal field rates is a relative rather than an absolute characteristic. Any herbicide is selective to a particular crop only within certain limits imposed by the complex interactions 265

Copyright 8 by Academic F’ress, Inc. All rights of reprodunion in any form reserved. ISBN 0-12-000736-3

266

W O N K. HATZIOS

between plants (crops and weeds competing with them), their environment, and the herbicide (Ashton and Harvey, 1971). Many of the currently available herbicides that are useful in managing difficult-to-control weeds are not sufficiently selective. In addition, the selective control of weeds that are botanically related to crops has always been a problem. To overcome these problems several approaches have been tried, with varying degrees of success. One approach, of course, is to develop new herbicides that are more selective than those currently available. Although this approach has been quite successful in the past, in recent years the high cost of developing new herbicides has imposed limits. It is encouraging, however, that some new compounds with strong activities and excellent selectivities have been introduced (Sanders, 1981). A second approach to overcoming the problem of the limited selectivity of some currently available herbicides is to confer crop tolerance on these herbicides, which can be achieved mechanically, genetically, or chemically. The mechanical approach of conferring crop tolerance on nonselective herbicides requires applying these chemicals in such a way as to avoid or minimize their contact with susceptible crops. Successful techniques to achieve this include directed sprays or critical timing of herbicide application to weeds prior to crop emergence. The success of this technique is reflected in the fact that for the past 10-15 years almost 70% of the area treated with herbicides in the United States employed preemergence rather than postemergenceherbicide applications (Adler et al., 1977). Conferring crop tolerance on nonselective herbicides genetically has challenged plant breeders for many years. To date, however, the success of this approach has been limited, although the discovery and study of weed biotypes resistant to triazine and bipyridylium herbicides has generated new interest in this field (LeBaron and Gressel, 1982). The potential use of plant breeding techniques for conferring crop tolerance on herbicides has been reviewed by Gressel et al. (1978), Gressel (1980), and LeBaron and Gressel (1982). Finally, an approach that has attracted considerable interest is based on the concept of enhancing crop tolerance to nonselective herbicides chemically with the use of herbicide antidotes.. According to Hoffman (1962), who introduced the idea of chemical enhancement of crop tolerance to herbicides, the appropriate use of herbicide antidotes could permit (1) the use of higher rates of herbicides with marginal selectivities resulting in more effective weed control; (2) the use of nonselective herbicides for selectiveweed control; (3) the use of herbicides under conditions where crop damage is liable to occur, such as with susceptible crop varieties or adverse weather or soil conditions; and (4) the protection of valuable crops that have been accidentally treated with a nonselective herbicide. In addition, herbicide antidotes could also be used as potential tools for elucidating sites and mechanisms of herbicide action.

HERBICIDE ANTIDOTES

267

The term antidote is used routinely in pharmacology to describe a remedy that counteracts the effects of a poison. The term herbicide antidote, as used by agronomists and weed scientists, refers to a chemical agent that selectively protects crop plants from herbicide injury without protecting weeds. In general, a drug antidote counteracts the action of a drug on humans at an internal site. In contrast, the site of the protective action of herbicide antidotes could be external or internal. Because of this, other terms, such as herbicide safener, herbicide antagonist, or crop protectant, have also been introduced and are used interchangeably to describe those chemical agents that enhance crop tolerance to herbicides. Crop protectants that wsrk outside the plant interfere mainly with herbicide absorption by acting as physical barriers or by competing for the sites of herbicide entry. The most successful of the externally acting crop protectants is activated carbon (charcoal), which has been used as a crop protectant against injury from soil-applied herbicides for a long time [for reviews, see Blair et al. (1976), Clapp (1974, 1975), and Gupta (1976)l. Activated carbon exhibits a great adsorptive capacity and acts as a physical barrier to herbicide uptake. In many publications, activated carbon has been referred to as a herbicide adsorbent rather than as a herbicide antidote because of its physical external action. Other examples of crop protectants that act externally by adsorbing herbicides include PC-671 (a formulated lignin by-product) and humic acids. Both of these polymeric substances have been partially successful in protecting soybeans against injury from metribuzin (Baumley et al., 1981; Mahoney and Penner, 1981) and atrazine (Dell’Agnola et al., 1981), respectively. The majority of chemical herbicide antidotes, however, counteract herbicide injury by working inside the plant rather than interfering with herbicide entry into the plant. To date, five compounds are marketed commercially as herbicide antidotes: naphthalic anhydride (NA), R-25788, CGA-43089, CGA-92194, and MON-4606. Chemical herbicide antidotes have been the subject of several earlier reviews (Blair et al., 1976; Gupta, 1976; Hoffman, 1978a,b; Gressel et al., 1982; Parker, 1983). In addition, specific aspects of herbicide antidotes have been the subject of two symposia organized by the American Chemical Society (173rd National Meeting, New Orleans, Louisiana, March 24, 1977) and the International Congress of Pesticide Chemistry (5th Meeting, Kyoto, Japan, August 29-September 4, 1982). The first of these two symposia resulted in the publication of the first book on herbicide antidotes (Pallos and Casida, 1978). The purpose of this article is to consolidate more recent contributions to our knowledge and understanding of the chemistry and development of herbicide antidotes, their practical applications, and the theories proposed about their mode of antidotal action with particular reference to the five chemicals mentioned. The herbicides and herbicide antidotes, designated by common names approved by the Weed Science Society of American or the British Standards Institution, are

KRITON K. HATZIOS

268

Table I Chemical Names of Herbicide Antidotes and Herbiades Mentioned by Common Name in the Text Common name or code name

AD-67 CDAA CGA-43089 CGA-92194 MON-4606 NA R-25788 R-28725 (AD-2) R-29148 s 4 9

Chemical name Herbicide antidotes N-Dichloroacetyl-1-oxa4aza-spiro-4,5decane N,N-Diallyl-2-chlolXm%mnl ‘& a-[(C yanomethoxy)imino]benzeneacetonitrile a-( 1,3-Dioxolan-2-yl-methoxy)iminobenzeneacetonitrile Benzyl-2-chloro-4-(trifluommethyl)-5-thiazole carboxylate 1,8-Naphthalic anhydride N,N-Diallyl-2,2-dichl~ceichloroacetamide 3-@ichloroacetyl)-2,2-dimethyl- 1,3-oxazolidine

2,2-Dmethyl-6-methyldichlomacetyloxaz~lidhe 4-Chloro-2-hydmxyiminoacetanilide Herbicides

Acetochlor Alachlor Amitrole Asulam Atrazine Barban Bensulide

Butachlor Butam Buthihle Butylate CDEC Chl-uat (CCC) Chlomitrofen Chlorsulfuron Cisanilide Cycloate DCPA Diallate Diclofop methyl Diethatyl Dimefuron Diphenamid Diuron Dowco 221 Eprom

2-Chloro-N-(ethoxymethyl)-6’-ethyl-~~totoluidide 2-Chloro-2‘ ,6’-diethyl-N-(methoxymethyl)acetanilide 3-Amino-s-triazole Methyl sulfanilylcarbamate

2-Chloro-4-(ethylamino)-6-( isopropylamino)-s-triazine 4-Chloro-2-butynyl-m-chlomarbanilate 0,O-Diisopropyl phosphomdithioate S-ester with N-(2-mercaptoethyl)benzenesulfonamide N-(Butoxymethyl)-2thlm-2‘,6’-diethylacetanilide 2,2-Dimethyl-N-(1-methylethyl)-N-(phenylmethy 1)propanamide 3-[5-( 1,l-Dimethylethyl)-l,3,4-thiadiazol-2-yl]-4-hydroxy-l-methyl-2-imidazolidmone S-Ethyl diisobutylthiocarbamate 2-Chlomallyl diethyldithiocarbamate 2-Chloroethyltrimethylammoniumchloride 4-Nitrophenyl 2,4,6-trichlorophenyl ether 2-Chloro-N-[ [(4-methoxy-6-methyl- 1,3,5-triazin-2-yl)ano]carbonyI]benzenesulfonamide cis-2,5-Dimethyl-N-phenyl1-pyrrolidinecarboxamide S-Ethyl N-ethylthiocyclohexanecarbamate Dimethyl tetrachlomtemphthalate S-(2,3-Dichloroallyl)diisopropylthiocarbamate 2-[4-(2,4-Dichlomphenoxy)phenoxy]pmpanoic acid methyl ester N-(Chloroacetyl)-N-(2,6-diethylphenyl)glycine 4-[2-Chloro-4-(3,3dimethylureido)phenyl]-2-rert-butyl-l,3,4-oxadiazolin5-one N,N-Dimethyl-2,2-diphenylacetamide 3-(3,4-Dichlomphenyl)-l,1-dimethylurea a-(2,2,2-Trichlomethyl)styrene N-Ethyl-N-propyl-3-propylsulfonyl1,2,4-triazole-1-carboxamide

HERBICIDE ANTIDOTES

269

Table I Continued Common name or code name EPTC Ethofumesate Fluazifop-butyl Glyphosate H-26910 Linuron MBR- 18337 MCPA Mefluidide Metolachlor Metribuzin Molinate Nitrofen NP 55 Paraquat Pebulate Pendimethalin Perfluidone Propachlor Propanil SD-58525 SD-91779 Sethoxydim Swep Terbutol Thiobencab Triallate Trifluralin UBI-S734 Vemolate Xylachlor 2,4-D 2,4,6-T

Chemical name S-Ethyl dipropylthiocarbamate (~)-2-Ethoxy-2,3-dihydro-3,3-dimethyl-5-benzofl methane sulfonate Butyl 2-[4-[5-(trifluoromethyl-2-pyridyloxy)]phenoxy]-proprionate N-(Phosphonomethy1)glycine N-Chloroacetyl-N-(2-methyl-6-ethylphenyl)glycine isopropyl ester 3-(3,CDichloropheny1)- 1-methoxy-1-methylurea N-[4-(Ethylthio)-2-(trifluoromethyl)pheny~]me~anesu~fonamide [(4-Chloro-o-tolyl)oxy]acetic acid N-[2,4-Dimethyl-5-[[(trifluoromethyl)sulfonyllamino]phenyl]-ace~ide 2-Chlbro-N-(2-ethyl-6-methylphenyl)-N-(2-methoxy1-methylethyl)acetamide 4-Amino-6-fert-butyl-3-(methylthio)-as-triazin-5(4~)-one S-Ethyl hexahydro-1H-azepine-1-carbothioate 2,4-Dichlorophenyl p-nitrophenyl ether 2-[N-Ethoxyamino)butyylidene]-5-(ethylthiopropy~~yclohexan1,3-dione ion 1,l ‘-Dimethyl4,4’-bipyridiurn S-Propyl butylethylthiocarbamate

N-(l-ethylpropyl)-3,4-dimethyl-2,6-dinitrobenzamine I , 1,I-Trifluoro-N-[2-methyl-4-(phenylsulfonyl)phenyl]methanesulfonamide 2-Chloro-N-isoprop ylacetanilide 3’ ,4’-Dichloropropionalide Experimental herbicide; chemistry has not been released Experimental herbicide; chemistry has not been released 2 4 l-(Ethoxyimino)butyl]-5-[2-(ethylthio)propyl]-3-hydroxy-2-cyclohexene-1one Methyl N-(3,4-dichlorophenyl)carbamate 2,6-Di-rert-butyl-p-tolylmethylcabamate S- [(4-Chlorophenyl)methyl]diethylcarbamothioate

S-(2,3,3-Trichloroallyl)diisopropy1thiocarbamate u,a,a-Tnfluoro-2,6-dinitro-N,N-dipropyl-p-toluidine 2 4 [ 1-(2,5-Dimethylphenyl)ethyl]sulfonyl]pyridine N-oxide S-Propyl dipropylthiocarbamate 2-Chloro-N-(2,3-dimethylphenyl)-N-( 1-methylethy1)acetamide 2,4-Dichlorophenoxyacetic acid 2,4,6-Trichlorophenoxyaceticacid

identified by chemical names in Table I. Botanical names of crops or weeds used, those of the Terminology Committee of the Weed Science Society of America, are given in Table 11.

270

IUtITON K. HATZIOS Table II Scientifk Names of Plants Mentioned by Common Name in Text Common name

Botanical name

Alexandergrass Alfalfa Barley Barnyardgrass Bean (field) Beet (sugar or red) Bluegrass (Kentucky) carrot Corn (maize) Cotton Flax Foxtail (green) Itchgrass Johnsongrass Lambsquarters (common) Millet (pmso) Oats Oats (wild) Onion Pigweed (redroot) Potato Purslane (common) Rice (cultivated and wild) Rye (common) Ryegrass (perennial) Shattercane Sorghum (grain) soybean Spinach sudangrass Timothy Tobacco Tomato Wheat

Brachiaria plantaginea (Link) A. Hitch. Medicago sativa L. Hordeum vulgare L. Echinochloa crus-gali (L.) Beauv. Phaseolus vulgaris L. Beta vulgaris L. Poa prafensis L. Daucus carota L. &a mays L. Gossypium hirsutum L. Linum usitarissimum L. Setaria viridis (L.) Beauv. Roettboelia enoltata L. Sorghum halepense (L.) Pen. Chenopodium album L. Panicum milliaceum L. Avena sativa L. Avena farua L. Allium cepa L. Amaranthus retroflexus L. Solanum tuberosum L. Portulaca oleracea L. Oryza sativa L. Secale cereale L. Lolium perenne L. Sorghum bicolor (L.)Moench Sorghum bicolor (L.) Moench Clycine mar (L.)Men. Spinacea oleracea L. Sorghum sudanense (piper) Stapf. Phlewn pratense L. Nicotiana tabacum L. Lycopersicon esculerum Mill. Triticum aesrivurn L.

II. DEVELOPMENT OF HERBICIDE ANTIDOTES A. HISTORY

The discovery and development of 2,4-Dduring World War II, and the subsequent introduction of a plethora of other organic herbicides, have revolutionized our approach to weed management in crop production. Quite early in the devel-

HERBICIDE ANTIDOTES

27 1

opment of weed science it was realized that the growth responses of a single plant species to combined applications of two or more herbicides could not always be predicted from the responses of the same plant to each chemical applied individually. Depending on the types of interactions of herbicides applied as a mixture, these responses have been described as synergistic, antagonistic, or independent (Putnam and Penner, 1974). Two chemical components of a mixture are said to interact synergistically when they complement the action of each other such that the total effect of their cooperative action is greater or more prolonged than the sum of the two components taken independently. Antagonism is a term used to describe the opposing action of two or more chemicals such that the action of one is impaired or the total effect of their cooperative action is smaller than the effect of the most active component alone. Two chemicals are said to interact independently when the total effect of their cooperative action is equal to the effect of the most active component of the mixture alone. Understanding antagonistic herbicides interactions is essential for the development of effective herbicide antidotes. It was the observation of the antagonistic interaction of two herbicides that was instnunental in the development of the concept of using chemical antidotes to increase crop tolerance to herbicides. In 1947 Hoffman noticed that 2,4,6-T, an inactive analog of the herbicide 2,4-D, antagonized the epinastic growth caused by sublethal doses of 2,4-D on tomato (Hoffman, 1953). Because of the structural similarity existing between 2,4,6-T and 2,4-D, a competitive inhibition at some common site of action was proposed as an explanation for the observed antagonism. Hoffman continued his efforts in che area of biologically important chemical interactions and in 1960 he reported that the injurious effects of the carbamate herbicide barban on wheat could be antagonized by 2,4-D (Hoffman et al., 1960). Because barban and 2,4-D are structurally different, the hypothesis of a competitive inhibition of these two chemicals at a common site of action was ruled out as an explanation for the observed antagonism. Instead, because 2,4-D accelerates plant growth and barban slows it down, Hoffman (1962) hypothesized that 2,4-D and barban may act at the same site with opposite effects. Indeed, 10 years later Beste and Schreiber (1970, 1972a,b) supported the hypothesis of Hoffman when they showed that whereas EFTC, another carbamate herbicide, inhibited ribonucleic acid (RNA) synthesis, 2,4-D enhanced it, even in the presence of EPTC. Efforts by Hoffman to make practical use of the antagonism of barban by 2,4-D on wheat failed because 2,4-D or other phenoxyacetic acid herbicides were too toxic to be useful as seed dressings, and foliar sprays with 2,4-D would also antidote barban effects on wild oats, a serious weed problem in wheat. Subsequent screening efforts by Hoffman using other chemicals that caused 2,4-D-like symptoms on tomato but were safe as dressings of grass crop seeds met with success. In 1962 Hoffman introduced the concept of chemically enhancing crop tolerance to herbicides by introducing the compound s-449 as an effective chemical antidote against barban injury to wheat (Hoffman, 1962).

272

IUUTON K. HATZIOS

Hoffman continued his efforts and in 1969, 21 years after his initial observation of the 2,4-D antagonism by 2,4,6-T on tomato, introduced the compound 1,8naphthalic anhydride (NA) as an effective protectant of corn against injury from the herbicide EPTC (Hoffman, 1969). 1,&Naphthalicanhydride became the first commercialherbicide antidote and in 1971it was patented by Gulf Oil Chemicals Company (Hoffman, 1971). At the same time, researchers of Stauffer Chemical Company discovered that dichloroacetamidederivatives were effective antidotes of thiocarbamateherbicides on corn, and the compound R-25788 was patented in 1972 as the second commercially developed herbicide antidote (Pallos ef al., 1972,1977). In 1974 the antidotal properties of oxime ethers were discovered in Switzerland by researchers at the Ciba-Geigy Corporation. After extensive field evaluation of these chemicals in the United States, CGA-43089, the most promising of these compounds to protect grain sorghum against injury from the herbicide metolachlor, was patented in 1978 as the third commercial antidote (Martin, 1978). A few years later, Monsanto Company introduced MON-4606,a derivative of 2,4-disubstituted 5-thiazolecarboxylicacid, as a new safening agent for alachlor injury to grain sorghum (Howe and Lee, 1980). MON-4606 is presently being developed as the fourth commercial herbicide antidote. Finally, CGA-92194, a chemical analog of CGA-43089, was introduced by the CibaTABLE III Major Events in the Development of Herbicidal Antidotes

YtW

Event

1947 1960 1%2

Hoffman observed the antagonistic interaction of 2,4,6-T and 2,4-D on tomato Hoffman demonstrated that barban injury to wheat could be antagonized by 2,4,-D Hoffman introduced the concept of a chemical herbicide antidote by introducing the chemical S-449 as an effective antidote of barban injury to wheat Hoffman introduced 1.8-naphthalic anhydride (NA) as an antidote against EFTC injury to corn Pallos er al. discovered R-25788 and other dichloroacetamides as effective chemical antidotes against thiocarbamate injury to corn 1,8-Naphthalic anhydride was patented as the first commercial herbicide antidote (U.S. Patent No. 3,564,768) R-25788 was patented as the second commercial herbicide antidote (Belgian Patent No. 782,120; U.S.Patent No. 4,021,224) Martin discovered the antidotal properties of oxime ethers against chlomacetanilide herbicide injury to grain sorghum CGA-43089 was patented as the third commercial herbicide antidote (U.S. Patent No. 4,070,289) Howe and Lee introduced 2,4-disubstituted 5-thiazolecarboxylic acids as effective antidotes against chlomacetenilide herbicide injury to grain sorghum (US.Patent No. 4,199,506) The Ciba-Geigy Corporation introduced CGA-92194 as an effective antidote against metolachlor injury to grain sorghum (U.S. Patent No. 4,269.775)

1%9 1970 1971 1972 1974 1978 1980 1982

HERBICIDE ANTIDOTES

273

Geigy Corporation (Dill et al., 1982) as the fifth commercially developed herbicide antidote. CGA-92194 is also a grain sorghum protectant against metolachlor injury. A brief summary of the major events that played a key role in the development of herbicide antidotes is given in Table III. B.

SEARCH FOR

HERBICIDE' ANTIDOTES

I . Important Considerations In general, the process of discovering and developing effective herbicide antidotes resembles very much that of the commercial development of herbicides. Both of these processes are quite lengthy and very expensive, primarily because of the increased governmental clearance requirements regulating pesticide registration. The use of random screening techniques has been long recognized as the preferred approach of the herbicide industry in finding and evaluating candidate chemicals as herbicides. The selection of candidate chemicals for inclusion in screening tests evaluating herbicidal activity can be based on three main methods, known as the empirical, imitative, and rational methods (Saggers, 1976). The empirical method is based on experiment and observation and includes the synthesis or acquisition of a large number of novel compounds with unknown biological properties, which are tested for possible activity. Commonly, one compound for each 12,000 or more screened is developed commercially as a herbicide (Krzeminski and Ryan, 1980). However, this method is very popular in spite of its low ratio of success because of its excellent chances for exclusive patentability of discovered active compounds. The imitative method is based on the synthesis of derivatives or analogs of existing compounds with known biological activity and selectivity. Obviously, this method has a much higher ratio of success than the empirical method, but its potential for exclusive patentability of discovered active chemicals is very limited. The third method, known as the rational method, is based on the selection of compounds that have been specifically synthesized to interfere with a desired biochemical or physiological plant process. Application of this method in herbicide development has been very limited. Random screening techniques based primarily on the empirical and to a lesser extent on the imitative method of selecting candidate chemicals also have been instrumental in the commercial development of herbicide antidotes. Herbicide antidoting, however, is very much dependent on the specific interactions of three main factors; the crop to be protected, the herbicide to be antidoted, and the potential antidote. Furthermore, because a desirable screening program for candidate antidotes must be economical, the program has to be selective as to the crops and herbicides that need to be considered. The primary screening tests for herbicide antidotes should involve most combinations of important herbicides,

274

W O N K. HATZIOS

possible antidotes, and major crops. On a worldwide basis, the major crops of importance include corn, wheat, soybeans, rice, grain sorghum, cotton, barley, oats, rye, sugar beets, potato, and alfalfa. A number of selective herbicides are already available for controlling problem weeds in these crops and there is no doubt that new ones will be developed in the future. However, as the weed complexes affecting any given crop are in continual change and new weed problems develop as existing problems are solved, the need for alternative herbicides to deal with these new weed problems continues. Alternative herbicides to control weeds that have developed resistance to triazine herbicides are very much needed at the present time for weed control programs in corn. Weeds that bear a very close botanical relationship to a given crop have always been difficult to control because the existing herbicides that are effective in controlling them are frequently injurious to the respective crops. The problem is exemplified by our limited success in the chemical control of wild oats in cultivated oats, shattercane in cultivated grain sorghum, and wild rice in cultivated rice. Herbicides and crops in this kind of situation need to be seriously considered for inclusion in screening programs of candidate antidotes. Most of the success in developing herbicide antidotes has been with antidoting such herbicides on several grass crops (Blair et al., 1976; Pallos and Casida, 1978; Parker, 1983). Candidate antidotes that are effective in protecting one or more major crops against one or more important herbicides are identified in the so-called primary antidote screen. This process includes laboratory and greenhouse multiple crop-multiple herbicide screening assays and a large number of candidate antidotes. The antidotes that show promise in the primary screen are further evaluated for their practical value under field conditions in the so-called secondary antidote screen. The most important properties of an antidote that are considered during this stage are selectivity, optimum rates for antidotal activity, antidote-toherbicide dosage ratios, suitability of active material for practical fomulations, and reliability of the antidote under field conditions. A herbicide antidote is said to be selective when it counteracts herbicides only on crop plants and not on weed species. In practice, the selectivity of herbicide antidotes is primarily the result of a selective placement, which usually is the coating of crop seeds with the antidote. Thus, coating of corn and grain sorghum seeds with the herbicide antidotes NA and CGA-43089 offers sufficient protection to these crops against injury from the herbicides EFTC and metolachlor, respectively, without protecting any weeds. In some cases, however, the selectivity of a herbicide antidote could be the result of a very specific crop-herbicide-antidote interaction such as occurs with corn-EFTC-R-25788. Broadcast application of the antidote R-25788 offers good protection against EFTC injury only to corn and not to any other grass or broad-leaved weeds that are present in the field (Stephenson and Chang, 1978). Eventually, it is the suitability of a candidate antidote for practical use in the field that determines whether or not this compound will be further developed commercially. Several of the important considerations for the evalua-

HERBICIDE ANTIDOTES

275

tion and development of candidate antidotes are discussed briefly in the next section.

2. Screening of Candidate Antidotes a. Antidotes for Chloroacetanilide and ThiocarbamateHerbicides. The first empirical screening tests of candidate herbicide antidotes were conducted by Hoffman (1962, 1978a), who found that with the exception of herbicides that inhibit photosynthesis at photosystem II, most of the important herbicides could be antidoted to some extent on some crop. By using multiple crop-multiple herbicide screening assays, Hoffman showed that the tolerance of all grass crops to chloroacetanilideor thiocarbamate herbicides could be enhanced chemically to some extent. In particular, the chloroacetanilide herbicide alachlor could be antidoted on many grass crops, including rice, grain sorghum, wheat, oats, barley, and rye, by more antidotes than any other available herbicide (Hoffman, 1978a). The highest ratio of success in Hoffman’s screening assays, however, was obtained with antidoting the thiocarbamate herbicide EFTC on corn. Hoffman screened over 4000 chemicals as candidate antidotes and observed that 40% (1600) of these chemicals could antidote EPTC injury to corn to some extent when applied as seed treatments (Hoffman, 1978a). The great majority of these chemicals were identified as amides or as ketone, acid, or amhe derivatives. Further evaluation led to the discovery and commercial development of NA as an antidote against EPTC injury to corn (Hoffman, 1969, 1971). In early studies (Giineyli, 1971), it was found that broadcast applications of NA to the soil were not practically effective because high rates (28 kg/ha) were required for antidoting EPTC injury to corn. Application of NA to the soil offered good protection to many weeds such as green foxtail (Stephenson and Chang, 1978). Therefore, in order to be selective NA must be applied as a seed treatment. Treatment of corn seeds with 5 g NA/kg of seed (0.5%) appears to be the optimum rate for antidotal activity, although rates as high as 2% have been reported as effective (Blair et al., 1976). As shown in Table IV, NA as a seed dressing is a very versatile herbicide antidote exhibiting limited botanical or chemical specificity; it can offer complete protection to corn, grain sorghum, rice, oats, and wheat against a number of herbicides such as the thiocarbamates, chloroacetanilides, barban, and perfluidone. In addition, NA is capable of offering partial protection to several grass crops against injury from a number of herbicides shown in Table IV. In practice, NA is not effective in protecting broad-leaved crops against herbicide injury. However, seed treatment with NA has resulted in partial protection against EPTC injury to field beans (Blair, 1979); DCPA, diphenamid, and trifluralin injury to tomato (Blumenfeld et al., 1973); and cisanilide injury to cotton (Holm and Szabo, 1974). In Tables IV-VIII, which summarize the main crop-herbicide

KRITON K. HATZIOS

276

Table N Eft'icacy of the Antidote NA as a Crop Protectant Pgoiost Herbicide Iqjury crops protected Herbicides counteracted

Complete protection

Reference

Partial protection

Reference

ChloroaCetgnilides

Alachlor

Sorghum

Rice

Catizone (1979); Hahn (1974); Jordan and Jolliffe (1971); Rains and Fletchall (1971); da Silva (1978); Spotanski and Bumside (1973); Truelove and Davis (1977); Whitwell and Santelman (1975) de Andrade (1981); Parker and Dean (1976)

Corn

Sorghum Kentucky bluegrass Timothy Bertges (1977)

Acetochlor

Corn

Butachlor

Sorghum Sorghum

Rice Diethatyl

Corn

H-26910 Metolachlor

Sorghum

Rice

Leavitt and Penner (1978a) Eastin (1972) Bertges (1977)

Leavitt and Penner (1978a) Eastin (1972) Ali and Mercado (1980) Parker and Dean (1976); Wirjahardja and Parker (1977) Leavitt and Penner (1978a) Leavitt and Penner (1978a)

Truelove and Davis (1977) Parker and Dean (1976) Bertges (1977) Kentucky bluegrass Bertges (1977) Timothy

Propachlor

Thiowbamates

m

Corn

Ahle and Cozart (1972); Burnside et al. (1971); Chang er al. (1973b); Giineyli (1971); Gupta and Niranwal(l976); Hoffman (1969); Jeffery and Connel (1973); Leavitt and

Jeffery et al. (1971); Lee et ~ l (1974b); . Wicks et ~ l (197lb) . Blumenfeld et al. Sorghum (1973) Field beans Blair (1979) Corn

277

HERBICIDE ANTIDOTES

Table IV Continued Crops protected ~

Herbicides counteracted

Butylate

Complete protection

Corn

Reference

Penner (1978a); Lee er al. (1974a); Peters and Dest (1971); Phatak and Bouw (1974); Rains and Fletchall (1971, 1973); Reeder (1970); Rceth (1973); Schmer et al. (1973); Schwartzbeck and Hoffman (1973); Wicks et al. (1971a) Roeth (1973) Corn

Sorghum

Cycloate Diallate

Sorghum

Mohate

Rice

Thiobencarb

Rice

Triallate

Jordan and Jolliffe Oats (1971) de Andrade (1981); Henry (1972); Parker and Dean (1976); Price and Merkle (1977); Smith (1971); Wirjahardja and Parker (1977) Henry (1972); Wyahardja (1979); Wyahardja and Parker (1977) Oats Wheat Corn

Vernolate Phenylcarbamates Barban

Partial protection

Corn Oats

Blair (1978) Ali and Stephenson

Corn

Reference

Burnside et al. (1971); Jeffery et al. (1971); Wicks et al. (1971a,b) Blumenfeld et al. (1973) Chang et al. (1974b)

Chang et al. (1974b) Blair (1979); Miller and Nalewaja (1980) Lee et al. (1974b); Schmer et al. (1973)

Ali and Stephenson (1979)

KRITON K. HATZIOS

278

Table N Continued crops protected ~~~~~

Herbicides counteracted

Complete protection

Wheat Terbutol

Amides Butam Cisani1ide

~

Reference

~

Partial protection

(1979); Chang et al. (1974b); Thiessen et al. (1980) Miller et al. (1978) Bertges (1977) Kentucky bluegrass Bertges (1977) Timothy Corn

Diphenamid

Corn Cotton Sorghum Tomato

Dinitroanilines Pendimethali

Sorghum

Trifluralin

sorghum

Tomato Miscellaneous Buthidazole Chlorsulfuron

Reference

corn Corn

Sorghum

Rice Barley Wheat DCPA

Tomato

Diclofop-methyl Dimefuron

Corn

Dowco 221 Epnmaz

Rice

Corn Rice

Richardson and Parker (1977) Holm and Szabo (1974) Holm and Szabo (1974) Holm and Szabo (1974) Blumenfeld et al. (1973) Ali and Mercado (1980) Blumenfeld et al. (1973) Blumenfeld er al. (1973) Hatzios and Penner (1980) Hatzios (1983b); Parker (1981); Parker et al. (1980); Richardson er al. (1981) Hatzios and Mauer (1983); Parker er al. (1980) Parker (1981); Parker et al. (1980) Parker et al. (1980) Parker et al. (1980) Blumenfeld et al. (1973) Parker (1981) Richardson and Parker (1977) Parker and Dean (1976) Parker and Dean (1976)

279

HERBICIDE ANTIDOTES

Table IV Continued Crops protected Herbicides counteracted

Complete protection

Reference

Ethofumesate Flumifop-butyl

Partial protection Rice Corn Sorghum

Mefluidide NP 55 Perfluidone

Corn

Corn Corn Sorghum

Sethoxydim

Blair and Dean (1976); Parker (1981) Parker (1981)

Rice

Corn Sorghum

Reference Parker and Dean (1976) Hatzios (1983b); Parker (1981) Hatzios and Mauer (1983) Parker (1 98 1) Parker (1981) Parker and Dean (1976)

Hatzios (1983b) Hatzios and Mauer (1983)

interactions that have been reported for the five commercially developed herbicide antidotes, the term complete protection denotes the protection offered by a given antidote that has been described in the literature as excellent, good, sufficient, significant, or effective, whereas the term partial protection refers to what has been described as moderate, limited, marginal, or low protection. The development of R-25788 and other dichloroacetamide derivatives as effective antidotes against thiocarbamate injury to corn appears to be an example of the application of the imitative method for selecting candidate antidotes. Support for this conclusion comes from the fact that chloroacetamides had been introduced as effective herbicides (Hamm and Speziale, 1956) and the potential of chloroacetanilide compounds as herbicide antidotes had been reported long before the discovery of R-25788 (Hoffman, 1962). In fact, the structural similarity of the antidote R-25788 to the chloroacetamide herbicide CDAA is remarkable, and the potantial activity of CDAA as a herbicide antidote against EPTC or other herbicide injury to corn has been documented (Chang et al., 1973b; Leavitt and Penner, 1978a; Hatzios and Penner, 1980). Of more than 500 N,N-substituted amides that showed initial promise as antidotes against thiocarbamate injury to corn, R-25788 was the most active and best suited for practical application (Pallos et al., 1977). Early studies with field applications of R-25788 showed that this compound was equally effective in protecting corn from EPTC whether applied as a seed treatment at a rate of 0.1% (w/w) or as a tank mixture with EPTC incorporated into the soil (Pallos et al., 1975). Soil applications of R-25788 did not' provide any protection to any grass or broad-leaved weeds

280

KRITON K. HATZOS

(Stephenson and Chang, 1978). Thus R-25788 exhibits a high degree of botanical specificity, being particularly effective only as a protectant of corn (Table V). Apart from its botanical selectivity, R-25788 is also chemically selective, as it is particularly effective in antidoting thiocarbamate and chloroacetanilide herbicides on corn (Table V). The chemical selectivity of R-25788 may be a result of the structural similarity between the antidote and the thiocarbamate and chloroacetanilide herbicides (Leavitt and Penner, 1978b). However, R-25788 has also been r e p r k d to effectively antidote barban and perfluidone on corn (Table V), and additionally to partially protect some grass crops against injury from a number of different herbicides (Table V). In practical terms, R-25788, like NA, is not effective in protecting broad-leaved crops from herbicide injury, but some preliminary studies showed that R-25788 was partially effective in protecting field beans against EPTC (Blair, 1979) and tomato against EFTC, cycloate, diphenamid, and trifluralin (Blumenfeld et al., 1973). Apart from the N,Ndisubstituted dichloroacetamides such as R-25788, other chemicals effective as thiocarbamate antidotes on grass crops include derivatives of N-dichloroacetyl-l,3-oxazolidine(Spotanski and Burnside, 1973, Leavitt and Penner, 1978a; Godig et al., 1982), phosphorus-containing compounds (Pallos and Baker, 1977), sulfide derivatives (Ameklev and Baker, 1977), thiobenzoic acid derivatives (Anonymous, 1977b), substituted azepines, diazepines, azabicycloalkanes and piperazines (Anonymous, 1977a), and the fungicide carboxin (Miller and Nalewaja, 1980). In particular, the N-dichloroacetyl-l,3-oxazolidinederivatives known by the designations R-29148, R-28725 (or AD-2), and AD-67 have shown promising activity for antidoting thiocarbamate herbicides on corn (Leavitt and Penner, 1978a, Gorog et al., 1982) and chloroacetanilideherbicides on sorghum (Spotanski and Burnside, 1973). Although specific details about the screening method of selecting oxime ether derivatives as candidate antidotes have not been revealed, it is probable that the antidotal properties of these compounds were discovered by means of empirical screening. Further evaluation of a large number of oxime ether derivatives led to the commercial development of CGA-43089 as an antidote against metolachlor injury to grain sorghum. Table VI summarizes the crop-herbicide interactions that have been reported for this antidote. As does R-25788, CGA-43089 exhibits a good degree of botanical and chemical selectivity; it appears to be a specific antidote counteracting chloroacetanilide injury to grain sorghum (Table VI). Good protection of grain sorghum was also offered by CGA-43089 against the herbicides ethofumesate, SD-58525, and SD-91779 (Leek and Penner, 1980; Hardcastle, 1982). CGA-43089 was also partially effective in protecting corn, rice, and grain sorghum from a number of herbicides (Table VI). Preliminary field studies with CGA-43089 showed that preemergence application of the antidote as a tank mixture with the herbicide improved the tolerance of grain sorghum to metolachlor, but the 4:l or greater antidote-to-herbicide ratio re-

28 1

HERBICIDE ANTIDOTES

Table V Efficacy of the Antidote R-25788as a Crop Protectant against Herbicide Injury Crops protected Herbicides counteracted Thiocarbamates EPTC

Butylate

Complete protection

Corn

Corn

Reference

Appleby er al. (1972); Carringer et al. (1974); Catizone (1979); Chang et al. (1972, 1973a,b); Dowler (1973); Elliot and Pumell (1976); Hammerton (1974); Heikes and Swink (1973); Herman et al. (1974); Jeffery and Connel (1973); Kennedy and Krueger (1978); Leavitt and Penner (1978b); Lee et al. (1974b); Martin and Burnside (1982); Meggitt et al. (1972); Michieka et al. (1978); Orr et al. (1978); Pallos er al. (1975); Phatak and Bouw (1974); Purnell and Bracey (1978); Rains and Fletchall (1971, 1973); Sagaral and Foy (1982); Schmer er d. (1973); Smith et d. (1973); Somody et al. (1 978); Wiese et al. (1979); Williams et al. (1973); Wright et 01. ( 1974) Chang et d. (1973a); Heikes and Swink (1973); Martin and Burnside (1982); Meggitt et al. (1972); Michieka et al.

Partial protection

Corn Barley Sorghum Field bean Tomato

Reference

Lee et al. (1974a) Lee et al. (1974a) Chang et al. (1972) Blair (1979) Blumenfeld er al. (1973)

(Continued)

282

KFUTON K. HATZIOS

Table V Continued Crops ptected Herbicides counteracted

Complete protection

Reference

Partial protection

Reference

(1978); Pallos et al. (1975) Cycloate

Corn Sorghum Tomato

Pallos et al. (1975) Blumenfeld et al. (1973) Blumenfeld et al. (1973) Chang et al. (1973a) Parker and Dean (1976)

Diallate MoIinate Pebulate CDEC Triallate

Corn Corn Corn

Vernolate

Corn

Chang et 01. (1973a); Carringer et al. (1974); Heikes and Swink (1973); Michieka et al. (1978); Orr er al. (1978); Pallos et al. (1975); Schmer et aI. (1973); Smith et al. (1973)

Corn Barley

Catizone (1979); Miller and Nalewaja (1980) Lee et al. (1974b) Lee et al. (1974a)

Acetanilides Alachlor

Corn

Chang er al. (1973a); Leavitt and Penner (1978a,b) Leavitt and Penner (1978a) Leavitt and Penner (1978a,b) Leavitt and Penner (1979a,b) Leavitt and Penner (1978a,b)

Barley Sorghum

Rao and Kahn (1975) da Silva (1978)

Corn

Leavitt and Penner (1978b)

Wheat

Miller et al. (1978)

Corn

Hatzios (1983b); Parker (1981); Parker et al. (1980) Parker et al. (1980)

Chang et al. (1973a) Chang et al. (1973a) Chang et al. (1973a)

Corn Rice

Wheat

Acetochlor

Corn

Diethatyl

Corn

H-26910

Corn

Metolachlor

Corn

Phenylcarbamates Barban

Corn

Blair (1978); Chang er a!. (1973a)

Miscellaneous

Chlorsulfuron

Barley

283

HERBICIDE ANTIDOTES

Table V Continued Crops protected Herbicides counteracted

Complete protection

Reference

Partial protection Sorghum Wheat Corn Tomato

Diclofop-methyl Diphenamid Linuron Peffluidone

Sethoxydim

Trifluralin

Corn

Blair and Dean (1976)

Corn Corn Barley Wheat Corn Sorghum Sorghum Tomato

Reference Hatzios and Mauer (1983) Parker er al. (1980) Parker (1981) Blumenfeld er al. (1 973) Chang er al. (1973a) Parker (1981) Parker er al. (1980) Parker er al. (1980) Hatzios (1983b) Hatzios and Mauer (1983) Blumenfeld er al. (1973) Blumenfeld er al. (1973)

quired was not practically efficient (Ellis et al., 1980). Because CGA-43089 improves the tolerance of selected grass weeds such as alexandergrass, itchgrass, Eleusine spp., and proso millet to metolachlor, the desirable method of CGA-43089 application in the field is seed dressing (Nyffeler er al., 1980). A CGA-43089 rate of 1.25-1.5 g/kg of sorghum seed is adequate for protecting sorghum from metolachlor applied at rates as high as 4 kg/ha (Ellis er al., 1980). CGA-43089 applied at rates exceeding 1.5 g/kg of seed was phytotoxic to sorghum (Davidson et af.. 1978; Ellis et al., 1980). To overcome the problem of CGA-43089 phytotoxocity at high application rates, Ciba-Geigy introduced CGA-92194, a chemical analog of CGA-43089, which is safer and more effective as a grain sorghum antidote against injury from the herbicide metolachlor (Dill et al., 1982). In field tests, CGA-92194 was found to be effective as a seed dressing at rates of 0.5-3.0 g/kg of seed against metolachlor injury to many varieties of grain, sweet, and yellow-endosperm sorghum. In addition, weed members of the subfamily Andropogoneae, such as itchgrass, johnsongrass, and sudangrass, were also protected to some extent by CGA-92194 against injury from metolachlor (Dill et al., 1982). Table VII shows that CGA-92194 was partially effective in protecting corn and grain sorghum against injury from herbicides other than the chloroacetanilides.

284

KRITON K. HATZIOS

Table VI Effkacy of the Antidote CGA-43089 as a Crop protectsnt against Herbicide wury crops protected Herbicides wuntecacted

Complete protection

Chloroacetanilides Alachlor sorghum

Partial Reference

protection

Reference

Abemathy and Keeling (1982); Chenault and Wiese (1980); da Silva and Ueda (1981); Foy and Witt (1982); Ketchersid et at. (1981); Leek and Penner (1980); Palmer and Jeffery (1982); Rhodes and Jeffery (1979); Simkins et al. (1980); Stahlman (1979); Winkle et al. (1980)

Acetochlor

Sorghum

Ketchersid and Merkle

Sorghum

Simkins et al. (1980)

(1982)

Metolachlor

Sorghum

Diethatyl

sorghum

Xylachlor

sorghum

Abemathy and Keeling Rice (1982); Boyd et al. (1979); Chang and Merkle (1982); Chenault and Wiese (1980); da Silva and Ueda (1981); Ellis et al. (1980); Foy and Witt (1982); Hardcastle (1979); Ketchersid el al. (1981); Ketchersid and Merkle (1982); Leek and Penner (1980); Palmer and Jeffery (1982); Rhodes and Jeffery’ (1979: Nyffeler et al. (1980); Simkins et al. (1980); Stahlman (1979); Turner et at. (1979); Winkle et al. (1980) Leek and Penner (1980); Simkins et al. (1980) Simkins er al. (1980)

Nyffeler et al. (1980)

HERBICIDE ANTIDOTES

285

Table VI Continued

Crops protected Herbicides counteracted

Complete protection

Reference

Partial protection

Reference

Thiocarbamates EPTC

Sorghum

Chang and Merkle (1982)

Miscellaneous Bensulide

Sorghum

Chang and Merkle (1982) Hatzios and Mauer (1983); Parker (1981) Hatzios (1983b) Parker (1981)

Chlorsulfuron

Sorghum

Corn Diclofop-meth yl Ethofumesate Fluazifopbutyl

Sorghum Sorghum

Leek and Penner (1980) Sorghum Corn Sorghum Sorghum

Perfluidone Sethoxydim

Corn SD-58525 SD-91779 UBI-S734 MBR-18337

Sorghum Sorghum

Hatzios and Mauer (1983) Hatzios (1983b) Parker (1981) Hatzios and Mauer (1983) Hatzios (1983b)

Hardcastle (1982) Hardcastle (1982) Sorghum Sorghum

Chang and Merkle (1982) Chang and Merkle (1982)

Screening assays by Howe and Lee (1980) resulted in the discovery of the antidotal properties of derivatives of 2,4-disubstituted 5-thiazolecarboxylic acids. These compounds are effective in antidoting alachlor effects on grain sorghum and butachlor effects on rice (Howe and Lee, 1980). Of more than 60 chemical analogs synthesized and tested by Howe and Lee (1980), MON-4606 is the only 2,4-disubstituted 5-thiazolecarboxylate that is currently developed as an antidote for alachlor injury to grain sorghum (Schafer et al., 1980). Table VIII summarizes the main crop-herbicide interactions that have been reported for this antidote. MON-4606 appears to be botanically and chemically selective, because it protects only grain sorghum against injury from chloroacetanilide herbicides. Partial protection against chloroacetanilideherbicides was offered by MON-4606 to corn, wheat, and rice (Table VIII), in addition, MON-4606 provided partial

KRITON K. HATZIOS

286

Table VII Eflkacy of the Antidote CGA-92194 as a Crop Protectant against Herbiade Iqjury Crops protected Herbicides counteracted Metolachlor

Chlorsulfumn EPTC and other thiocarbamates Fluazifop-butyl Sethoxydim

Complete protection

Reference

Partial protection

Reference

Corn Sorghum Sorghum

Hatzios (1983b) Hatzios and Mauer (1983) Dill et al. (1982)

Corn Sorghum Corn Sorghum

Hatzios (1983b) Hatzios and Mauer (1983) Hatzios (1983b) Hatzios and Mauer (1983)

Turner et al. (1982); Dill et al. (1982) Sweet sorghum Dill et al. (1982) Yellow Dill er al. (1982) endospenn sorghum Itchgrass DIll et al. (1982) Johnsongrass Dill et al. (1982) sudangrass Dill et al. (1982) Sorghum

protection to grain sorghum and corn against some thiocarbamate herbicides (Schafer et al., 1980; Hatzios, 1982a). MON-4606 was totally inactive in protecting broad-leaved crops such as soybeans, sugarbeets, and cotton against any herbicide or in antidoting the effects of dinitroanilines,triazines, diphenyl ethers, bipyridiliums, phenoxyacetic acids, and glyphosate on any grass or broad-leaved crop examined (Schafer et al., 1980). Preliminary studies in the field showed that the preferred application methods of MON-4606 are seed treatment at 1.25 g/kg of seed or in-furrow application at 0.14 kg/ha (Schafer et al., 1980). Tank mixtures of alachlor plus MON-4606 applied as preplant-incorporated or preemergence treatments did not increase the tolerance of several grass or broadleaved weeds, such as green foxtail, barnyardgrass, redroot pigweed, common lambsquarters,and common purslane, to alachlor (Schafer et al., 1980). However, weeds of the genus Sorghum could be protected by MON-4606 against alachlor injury (Schafer et al., 1981). b. Antidotesfor Other Herbicides. Attempts to antidote diuron or other photosynthesis-inhibiting herbicides were totally unsuccessful although more than 6OOO chemicals were screened as candidate antidotes (Hoffman, 1978a). Attempts by other investigators to antidote photosynthesis-inhibiting herbicides such as the urea or triazine derivativeson several crops have either failed or been

HERBICIDE ANTIDOTES

287

Table VIII Efficacy of the Antidote MON-4606as a Crop Protectant agaioSt Herbicide Injury Crops protected Herbicides counteracted

Complete protection

Chlomacetanilides Alachlor

Sorghum

Acetochlor

Sorghum

Butachlor Metolachlor

Rice Sorghum

Thiocarbamates EPTC and other thiocarbamates

Reference

Abemathy and Keeling (1982); Brinker et al. (1982); Foy and Witt (1982); Gingerich er al. (1981); Howe and Lee (1980); Ketchersid and Merkle (1982); Palmer and Jeffery (1982); Moshier and Russ (1980); Schafer et al. (1980); Worsham (1981) Abemathy and Keeling (1982); Brinker et al. (1982); Ketchersid and Merkle (1982) Howe and Lee (1980) Abemathy and Keeling (1982); Foy and Witt (1982); Howe and Lee (1980); Ketchersid and Merkle (1982); Moshier and Russ (1980); Palmer and Jeffery (1982); Worsham (1981)

Partial protection

Com Wheat Rice

Corn Wheat

Sorghum Corn

Reference

Hatzios (1982a); Schafer er al. (1980) Schafer ei al. (1980) Schafer et al. (1980)

Hatzios (1982a); Schafer et al. (1980) Schafer et al. (1980)

Schafer er al. (1980) Hatzios (1983a); Schafer et al. (1980)

only partially successful. A number of diverse chemicals were tested as candidate antidotes in these studies, including N-knzenesulfonyl carbamate derivatives (Gaughan et al.. 1979), a,o-diaminoalkanes (Okii et al., 1979a-e), diazosulfonates (Phillips and Bhagsari, 1978), ethoxylated aliphatic derivatives (Street et al., 1980a,b), substituted rnethoxyanilides (Pallos, 1977), phenylarsonic acid (Kadous et al., 1977), and the growth retardant chlormequat (CCC) (Kirtland, 1973; Ohali et al., 1979). In addition, the antagonism of urea and

288

KRITON K. HATZIOS

triazine herbicides by dinitroaniline herbicides on some broad-leaved crops has also been reported (Laddlie el al., 1977; Friesen, 1979; Malefyt and Duke, 1981). The herbicides 2,4-D, paraquat, and trifluralin were also among those that Hoffman (1978a) attempted to antidote in his empirical multiple crop-multiple herbicide screening assays. Other than his observation of the antagonism of 2,4D by 2,4,6-T on tomato, Hoffman did not reveal any other chemical that showed promise as a 2,4-D antidote. The antidote-to-herbicideratio in the case of 2,4-6T and 2,4-D antagonism was 1OO:l or greater, and 2,4,6-T was not effective as a tomato protectant against lethal doses of 2,4-D; therefore, 2,4,6-T was not developed further as an antidote (Hoffman, 1978a). Antidoting 2,4-D or other phenoxyacetic acid herbicides on broad-leaved crops has been the concern of other investigators. Some degree of crop injury prevention from sublethal doses of these herbicides has been obtained with a number of candidate antidotes, including a-aminooxyacetic acid (Amrhein and Schneebeck, 1980), substituted akoxycarboxycarbonylmethoxy-2,1,3-benzothiadiazoles (Van Daalen and Daams, 1970), the vitamin riboflavin (Kezeli et al., 1973), naphthenate and cyclohexanecarboxylatepotassium salts (Peirson et al., 1976), and the element titanium (Farkas et al., 1981a,b). Hoffman’s screening tests also revealed that the herbicide paraquat could be antidoted by oxidation-reduction compounds when both the antidote and paraquat were applied as foliar sprays. Thus, ferrous sulfate in a 124:l ratio could antidote paraquat on wheat and oats (Hoffman, 1978a). Seed application of ferrous sulfate provided minimum protection to wheat indicating a low suitability of this antidote for practical uses. Hoffman (1978a) further reported that trifluralin effects on corn could be antidoted in the laboratory but not in soil tests. However, he did not reveal the chemistry of the compound(s) that were active as trifluralin antidotes in his studies. It has been reported that exogenous applications of the vitamin D-(Ytocopherol or of other lipid substances offered partial protection to oats, tobacco, and other crops against trifluralin injury (Hilton and Christiansen, 1972; Camper and Carter, 1974; Christiansen and Hilton, 1974; Huffman and Camper, 1978). In addition, soil-incorporated applications of trifluralin and the amino acids cysteine resulted in a reversal of trifluralin inhibition of secondary root development of corn and cotton (Shahied and Giddens, 1970). Ascorbic acid and 2,3dimercaptopropanol also have been reported as potential antagonists of trifluralin, as they reduced root swelling caused by trifluralin on corn seedlings (Lignowski, 1970; Lignowski and Scott, 1972). The mechanism by which cysteine, 2,3-dimercaptopropanol, and ascorbic acid counteract the inhibiting action of trifluralin is not known. Finally, antagonism of trifluralin effects on cotton by the organophosphateinsecticides phorate and disulfoton has been reported (Arle, 1968; Hassawy and Hamilton, 1971). The vitamin riboflavin also has been implicated in reducing damage to corn

HERBICIDE ANTIDOTES

289

from the herbicide amitrole (Castelfranco et al., 1963); treatments of rice seeds with the fungicide HMI (3-hydroxy-5-methylisoxazole)reduced the phytotoxicity of the herbicides simetryne, nitrofen, chlornitrofen, propanil, and a combination of swep plus MCPA (Ogawa and Ota, 1976). 3 . Structure-Activity Relationships

In addition to being very useful for the detection of antidotal activity of candidate chemicals, the primary screen serves a second important function by providing information that could be used in structure-activity correlation studies. Structure-activity correlations are very important in antidote research because they could provide organic chemists or biologists with useful information about the chemical substituents that are necessary for antidotal activity. However, detailed structure-activity correlations have been conducted mainly with various amide antidotes that protect corn from thiocarbamate herbicide injury (Pallos etal., 1975, 1977; Stephenson et al., 1978, 1979; Dutka et al., 1979). Of several hundred amide derivatives tested as candidate antidotes for thiocarbamate injury on corn, the most effective compounds were either N,N-disubstituted acetamides (Pallos et al., 1975, 1977; Stephenson et al., 1978, 1979) or substituted N-acetyl-l,3-oxazolidines (Dutka et al., 1979; Gorog et al., 1982). Structure-activity correlations with these chemicals revealed that the acetyl group of either the N,N-disubstituted acetamides or the N-acetyl-1$oxazolidines was essential for antidotal activity, which suggested that a transacylation reaction may be involved in the biological activity of these antidotes (Dutka et al., 1979). The biological activity of both of these classes of amide antidotes was found to be enhanced significantly by the introduction of electronegative chlorine atoms into the acetyl group. However, the dichloroacetamides or the dichloroacetyl oxazolidines were more effective antidotes than their monochloro or trichloro analogs (Pallos et al., 1977; Stephenson et al., 1978, Dutka et al., 1979). Comparative studies with N,N-disubstituted dichloroacetamides which differed in the length of the N,N-substituent carbon chains showed that the most active antidotes for a series of thiocarbamate herbicides were the dichloroacetamides most similar to the herbicides with respect to the N,N-disubstituted alkyl groups (Stephenson et al., 1979). Thus N,N-dipropyl-2,2-dichloroacetamide was the most active antidote for EPTC and vernolate on corn, whereas N,N-diisobutyl-2,2-dichloroacetamidewas the most active antidote for butylate. Similar analogs with N ,N-disubstituted, three-carbon chains containing double or triple bonds were less active than N,N-dipropyl-2,2-dichloroacetamide as EF'TC antidotes (Stephenson et al., 1979). These studies, however, were carried out in a soil-free bioassay system. Under field conditions R-25788 was found to be the most active dichloroacetamide antidote suited for practical use (Pallos et al., 1975).

290

KIUTON K. HATZIOS

Structure-activity correlation studies of several derivatives of 2,4-disubstituted 5-thiazolecarbozylic acids have been conducted by Howe and Lee (1980). They reported that of more than 60 2,4-disubstituted 5-thiazolecarboxylic acids examined, those with the highest activity were the thiazole alkyl esters of the thiazolecarboxylic acids that had a chlorine atom in the 2-position and a trifluoromethyl group in the 4-position of the thiazole ring. Detailed studies on structure-activity relationships of oxime ether antidotes have not been reported. Chang and Merkle (1982) showed that CGA-43089 was by far the most active of eight oximes tested as potential antidotes for metolachlor on grain sorghum. The other oximes tested, in terms of decreasing effectiveness as antidotes, were dimethylglyoxime > benzaphenone oxime > pyridine-Zaldoxime > benzoin-aoxime > methyl thioacetahydroxamate > pyridine-Zaldoxime methiodide > 5nitro-2-furancarboxyaldehyde. Furthermore, a study on the structure-activity relationships of oximes as seed safeners of grain sorghum against metolachlor injury revealed that the number of nucleophilic sites in the molecule of an oxime significantly affects its safening activity (Chang and Merkle, 1983). An oxime with two nucleophilic sites in its chemical structure is more effective as a seed safener than is an oxime with only one. Thus, pyridinealdoxime 0-ethers were the most effective and monoximes were the least effective group of several oximes evaluated as sorghum protectants against metolachlor. Dioximes and pyridinealdoximes were intermediate in their safening effect. Although the 0ether form of an oxime often resulted in increasing safening activity, the 0-ether forms of certain oximes were more phytotoxic to sorghum seeds (Chang and Merkle, 1983). 4 . Novel Approaches in Antidote Research

Thus far, most of the success in discovering herbicide antidotes has resulted from random screening tests based primarily on the empirical method of selecting a large number of chemicals as candidate antidotes. A more novel approach, however, could select candidate antidotes by considering available information on modes of herbicidal action or modes of crop tolerance to herbicides. This rational approach calls for an examination of the potential antidotal activity of all chemicals that could stimulate or inhibit any enzymatic or other physiological systems involved in the expression of herbicidal activity on plants or the detoxification of these toxicants by plants. Although our knowledge and understanding of the exact mechanisms of herbicide actions or herbicide degradation by plants are far from complete, recent advances in these fields are helpful in demonstrating the potential application of the rational approach in herbicide antidote development. Some examples are discussed briefly here. Studies by Veerasekaran et al. (1981a,b) showed that the herbicide asulam, a

HERBICIDE ANTIDOTES

29 1

sulfanilamide derivative, exerts its phytotoxicity by inhibiting the biosynthesis of the vitamin folic acid. Exogenous application of 4-aminobenzoic acid, a precursor of folic acid biosynthesis in plants, or of folic acid resulted in the reversal of the inhibitory action of asulam on several plant species such as wheat, wild oats, and carrots (Killmer et al., 1980; Veerasekaran et al., 1981a). Recent advances in our understanding of the mechanism of plant tolerance to bipyridilium herbicides indicate that the tolerance of perennial ryegrass to paraquat is closely associated with increased levels of the enzyme superoxide dismutase, which detoxifies superoxide oxygen (0,- ) produced by the herbicide (Harper and Harvey, 1978). The hydrogen peroxide (H202) that is generated through the action of superoxide dismutase on superoxide oxygen is detoxified by catalase and peroxidase. Theoretically, if a compound stimulating superoxide activity in plants, or a compound with superoxide activity of its own, could be synthesized, it could serve as a crop protectant against bipyridilium herbicides. A report by Youngman et al. (1979) supported this hypothesis by demonstratingthe antidotal activity of D-penicillamine copper complex, a chelate with superoxide dismutase activity, against paraquat injury to flax cotyledons. Although the exact mechanism of glyphosate phytotoxicity is not fully understood at the present time, authors of several studies have proposed that it interferes with aromatic amino acid metabolism (Jaworski, 1972; Haderlie et al., 1977; Duke et al., 1979, 1980; Gresshoff, 1979; Hoagland et al., 1979). Reversal of the inhibitory action of glyphosate by exogenous applications of several aromatic amino acids, such as phenylalanine and tyrosine, has been observed in some of these studies and has been used to support the hypothesis of glyphosate interference with aromatic amino acid biosynthesis (Jaworski, 1972; Haderlie et al., 1977; Gresshoff, 1979). Other investigators have disputed these results by reporting that exogenous applications of aromatic amino acids did not reverse glyphosate action on plants (Duke et al., 1979). Killmer et al. (1981) proposed that glyphosate may act by depleting respiratory substrates, and they demonstrated that exogenous applications of the amino acids aspartate and glutamate, and of organic acids of the tricarboxylic acid cycle (a-ketoglutarate, succinate, or malate), reversed the inhibitory action of glyphosate on plants. Finally, an additional innovative approach to antidote herbicides on crops has been developed by Geller and Nikolaenko (1972), who reported that EPTC injury to sugar beet seedlings could be significantly decreased under greenhouse conditions with the use of the bacterial fertilizers azotobacterin and phosphobacterin applied as seed dressings. Azotobacterin is a preparation of the Azotobacter strain 57, and phosphobacterin is a preparation of phosphate-dissolving bacteria. The potential of other soil microorganisms for reducing the action of specific herbicides on plants has also been reported (Sobiesczanski, 1975; Wegrzyn, 1975).

292

KRITON K. HATUOS

Although these novel approaches in searching for herbicide antidotes have not yet resulted in the commercial development of any antidotes, they should prove useful and perhaps more successful in the future.

Ill. CHEMISTRY OF HERBICIDE ANTIDOTES A. SYNTHESIS The five chemicals that have been developed commercially as herbicide antidotes can be classified chemically as (1) derivatives of acenaphthene (e.g., 1,8naphthalic anhydride); (2) dichloroacetamides (e.g .,R-25788); (3) oxime ethers (e.g., CGA-43089 and CGA-92194);and (4) derivatives of 2,4-disubstituted 5thimlecarboxylic acids (e.g., MON-4606). The chemical names and structures of these five antidotes are given in Fig. 1 . A brief discussion of the chemical methods available for the preparation of these antidotes is presented here. 1. I &Naphthalic Anhydride

1 ,%Naphthalic anhydride, often abbreviated as NA, is prepared commercially by catalytic air oxidation of acenaphthene (Riden, 1976). In the laboratory, NA is synthesized by the oxidation of acenaphthene with various oxidizing agents (Mullison, 1979). 1,&Naphthalic anhydride can be purified by crystallization from nitric or acetic acids.

2 . R-25788 and Other Dichloroacetamides R-25788 is prepared according to the method described by Pallos ef al. (1977). Dichloroacetyl chloride is dissolved in methylene chloride and the solution is cooled in an ice bath. Diallyl amine is added dropwise to the solution while the temperature is maintained below 10°C. The mixture is stirred at room temperature for about 4 hr and then washed twice with water, dried over magnesium sulfate (MgSO,), filtered, and recovered under vacuum distillation as a colorless liquid. A second method for the synthesis of R-25788has been described by Stephenson ef al. (1978). This method is also based on the mixing of diallyl mine and dichloroacetyl chloride but differs from the previous method in the way the reaction is carried out and in the subsequent steps for the recovery of R-25788. Instead of the code name R-25788, the name DDCA has been used in the literature by some investigators in referring to this antidote (Ezra and Gressel, 1982). Either of the aforementioned methods also can be used for the preparation of other dichloroacetamide antidotes if the appropriate acid chloride is mixed with an appropriate amine.

HERBICIDE ANTIDOTES

293

NA 1,8-naphthalic H

CI

anhydride

0 C%-CH=CH2 II C- N CH2- CH =

- C-I

(

CII

CH2

R- 2 5 7 8 8 ~,~-diallyl-2,2-dichloroacetamide

QC-

II

CN

N- 0 - CH2- CN

CGA-43089 a-[(cyanomethoxy)imino]benzeneacetonitrile

0 fl-

CN

N-

0

- CH2

CGA-92194 a-[(1.3-dioxolan-2-yl-methoxy)imino]benzeneacetonitrile 0 II

F ~ F s- r c - o - c H 2 N\

v Cl

MON-4606 5 - t h i a z o l e c a r b o x y l i c acid, benzyl ester, 2 - c h l o r o - 4 - t r i f l u o r o m e t h y l

FIG.1. Chemical structures and names of five commercially developed herbicide antidotes.

294

KRITON K. HATZIOS

3 . CGA-43089 and CGA-92194

CGA-43089 is synthesized as follows (Martin, 1978). Phenylglyoxylonitrile-2-oxime (sodium salt) is suspended in acetonitrile in a sulphonating flask and chloracetonitrile (dissolved in acetonitrile) is added dropwise to the suspension, resulting in a slight increase in temperature. The suspension is subsequently refluxed with stirring for 3 hr, during this process the color of the reaction mixture turns to a light green. After cooling to room temperature, the formed sodium chloride is filtered off and the filtrate is concentrated in a rotary evaporator. The residue is then dissolved in acetonitrile and the solution is stirred with charcoal and filtered until a clear concentration of the filtrate in the rotary evaporator yields the oxime ether. Recrystal;ization of the oxime ether residues from isopropanol yields pure CGA-43089 as a white solid. Oximes exist in two stereoisomeric forms, the syn and the anti forms, and CGA-43089 therefore can exist in both forms and as a mixture of them. In the literature, CGA-43089 is sometimes referred to by the name cyoxmetrinil. CGA-92194, an analog of CGA-43089, is prepared by a similar procedure. 4. MON-4606

The preparation of MON-4606 is based on the reaction of ethyl 2-chloro-4trifluoromethyl-5-thimlecarboxylicacid with benzyl alcohol (Howe and Lee, 1980). The reaction is refluxed for 16 hr and excess benzyl alcohol is removed under reduced pressure. The residue is then dissolved in ether and the ether solution is washed with sodium bicarbonate, dried with MgSO,, and concentrated under reduced pressure. Distillation of the crude residue yields MON-4606 as a white solid. The ethyl 2-chloro-4-trifluoromethyl-5-thiazolecarboxylic acid can be synthesized by two principal methods (Howe and Lee, 1980). The first method is based on the reaction of P-aminoacrylates with chlorocarbonylsulfenyl chloride, whereas the second method is based on the reaction of equimolar quantities of ketoesters (dissolved in chloroform) with sulfuryl chloride. Through a series of specific steps, both methods lead to the preparation of specific 2,4-disubstituted 5-thiazolecarboxylic acids. These steps were described in detail by Howe and Lee (1980). B . CHEMICAL, PHYSICAL, AND TOXOCOLOCXAL PROPERTIES

The chemical, physical, and toxicological properties of the five commercially developed herbicide antidotes are presented in Table IX.With the exception of R-25788, these chemicals are solids with low water solubility and low vapor pressure; R-25788 is liquid with good water solubility and relatively high vapor

Table IX Chemical, Physical, and Toxicological Properties of Commercially Developed Herbiade Antidotes ~

Antidote

h m

1,s-Naphthalic acid R-25788 CGA-43089 CGA-92194 MON-4606

198.2 208.09 185.2 232.24 321.72

Physical state Solid Liquid Solid Solid Solid

Color Light tan Colorless White Colorless White

~~~~~~~

Solubility in water at 20°C (ppm)

Melting point ("C)

c2 5000 95 20

27G274

LOW

-

55-56

77.7 56-58

Vapor pressure (mm Hg) 6x

-

(25°C)

3.9 x 10-6 (20°C) (25°C) 3.8 x

Acute oral LDm in rats (mgkg) 12,300 2,000 2,277 >3,000 7,700

296

KRITON K. HATziOS

pressure. 1,8-Naphthalic anhydride and MON-4606 are relatively nontoxic and R-25788, CGA-43089, and CGA-92194 slightly toxic to rats. C. ANALYTICAL PROCEDURES

Several analytical methods are available for the analysis of crop residues or of the formulations of the currently marketed herbicide antidotes. According to Riden (1976), infrared spectroscopy (IR) is the method of choice when a formulation analysis of 1,s-naphthalic anhydride is needed. The NA residues in crop samples can be detected by gas chromatography using an electron capture detector (Riden, 1976); NA can be extracted with methanol and cleaned up on a column of anion-exchangeresin. W e d descriptions of these techniques can be found in Riden (1976). Residues of R-25788 in plants, soils, and rats can be detected according to the methodology described by Miaulis et al. (1978). R-25788 residues can be recovered from plant tissues by extraction with a mixture of ethanollwater (7:3), filtration, rotary evaporation of the fdtrate, and partitioning of the concentrate between chloroform and water. The aqueous phase is then purified by means of column chromatography, and the R-25788 can be identified by gas chromatography with a flame ionization detector or by means of a combined system of gas chromatography and mass spectrometry (Miaulis et al., 1978). The application of other spectroscopictechniques, such as IR, nuclear magnetic reasonance (NMR) spectroscopy, and mass spectrometry, for the characterization of R-25788 has been described by Stephenson et al. (1978). Analytical procedures for the characterization of the oxime ether antidotes CGA-43089 and CGA-92194 have been described by CibaGeigy.' These procedures include thin layer chromatography (TLC) and gas chromatography with a flame ionization detector. Analytical methodology for residue analysis of the antidote CGA-43089 in soil or sorghum samples has also been described by Egli (1982). Detailed analytical methods for residue analysis of MON-4606 have not yet been described.

IV. FIELD PERFORMANCE OF HERBICIDE ANTIDOTES A. APPLICATIONS OF HERBICIDE ANTIDOTES A summary of the agricultural uses, formulations, and methods of application of the five herbicide antidotes that are presently marketed in the United States is given in Table X. All five antidotes can be applied as seed dressings, but it is 'Ciba-Geigy (1981). Status Report of CGA-92194. Agricultural Division, Basle, Switzerland.

Table X Agricultural Uses, Formulations, and Application Methods of Five Commercial Herbicide Antidotes ~~~~~

Antidote

Herbicide Crop protected counteracted

1,8-Naphthalic acid Corn

R-25788

Corn

Wheat

EPTC, butylate, vemolate EPTC Butylate Vemolate Triallate

Formulation

Trademark

95% seed

PROTECT" protectant powder Tank mixture, ERADICANE" with herbicide SUTAN+@ VERNAM+@ 20% seed protectant powder 2.09 S CONCEP I@

-

CGA-43089

Grain sorghum Metolachlor

CGA-92194

Grain sorghum Metolachlor 70 WP or 50 SD CONCEP 11"

MON-4606

Grain sorghum Alachlor

80 wp 1% Granular

SCREEN"

Application method Seed dressing

Recommended rate 0.5% by weight

Replant 0.56 kgha incorporation 0.25 kgha 0.56 kg/ha Seed dressing 0.5% by seed weight Seed dressing Seed dressing Seed dressing In-furrow application

Storage stability of treated seed Manufacturer

More than 1 year Gulf Oil -

Stauffer

More than 1 year Stauffer

0.125% by More than 1 year seed weight 0.054.3% by More than 1 year seed weight 0.06-0.25% by More than 1 year seed weight 0.07-O.25 kg/ha

Ciba-Geigy Ciba-Geigy Monsanto Monsanto

298

KIUTON K.HATZIOS

preferable to apply R-25788 preplant incorporated into the soil as a tank mixture with thiocarbamate herbicides. Under practical conditions, the application of herbicide antidotes in the field does not involve any extra operation, because the respective companies that manufacture herbicide antidotes market them either as prepackaged tank mixtures with the herbicide, as in the case of R-25788, or as crop seeds dressed with the antidote, as in the case of CGA-43089 and CGA-92194. To be effective in the field, a herbicide antidote that is used as a prepackaged tank mixture with a herbicide must have several features in common with the herbicide. According to Prochnow (1978), the most important of these features are (1) the antidote should not be more soluble than the herbicide, to avoid leaching of the antidote from the zone of protection under conditions of heavy rainfall or irrigation; (2) the antidote should remain in the protection zone as long as the crop is susceptible to the herbicide; (3) the antidote should not interfere with the effectiveness of the herbicide on target weeds; and (4) the antidote should be compatible with other herbicides or insecticides that may be added in the spray tank or applied to the soil control zone when the herbicide plus antidote mixture is present. The tank mixtures of R-25788 with the thiocarbamate herbicides EPTC, butylate, and vernolate are marketed by Stauffer under the trade names Eradicane, Sutan+ , and Vernam+ , respectively. The herbicide-to-antidote ratio is 12:l in Eradicane and Vernam+, and 24:l in S u m + . R-25788 can also be used as a seed treatment on wheat to protect it from triallate injury. 1,8-Naphthalic acid is applied as a seed dressing to corn, but because its effectiveness in protecting corn from thiocarbamate herbicide injury is less than that offered by R-25788 under field conditions (Stephenson and Chang, 1978), its practical use is not as extensive now as it was earlier. In fact, Sanders (1981) revealed that NA is no longer made by Gulf Oil Chemicals Company and that other companies are testing NA for possible licensing from Gulf. All three grain sorghum protectants against chloroacetanilide herbicide injury are applied in the field primarily as seed dressings. In addition, MON-4606is effective as an in-furrow application (Schafer et al., 1980). With seed dressings, the duration of the safening activity in stored seeds is a factor that needs to be considered. Tests with all five antidotes have shown that storage for more than 1 year of crop seeds dressed with antidotes does not result in a loss of safening activity. However, it is recommended that in every planting season, crop seeds should be used that have been recently treated with antidote. Dressing of crop seeds with herbicide safeners could be done in two ways (Muller and Nyffeler, 1981): (1) as a sequential treatment after the seed has received the usual fungicidelinsecticideprotection, or (2) as a “one-shot treatment” together with the usual fungicide/insecticide seed dressing. It is obvious that, in either method, if an antidote is to be effective it should be compatible with the fungicides or insecticides that are also used as seed dressings.

HERBICIDE ANTIDOTES

299

B. FACTORSAFFECTINGFIELDPERFORMANCEOF HERBICIDE ANTIDOTES

Several environmental factors such as temperature, light, soil moisture, and soil type have been reported to influence the efficacy of herbicide antidotes in the field. In addition, the timing of the antidote and herbicide applications as well as varietal differences are also important. The activities of the antidotes NA and CGA-43089 against herbicide injury on crops were higher at high than at low temperatures (Guneyli, 1971; Leek and Penner, 1981). Soil moisture appears to be very important for the antidotal activity of CGA-43089 under field conditions. Ketchersid et al. (1981) reported that CGA-43089 was not effective in counteracting metolachlor injury to grain sorghum under extremely wet conditions in the field, In contrast, Nyffeler et al. (1980) showed that under greenhouse conditions, the safening effect of CGA-43089 was greater in wet soils, when crop injury from metolachlor is greater. Variations in the antidotal activity of CGA-43089 as a result of variations of soil moisture levels also have been reported by Simkins et al. (1979). Muller and Nyffeler (1981) reported that temperature and soil moisture changes do not affect the safening activity of CGA-43089 directly, but rather they influence the response of grain sorghum to treatments with metolachlor. Soil moisture did not appear to affect the antidotal activity of NA (Hahn, 1974), R-25788 (Burt and Akinsorotan, 1976), and MON-4606 (Gingerich ef al., 1981). However, some reports indicated that R-25788 is more mobile than the herbicide EPTC in wet soils (Buzio and Burt, 1977; Burt and Buzio, 1978). This separation of EPTC and R-25788 following the application of Eradicanea to wet soils has been proposed as a possible reason for the sporadic toxicity of this herbicide/antidote mixture on corn (Burt and Buzio, 1979). The EPTC/R-25788 ratio could be restored to normal upon drying of the soil (Buzio and Burt, 1977). Soil type has also been associated with antidote efficacy in the field. For example, NA was highly effective as a barban antidote in oats grown in light sandy soils but was much less effective in silty clay soils (Thiessen et al., 1980). The antidotal activity of NA against barban on wheat was not dependent on changes in soil fertility (Thiessen et al., 1980). Similarly, in light soils R-25788 was less effective as an EPTC antidote on corn (Crook, 1975). The antidotal activity of MON-4606, however, was not highly dependent on soil type because this antidote was equally active on soils with textures ranging from silt loam to muck (Schafer et al., 1981). The timing of the antidote application with respect to the herbicide application can also affect the safening activity; NA was most protective to corn when applied on the same day or 1 day after treatment with EPTC (Guneyli, 1971). Protection of corn by NA was not complete if the antidote was applied 2 or more days after treatment with the herbicide. The influence of critical timing for the

300

KRITON K. HATZIOS

effectiveness of the antidote R-25788 as a protectant of corn against EPTC injury has been demonstrated in a series of experiments conducted by Donald and Fawcett (1976). They found that R-25788 was able to prevent EPTC injury to corn when applied as a soil drench at any time prior to coleoptile emergence. If the antidote was applied to EFTC-treated corn after coleoptile emergence, various degrees of corn injury were evident. Thus when R-25788 was applied after the second and third leaves of EPTC-treated corn had bent and failed to uncurl, it was completely ineffective as an EPTC antidote. If the injured second and third leaves were removed, the higher order leaves still did not expand, even when the antidote was supplied (Donald and Fawcett, 1976). In another study, corn was found to be more susceptible to EPTC plus R-25788 at 4 weeks after planting, indicating that application of the mixture at earlier dates was effective (Burt and Buzio, 1979). Timing appeared to be less critical for the effectiveness of CGA-43089; this antidote, in addition to its effectiveness against preplant-incorporated and preemergence applications of metolachlor, was also effective against early postemergence applications to grain sorghum seedlings (Leek and Penner, 1981). Finally, intraspecific differential responses of several crop cultivars or genotypes to combinations of herbicides and antidotes have been reported. For example, the corn cultivar ‘TXS114‘ was highly tolerant to EPTC at 6.7 kg/ha with or without the antidote R-25788 (Sagaral and Foy, 1982). Furthermore, although R-25788 was able to protect many of the EPTC-susceptible corn cultivars examined in the previous study, it did not alleviate EF’TC injury to ‘XL55’and ‘XL379’ corn cultivars. Similarly, although MON-4606 offered good protection to 12 out of 13 grain sorghum hybrids against alachlor injury, it was not effective in protecting the ‘G623GBR’ hybrid of sorghum (Schafer et al., 1980). Also, despite pretreatment with NA, three cultivars of oats did not recover from postemergence applications of barban (Thiessen et al., 1980). Of the many grain sorghum hybrids tested, no hybrid was found that had insufficient tolerance to metolachlor in the presence of CGA-43089 (Davidson ef al., 1978). In general, the crop tolerance offered by herbicide antidotes against herbicide injury in the early stages of crop growth persists throughout the life of the crop. An exception is seen in the work of Wright et al. (1974), who reported that the protection offered by R-25788 to many corn genotypes against EPTC injury early in the season was greatly reduced, in many of the genotypes, symptoms of EF’TC injury appeared later in the season. C. ADVEFSE

E m s OF HERBICIDE ANTIDOTES

In all investigations reported, treatment of crop seeds with the optimum rates of herbicide antidotes did not result in any adverse phytotoxic effects on crop

HERBICIDE ANTIDOTES

301

growth. However, NA applied alone as a seed dressing to corn at 0.5% by seed weight has been reported to interfere slightly with corn growth (Giineyli, 1971; Hickey and Krueger, 1974a). Eastin (1972) reported that the same treatment of NA (0.5% by seed weight) was injurious to sorghum in the absence of herbicide treatments. Although NA provided some protection to field beans against injury from EPTC when applied as seed dressing at 0.5%, it caused marked chlorosis to the bean foliage (Blair, 1979). Chlorotic effects of bean foliage were also caused by R-25788 in the absence of EPTC when R-25788 was applied as seed treatment at 2% by seed weight (Blair, 1979). Reports on adverse effects of R-25788 to corn or any other grass crop are not available. Adverse effects of CGA-43089 on the seed viability of sweet and yellow endosperm sorghum have been observed (Davidson et al., 1978; Dill et al., 1982), and the application of CGA-43089 as a seed dressing at rates higher than 1.88 gtkg of sorghum seeds was found to be phytotoxic even to grain sorghum (Davidson et al., 1978). In contrast to CGA-43089, its chemical analog CGA-92194 is not phytotoxic to grain sorghum and does not adversely affect the seed viability of sweet or yellow endosjwm sorghum (Dill et al., 1982). Similarly, MON-4606 has not been found to cause any adverse effects on the seed viability or growth of grain sorghum (Schafer et al., 1980, 1981). However, Ketchersid and Merkel (1983) observed measurable effects of the protectants CGA-92194 and MON-4606 on the growth and respiration of sorghum seedlings during imbibition and early stages of germination. These effects of CGA-92194 and MON-4606 were less than those caused by CGA-43089 under similar conditions.

V. MODE OF ACTION OF HERBICIDE ANTIDOTES The exact mechanisms by which the currently available herbicide antidotes protect grass crops against injury from chloroacetanilide and thiocarbamate herbicides are not fully understood. Our limited understanding of the mechanisms of the phytotoxic action of the chloroacetanilideand thiocarbamate herbicides at the biochemical or physiological level is partly responsible for this. Shoot absorption through the emerging coleoptile of grasses has been shown to be more important than root uptake for both of these herbicidal classes (Gray and Joo, 1978; Ashton and Crafts, 1981). The meristematic region of grass shoots also appears to be the site of action of these herbicides and, in general, chloroacetanilideand thiocarbamate herbicides are considered to act as inhibitors of early shoot growth of germinating grass seedlings (Ashton and Crafts, 1981). Following soil applications of thiocarbamate or chloroacetanilide herbicides, susceptible grass seedlings that emerge from the soil are either greatly stunted or seriously deformed, but they are not killed. The symptoms of chloroacetanilide and thiocarbamate

302

KRITON K. HATZIOS

herbicide injury to corn or sorghum seedlings involve leaf or shoot deformations such as leaf twisting or rolling, and at high rates leaves fail to emerge through the coleoptile (Leavitt and Penner 1978b). These effects of chloroacetanilide and thiocarbamate herbicides on grass crops are counteracted by the currently available herbicide antidotes. Advances in our understanding of the modes of action of herbicides antidotes are also complicated by the fact that some of the studies examining the phytotoxic effects of chloroacetanilideand thiocarbamate herbicides on plants, as well as their counteraction by herbicide antidotes, have been conducted with broadleaved plants that are not protected against these herbicides under field conditions. Isolated spinach chloroplasts (Wilkinson and Smith, 1975), red beet disks (Wilkinson and Smith, 1976), and tobacco suspension cultures (Rennenberg et al., 1982) have all been used as plant systems to study the phytotoxic effects of thiocarbamate herbicides and their counteraction by the antidotes NA and/or R-25788. The significance of the results of these studies is difficult to assess when attempting to explain the protective action of herbicide antidotes on grass crops. Extensive research on the mode of antidotal action of herbicide safeners has been conducted primarily with the antidoteR-25788 and to a lesser extent with the antidotes NA and CGA-43089. The findings of all these studies have resulted in a plethora of proposed hypotheses about the mechanisms of the antidotal action of herbicide safeners. Herbicide antidotes could protect grass crops from chloroacetanilide or thiocarbamate herbicide injury by (1) interfering with herbicide uptake and/or translocation in the protected plant, (2) counteracting herbicide phytotoxicity through a competitive inhibition at some common site within the protected plant, (3) stimulatingherbicide degradation by the protected plant, and (4) combinations of mechanisms (1)-(3). The status of our current knowledge on the antidotal action of the herbicide safeners NA, R-25788, and CGA-43089 is discussed briefly in the following sections. Studies on the mode of action of the antidote MON-4606 are not yet available, whereas the antidotal action of CGA-92194 probably resembles that of its chemical analog CGA-43089. A. MODEOF ANTIDOTALA ~ O OF N ~,~-NAP€ITHALIC ANHYDRIDE

Early studies on the mode of antidotal action of NA revealed that its protective effect against alachlor injury to sorghum was physiological in action and not the result of a physical deactivation of the herbicide (Hickey and Krueger, 1974a). Also, it was shown that the counteraction of alachlor effects on sorghum by NA was more apparent in sorgbum shoots than in sorghum roots (Jordan and Jolliffe, 1971). Subsequent reports proposed that the protective effect of NA against metolachlor injury to sorghum was partly caused by decreased herbicide uptake

HERBICIDE ANTIDOTES

303

and translocation (Ahrens and Davis, 1978). In other studies, however, NA was not found to interfere with herbicide uptake and translocation in the protected plants (Murphy, 1972; Holm and Szabo, 1974). Because corn seedlings treated with NA absorbed more ['TIEPTC than nontreated ones, NA was reported to act as a stimulator rather than as an inhibitor of herbicide uptake (Guneyli, 1971). The NA-induced stimulation of [14C]EPTCuptake by corn cell suspensions was found to be concentration dependent (Ezra et al., 1982). In a number of other studies, interference of thiocarbamate and chloroacetanilide herbicides with plant physiological systems has been shown to be counteracted by NA treatments. Thus, the EPTC-induced inhibition of fatty acid synthesis in spinach chloroplasts and red beets disks was reversed by treatments with NA (Wilkinson and Smith, 1975, 1976). Hickey and Krueger (1974b) showed that alachlor alone increased the force needed for leaf emergence from corn coleoptiles, but treatment with NA in a 6:l (antidotelherbicide) ratio reduced this force significantly. Hoffman (1978a) suggested that NA may act by preventing the precocious bud dormancy of corn seedlings which is induced by thiocarbamate and chloroacetanilide herbicides. This hypothesis was supported by the fact that use of other well-known bud-dormancy-breaking chemicals such as 1,Zdibromaethane resulted in the protection of corn from EPTC injury (Hoffman, 1978a). In studies monitoring the loss of 32P from onion roots it was found that the mode of action of metolachlor involved membrane damage and that NA was capable of protecting onion roots from permeability changes induced by this herbicide (Mellis et al., 1982). Because metolachlor, even at high concentrations, did not induce any loss of 3*P from corn roots (Mellis et al., 1982), such a mechanism could not explain the protective action of NA in counteracting metolachlor or other chloroacetanilide herbicide injury to corn under field conditions. An alternative mechanism proposed for explaining the antidotal action of NA suggests an antidote-induced increase in the rate of metabolic detoxification of chloroacetanilide or thiocarbamate herbicides in the protected plants. Guneyli (1971) proposed that NA acts by activating the enzyme system(s) responsible for EPTC breakdown in corn seedlings, whereas Holm and Szabo (1974) reported a marked enhancement in the rate of metabolic breakdown of the herbicide cisanilide in NA-treated corn seedlings. The significance of this mechanism, however, has been disputed by other investigators who failed to detect significant differences in the patterns of herbicide metabolism by NA-treated and nontreated plants (Murphy, 1972; Hahn, 1974). Furthermore, NA was not effective in enhancing the glutathione-S-transferaseactivity and gluthathione content in roots of corn seedlings (Lay and Casida, 1976; Fedtke, 1981). As it will be discussed in the next section, enhancement of the gluthathione content in corn roots has been correlated with the protective effect of the antidote R-25788. From the previous discussion, it is apparent that our understanding of the

304

KRTTON K. HATZIOS

antidotal action of NA is far from complete. In spite of many conflictingreports, NA most probably exerts its antidotal action either through a stimulation of herbicide metabolism or through a counteraction of the phytotoxic effects of herbicides at a common site within the protected plant. The proposal that NA may act by interfering with herbicide uptake and translocation in the protected plants has not gained credence among researchers active in this field. However, because of its limited botanical and chemical specificity, NA may have more than one mode of action or it may act through a combination of all of the proposed mechanisms. Future research is needed for a better understanding of the antidotal action of NA. B. MODEOF ANTIDOTAL ACTIONOF R-25788

In contrast to NA, R-25788 represents a physiologically selective antidote that protects corn from thiocarbamate and chloroacetanilide herbicide injury (Pallos et al., 1975; Leavitt and Penner, 1978b). Extensive research with the antidote R-25788 during the last decade has resulted in many proposed mechanisms for explaining its antidotal action. Early studies on the physiological behavior of R-25788 showed that the coleoptiles of corn seedlings were the common site of uptake and action of both the thiocarbamate and chloroacetanilideherbicides and the antidote R-25788 (Donald and Fawcett, 1976; Gray and Joo, 1978). Because of this, the possibility that R-25788 may protect corn from herbicide injury by preventing the uptake and translocation of chloroacetanilide and thiocarbamate herbicides has been examined. On the basis of present evidence, R-25788 does not appear to prevent EPTC injury to corn seedlings by inhibiting herbicide uptake or by altering herbicide translocation in corn (Chang et al., 1974a; Marton et al., 1978; Sagaral, 1978). On the contrary, an antidote-induced stimulation of EPTC uptake has been reported to occur in some corn cultivars that were treated with EPTC and R-25788 (Carringer et al., 1974; Sagaral, 1978). It was shown, however, that simultaneous applications of R-25788 and [14C]EPTCto corn cell suspensions resulted in a rapid reduction of EPTC uptake by the corn cells (Ezra et al., 1982). Kinetic analysis of this data indicated the existence of a competitive inhibition of EPTC uptake by the antidote R-25788 on corn. This competition for uptake is considered to be the first step in a series of interactions between EPTC and R-25788 rather than a major mechanism involved in the protective effect of R-25788 (Ezra et al., 1982). The similarity between thiocarbamateand chloroacetanilideherbicide injury to corn combined with the efficacy of R-25788 as an antidote for both herbicidal classes on corn suggested that these two herbicidal classes have similar modes of herbicidal action (Leavitt and Penner, 1978b). Furthermore, because R-25788 structurally resembles both of these herbicidal classes, the possibility that

HERBICIDE ANTIDOTES

305

R-25788 may counteract the effects of chloroacetanilide as well as thiocarbamate herbicides on corn through a competitive inhibition at some active site specific to corn has been investigated. In early studies, Wilkinson and Smith (1975) demonstrated that R-25788 could reverse an EPTC-induced inhibition of lipid synthesis in isolated spinach chloroplasts, which suggested that R-25788 may compete with EPTC at a site of lipid synthesis. This hypothesis was supported by studies with corn-isolated protoplasts (Sagaral, 1978) and corn cell suspensions (Ezra and Gressel, 1982) that demonstrated that lipid synthesis was indeed a target site involved in the early action of EPTC. When R-25788 was added simultaneously with EF'TC to corn protoplasts or cell suspensions, it partially prevented the EPTC-induced inhibition of lipid synthesis (Sagaral, 1978; Ezra and Gressel, 1982), probably through a stimulation of the incorporation of [14C]acetate into the plant tissues (Ezra and Gressel, 1982). However, the effects of EPTC as well as of the chloroacetanilide herbicide metolachlor on lipid synthesis of corn have been disputed by other investigators (Leavitt and Penner, 1979a). They suggested that R-25788 may act by preventing an EPTC-induced aggregation of the epicuticular wax layer of corn rather than by overcoming a blocking of lipid synthesis by the herbicide. The involvement of such a mechanism in the antidotal action of R-25788 has been further supported by Gorog et al. (1982), who demonstrated that the antidotes R-25788, R-28725, and AD-67 protect corn from EPTC injury by preventing a herbicide-induced aggregation of epicuticular wax. Through this mechanism, R-25788 prevents the formation of areas where the underlying cuticle layers are exposed, resulting in a decreased transpiration (Leavitt and Penner, 1979a; Gorog et al., 1982). Apart from lipid synthesis or epicuticular wax distribution in corn, other active sites that may be involved in the competitive inhibition of thiocarbamate and chloroacetanilide herbicide effects by the antidote R-25788 include membrane function (Bujtas, 1978), peroxidase activity (Harvey et al., 1975), and polyribosome formation (Rao and Kahn, 1975). The EPTC-induced enhancement of membrane permeability of sugar beet root disks was counteracted by R-25788 (Bujtas, 1978), and the EPTC-induced stimulation of peroxidase activity in corn seedlings was annulled by R-25788 (Harvey et al., 1975). Similarly, the alachlor-induced inhibition of polyribosome formation in barley roots was reversed by R-25788, probably by a competition with the herbicide for this active site (Rao and Kahn, 1975). In contrast to all of the aforementioned reports that specify the plant active sites for which the herbicides and the antidote may be competing, Stephenson et al. (1978, 1979) proposed that R-25788 or other dichloroacetamide antidotes may apt as herbicidally inactive competitive inhibitors at an unknown site of thiocarbamate herbicide action that is specific to corn. An additional hypothesis that offers a probable explanation for the mechanism of antidotal action of R-25788 is the enhanced breakdown of thiocarbamate or chloroacetanilide herbicides in antidote-treated corn plants. In early studies by

306

KRITON K. HATZIOS

Wright et al. (1973) and Chang et al. (1974a) it was concluded that R-25788 increased the rate of butylate and EPTC metabolism in treated corn seedlings. Because of the lack of protection by R-25788 against herbicide injury to other grass crops, Wright et al. (1973) proposed that an alternate pathway for the degradation of thiocarbamate herbicides could be present in corn but not in other grasses. Lay et al. (1975) were the fmt to demonstrate that such an alternate pathway for thiocarbamate herbicide degradation in corn indeed exists and involves two steps. In the first step the thiocarbamate herbicides are converted through an oxidation reaction to their respective sulfoxides, which are then conjugated to gluthathione (GSH) in the second step. The sulfoxidation of thiocarbamate herbicides has been viewed as a bioactivation reaction because the thiocarbamate sulfoxides were found to be phytotoxic to many plants but not to corn (Casida et al., 1974). However, it was shown that a two-step oxidation of EF'TC (EPTC --* EPTC sulfoxide --* EPTC sulfone) rather than a single-step oxidation (EPTC + EPTC sulfoxide) was necessary for the conjugation of EFTC to GSH (Horvath and Pulay, 1980). In addition, Komives and Dutka (1980) demonstrated that the phytotoxicity of EPTC to corn does not result from the action of the EPTC sulfone but is partly the result of the action of the EPTC sulfoxide. They suggested that both the sulfoxidation of EPTC and the subsequent conjugation to GSH are equally important for its detoxication in corn. In a series of reports Casida and co-workers (Lay et al., 1975; Lay and Casida, 1976; 1978; Hubbel and Casida, 1977) emphasized that the conjugation of thiocarbamate sulfoxides to GSH was the most important step in thiocarbamate herbicide detoxication by corn and proposed that R-25788 and other dichloroacetamide antidotes protect corn by increasing the rate of thiocarbamate sulfoxide conjugation to GSH. This antidotal action of R-25788 is probably the result of an antidote-induced elevation of the GSH content and GSH-S-transferase activity in the roots of corn seedlings (Lay et al., 1975; Lay and Casida, 1976). Subsequent studies by other investigators demonstrated that pretreatments with R-25788 do indeed result in an elevation of GSH content of corn seedlings (Carringer et al., 1978b; Stephenson et al., 1980), corn cell suspensions (Ezra and Gressel, 1982), and tobacco cell suspensions (Rennenberg et al., 1982). However, the existence of a GSH-S-transferase enzyme in corn roots that catalyzes the conjugation of thiocarbamate sulfoxides with GSH has been disputed by some investigators (Carringer et al., 1978a; Leavitt and Penner, 1979b) who 'reported that the GSH-EPTC sulfoxide conjugation proceeds spontaneously rather than enzymatically. It is quite possible that R-25788 protects corn from EFTC or other thiocarbamate injury by increasing GSH production and not by stimulating the activity of a GSH-S-transferase. This antidote-induced elevation of GSH content in corn mots is probably the result of a direct activation of GSH synthetase by the antidote R-25788 (Carringer et af., 1978b). However, Rennenberg et al. (1982) proposed that the stimulation of GSH synthesis by R-25788 in

HERBICIDE ANTIDOTES

307

heterotrophically grown tobacco cell suspensions is not caused by a direct activation of preexisting enzymes but by an enhancement of the amount of enzymes involved in this process. According to Adams and Casida (1981), R-25788 elevates GSH content in corn by acting at an early stage in the sulfate + GSH biosynthetic pathway. The validity of the theory that R-25788 protects corn by increasing GSH content has been questioned by some researchers (Ezra and Gressel, 1982; Gressel et al., 1982). Thus Ezra and Gressel(l982) showed that although most of the [l4C]EPTC was rapidly biotransformed by corn cell suspensions within 8 hr, measurable increases in GSH following treatment with R-25788 began after 12 hr. In addition, they showed that R-25788 added simultaneously to corn suspension cells with EPTC reversed the inhibition of lipid synthesis induced by EPTC. Based on these observations, Ezra and Gressel (1982) proposed that R-25788 protection to corn may involve more than one mechanism, such as an initial rapid effect on lipid synthesis followed by a slower effect that results in elevations of cellular GSH content. The limited significance of the antidoteinduced elevation of GSH content for the protection of corn by R-25788 was also emphasized in a report by Fedtke (1981), who showed that apart from R-25788 a number of other herbicides and plant-growth regulators significantly increased GSH content in corn and soybeans. However, none of these herbicides or growth regulators protects corn against EPTC injury. The conclusion that the observed elevation of GSH content of corn is not critical for the antidotal action of R-25788 against EPTC is further supported by Casida et al. (1974), who reported that in the absence of R-25788, corn was injured by EPTC at 3.4kg/ha but could tolerate EPTC sulfoxide applications of as high as 27 kg/ha without significant injury. Leavitt and Penner (1979b) proposed that the protective effect of R-25788 against EPTC injury to corn may have a different basis than its protective effect against chloroacetanilide herbicide injury. They suggested that an antidote-induced increase in the GSH content of corn may be important for the safening effect of R-25788 against chloroacetanilide injury because the rate of the alachlor-GSH conjugation was less efficient than that of the GSH-EPTC sulfoxide conjugation. The significance of GSH conjugation in the metabolism of chloroacetanilideherbicides has been reviewed by Hatzios and Penner (1982). In addition, Leavitt and Penner (1979b) proposed that R-25788 may protect corn from thiocarbamate herbicide injury by stimulating the rate of their sulfoxidation. The thiocarbamate sulfoxides are subsequently detoxified by conjugation to GSH. A similar hypothesis for the mechanism of action of R-25788 has also been proposed by Horvath and Pulay (1980). The sulfoxidation of thiocarbamate herbicides in corn is believed to be an enzymatic reaction mediated by mixed-function oxidases (Hubbel and Casida, 1977). Studies by Komives and Dutka (1980) showed that the insecticide syner-

308

KRITON K. HATZIOS

gists and mixed function oxidase inhibitors, piperonyl butoxide and SKF-525A, synergized the phytotoxic action of EPTC on corn, indicating that EPTC sulfoxidation plays a key role in the antidotal action of R-25788. Similar results have been observed by other researchers who demonstrated that in the presence of R-25788, EPTC interacted synergistically with the herbicide tebuthiuron (Hatzios, 1981) or with the antioxidants, piperonyl butoxide and propyl gallate (Hatzios, 1982b). Other studies have also demonstrated a synergistic interaction of EPTC with the herbicide 2,4-D (Martin and Burnside, 1982) or the insecticide fonofos (Freeman, 1978)on corn in the presence of the antidote R-25788. At the present time, the significance of these last results cannot be easily assessed with reference to the mode of antidotal action of R-25788. The involvement of the oxidative metabolism of EPTC in the antidotal action of R-25788 has been disputed by Taft (1976), who reported that mixed-functionoxidase inhibitors did not affect either the EPTC injury to corn or the protective action of R-25788. From the previous discussion, it is evident that our current understanding of the antidotal action of R-25788 is not very clear. None of the presently available theories about the mode of antidotal action of R-25788 is unequivocally accepted, and additional studies obivously are needed to elucidate the exact mode of action of this antidote.

c. MODEOF ANTIDOTALACMONOF CGA-43089 Because of the recent introduction of CGA-43089 as a sorghum protectant against metolachlor injury, studies on the mode of its antidotal action are limited. Because the main entry point of both the herbicide metolachlor and the antidote CGA-43089 is the coleoptile of sorghum seedlings (Nyffeler et al., 1980, Ketchersid and Merkel, 1981a), interference of CGA-43089 with metolachlor uptake by sorghum coleoptiles could explain the protective effect of this antidote. Such a hypothesis appeared to be supported by the results of studies conducted by Ketchersid and Merkel (1981b) and Ketchersid et al. (1982). However, the apparent competitive effect of CGA-43089 on the absorption of metolachlor was most evident in the roots rather than in the coleoptiles of grain sorghum seedlings (Ketsersid el al., 1982). In addition, the decreased rate of metolachlor uptake in the presence of CGA-43089 did not appear to be related directly to changes in cell permeability. Ebert (1982) proposed that the safening action of CGA-43089 against metolachlor injury to grain sorghum may result from its ability to prevent a metolachlor-induced loss of cuticular integrity in sorghum plants which greatly reduces the penetration of metolachlor.. Other investigatorshave disputed the theory that reduced uptake of metolachior by sorghum seedlings in the presence of CGA-43089is the reason for the antidotal effect of this protectant (Winkle et al., 1980; Christ, 1981). Christ (1981)

HERBICIDE ANTIDOTES

309

proposed that CGA-43089 is able to reduce the active amount of metolachlor at a site of action specific to sorghum. An antidote-induced increase in the rate of the metabolic detoxication of metolachlor by sorghum seedlings could explain such an action of CGA-43089. More recent studies, however, demonstrated that CGA-43089 did not influence the ability of sorghum tissues to metabolize the herbicide metolachlor (Winkle et ul., 1980; Ketchersid and MerkIe, 1981b; Leek and Penner, 1982). It has been observed that CGA-43089 fails to counteract metolachlor injury to sorghum grown in nutrient solution (Leekand Penner, 1981) or under conditions of excessive soil moisture (Ketchersid et al., 1981; Leek and Penner, 1981). In addition, the results of another study showed that in the presence of CGA-43089, metolachlor interacted synergistically with ozone and the antioxidants piperonyl butoxide and propyl gallate on sorghum under greenhouse conditions (Hatzios, 1983a). The results of this study combined with the observation that CGA-43089 fails to protect sorghum against metolachlor under extremely wet (anaerobic) conditions indicate that CGA-43089 may act by stimulating the activity of a biological oxidation system (possibly a mixed-function oxidase) that could be involved in the metabolic detoxication of metolachlor in sorghum. Although such a mechanism for the protective action of CGA-43089 has been postulated (Hatzios, 1983a), further studies are needed to establish it as a viable theory explaining the antidotal activity of this compound. Finally, CGA-43089 was not active in increasing the GSH content of corn or soybean tissues (Fedtke, 1981).

VI. DEGRADATION OF HERBICIDE ANTIDOTES IN PLANTS Studies on the degradation of herbicide antidotes in plants are also limited. Laboratory investigations with 14C-labeledNA showed that this antidote was not metabolized or bound in corn plants, that no volatile “T metabolites were formed, and that no residues for NA were found in mature corn plants beyond the fifth or sixth week after emergence (Riden and Asbell, 1975). Corn seedlings grown in soil treated with I4C-labeled R-25788 liberated 6% of the absorbed radioactivity as 14C02in a 10-day period (Murphy et ul., 1974). Approximately 80-85% of the absorbed radioactivity could be extracted with ethanol. Four metabolites of R-25788 were separated in corn tissues, identified as N-allyl-2,2dichloroacetamide, N,N-diallylglycolamide, N,N-diallyloxamic acid, and the glycoside of N,N-diallylglycolamide (Murphy et al., 1974). Studies on the metabolism of the antidotes CGA-43089 and MON-4606 in sorghum plants have not been conducted.

3 10

IUUTON K. HATWOS

VII. SUMMARY

The concept of using herbicide antidotes offers a potential alternative for increasing the selectivity of currently available herbicides. A desirable herbicide antidote is a chemical agent that selectively protects crops from herbicide injury without protecting weeds. This selectivity is the result of either a very specific crop-herbicide-antidote interaction or a selective treatment such as the dressing of crop seeds with the antidote. Herbicide antidotes are developed primarily through random screening techniques that involve most combinations of important herbicides, major crops, and candidate antidotes. Chemicals that are currently used as herbicide antidotes include NA, R-25788, CGA-43089, CGA-92194, and MON-4606.These antidotes offer adequate protection to grass crops that are damaged but not killed by specific herbicides. Thus the presently available herbicide antidotes can counteract, to some extent, the effects of chloracetanilide and carbamate herbicides on grass crops such as corn and grain sorghum. Several environmental factors such as temperature, soil moisture, and soil type may affect the field performance of herbicide antidotes and need to be seriously considered. In addition, the timing of herbicide and antidote applications to the crop as well as the differential intraspecific tolerance of crop cultivars to combinations of herbicides plus antidotes need to be established for optimum effectiveness of herbicide antidotes in the field. The mode of antidotal action of the presently available herbicide safeners is not fully understood. It is believed that rather than merely preventing the entry of herbicides into the plant, herbicide antidotes work inside the plant to counteract the actions of herbicides either by competing with them for a common site of action or by stimulating their metabolic detoxication in the protected crops. The development of effective antidotes that could protect broad-leaved crops against injury from herbicidal photosynthetic inhibitors represents the greatest challenge of the pesticide industry in the near future. Advances in our understanding of herbicide action and degradation by plants may lead to the development of more effective herbicide antidotes in the future. REFERENCES Abemthy, J. R., and Keeling, J. W. 1982. Abstr. Weed Sci. Soc. Am., No. 34. Adams, C.A., and Casida. J. E. 1981. Abstr. Papers Am. Chem. Soc. 182nd Meet., PEST, p. 72. Adler, E. F.,Wright, W. L., and Klingmim, G. C. 1977. In “Pesticide Chemistry in the 20th Century” (1. R. Plimmer, ed.), pp. 39-55. h e r . Chem. Soc., Washington,D.C. Ahle, J., and Cozart, E. 1972. Proc. North Cent. Weed Control Cot& 27,23. Ahrens, W.H., andDavis, D. E. 1978. Proc. Souzh. Weed Sci. Soc. 31, 249. Ali, A., and Stephenson, G. R. 1979. Abstr. Weed Sci. Soc. Am., No. 1. Ali, M.,and Mercado, B. L. 1980. Philipp. 1. Weed Sci. 7, 26-32.

HERBICIDE ANTIDOTES

31 1

Amrhein, N., and Schneebeck, D. 1980. Physiol. Planr. 49,62-64. Anonymous. 1977a. Weed Absrr. 26, 310. Anonymous. 1977b. Weed Absn-. 26, 2873. Appleby, A. P., Giineyli, E., and Colbert, D. R. 1972. Res. Prog. Rep. West. Soc. Weed Sci., p. 80. Arle, H. F. 1968. Weed Sci. 16,430-435. Ameklev, D. R., and Baker, D. R. 1977. Chern. Absn-. 87, 1732. Ashton, F. M., and Crafts, A. S. 1981. “Mode of Action of Herbicides,” 2nd ed. Wiley, New York. Ashton, F. M., and Harvey, W. A. 1971. Circ. Calif. Agric. Exp. Stn.. No. 588. Baumley, J. J., Vitolo, D. T., and Ilnicki, R. D. 1981. Proc. Northeust. Weed Sci SOC. 35,52-55. Bertges, W. J., Jr. 1977. Diss. Abstr. Inr. B 37, 4795. Beste, C. E., and Schreiber, M. M. 1970. Weed Sci. 18, 484-488. Beste, C. E., and Schreiber, M. M. 1972a. Weed Sci. 20, 4-7. Beste, C. E., and Schreiber, M. M. 1972b. Weed Sci. 20, 8-1 1. Blair, A. M. 1978. Weed Res. 18, 77-81. Blair, A. M. 1979. Ann. Appl. Biol. 92, 105-111. Blair, A. M., and Dean, M. L. 1976. Weed Res. 16,47-52. Blair, A. M., Parker, C., and Kasasian, L. 1976. PANS 22, 65-74. Blumenfeld, T.,Horowitz, M., and Herzlinger, G. 1973. Phytoparusiticu 1, 82. Boyd, J. W., Mason, J. F., and Murray, D. S. 1979. Proc. South. Weed Sci. Soc. 32, 56. Brinker, R. J., Schafer, D. E., Hakes, H. R., and Radke, R. 0. 1982. Proc. South. Weed Sci. Soc. 35,290. Bujtas, G. 1978. Novinytermelis 27, 223-229. Bumside, 0. C., Wicks, G. W., and Fenster, C. R. 1971. Weed Sci. 19, 565-568. Burt, G. W., and Akinsorotan, A. 0. 1976. Weed Sci. 24, 319-321. Burt, G. W., and Buzio, C. A. 1978. Proc. Northeust. Weed Sci. SOC. 32, 62. Burt, G. W., and Buzio, C. A. 1979. Weed Sci. 27,460-463. Buzio, C. A., and Burt, G. W. 1977. Proc. Norrheust. Weed Sci. SOC. 31, 115. Camper, N. D., and Carter, G. E., Jr. 1974. Proc. South. Weed Sci. Soc. 27, 359. Carringer, R. D., Rieck, C. E., and Poneleit, C. G. 1974. Proc. North Cent. Weed Control Conf. 29, 32. Carringer, R. D., Rieck, C. E., and Bush, L. P. 1978a. Weed Sci. 26, 157-160. Carringer, R. D., Rieck, C. E., andBush, L. P. 1978b. WeedSci. 26, 167-171. Casida, J. E., Gray, R. A., and Tilles, H. 1974. Science (Washington, D.C.) 184, 573-574. Castelfranco, P., Oppenheim, A., and Yamaguchi, S. 1963. Weeds 11, 111-115. Catizone, P. 1979. Dijesu Pianre 2, 283-299. Chang, F. Y., Bandeen, J. D., and Stephenson, G. R. 1972. Cun. J . Plant Sci. 52, 707-714. Chang, F. Y., Bandeen, J. D., and Stephenson, G. R. 1973a. Weed Res. 13, 399-406. Chang, F. Y., Stephenson, G. R., and Bandeen, J. D. 1973b. Weed Sci. 21, 292-295. Chang, F. Y., Stephenson, G. R., and Bandeen, J. D. 1974a. J. Agric. Food Chern. 22, 245-248. Chang, F. Y., Stephenson, G. R., Anderson, G. W., and Bandeen, J. D. 1974b. Weed Sci. 22, 546-548. Chang, T. S., and Merkle, M. G. 1982. Weed Sci. 30, 70-73. Chang, T. S., and Merkle, M. G. 1983. Proc. South. Weed Sci. SOC. 36, 383. Chenault, E. W., and Wiese, A. F. 1980. Proc. South. Weed Sci. SOC. 33, 69. Christ, R. A. 1981. WeedRes. 21, 1-8. Christiansen, N. M.,and Hilton, J. L. 1974. Crop Sci. 14, 489-490. Clapp, K . E. 1974. Proc. West. Soc. Weed Sci. 27, 40-42. Clapp, K . E. 1975:Proc. West. Soc. Weed Sci. 28, 31-34. Crook, J. J. 1975. Proc. Norrheust. Weed Sci. Soc. 29,49-54.

312

KRI”ON K. HATZIOS

da Silva, J. B. 1978.Diss. Abstr. Int. B 39, 522. da Silva, J. B. and Ueda, A. 1981. Weed A b m . 30, 3022. Davidson, W. E., Cagnon, S. A., Christensen, M. D., and Don,J. E. 1978.Proc. Wesr. Soc.Weed Sci. 31, 123-129. de Andrade, V. A. 1981. Weed Abstr. 31,2277. Dell’Agnola, G., Ferrari, G., and Nardi, S. 1981.Pestic. Biochem. Physiol. 15, 101-104. Dill.T. R.,Turner. W. E., Nyffeler, A., and Quadranti, M. 1982.Abstr. WeedSci. Soc.Am., No. 33. Donald, W. W., and Fawcett, R. S. 1976.Abstr. Weed Sci. Soc.Am., No. 169. Dowler, C. C. 1973. Proc. Sourh. WeedSci. Soc. 26, 143. Duke, S. O.,Hoagland, R. E., and Elmore, C. D. 1979. Physiol. Plont. 46, 307-317. Duke, S. O.,Hoagland, R. E., and Elmore, C. D. 1980. Plum Phyiol. 65, 17-21. Dutka, F., Komives, T., Marton, A. F., Hulesch, A., Fodor-Csobra, K., and Kappati, K. 1979. Proc. Hung. Meer. Biochem. 19th. pp. 253-256. Eastin, E. F. 1972.Agron. J. 64, 556-557. E M , E. 1982. Weed Res. 22, 305-311. Egli, H. 1982.J. Agric. F d C h e m . 30,861-866. Elliott, R. A., and Purnell, T. J. 1976.Proc. Br. Crop Prot. COP$Weeds 2, 533-540. Ellis, J. F., Peek, J. W., Boehle, J., Jr., and Miiller, G. 1980. Weed Sci. 28, 1-5. Ezra, G., and Gressel, J. 1982.Pesric. Biochem. Physiol. 18, 107-112. Ezra, G., Krochmal, E., and Gressel, J. 1982.Pesric. Biochem. Physiol. 18, 107-112. Farkas, E., Richnovszky, A., and Pais, I. 1981a. Acta Agron. Acad. Sci. Hung. 30, 371-376. Farkas, E.,Toth, A., and Pais, I. 1981b.Acta Phyropathol. Acad. Sci. Hung. 16, 259-261. Fedtke, C. 1981. Z . Pj7anzenkr. P&nzenschurz 9, 141-146. Foy, C. L.,and Witt, H. L. 1982. Proc. South. Weed Sci. Soc. 35, 53. Freeman, J. A. 1978. Can.J. PlanrSci. 58, 1119-1121. Friesen, G. H. 1979. Can. J. Plant Sci. 59, 535-537. Gaughan, E. J., Mahoney, M. D., and Pallos, F. M. 1979. Chem. Abstr. 91, 20565On. Geller, I. A., and Nikolaenko, I. Z.1972. Sakh. Svekla 17, 25-26. Gingerich, L. L.,Schneider, R. P., Schultz, G. E., and Rieck, W. L. 1981.Absrr. Weed. Sci. Soc. Am., No. 54. G M g , K., Muschinek, G. Y.,Mustardy, L. A., and Faludi-Daniel, A. 1982.WeedRes. 22,27-33. Gray, R. A., and Joo, G. K. 1978.I n “Chemistry and Action of Herbicide Antidotes” (F. M. Pallos and J. E. Casida, eds.), pp. 67-84. Academic Press, New York. Cressel, J. 1980.In “Plant Cell Cultures: Results and Perspectives” (F. Sala, B. Parisi, R. Cella, and 0.Cifferi, eds.), pp. 379-388. Elsevier, Amsterdam. Gressel,J., Zilkah,S., and Ezra, G. 1978.I n “Frontiers of Plant Tissue Culture” (T. A. Thorpe, ed.), pp. 427-436. Int. Assoc. Plant Tissue Culture, Calgary. Gressel, I., Ezra, G., and Iain, S. M. 1982. I n “Chemical Manipulation of Crop Growth and Development” (J. D. McLaren, ed.), pp. 79-91. Butterworth, London. Gresshoff, P. M. 1979.Ausr. J. Plant Physiol. 6, 177-185. Giineyli, E. 1971. Diss. Abstr. Int. B 32, 1957-1958. Gupta, 0.P. 1976. World Crops 28, 134-138. G ~ p t a0. , P., and Niranwal, N. K. 1976. PANS 22, 86-89. Haderlie, L. C., Widholm, J. M., and Slife, F. W.1977.Plant Physiol. 60,40-43. Hahn, R. R. 1974. Diss. Absrr. Inr. B 35, 1483. Hamm,P. C.,and Speziale, A. J. 1956. J . Agric. Food Chem. 4,518-522. Hammerton, J. L. 1974. PANS 20,429-436. Hardcastle, W . S. 1979. Proc. South. WeedSci. Soc. 32, 386. Hardcastle, W. S. 1982.Proc. Sourh. Weed. Sci. Soc. 35, 54.

HERBICIDE ANTIDOTES

313

Harper, J. B., and Harvey, B. M. R. 1978. Plant Cell Environ. 1, 211-215. Harvey, B. M. R., Chang, F. Y., and Fletcher, R. W. 1975. Can. J. Bor. 53, 225-230. Hassawy, G. S., and Hamilton, K. C. 1971. Weed Sci. 19, 166-169. Hatzios, K. K. 1981. WeedSci. 29, 601-604. Hatzios, K. K. 1982a. Proc. South. WeedSci. SOC. 35, 20-21. Hatzios, K. K. 1982b. Virginia J. Sci. 33, 77. Hatzios, K. K. 1983a. Weed Sci. 31, 280-284. Hatzios, K. K. 1983b. Proc. South. WeedSci. SOC. 36, 43. Hatzios, K. K., and Mauer, C. M. 1983. Abstr. Weed Sci. Soc. Am., No. 70. Hatzios, K. K., and Penner, D. 1980. Weed Sci. 28, 273-276. Hatzios, K. K., and Penner, D. 1982. “Metabolism of Herbicides in Higher Plants.” Burgess, Minneapolis, Minnesota. Heikes, E. P., and Swink, J. F. 1973. Proc. West. SOC. Weed Sci. 26, 32-34. Henry, C. S. 1972. Proc. South. WeedSci. SOC. 25, 192. Herman, D. J., Smith, W. F., and Ilnicki, R. D. 1974. Proc. Northeast. Weed Sci. SOC. 28, 24-28. Hickey, J. S., and Krueger, W. A. 1974a. Weed Sci. 22, 86-90 Hickey, J. S., and Krueger, W. A. 1974b. Weed Sci. 22, 250-252. Hilton, J. L., and Christiansen, M. N. 1972. Weed Sci. 20, 290-294. Hoagland, R. E., Duke, S. O., and Elmore, C. D. 1979. Physiol. Plant. 46,357-366. Hoffman, 0. L. 1953. Plant Physiol. 28, 622. Hoffman, 0. L. 1962. Weeds 10, 322-323. Hoffman, 0. L. 1969. Absn. Weed Sci. Soc.Am., No. 12. Hoffman, 0. L. 1971. U.S. Patent No. 3,564,768. Hoffman, 0. L. 1978a. In “Chemistry and Action of Herbicide Antidotes” (F. M. Pallos and J. E. Casida, eds.), pp. 1-13. Academic Press, New York. Hoffman, 0. L. 1978b. CHEMTECH 8, 488-492. Hoffman, 0. L., Gull, P. W., Zeisig, H. C., and Epperly, J. R. C. 1960. Proc. North Cent. Weed Control Conf. 17, 20. Holm, R. E., and Szabo, S. S. 1974. Weed Res. 14, 119-122. Horvath, L., and Pulay, A. 1980. Pesric. Biochem. Physiol. 14, 267-270. Howe, R. K., and Lee, L. F. 1980. U.S. Patent No. 4,199,506. Hubbel, J. P., and Casida, J. E. 1977. J. Agric. Food Chem. 25, 404-413. Huffman, J. B., and Camper, N. D. 1978. Weed Sci. 26, 527-530. Jaworski, E. G. 1972. J. Agric. Food Chem. 20, 1195-1198. Jeffery, L. S., and Connel, J. T. 1973. Proc. South. Weed Sci. SOC. 26, 145. Jeffery, L. S., Connel, J. T., and Morgan, T. H. 1971. Proc. South. Weed Sci. SOC. 24, 156. Jordan, L. S., and Jolliffe, V. A. 1971. Bull. Environ. Conram. Toxicol. 6, 417-421. Kadous, E. A., Din, Tag-El, Khalifa, M. A. S., Lamie, O., El-Desb, S. T., and El-Nawawy, S. A. 1977. WeedAbsrr. 26, 1426. Kennedy, J. M., and Krueger, W. A. 1978. Proc. South. Weed Sci. SOC. 31, 66. Ketchersid, M. L., and Merkle, M. G. 1981a. Proc. South. Weed Sci. Soc. 34, 44. Ketchersid, M. L., and Merkle, M. G. 1981b. Absn. Weed Sci. SOC.Am., No. 215. Ketchersid, M. L., and Merkle, M. G. 1982. Proc. South. Weed Sci. SOC. 35, 368. Ketchersid, M. L., and Merkle, M. G. 1983. Proc. Sourh. Weed Sci. Soc. 36, 384. Ketchersid, M. L., Norton, K., and Merkle, M. G. 1981. Weed Sci. 29, 281-287. Ketchersid, M.L., Vietor, D. M., and Merkle, M. G. 1982. J. Plant Growth Regul. 1, 285-294. Kezeli, T. A., Gvamichava, N. E., Tarazashvhi, K. M., and Piranishvhi, N. S. 1973. Soobshcheniya Acud. Nauk Gruzinskoi SRR 69, 145-148. Killmer, J. L., Widholm, J. M., and Slife, F. W. 1980. Plant Sci. Len. 19, 203-208. Killmer, J. L., Widholm, J. M., and Slife, F. W. 1981. Plant Physiol. 68, 1299-1302.

314

KRITON K. HATZIOS

Kirkland, K. 1973. Diss. Abstr. Int. B 33, 40664067. Komives, T., and Dutka, F. 1980. Cereuf Res. Commun. 8, 627-633. Kneminski, S., and Ryan, J. 1980. Agrichem. Age 24, 6-8. Laddlie, J. S., Meggitt, W. F., and Penner, D. 1977. Weed Sci. 25,88-93. Lay, M. M., and Casida, J. E. 1976. Pestic. Biochem. Physiol. 6, 442-456. Lay, M. M., and Casida, J. E. 1978. In “Chemistry and Action of Herbicide Antidotes” (F. M. Pallos and J. E. Casida, eds.), pp. 151-160. Academic Press, New York. Lay, M. M.,Hubbel, J. P., and Casida, J. E. 1975. Science (Washington, D.C.) 189, 287-289. Leavitt, J. R. C., and Penner, D. 1978a. Weed Res. 18, 281-286. Leavitt, J. R. C., and Penner, D. 1978b. Weed Sci. 26, 653-659. Leavitt, J. R. C., and Penner, D. 1979a. Weed Sci. 27, 47-50. Leavitt, J. R. C., and Penner, D. 1979b. J. Agric. Food Chem. 27, 533-536. LeBaron, H., and Gressel, J. 1982.“Herbicide Resistance in Plants.” Wiley, New York. Lee, G. A., Alley, H. P., and Gale, A. F. 1974a. Wyoming Agric. Res. Stn. Res. J. 83, 47-59. Lee, G. A., Alley, H. P., and Gale, A. F. 1974b. Wyoming Agric. Res. Stn. Res. J . 83, 67-76. Leek, G. L., and Penner, D. 1980. Absrr. Weed Sci. SOC. Am., No. 182. Leek, G. L., and Penner, D. 1981. Abstr. WeedSci. SOC. Am., No. 216. Leek, G. L., and Penner, D. 1982. Absrr. Weed Sci. SOC. Am., No. 185. Lignowski, E. M. 1970. Dim. Absrr. Int. B 31, 992-993. Lignowski, E. M., and Scott, E. G. 1972. Weed Sci. 20, 267-270. Mahoney, M. D., and Penner, D. 1981. Proc. North Cent. Weed Control Conf. 36, 103. Malefyt, T., and Duke, W. B. 1981. Proc. Northeast. Weed Sci. Soc. 35, 59. Martin, A. R., and Burnside, 0. C. 1982. WeedSci. 30, 269-272. M h n , H. 1978. U.S.Patent No. 4,070,389. Marton, A. F., Aprokovacs, A. V.,Komines, T., Fodor-Csobra, K., and Dutka, F. 1978. Proc. Hung. Annu. Meer. Biochem. IBth, pp. 111-112. Meggitt, W. F., Kern, A. D., and Armstrong, T. F. 1972. Proc. North Cent. Weed Control Conj. 27, 22-23. Mellis, J. M . , Pillai, P., Davis, D. E., and Truelove, B. 1982. Weed Sci. 30, 399-404. Miaulis, J. B., Thomas, V. M., Gray, R. A., Murphy, J. J., and Hollingworth, R. M. 1978. In “Chemistry and Action of Herbicide Antidotes” (F.M. Pallos and J. E. Casida, eds.), pp. 109-131. Academic Press, New York. Michieka, R. W., Somody, C. N., and Ilnicki, R. D. 1978. Proc. Northeusr. Weed Sci. Soc. 32, 4-6. Miller, S . D., and Nalewaja, J. D. 1980. Agron. J . 72, 662-664. Miller, S. D., Nalewaja, J. D., and Pudelko, J. 1978. Weed Sci. 26, 116-1 18. Moshier, L. J., and Russ, 0. G. 1980. Proc. North Cent. Weed Conrrol Conj. 35, 71. Miiller, G., and Nyffeler, A. 1981. Actu Phytoputhol. Acud. Sci. Hung. 16, 245-248. Mullison, W. R. 1979. “Herbicide Handbook,” 4th ed. Weed Sci. Soc.Am., Champaign, Illinois. Murphy, J. J. 1972. Chem. Biol. Inreruct. 5, 284-286. Murphy, J. J., Miaulis, J. B., and Gray, R. A. 1974. Abstr. Weed Sci. SOC. Am., No. 3. Nyffeler, A., Gerber, H. R., and Hensley, J. R. 1980. Weed Sci. 28,6-10. Ogawa, M., and Ota, Y. 1976. Proc. Asiun-PaciJic Weed Sci. Soc. Con$ Srh, pp. 303-306. Ohali, Y., Eshel, Y.,Marani, A,, and Rubin, B. 1979. Phyropurasiticu 7, 139. Okii, M., Teranishi, M.,Konnai, M., and Takematsu, T. 1979a. Weed Res. Jpn. 24, %-loo. Okii, M., Matsukuma, I., Konnai, M., and Takemutsu, T. 1979b. Weed Res. Jpn. 24, 101-106. Okii, M., Watanabe, K., K o ~ a i M., , and Takematsu, T. 1979c. Weed Res. Jpn. 24, 194-198. Okii, M., Onitake, T., Konnai, M., and Takematsu, T. 1979d. Weed Res. Jpn. 24, 221-225. Okii, M., Nishimura, T., Konnai, M.,and Takemutsu, T. 1979e. Weed Res. Jpn. 24, 226-232.

HERBICIDE ANTIDOTES

315

On,J. D., Martella, D., and Lorentzen, L. 1978. Proc. Calif. Weed Conf. 30, 133-147. Pallos, F. M. 1977. Chem. Abstr. 86, 101934~. Pallos, F. M., and Baker, D. R. 1977. Chem. Abstr. 87, 113124h. Pallos, F. M., and Casida, J. E. 1978. “Chemistry and Action of Herbicide Antidotes.” Academic Press, New York. Pallos, F. M., Brokke, M. E., and Ameklev, D. R. 1972. Belgian Patent No. 782,120. Pallos, F. M., Gray, R. A., Ameklev, D. R., and Brokke, M. E. 1975. J. Agric. Food Chem. 23, 82 1-822. Pallos, F. M., Brokke, M. E., and Ameklev, D. R. 1977. U.S. Patent No. 4,021,224. Palmer, G. K., and Jeffery, L. S. 1982. Proc. South. Weed Sci. SOC. 35, 55. Parker, C. 1981. Trop. Pesr Manag. 27, 139-140. Parker, C. 1983. Pesric. Sci. 14, 40-48. Parker, C., and Dean, M. L. 1,976. Pestic. Sci. 7, 403-416. Parker, C., Richardson, W. G., and West, T. M. 1980. Proc. Br. Crop Pror. Conf. Weeds 1,15-21. Peirson, D. R., Borden, J. H., and Vakenti, J. M. 1976. Can. J . Phnr Sci. 56, 119-125. Peters, R. A., and Dest, W. M. 1971. Proc. Northeast Weed Sci. SOC. 25, 39-41. Phatak, S. C., and Bouw, W. J. 1974. HortScience 9, 34. Phillips, B. A., and Bhagsari, A. S. 1978. In “Chemistry and Action of Herbicide Antidotes” (F. M. Pallos and J. E. Casida, eds.), pp. 21-34. Academic Press, New York. Price, J. B., and Merkle, M. G. 1977. Proc. Sourh. Weed Sci. Soc. 30, 68. Prochnow, C. L. 1978. Proc. West. SOC. Weed Sci. 31, 62-63. Purnell, T. J., and Bracey, D. J. 1978. Proc. East Afr. Weed Sci. Cot$ 6th. pp. 90-96. Putnam, A. R., and Penner, D. 1974. Residue Rev. 50, 73-110. Rains, L. J., and Fletchall, 0. H. 1971. Proc. North Cent. Weed Control Conf. 26, 42-44. Rains, L. J., and Fletchall, 0. H. 1973. Absrr. Weed Sci. Soc. Am., No. 13. Rao, V. S., and Kahn,A. A. 1975. Abstr. Weed Sci. SOC.Am., No. 175. Reeder, J. 1970. Crops Soils 23, 15-16. Rennenberg, H., Birk, C., and Schaer, B. 1982. Phyrochemisriy 21, 5-8. Rhodes, G. N., and Jeffery, L. S. 1979. Proc. Sourh. WeedSci. SOC.32, 310. Richardson, W. G., and Parker, C. 1977. Tech. Rep. ARC Weed Res. Organ., No. 43. Richardson, W. G., West, T. M.,and Parker, C. 1981. Tech. Rep. ARC WeedRes. Organ., No. 62. Riden, J. 1976. In “Analytical Methods for Pesticides and Plant Growth Regulators (G. Zweig, ed.), Vol. 8, pp. 483-489. Academic Press, New York. Riden, J., and Asbell, W. J. 1975. Environ. Q w l . Suf. (Suppl. 111), 175-179. Roeth, F. W. 1973. Weed Sci. 21, 474-476. Sagaral, E. G. 1978. Dis. Abstr. Inr. B 39, 5174. Sagaral, E. G., and Foy, C. L. 1982. Weed Sci. 30,64-69. Saggers, D. T. 1976. In “Herbicides: Physiology, Biochemistry, Ecology” (L. J. Audus, ed.),Vol. 2, 2nd ed., pp. 447-473. Academic Press, New York and London. Sanders, H. J. 1981. Chem. Eng. News59, 20-35. Schafer, D. E., Brinker, R. J., and Radke, R. 0. 1980. Proc. North Cent. Weed Control Cot$ 35, 67-71. Schafer, D. E., Brinker, R. J., and Radke, R. 0. 1981. Absrr. Weed Sci. Soc.Am., No. 53. Schmer, D. A., Lee, G. A., and Alley, H. P. 1973. Proc. West. Soc. Weed Sci. 26, 38-39. Schwartzbeck, R. A., and Hoffman, 0. L. 1973. Proc. Northem. Weed Sci. SOC.27, 56. Shahied, S. I., and Giddens, J. 1970. Agron. J . 62, 306-307. Simkins, G. S., Moshier, L. J., and Russ, 0. G. 1979. Proc. North Cent Weed Control Cog. 34, 45. Simkins, G. S., Moshier, L. J., and Russ, 0. G. 1980. Weed Sci. 28, 646-649.

316

KRITON K.HATZIOS

Smith, R. 1. 1971. Weeds Today 2, 12-13. Smith, W. F., McAvoy, W. L., Lay, M. M., and Ilnicki, R. D. 1973. Proc. Northemt. Weed Sci. SOC. 27, 57-60. Sobiescmski, J. 1975. Rocz. Glebozn. 26, 79-89. somody, C. N., Michieka, R. W., Ilnicki, R. D., and Somody, J. 1978. Proc. Norfhemt.Weed Sci. SOC. 32, 129-133. Spotanski, R. F., and Burnside, 0. C. 1973. Weed Sci. 21, 531-536. Stahlman,P. W. 1979. Proc. North Cent. Weed Conrrol Conf 34, 59. Stephenson, G. R., and Chang, F. Y. 1978. I n “Chemistry and Action of Herbicide Antidotes” (F. M. Pallos and 1. E. Casida, eds.), pp. 35-61. Academic Ress, New York. Stephenson, G. R., Bunce, N.I., Makowski, R. I., and Curry, 1. C. 1978. J . Agric. Food Chem. 26, 137-140. Stephenson, G. R., Bunce, N. H., Makowski, R. I., Bergsma, M. D., andcurry, 1. C. 1979. J. Agric. Food Chem. 27, 543-547. Stephenson, G. R., Ah, A., and Ashton, F. M. 1980. Absrr. Weed Sci. SOC.Am., No. 180. Street, 1. E., Walker, R. H.,Jolley, E. R., and Buchanan, G. A. 198Oa. Proc. South. Weed Sci. SOC. 33,237. Street, 1. E.,Walker, R. H.,Jolley, E. R., and Buchanan, G. A. 198%. Absrr. Weed Sci. Soc.Am., No. 22. Taft,L. G . 1976. Diss. Abstr. Int. B 36, 5387. Thiessen, E. P., Stephenson, G. R., and Anderson, G. W. 1980. Can. J . PiuntSci. 60,1005-1013. Truelove, B., and Davis, D. E. 1977. Proc. Sourh. Weed Sci. Soc. 30, 364. Turner, W. E., Holt, B. E.,McMahon, A., and Thomas, 1. M. 1979. Proc. South. Weed Sci. Soc. 32, 124. Turner, W. E.,Clark, D. R., Helseth, N.,Dill, T. R., King, J., and Seifried, E. B. 1982. Proc. South. WeedSci. Soc. 35, 388. Van Daalen, J. J., and Daams, J. 1970. Naturwissenschufren 8, 395. Veerasekaran, P., Kirkwood, R. C., andPowell, E. W. 1981a. Pesric. Sci. 12, 325-329. Veerasekaran, P., Kirkwood, R. C., and Powell, E. W. 1981b. Pestic. Sci. 12, 330-338. Wegrzyn, T. 1975. Rocz. Gleborn. 26,79-89. Whitwell, T.,and Santelman, P. W. 1975. Proc. South. Weed Sci. Soc. 28, 38. Wick, G. A., Burnside, 0. C., and Fenster, C. R. 1971a. Crops Soils 24, 10-1 1 . Wicks, G. A., Burnside, 0.C., and Fenster, C. R. 1971b. Absfr. Weed Sci. SOC. Am., No. 50. Wiese, A. F., Collins, J. W.,Lavane, D. E., Chenault, E. W., andDavis, 1. L. 1979. Proc. South. Weed Sci. Soc. 32, 121. Wilkinson, R. E., and Smith, A. E. 1975. Weed Sci. 23, 90-92. Wilkinson, R. E., and Smith, A. E. 1976. Weed Sci. 24, 235-237. Williams, 1. L.,Jr., Ross, M. A., and Bauman, T.T. 1973. Proc. North Cent. Weed Control Con$ 28, 84-85. Winkle, M. E., Leavitt, 1. R. C., and Bumside, 0. C. 1980. Weed Sci. 28, 699-704. Wirjahardja, S. 1979. BIOTROP Newsl. 28, 4. Wirjahardja, S., and Parker, C. 1977. Proc. Asian-Pacific Weed Sci. Soc. Conf. 6th 1, 316-321. Worsham, A. D. 1981. Proc. South. WeedSci. Soc. 34, 43. Wright, T. H., Rieck, C. E., and Thompson, L., Jr. 1973. Proc. South Weed Sci. Soc. 26, 383. Wright, T.H., Rieck, C. E., and Poneleit, C. G. 1974. Abstr. Weed Sci. SOC.Am., No. 1 . Youngman, R. J., Dodge, A. D., Longfelder, E., and Elstner, E. F. 1979. Experientia 35, 1295- 12%.

ADVANCES IN AGRONOMY, VOL. 36

BUFFALO GOURD AND JOJOBAPOTENTIAL NEW CROPS FOR ARiD LANDS LeMoyne Hogan and William P. Bemis Department of Plant Sciences, University of Arizona Tucson, Arizona

.......................................................... ................ Introduction ...................................................... Domestication ........................ ....................... SeedRoduction ...................... ....................... Product Evaluation.. .............................................. Recent Developments. .............................................

I. Introduction

II. Buffalo Gourd Cucurbitufoetidissima HBK.. . . . . . . . . . . A.

B. C. D. E. 111. Jojoba: Simmondsia chinensis (Link) Schneider . . ................... A. Introduction ................................. B. Improvement of Natural Stands. . . . ............................ C. Plantation Culture.. ............................................... D. VegetativePropagation............................. E. Diseases and Pests ..... ................................ F. Harvesting.. .......... ..................... G. Yields ................... ............................. H. Product Uses.......................... .................. IV. Conclusion ........................................... References .............. ......................................

317 319 319 321 324 328 331 332

335 343 346 347

1. INTRODUCTION

The recent rise of the soybean as a major American crop is an example of the value of research on new crops. Although it was introduced into the United States in 1850, its production remained relatively minor until about 1960. It was not until several developments occurred, such as characterization of the oil, utilization of the oil and meal, and the development of adapted varieties and optimum cultural practices, that the soybean reached its current level of production. In the mid 1970s, concern and support for research on potential new crops in 317

copright 0 by Academic Press, Inc. All rights of reproduction in any form reserved. ISBN 0-12JXMJ736-3

318

LEMOYNE HOGAN AND WILLIAM P. BEMIS

the United States began to be expressed. In a report prepared for the National Science Foundation (NSF) (Theisen ef al., 1978), the following four basic concerns for new crops research were listed: (1) the genetic pool of agricultural crops may be vulnerable unless enriched and diversified with new genetic material; (2) present crops are either very demanding environmentally or narrowly adapted ecologically; (3) present crops require very high energy inputs in the fom of fuel, fertilizers, pesticides, processing, and irrigation; and (4) the lack of production alternatives in diversified markets exposes producers unduly to price instabilities that are often induced by demand factors external to the U.S. economy. Other papers such as “Food and Agriculture” (Wortman, 1976) considered the problem of feeding the world population. Harlan (1976) points out that as agriculture evolved it changed from being highly diversified, involving thousands of species of plants of which relatively few were ever domesticated, into an agriculture so highly specialized that most of the world population is absolutely dependent on a handful of species. Harlan lists 30 crop species having an annual world production of 10 million metric tons or greater. These 30 crop species have a total yield of 2360 million metric tons and the top four crop species (i.e., wheat, rice, maize, and potato) account for 1280 metric tons or 54% of the total. The failure or partial failure of any one of these species could mean automatic starvation for millions of people. In the early 1800s Ireland developed a nearmonoculture agriculture based on the potato. Total crop failure in 1845and 1846 because of a fungal disease resulted in at least 1 million persons dying of starvation as a direct result of the 2-year famine (Salaman, 1949). There is, however, good reason that a few major crop species dominate world agriculture. These crops are profitable to produce on a very large scale and are well adapted to mechanization. The grain crops are not particularly perishable and can be stored and transported with relative ease. Each species is particularly adapted to large growing areas where they produce excellent yields. For example, the “corn belt” of the United States is particularly well adapted to producing corn (maize) (Loomis, 1976);the region has large expanses of relatively level land and the soil is fertile and well drained. The growing season includes at least 120 days when the temperature is above 10°C, and ample rainfall is normally well distributed over the growing season. Loomis (1976) states that in the western plains of the UNted States the supply of moisture is inadequate for corn, which is therefore grown only where irrigation is possible. The wheat plant, which dominates the semiarid croplands of the world, fills the need in these area for a cultivated crop with a lower demand for water and a greater tolerance for drought. Rice, a grain crop adapted to growing in fields flooded by water and requiring a long growing season to mature, is the staple grain of much of the world’s tropical agriculture. These and other major staple world crops are grown in the most desirable

BUFFALO GOURD AND JOJOBA: NEW AFUD CROPS

319

agricultural areas as determined by soil and climate. There are, however, vast areas of semiarid and arid regions of the world that are virtually unused as croplands because existing crop species are not sufficiently well adapted to xeric agriculture. The selection and development of new crops adapted to such regions could bring large areas of land presently unused into production. As Harlan (1976) has pointed out, agriculture as we know it today has evolved from a subsistence agriculture involving thousands of edible plant species into commercially produced crops developed from only a few hundred of those species. One might argue that through this evolutionary process, all potential crop species have been tested and eliminated, and there may be some validity to this argument, particularly with respect to the production areas defined as prime by soil and climate. New crops have become highly successful after being introduced and developed for adaptation into new areas. The potential for any new crop to achieve an established position in the highly competitive agricultural and industrial economy of the United States requires outstanding attributes as well as a long, adequately funded period of research and development. The purpose of this article is to review the progress made in developing two potential crops for arid lands: buffalo gourd and jojoba.

II. BUFFALO GOURD: Cucurbita foetidissima HBK A. INTRODUCTION

This feral species of Cucurbitu, the genus that includes the cultivated squash and pumpkins, is a New World species indigenous to the arid and semiarid regions of western North America (Bemis et ul., 1978a). The origin of the diversification of Cucurbitu is southern Mexico to Guatemala, according to Whitaker and Bemis (1975). The genus evolved into many species-tropical, mesophytic, and a group of highly specialized species adapted to xerophytic environments; the buffalo gourd was one of these. The buffalo gourd is perennial by virtue of its exceedingly large fleshy storage roots. A single root may reach fresh weights of over 40 kg in three or four seasons of growth. The frost-sensitive vines are killed by temperatures below O'C, but the roots may survive winter air temperatures as low as -25'C, particularly when the soil has the insulation of snow cover. The primary mode of reproduction is asexual by the development of adventitious roots produced at the nodes of the vines. Large homogeneous colonies of plants are produced in this manner, and annual vine growth is extensive. The large unisexual flowers (pistillate or staminate) are borne singly at most

320

LEMOYNE HOGAN AND WILLIAM P. BEMIS

nodes of the vine. The predominant sex expression of the plant is monoecious; pistillate and staminate flowers are produced on the same plant. However, a dominant mutant gene causing abortion of the male flower buds resulting in gynoecious (all female) plants was found to be widespread in native populations. The fruit (pepos) are usually round with diameters of 5-7 cm. A single plant is capable of producing as many as 200 fruits in a single growing season. The number of seeds per fruit ranges from 200 to 300, with an average weight of 4 g/100 seeds; the seed contains 3 0 4 0 % edible oil and 30-3596 protein. The vine growth is extremely prolific. A single root in its second season of growth will produce from 6 to 20 vine initials that are capable of producing vines 2 6 m in length. These initial vines also branch at many of the nodes resulting in a dense mat of vine growth. The vines are ground cover rather than climbing vines; their harsh, sandpaper-like leaves are usually entire, ovate to sagittate, with a base width of 10-13 cm and a midrib length of 20-25 cm. The photosynthetic capability of these plants must be tremendous. The association between the aboriginal Americans and the buffalo gourd has existed for as long as 9ooo years (Cutler and Whitaker, 1961). Although it was never a cultivated plant, it was an excellent “camp follower,” being well adapted to perennial growth on trash heaps around centers of habitation. It can be found today growing on dump heaps around the mesa top pueblos of the Hopi Indians in some of the oldest continuously inhabited settlements in America (Nabhan, 1980). The Indians used the buffalo gourd for food, utilizing the seeds which were usually ground up and eaten in the form of a cooked mush. All parts of the buffalo gourd plant are extremely bitter because of water-soluble glycosides known as cucurbitacins. The seeds from the fruit do not contain this bitter chemical, but unless they are washed well they will be coated with it because it is present throughout the pulp of the fruit. Besides eating them cooked, the Indians extracted oil from the ground up seeds by rinsing them in hot water. The turgid pepos, and to a lesser extent the root, were used domestically as hand and laundry soap, shampoo, and stain remover; a turgid pep0 cut in half and added to laundry results in a detergent action. The root of the buffalo gourd was used extensively by the Indians for medicinal purposes. It was used to cure saddle sores on their horses, as a laxative, and as a treatment for hemorrhoids. The root is so extremely bitter that it was not used as a food. Thus for more than 9OOO years, the buffalo gourd evolved under the passive influence of the aboriginal American who provided a trash heap culture with additional waste water for the continued existence of the buffalo gourd. During this period of its evolution, the buffalo gourd may have become dependent upon man for its survival (Bohrer, 1975), as it is not found mainly on disturbed soil areas such as abandoned cultivated fields, pastures, fence rows, and railroad and highway right-of-ways. The program at the University of Arizona in domesticating the buffalo gourd, discussed in the following section, illustrates the difficulties and problems encountered in research on a wild plant species.

BUFFALO GOURD AND JOJOBA: NEW ARID CROPS

32 1

B . DOMESTICATION

The early use of the buffalo gourds by the American Indians was described in the previous section. Curtis (1946) described the potential of this feral species, based on native colonies of this species: (a) the plants are perennial; (b) they grow on wastelands in regions of low rainfall; (c) they can produce an abundant crop of fruit containing seeds rich in oil and protein; and (d) the fruit lends itself to mechanical harvesting. Curtis states in conclusion that it is ironic that the answers to some of the problems of our undernourished populations may be growing widely in their immediate vicinities as neglected weeds. The first physical and chemical analysis of buffalo gourd seed oil was reported by Wood and Jones (1943). The first detailed chemical analysis of the seed oil was reported by Shahani et al. (1951). The fatty acid composition was acceptable; the major components are linoleic (65.3%), oleic (23.0%), palmitic (6.13%), and stearic (2.22%) acids. They reported that the crude oil was dark in color and exceedingly resistant to bleaching. Refining, bleaching, and deodorization, however, gave a bland oil with good stability and no tendency to revert to poor flavor. A more extensive publication by Bolley et al. (1950) on the utilization of the buffalo gourd concluded that the wild gourds should continue to receive serious consideration as an oil seed crop, particularly for arid and semiarid land. In 1952, Paur studied some small plots of buffalo gourds in New Mexico where the plants grew well on a light mesa soil, had low water requirements, and were free from diseases and insect injury. There was little research activity on the domestication of the buffalo gourd, however, until an extensive report was published by Curtis and Rebeiz (1974) covering 6 years of field research to domesticate the buffalo gourd as part of a project sponsored by the Ford Foundation in Tel Amara, Lebanon. (The program was terminated with the retirement of L. C. Curtis and the subsequent civil war in Lebanon.) The initial germ plasm source upon which the work in Lebanon depended came from a restricted genetic base of a single collection from a site in Texas. Despite this narrow genetic base, the results indicated a vast store of genetic variation within this species. For example, fruit yields from 712 single, 2-year-old plants, spaced 3 X 3 m, were as follows: 60 plants produced 0 fruit, 533 plants produced from 1 to 100 fruits, 104 produced from 101 to 200 fruits, and 15 produced from 201 to 300 fruits. The highest yielding plant produced 271 fruits. The crude fat from seed of 50 selected plants ranged from 25.6 to 42.8%, and the protein ranged from 25.9 to 35.0%. Many plants with unusual characters such as vine habit and sex expression were observed, again indicating the enormous amount of genetic variation in this species. An additional paper by Curtis and Gomez (1974) was published in Mexico where a program cooperative with the one in Lebanon was continuing. Cucurbitufoetidissima has been part of a research project at the University of

322

LEMOYNE HOGAN AND WILLIAM P. BEMIS

Arizona since 1963, but only as one of several species of Cucurbitu used to study the genetic barriers to interspecific hybridization. In 1973, at the urging of L. C. Curtis and others, a movement commenced at the University of Arizona to explore the possibility of initiating a full-scale research project on the domestication and utilization of the buffalo gourd. A committee appointed in the College of Agriculture to study the feasibility of this investigation recommended the initiation of such a study. With the assistance of the Agency for International Development, a technical series bulletin was published (Bemis et al., 1975). The project was conceived as covering a broad spectrum including breeding, domestication, and utilization. Obviously, such a broad-based program called for an interdisciplinary approach to these problems. The initial major disciplines involved were genetics and plant breeding, biochemistry, and nutrition and toxicology (Bemis et al., 1978b). To study the agronomics of the buffalo gourd as a new field crop, the first major objective was the creation of a relatively homogeneous seed source from which cultural plots would be established, the seed originally collected from wild colonies was extremely heterogeneous. In addition to this requirement, the test seed had to be representative of the best available germ plasm for crop production. I . Plant Collections

The current distribution of the feral buffalo gourd extends over nearly 3 million kmz in North America. The north-south axis of its range extends from Guanajuato in central Mexico to near the southern South Dakota border in the United States, a distance of about 2700 km. Presumably the major stands of plants extend from the Chihuahuan plateau in Mexico northward along the eastern base of the Rocky Mountains. A westward branch extends through New Mexico into south central Arizona. The eastern extension of its range, almost to the Mississippi River, is probably recent because the plants are found mainly along highways or railroad right-of-ways. Its most westward advance into the coastal mountains of southern California is also probably recent. It can be considered both a ruderal (disturbed areas, roadsides) and an agrestal colonizer (agricultural weed). The mature gourds are attractive and were probably carried along by people in their travels, which greatly extended the current native range of the buffalo gourd. Regardless of how its current range was developed, it represents the genetic variation that exists in this species. An effort was made to bring a cross section of this genetic diversity into one location through plant collection trips and the establishment of germ plasm nurseries (GPN). Table I shows the number of accessions collected from three trips made in 1975, 1976, and 1977, and the three germ plasm nurseries established from the collected accessions. (Some of

323

BUFFALO GOURD AND JOJOBA: NEW ARID CROPS

Table I Accessions Used in Seeding the Three Germ Plasm Nurseries (GPN) Tabulated by Collection Site Collection site

Number of accessions GPN-76

GPN-77

GPN-78 ~~

Arizona New Mexico Texas California Mexico Nebraska Kansas Oklahoma Colorado Utah Illinois"

43 2 6

1

17 10 1

9 6 6 6 -

Total ~

47 23 21 11 10 9 6 6 6

3 4 5 10

-

5

1

10 -

67

1 -

I45

"Five adventitious mots sent from Illinois have been established in GPN-77.

the accessions shown in Table I were not from the collection trips per se, but were sent in by interested cooperators.)

2 . Germ Plasm Nurseries The purpose of the germplasm nurseries (GPN) is to establish a population of buffalo gourd plants that would represent a cross section of the genetic diversity found in the range of its native habitat. It is obvious that growing the GPN in a single location will not give the full extent of inherent genetic diversity because the environment is held as a constant. However, economic limitations restricted the study to a single location. The degree of variation found in some of the original collections is shown in Table II. The data in Table I1 are from fruit representing 85 accessions or collection sites (29 in Arizona, 19 in New Mexico, 12 in Texas, 7 in Nebraska, and 6 each in Colorado, Kansas, and Oklahoma) (Scheerens et al., 1978). The GPNs were established at the Agricultural Experiment Station of the University of Arizona, Tucson, Arizona. Seed for each accession was hand planted in 15 hills 1 m apart in rows 2 m apart, and subsequently thinned to 1 plant per hill. A total of 145 accessions were seeded in this manner. These GPNs now serve as a source of material to initiate breeding programs. Because the buffalo gourd is a perennial species, it presented the problem of when (Le., at what plant age) selections for improved plant types should be

324

LEMOYNE HOGAN AND WILLIAM P. BEMIS

Table I1

Means,Ranges, and Coefficients of Variation (CV) of Seven Fruit and Seed Chpraetws of 85 Buffalo G o d Accessions charactelistic

Mean

Range

cv

Fruit diameter (cm) Seed weightl100 seeds (g) Seed weightlfmit (9) Seed number/fmit Embryo in seed (76) Crude fat in seed (8) Crude protein in seed (%)

6.5 3.8 8.4 225 67.4 32.9 30.7

5.2-7.7 1.1-5.5 2.2-18.8 87-386 42.1-76.3 21.143.1 19.5-35.4

9.1 20.0 36.9 38.3 9.7 14.3 8.7

made. In order to expedite the program, plant growth and fruit production were recorded in the first season, but the selection of inidividual plants for breeding stock was delayed until the second season of growth. The criteria for plant selection were primarily plant vigor and fruit production, but a chemical analysis for seed oil quantity and quality was also made. Plant selections were made in the GPNs and a final selection from within eight accessions is shown in Table III. With the exception of Accession 300, all were from populations with segregating monoecious and gynoecious plants. C.

SEED P R O D U ~ I O N

The buffalo gourd has been classified as monoecious, which is consistent with all of the known species of Cucurbiru. Observations of plants in GPN-76 during Table III Accession Numbers and Collection Locations of Lines Selected for Breeding and Seed Production Accession number

Collection location

140 142 156

Forest Dale, Arizona Jerome, Arizona Greaterville, Arizona Patagnia, Arizona Lochiel, Arizona Pearce, Arizona Hatch, New Mexico Saltillo, Mexico

158

162 185 250 300

BUFFALO GOURD AND JOIOBA: NEW ARID CROPS

325

their second growing season indicated that many of the accessions were segregating into two types of sex expression. Forty-seven accessions, each having a population of 10 or more plants, were classified as having all monoecious plants (pistillate and staminate flowers) or as segregating into monoecious and gynoecious plants (pistillate flowers only). Twenty-five accessions were entirely monoecious; however, 22 accessions were segregating into monoecious and gynoecious plants and were randomly scattered throughout the collections. The total count for sex type of the 22 segregating accessions was 153 monoecious to 118 gynoecious plants. Gynoecious expression is conditioned by a dominant gene restricted to the heterozygous state, and monoecious expression is determined by the homozygous recessive state (i.e., Aa is gynoecious sex, and aa is monoecious sex) (Dossey et al., 1981). A suggested term for this type of sexual expression is monogymdioecy , indicating populations that contain monoecious and gynoecious plants in approximately equal numbers. Curtis and Gomez (1974) observed a similar, or possibly the same, type of sex expression in their buffalo gourd populations in Lebanon. They referred to this mutant type as antherless and noted that it segregated in a 1: 1 ratio when crossed with a monoecious plant. They also reported that the mean fruit yield per plant for the 10 best antherless plants was 272 fruits; the mean fruit per plant of the 6 best monoecious plants was only 209 fruits. Because a major objective of the domestication program was the rapid production of a relatively homogeneous seed supply, it was decided to produce “hybrid seed” utilizing the gynoecious character. After only one generation of selection and selfing (in breeding), the experimental hybrids shown in Table IV with the Table IV

Seed Source of 10 Experimental Entries of Buffalo Gourds in a Replicated Yield Test at Marana, Arizona Entry number

Designation

Seed source accession numberso

1 2

Experimental hybrid Experimental hybrid Experimental hybrid Experimental hybrid Experimental hybrid Experimental hybrid Experimental hybrid OP selectionb OP selection OP selection

158 X 142 142 X 158 140 X 142 162 X 142 184 X 142 250 X 300 (140 X 156) X 300 142 158 300

3 4

5 6

I 8 9 10

“Refer to Table ID for accession number identification open polliiation.

LEMOYNE HOGAN AND WILLIAM P. BEMIS

326

Table V First-season Fruit, Seed, and Projected Seed Oil Yields from the Marana Yield Triala

Entry number* 4 7 6 3 1

5 2 10 8 9

FruiVplot (84 mZ)

464 474 344 270 202 169 176 252 75 14

Seed weighVplotC (kg) 7.57 4.84 4.31 3.34 2.92 2.66 2.36 1.91 1.17 0.17

(a)

(b) (bc)

Seed yield (kg/ha)

Projected oil yieldd (kg/ha)

908.4 580.8 512.4

376.9 222.6 200.4 158.5 135.3 125.4 11.3 81.2 54.6 8.2

(bcd)

400.8

(cde) (cde) (cde)

350.4 319.2 283.2 229.2 140.4 20.4

(de)

(ef) (f)

“Data are means of four replications. *Refer to Table IV for entry identification. ‘Meansfollowed by the same letter are not significantly different at the 0.05%level. dBased on 100% extraction of seed oil.

three pollinator lines were tested in a replicated yield trial. The yield test was seeded in dry beds at the Agricultural Experiment Station, Marana, Arizona, on May 7, 1979, and irrigated until May 10, 1979. The individual plots were 4 rows 1 m wide and 21 m long, giving each plot an area of 84 m2. The plots were separated by borders 2 m wide. The plants were thinned to single plants 70 cm apart in the rows. The plant stand was good and there were approximately 120 plants in each plot. The plots were replicated four times. The vines were periodically turned back onto the plots to keep them from intertwining. Yield data for the first season are presented in Tables V-VII (Vasconcellos et al., 1981). Table V shows the fruit and seed yields per plot and the projected yields of seed and seed oil in kilograms per hectare. The buffalo gourd is a perennial, and many of the wild collections fail to flower in the first season from seeding. Thus these data are limited and do not reflect the ultimate potential of buffalo gourd yields. The data do, however, show the tremendous genetic variation of these highly selected lines. It is of interest to plant breeders and geneticists that the seven experimental hybrids were superior in first-season seed yield over the three highly selected open-pollinated parental lines (i.e., Accessions 142, 158, and 300). The first-season root and projected starch yields are shown in Table VI. The plants were thinned to an in-row spacing of about 70 cm between plants, which was designated to give maximum root yields after the second (see Table IX) or third season of growth; thus the root yields are relatively low for the first season. For example, the calculated plant density at the spacings in the yield trials is

327

BUFFALO GOURD AND JOJOBA: NEW ARID CROPS

Table VI First-Season Root and Projected Starch Yieidsu

Entry numberb 10 7 6 5 8 2 3 9 4 1

Fresh weightC Root yields (kgplot) &@a) 23.5 22.0 21.8 21.3 19.7 19.4 18.6 17.4 17.3 14.9

(a) (ab) (ab) (ab) (abc) (bc) (bc)

(cd) (cd) (d)

Projected starch yield (kg/ha)

11,162 10,459 10,402 10,117 9,357 9,215 8,835 8,265 8,270 7,077

,

1997 1808 1976 1821 1712 1769 1599 1603 1709 1274

"The data are from a single row 21-m long containing approximtely 30 plants; the data are means of four replications. "Refer to Table IV for entry identification. 'Means followed by the same letter are not significantly different at the 0.05% level.

14,285 plants/ha and the mean fresh weight of first-season roots is 655 g, giving a fresh root yield of 9357 kg/ha. For first-season high root yields the plant density must be increased. A trial planting at Mesa, Arizona, in which the plant density was 330,000 plants/ha, produced first-season root yields of 28,420 kg/ha. In these high-density plant populations, the mean fresh weight per root Table VIl The Quantity and Fatty Acid Spectra of Crude Seed Oils from First-Season Plants Fatty acid spechum

(a)

Entry numbera

crude oil

160

18:O

18:l

18:2

1 2 3 4 5 6 7 8 9 10

38.62 29.29 29.55 41.49 39.28 28.48 39.12 38.89 40.46 35.43

8.1 7.0 8.2 7.4 8.5 8.1 7.4 7.4 10.2 8.4

3.8 3.9 4.6 3.4 3.5 3.0 3.0 3.0 3.0 3.0

24.9 30.4 29.6 27.3 27.1 24.7 25.5 21.1 25.6 26.5

63.3 58.7 57.6 61.8 60.8 64.1 64.2 68.5 61.2 62.1

"Refer to Table IV for entry identification.

L.EM0YNE HOGAN AND WILLIAM P. BEMIS

328

Table VIII

Seed Yield per Replication Exprsrscd as k g b for the Three Best Yielding Entries over Three !haom of Yielding@ Entry number 4

Entry number 6

Entry number 7

Replication

1979

1980

1981

1979

1980

1981

1979

1980

1981

A

1172 923 918 620

3274 1327 491 692

24% 2325 323 326

513 439

339 993 1028 234

222 1%5 1822 113

352 984 622 368

495 1814 1220 334

365 1746 363 422

B C

D

558 501

"The data for 1980 and 1981 mflect herbicide injury to the plants.

was only 86 g and the percentage starch was slightly lower than for larger roots. The vines, under high plant densities, failed to produce fruit. The concept of high plant densities for annual toot production is still being studied. The quantity and quality of the crude seed oil for the 10 entries in the variety trial are shown in Table VII. The experimental hybrids and the three openpollinated selections are very consistent in their percentage crude seed oil and fatty acid spectra. They all have relatively high levels of linoleic acid (18:2), which makes the vegetable oil desirable from a nutritional viewpoint. The mean crude seed oil for 85 wild accessions was 32.9% (Table 11), whereas the 10 entries in the trial test had a mean of 39.0%. This appears to be a selective improvement over the mean of the original population. In 1980, the second season of growth of the yield plots at Marana, the roots apparently penetrated a zone of residual herbicides and most of the plants were adversely affected, rendering the plots so highly variable that for the most part they were meaningless. Table VIII illustrates the problem. The speculation of seed yields of 2000 kg/ha remains a speculation at this point in time. D. P R O D UEVALUATION ~

The primary agricultural products of the buffalo gourd are vegetable oil and protein from the seed, starch from the root, and possibly forage from the vines (Bemis er al., 1979). Seeds account for about one-third of the dry weight of a buffalo gourd fruit and are composed of one-third seed coat and two-thirds embryo. The storage cotyledons of the embryo contain the oil and protein. The oil in the seed can be readily isolated by solvent extraction or by mechanical pressing. The quantitative and qualitative values of the seed oil are shown in Table VII. Unrefined oil has been evaluated in a feeding study with weanling mice (Bemis et al., 1977)by incorporation in isocaloric isonitrogenous diets in

BUFFALO GOURD AND JOJOBA: NEW ARID CROPS

329

Table M Second-Season Root Yields“

Entry numberb

Fresh weightC (kg/plot)

Root yields Wha)

7 6 8 9 10 2 3 5 4 1

37.1 36.1 34.5 31.5 31.4 31.1 29.9 29.7 27.0 26.8

17,622 17,147 16,387 14,962 14,915 14,772 14,202 14,107 12,825 12,730

_

_

_

~

“Data from a single row 21 m long containing approximately 30 plants; means of four replications. bRefer to Table IV for entry identification. ‘No significant differences at the 0.05%level.

amounts ranging up to 11% of the total diet. Even at the highest level, normal growth occurred with no evidence of deleterious effects. Buffalo gourd seed oil appears to be a satisfactory vegetable oil, having properties similar to corn oil. Approximately one-third of the buffalo gourd seed is protein. The essential amino acid composition as percentage of the total protein and compared with whole egg protein is shown in Table X. Buffalo gourd seed meal protein is typical of most plant proteins, that is, to be an efficient food or feed for monogastric animals it must be supplemented with additional amino acids. A recent feeding study with mice has been completed (Thompson et al., 1978), and the Table X Essential Amino Acid Composition as Percentage of Protein Amino acid

Whole egg

Buffalo gourd

Lysine Histidine Arginine Threonine Valine Methionine + cysteine Isoleucine Leucine Phenylalanine + tyrosine

6.3 2.2 6.7 4.8 7.6 7.6 8.7 9.5 10.4

4.5 2.2 13.0 2.0 3.8 1.7 3.3 5.7 7.4

330

UMOYNE HOGAN AND WILLIAM P. BEMIS

Essentipl Amino

Table X Acid Composition as Percentage of Protein

Amino acid

Whole egg

Lysine Histidine Arginine Tbreonine Valie Methionine + cysteine Isoleucine Leucine Phenylalanine + tyrosine

6.3 2.2 6.7 4.8 7.6 7.6 8.7 8.5 10.4

Buffalo gourd

4.5 2.2 13.0 2.0 3.8 1.7 3.3 5.7 1.4

results are summarized in Table XI. Body weight gains of mice fed buffalo gourd seed supplemented with lysine, methionine, threonine, and valine were equivalent to those of mice fed whole egg protein. There is no evidence that any toxic material is associated with the oil or protein from buffalo gourd seed. Starch is the major component of the large storage roots of the buffalo gourd (Tables VI and M).Starch is readily isolated from the roots in the laboratory (Berry er al., 1975). Washed roots are ground in water to a slurry which is filtered through a 150 mesh screen to remove most of the fiber content of the root. The starch is allowed to settle from the slurry and is centrifuged to remove excess fibers. Isolation of starch in this manner provides a product which is free Table XI Body Weight Gain,Protein EfFiciency Ratio (PER), and Net Protein Retention (NPR) of Mice Fed Buffalo Gourd Seed Supplemented with Amino AcidsD Treatment Whole egg Buffalo gourd whole seed (BG) BG + Met BG + Lys BG + Val BG + Thr BG + Met + Lys BG + Met + Thr BG + Met + Thr + Lys + Val

Body weight gain (g/day) 0.44 0.22 0.24 0.21 0.21 0.31 0.25 0.26

0.46

'Values with different letters are significantly different at the 0.05% level.

PER 2.50 1.21 1.24 1.17 1.07 1.45 1.28 1.35 2.04

(c) (a) (a) (a) (a) (a) (a) (a) (b)

NRP

4.02 2.31 2.56 2.32 2.28 2.48 2.40 2.72 3.00

(c) (a) (ab) (a) (ab) (ab) (ab) (ab) (b)

BUFFALO GOURD AND JOJOBA NEW ARID CROPS

331

Table XU

Properties of Buffalo Gourd Starch and Other Starches Analysis Protein (9%) Lipid (%) Ash (966) Phosphorus (%) Granule mean diameter (pm) Granule diameter range (pm) Gelatinization ("C) Amylose (blue value %) Iodine affiity

Buffalo gourd 0.85 0.57 0.10 0.022 6 2-17 57.0-60.5 26.0 5.06

Potato 0.05 0.50 0.35 0.075 50 15-100

58.0-66.8 24.0 4.61

Tapioca 0.19 0.99 0.18 0.01 17 5-35 58.5-70.0 20.0 4.43

Corn 0.83 0.61 0.07 0.018 18 10-25 62.0-70.8 20.3 4.02

of cucurbitacins, the extremely bitter glycosides that are present in all parts of the buffalo gourd plant except the seeds. Buffalo gourd root starch has unique rheological properties that make it attractive as a food additive. Similar to cassava root starch, it may have an application in the food industry. It has excellent viscosity and stability properties at prolonged periods of high temperatures. The general properties of buffalo gourd starch are compared with other starches in Table XU. One of the striking features of the buffalo gourd is its ability to produce large amounts of foliage. The prodigious vine growth represents a potential source of feed for ruminants. Harvesting the vegetation prior to frost exposure gives a material which is nearly 60% digestible with a crude protein level in the range of 10-13% (dry weight basis). Experimental diets based on such material have a digestible energy estimated at 2 kcal/kg dry matter (Waymack et al., 1979). E. RECENTDEVELOPMENTS

The buffalo gourd is currently being grown in Australia where several hundred acres have been seeded under dry-land agriculture. The initial seeding was made in November, 1981. The Australian venture is being promoted by a private corporation, Primary Energy Australia Pty. Ltd., which is working in close association with the University of Arizona. Two tons of buffalo gourd roots have been processed for starch extraction at a commercial starch plant in Monte Vista, Colorado, operated by the A. E. Staley Mgf. Co. If a new, feral species is to be developed into an agricultural crop, it must have the support of the financial sector for large-scale testing. The buffalo gourd is now on the threshold of such a program.

332

LEMOYNE HOGAN AND WILLIAM P. BEMIS

111. JOJOBA: Simmondsia chinensis (Link) Schneider A. INTRODUCTION

Jojoba, Simmondsiu chinensis (Link) Schneider, is a large evergreen shrub which is native only to the Sonoran Desert of Arizona, California, and northwestern Mexico in locations with specific environmental and soil conditions (Gentry, 1958). Since 1972, there has been worldwide interest by many research and commercial groups in the domestication of this plant as a new crop for arid regions. The primary factor contributing to this interest is the production, under relatively low moisture conditions, of seed containing from 40 to 60% high quality liquid wax. The composition and physical properties of jojoba wax closely resemble those of sperm whale oil, which has been widely used in cosmetics and lubricants because of its oiliness, metallic wetting properties, and nondrying characteristics. Natural jojoba stands have provided commercial quantities of seed for research, product development, and the development of markets. Although no accurate data are available, it is estimated that between 10,OOO and 16,000 metric tons of clean, dry seed are produced by wild plants each year in the United States and Mexico (National Academy of Sciences, 1977). Only a relatively small amounts of this seed is actually harvested because of inaccessibility of stands, costs of harvesting, and land ownership. The seed yield of wild plants fluctuates greatly from year to year depending on local climatic conditions. This annual variation in availability of seed from wild plants and fluctuating prices has seriously affected the availability of seed and wax to consumers during the past several years (Katoh and Kunimoto, 1983). The f m t plantation-grown seed was marketed in 1982, and as additional plantings come into production it is believed that stabilized supplies of seeds will stimulate additional markets. I . Plant Description

Although jojoba was assigned to the Buxaceae family by early plant taxonomists, there remains controversy as to whether or not it should be placed in a new monotypic family, Simmondsiaceae, as it has few characteristics of other members of the boxwood family (Gentry, 1958). Studies have shown that 2 N = 52 chromosomes in jojoba pollen mother cells. There is also evidence that jojoba is a polyploid (Sherman and Ray, 1978). Jojoba is believed to be extremely long lived, from 100 to 200 years under wild conditions. (Ferguson, 1977). The size of mature plants is influenced by soil type, temperature, available moisture, and genetic characteristics. Mature plant height varies from 0.6 m in poor sites to more than 4.57 in good sites. Its opposite leaves usually live through two or more seasons and continue growth

BUFFALO GOURD AND JOJOBA NEW ARID CROPS

333

during the second year. The leaves vary considerably in pubescence, thickness, color, and size on different plants (Gentry, 1958); they are usually soft and greygreen the first year, changing to pale yellow-green after the second year. In deep soils, multibranched roots as long as 9 m are common. Jojoba is dioecious (i.e., male and female flowers occur on different plants). Seedling populations initially contain approximately equal numbers of male and female plants. Studies have shown that under more stressful conditions, such as on south-facing slopes, more males than females are found (Haase, 1976). Seedling jojoba plants vary greatly in time of flower initiation, pollen release by males, and the receptivity of female flowers. The time of seed development and maturity also varies for individual plants and at different sites. The majority of jojoba plants in native habitats initiate flower buds in late summer or fall. Flower bud formation is largely associated with, and dependent on, new vegetative growth which occurs in response to temperature and moisture. Female flowers are green, urn-shaped, and relatively inconspicuous. The pistils grow from the tip of a green ovary; each ovary contains ovules, but usually only one seed develops per ovary. Male flowers are found in clusters varying in number from three to several hundred. Most pollination and fruit development occurs in early or late spring; however, some pollination may take place in the fall with some fruit development occurring in early winter. In the United States, seed maturity usually occurs during a 3-month period from the middle of June to the middle of September (Sherbrooke, 1978). The fruiting pattern varies from plant to plant, but the normal pattern is the formation of a single fruit at alternate nodes on new growth. Plants are also found that produce a single fruit at each node or multiple fruits in clusters. Mature jojoba seed has an outer tough, leathery, dark-brown seed coat. Seeds vary in color, shape, and size among individual plants and populations. Seed size is controlled both by genetic and environmental factors and ranges from 700 to 5300 seeds/kg (Gentry, 1958). The large cotyledons make up the bulk of the seed and are filled with liquid wax. Jojoba is wind pollinated. The flowers of both sexes lack petals, nectaries, and scent glands (Sherbrooke, 1978). Large quantities of pollen, as much as 15 g/plant, are produced by male plants. Jojoba pollen is relatively heavy and settles out of the air rapidly, falling within a few meters of the point of origin. Jojoba is one of the earliest Sonoran Desert plants to flower in the spring (Buchmann et al., 1983). Honey and other native bees remove large quantities of pollen for food before other pollen sources are available. Pollen can be stored for as long as 5 months under liquid nitrogen at - 197°C (Lee eb al., 1980). 2 . Natural Distribution

Natural populations of jojoba are distributed over about 38,620 km2 of the Sonoran Desert of Mexico and the United States between latitudes 25" and 31"N

334

LEMOYNE HOGAN AND WILLIAM P. BEMIS

and longitudes 101" and 117" W (Gentry, 1958). Arizona contains the largest concentration of native jojoba in the United States. Very large native populations are found east of Phoenix near Globe, Superior, and Winkleman, and other large stands are found surrounding Tucson and southwest of Tucson adjacent to and on the Papago Indian reservation. The Kofa Mountains near Yuma also contain large populations. Jojoba stands found in California along the southern coast and inland deserts; 11 distinct populations have been identified in southern California, and 2 relatively large populations are found at Twentynine Palms and Aguanga. The other 9 California populations are small, each consisting of only a few hundred plants (Yermanos, 1974). The largest native stands in Mexico are found in northern and southern Baja California and coastal Sonora. Plant population density varies from a few individual shrubs in the Dos Cabezas Mountains east of Wilcox, Arizona, to as many as 500 plants/ha in the Tucson Mountains. Near heavily populated areas such as Tucson, some natural stands are gradually being replaced by urban development. Wild stands of jojoba are found from sea level, along the coast of Mexico, to at least 1524 m, at Wasson Peak in the Tucson Mountains. The distribution of jojoba at different elevations is strongly correlated with low temperatures. Native stands at the higher elevations in Arizona are usually found on slopes that have excellent air drainage; they are not found in the lower valleys which receive the cold air drainage from nearby mountain slopes (Hogan, 1979). The densest stands are usually located on the north, northeast, and northwest slopes (Sherbrooke, 1978). At lower latitudes with warmer winters, as in Sonora and Baja California, they can be found at low elevations on relatively level land. Mature jojoba plants will withstand low temperatures of -9.5"C and young hardened plants will withstand low temperatures of -4°C for short periods (Gentry, 1958). Although mature plants will survive relatively low temperatures, fruit production and vegetative growth can be severely restricted (Hogan, 1979). Jojoba is a drought- and heat-tolerant species. It is capable of withstanding extremely high temperatures and severe water stress (Al-Ani et al., 1972). It can tolerate highs of 43-46°C in the shade. In Sonora and Baja California, mature jojoba has been observed to defoliate for several months under extreme drought conditions and resume growth after receiving rains. The most productive wild jojoba is that receiving 38-46 cm of moisture per year (Gentry, 1958). Native stands of jojoba are usually restricted to well-drained, coarse desert soils and coarse mixtures of gravels and clays (Gentry, 1958). It has been found that some individual plants are very tolerant of salinity although others are sensitive (Rasoolzadegan, 1980). 3. Germ Plasm

Wild jojoba is extremely heterogeneous when grown from seed. Extreme variations have been found in all characteristics studied, including growth habit,

BUFFALO GOURD AND JOJOBA: NEW ARID CROPS

335

seed size and shape, time of flowering, pollination, fruit maturity, yield, and tolerance to salinity. The first significant jojoba seed germ plasm collection was made by Gentry in 1957 (Forti, 1973). This collection was made available to the University of Arizona to establish germ plasm nurseries. The largest collection of vegetatively propagated germ plasm material was made by the University of Arizona and is currently being used for selection and plant breeding (Hogan and Palzkill, 1983). B. IMPROVEMENTOF NATURALSTANDS

There has been continued interest in the United States and Mexico in the manipulation and management of wild populations by removing competing species and improving the availability of moisture through water harvesting. Ehrler and Fink (1978) have increased yields by water harvesting, but it remains to be seen if these procedures will be economically feasible considering the terrain in which the native plants grow and the necessity for hand-harvesting such areas. C. PLANTATION CULTURE

If jojoba is to become an important commodity in world trade, it must be domesticated and grown under cultivation (Mirov, 1952). The limited quantity and unpredictability of supplies and the prices of seed seriously impact the development of new jojoba products and markets (Libby, 1980; Katoh and Kunimoto, 1983). The first commercial plantations in the United States, consisting of 130 ha, were planted in the fall of 1978. The planting of commercial jojoba plantations (McKeon, 1983) is proceeding at a very fast pace both in the southwestern United States and in other arid regions of the world. A survey of commercial plantings by Whittaker (1983) showed that 6667 ha in Arizona, 4176 ha in California, and 71 ha in Texas had been established in the United States by October, 1982. An additional 7400 ha are scheduled to be planted in the United States in 1983. Successful plantation establishment depends on the following: (1) the availability of superior uniform cultivars; (2) the development of suitable cultural practices; and (3) the design and development of a low cost harvesting system (National Academy of Sciences, 1977). The nonavailability to growers of superior, uniform cultivars is presently a major limiting factor to higher yields and additional plantings. Commercial growers have proceeded by stages to an understanding of the great variability found in jojoba seedlings. Seedling transplants were used to establish the first plantings, but it was soon realized that this was not cost effective as it did not allow growers to economically eliminate excess male and low-producing female plants. Transplanted seedlings are still used to a limited extent to fill gaps in direct-seeded fields. Approximately 356 ha of

336

LEMOYNE HOGAN AND WILLIAM P. BEMIS

seedling transplants have been established in the United States (Whittaker, 1983). Direct seeding became the standard planting practice about 1978. Various types of row-crop planters have been used by growers with both good and poor results. The most uniform stands have been obtained by growers using vacuumplate planters. Direct seeding of jojoba requires good control of several factors: viable seed, temperature, seedbed preparation, moisture, and seed depth. Although jojoba seed can be stored for several years under good environmental conditions and still have a high germination rate, it must be remembered that seed from any one source is extremely variable (Hogan, 1979). All sources of seed have been taken from many extremely variable wild or cultivated mother plants, thus these seeds have neither the same germination requirements nor the same germination rate. Therefore, the selection of any particular environment for planting will not be optimum for all the seeds planted at that location. Some seeds will require a longer germination than others, some will germinate under very saline conditions, and some will not tolerate salinity.(Rasoolzadegan, 1980). During recent work at the University of Arizona it was found that although 30% of seed from one mother plant germinated at 3.5% total soluble salt, seed from other sources could not tolerate lower levels of salinity. Some seedlings are vigorous whereas others are very slow growing. Preliminary studies have shown serious variations in the factors affecting jojoba seed germination which thus emphasizes the need for more careful research in this area. One has only to study a typical direct-seeded field soon after germination to understand the importance of the problem. If a good stand is obtained initially and if the seedlings grow vigorously, many common cultural problems are reduced. Many of the unexpected establishment and cultural costs growers encounter are a direct result of an initially poor stand of plants. If a grower is planting late in the spring when soil temperatures are too high for good germination, a poor stand will result, and he must then wait until late summer or early fall when suitable soil temperatures again occur to replant the “skips.” If he waits too late into the fall to plant, soil temperatures will be too low for good germination. Unevenly aged plants are much more difficult to maintain in a planting than evenly aged plants, and initial seed production will also be delayed. The optimum depth of seeding varies according to soil type, soil temperature, moisture management, and wind. Many inexperienced growers find that the arid desert soils suitable for jojoba are difficult to manage because of sudden and drastic changes in weather. The optimum depth of planting is easier to determine if a grower is very familiar with his particular property. More experienced growers have learned when they can expect strong wind storms that will uncover seeds, how fast their soils dry, and the land area that can be kept moist with their irrigation system for seed germination. They also will know when and for how long soil temperatures will be suitable for the best jojoba seed germination. If the

BUFFALO GOURD AND JOJOBA: NEW ARID CROPS

337

soil temperature is 21"-27°C for at least 3-4 weeks, the seed is kept moist, and the wind neither uncovers nor burys it too deep; a planting depth of 1.90-2.54 cm usually gives good results. Weed control has been one of the most costly operations for most growers. Many of the weed problems have been associated with poor initial stand establishment, followed by several attempts to replant. In many instances soils have been kept too wet for too long in an attempt to germinate seeds. This promotes fast weed establishment and growth. Jojoba seedlings are relatively slow in growth the first year. Since jojoba is a new crop, little has been done to develop effective herbicides and to obtain clearance for their use. 1 . Cold Tolerance Jojoba plants are very susceptible to low temperatures the f ist winter (Hogan,

1979). As the plants age, vegetative growth is less susceptible to freezing. Young plants have been severely damaged or completely killed at -7.2"C for 3 hr. Damage to new growth was found at -2°C. Flower buds are formed one year, overwinter, and produce fruit the following summer. If jojoba is planted in areas where periodic winter temperatures are lower than -4 or -5"C, a grower can expect severe flower damage and little seed production the following year (Sherbrooke, 1978;Yermanos, 1978).Therefore, no commercial planting should be made where winter temperatures drop below -4°C. 2 . Moisture Requirements Jojoba is an attractive new crop for arid areas of the world because of its relatively low requirement for water. Positive net photosynthesis has been found on wild jojoba plants with water potentials of as low as -7000 kPa (Al-Ani el al., 1972). Although jojoba is native in areas receiving less than 12 cmlyear, wild plants grow best and produce more seed in areas receiving 38-50 cm/year (Gentry, 1958). Establishedjojoba plants can survive under very arid conditions, as in parts of Sonora and Baja California, which often do not receive rainfall for several months. Jojoba can drop all its leaves and remain leafless for several months in response to drought, but after a good, penetrating rain will resume active growth (Gentry, 1958). To establish and maintain a commercial jojoba planting an adequate, dependable supply of water must be available to the grower. Halderman (1983)suggests that the water supply, under Arizona conditions, should be no less than 14 to 18.7 liters/min/ha to meet projected plant needs during May to avoid the development of soil moisture stress and consequent yield reduction. In Arizona, it appears that the peak water demand for a mature jojoba plantation will occur in

338

LEMOYNE HOGAN AND WILLIAM P. BEMIS

May when the temperature is high, the humidity is low, and the jojoba fruit are developing and maturing. All locations in Arizona and California with suitable temperature relations for jojoba cultivation will require supplemental irrigation for significant, reliable seed yields. In the United States, three methods of irrigation are in general use for jojoba: drip, furrow, and sprinkler. In October, 1982,5136 ha of jojoba were irrigated by drip, 3814 ha by furrow, and 838 ha by sprinkler (Whittaker, 1983). Halderman (1983) outlines 10 site conditions and situation factors that affect the selection of an irrigation method for jojoba. These are as follows: 1. Field topography, slope, shape 2. Soil: texture, depth, intake rate, variability 3. Water supply: well pumping rate or canal delivery, schedule, quality (dissolved salts and suspended particles), cost 4. Energy: availability, cost 5 . Cost: first cost, amortized cost, operating cost, salvage value 6. Existing irrigation system, if any 7. Labor: labor requirements, skill level 8. Reliability; maintenance; longevity; mechanical, pest, or vandalism hazard; mechanics of repair and replacement; sales and service support 9. Noninigation uses: fertilizer application, pesticide application, frost protection, control of wind erosion 10. Operator experience and preference

Halderman (1983) suggests that surface irrigation appears to be the most predictable and widely adapted of the three methods unless specific factors are identified which either render surface irrigation impractical or especially favor sprinkler or drip irrigation. Linear-move sprinkler systems work best on highintake-rate soils. The undetermined longevity of drip systems, the high maintenance level required, and the lack of experience with installing replacement systems reduce the certainty of sustained and reliable performance. Irrigation water scheduling will vary with the stage of plantation development (i.e., establishment, plant development, and production). Frequent light irrigations are needed during the seed germination period; drip systems are ideal during the establishment period. During the plant development stage, at least two heavy irrigations per year are required (spring and fall); more may be required for very sandy soils with little reserve soil moisture. Increased seed yields have been obtained (Benzione et al., 1980) with additional applications of water on a mixed seedling population. Moisture responses are difficult to measure with seedling populations because of plant variability. Moisture studies utilizing clonal plant material are needed to determine optimal moisture requirements at various plant ages, for different soils, and for environmental differences.

BUFFALO GOURD AND JOJOBA: NEW ARID CROPS

339

3 . Fertilizers Studies investigating the mineral nutrition and growth of jojoba have been limited and have primarily focused on nitrogen. Meneley (1975) failed to show a significant difference between six treatments of NH4N0, ranging from 0 to 400 ppm during the first 124 days of seedling growth in greenhouse studies. Benzione el al. (1980) found that seedling jojoba plants established in 1966 responded to additional irrigation and fertilizers during a 4-year period (19781981) under Israeli conditions. Control plants averaged 896 g of seed/plant for the 4-year period; those receiving 0.5 m3/plant of water and no fertilizer at monthly intervals produced 2075 glplant; those receiving monthly irrigations at the same rate but also receiving a low rate (a mixture of 109g NH4N0,, 30.1 g 20-20-20 (N-P-K) fertilizer, and 449 g (KNO,) of fertilizer produced 3769 g of seed. Monthly irrigations and more fertilizer (three times the low rate) produced 4085 g. Adams et al. (1976) found no significant response to additions of N-P-K fertilizer in nonirrigated field experiments. Observable differences in growth between seedlings fertilized with N-P-K plus minor elements and unfertilized seedlings grown on deep sand near Blythe, California have been observed. It is agreed that jojoba does respond to the application of fertilizer in those situations where one or more nutrients are limiting; however, definitive studies have not been conducted to show the levels required for each of the elements under different soil, moisture, and climatic conditions. In noncultivated coastal California plants, tissue nitrogen (N) levels of 3% were associated with highest growth, whereas yellow, chlorotic, N-deficient leaves had tissue N concentrations of less than 1% in greenhouse study by Adams et al. (1976). In Western Australia, leaves of plants classed as healthy had 2.06% N as compared to 1.16% N for yellow, chlorotic leaves. Highest growth rates for wild plants in coastal California were associated with tissue phosphorus (P) levels greater than 0.3% (Adams et al., 1976). In Western Australia, healthy and yellow leaves had P levels of 0.12 and 0.59% respectively. 4 . Photosynthesis and Respiration

Rates of apparent photosynthesis (AP) for well-watered jojoba are normally in the range of 7-1 1 mg C0,/dm2/hr, (Collatz, 1977; Glat, 1979; Rasoolzadegan, 1980). Collatz (1977) found that recently matured leaves under relatively high light intensities (1500 pE/m2/sec) had an AP rate of 13.3 mg C0,/dm2/hr. Large seasonal differences in photosynthetic rates have been reported for plants in the wild (Al-Ani et al., 1972; Woodhouse, 1978). Low leaf water potentials have a markedly depressing effect on photosynthesis, but jojoba can continue CO,

340

LEMOYNE HOGAN AND WILLIAM P. BEMIS

uptake at surprisingly low levels of leaf water potential (Al-Ani et al., 1972; Bunce and Miller, 1976; Adams et al., 1976; Collatz, 1977; Glat, 1979). Active photosynthesis has been monitored at leaf water potentials of as low as -70 bars (Al-Ani et al., 1972; Woodhouse, 1978), but large reductions in photosynthesis (5040% of maximum) are commonly encountered when leaf water potentials reach levels in the range of -30 to -40 bars (Bunce and Miller, 1976; Glat, 1979). Variation in Ap among individual rooted cuttings from the same mother plant has been reported to be as large as that between cuttings from different mother plants (Glat, 1979; Rasoolzadegan, 1980). Rates of dark respiration for well-watered jojoba are normally in the range of 1.5-2.5 mg CO2/dm2hr (Glat, 1979; Rasoolzadegan, 1980). Glat (1979) found that dark respiration rates were steady during the course of a day, showing a slight upswing in the later afternoon. Dark respiration rates also fluctuate seasonally, with higher rates in the summer and lower rates in the winter for California. The opposite is true near Tucson, Arizona (Al-Ani et al., 1972).

D. VEGETATIVEPROPAGATION

Because jojoba is very heterogeneous when grown from seed, it is now considered essential that future plantations be established from high-yielding selected cultivars propagated by vegetative means if economical yields are to be obtained (Hogan, 1979). Asexual propagation also en%uresthat growers can plant to the desired stand with sexed plants. Possible asexual propagation methods for jojoba are grafting (Yermanos, 1974), stem cuttings, and tissue culture (Hogan et al., 1978; Birnbaum, 1980). Studies at the University of Arizona have shown that jojoba stem cuttings can be economically propagated in commercial quantities. More than 150 ha of rooted jojoba cuttings had been established by commercial growers by October, 1982. Jojoba is not a difficult plant to propagate by cuttings, provided certain procedures are followed. Significant differences in rooting potential between different mother plants have been found by several workers (Abramvitch, 1977; Cardran, 1980; Low and Hacket, 1981). Seasonality is also an important factor affecting the rooting potential of cuttings. In Israel, spring and summer cuttings rooted at the highest percentage. In Arizona, cuttings root equally well from the same plants except in January and February. High rooting percentages have generally been found to coincide with periods of active vegetative growth. Other plant factors affecting rooting are vigor of the stock plant, developmental stage of the cuttings, and susceptibility to defoliation under propagation conditions. Vigorous, unstkssed stock plants should be used as the source of cuttings (Hogan ef al., 1978). Older, woody tissue is more difficult to root than younger, newly hardened shoots (Hogan and Maisari, 1976). Defoliation of

BUFFALO GOURD AND JOJOBA: NEW ARID CROPS

34 1

cuttings during mist propagation is strongly related to poor rooting (Hogan, 1979). Jojoba has been successfully rooted from cuttings with one node (exclusive of the terminal node) (C. W. Lee, unpublished), three nodes (Low and Hackett, 1981), four nodes (Abramvitch et al., 1978), and five to seven nodes (Cardran, 1980). Maisari (1966), Hogan (1979), and Reddy (1980) were successful using cuttings of 10-15 cm in length. Gentry (1958) reported that indoleacetic acid (IAA) was used on cuttings that rooted successfully (5040%).In Arizona Maisari (1966) found that cuttings dipped for 3 seconds in 3000, 4000, or 5000 ppm indolebutyric acid (IBA) rooted significantly better than controls, and those rooted with 4000 or 5000 ppm IBA had significantly better root systems. Hogan et al. (1978) looked at IBA concentrations (in 50% water, 50% ethanol) from 0 to 20,000 ppm and found that even though maximum rooting (61.7%) took place at 8000 ppm, stem damage occurred at that level and higher, and they recommended 4000-6000 ppm as safe and effective. In Israel, cuttings were dipped from 1 to 2 min in water-soluble IBA at concentrations of 0-20,000 ppm. Forty-five percent rooting was obtained from controls, 55% at lo00 ppm, 60% at 5000 ppm, and 80% at both 15,000 and 20,000 ppm. Auxins failed to promote rooting in periods of low growth and low rooting potential (Low and Hackett, 1981).

1. Handling and Storage of Cuttings Only one controlled experiment (Feldman, 1982) has been conducted on the effect of different methods for the storage and transport of unrooted cuttings. Other workers have utilized the following methods with varying degrees of success: (1) transport in wet cloth bags and storage for 2 days at 40°F (Maisari, 1966); (2) transport of water-sprinkled cuttings in sealed plastic bags inside an ice chest without direct contact between bags and ice, with storage up to 4 days (Hogan et al., 1978); (3) cuttings rinsed, rolled in moist toweling, and transported by truck in cool ice chests for a distance of 500-600 miles (Reddy, 1980); and (4) cuttings wrapped in moist newspaper, packed in a cardboard box, and sent from San Diego to Davis, California via Sacramento by air freight (Low and Hackett, 1981). Unr00ted cuttings from Arizona were placed dry in plastic bags which were placed in a Styrofoam cooler containing “blue ice” enclosed in rigid polyethylene containers with a cardboard partition to prevent contact between the cuttings and the ice. This was enclosed in a cardboard box and shipped by air to Perth, Western Australia; it was in transit for 6 days. Upon arrival, the cuttings were removed, fumigated with methyl bromide, and rooted under mist. Rooting percentages for clones were equivalent to the percentages usually obtained in Arizona for the same clones. Feldman (1982) found that cuttings wrapped in moist paper towels and transported in plastic bags averaged shoot water potentials of - 14 bars at the time of “sticking,” as opposed to less than - 14 bars for

342

LEMOYNE HOGAN AND WILLIAM P. BEMlS

those transported in paper bags. Cuttings transported in paper bags rooted at significantly higher rates than those transported in plastic bags. Rooted stem cuttings often grow relatively slowly after lining out. Cuttings have been successfully stored for as long as 3 weeks in Arizona when placed dry in plastic bags and kept in a cool, shady location.

2 . Rooting Media Many different rooting media have been used with jojoba cuttings. Maisari

(1966)looked at the effectiveness of concrete sand, vermiculite, and sand plus vermiculite (1:1); he found that in terms of both rooting percentage and root quality vermiculite was significantly more effective than either sand or a mixture of sand and vermiculite. Abramvitch et al. (1978)found that a mixture of peat and polystyrene (1:1) was better than peat and vermiculite (1 :l), sand, or a liquid media of 50% Hoagland's solution. Other media used successfully include perlite and vermiculite (1:l)(Hogan, 1979) and 100% perlite (Low and Hackett, 1981). Most jojoba cuttings have been rooted in flats filled with media (Maisari, 1966; Hogan et al., 1978;Cardran, 1980;Low and Hackett, 1981), but some cuttings have been rooted in individual containers to facilitate transplanting and to prevent transplant shock. If individual containers are used for rooting, clones with high rooting percentages are necessary for the process to be economical. Careful attention to the choice of media is also necessary to provide adequate drainage, as well as providing a media that allows roots to become easily established in the soil of the field. At the University of Arizona, single-node cuttings in Oasis cubes had a rooting success of 50%, as compared to 75% for those in flats containing vermiculite/perlite (1:1). Interrupted mist generally has been used in the environmental control of jojoba cuttings (Hogan et al.,, 1978;Reddy, 1980;Low and Hackett, 1981), and mist intervals vary with time of year and location. Bottom heat is beneficial for rooting jojoba cuttings. This can be provided by any source, such as hot water pipes, steam, or electric heating cables. Temperatures at the base of the cutting should be maintained between 25" and 30°C. The time required to satisfactorily root jojoba cuttings varies with season, the condition of the cuttings, and the care given to the cuttings during the rooting period. In general, cuttings root between 30 and 60 days (Maisari, 1966;Hogan, 1979;Reddy, 1980).

3. Nutrition during Rooting To promote root growth, the application of a dilute fertilizer solution to the rooting media after rooting had commenced was recommended by Hogan (1979). In Israel, Osmocote was incorporated into the rooting medium and resulted in increased shoot growth associated with poor root growth. Feldman

BUFFALO GOURD AND JOJOBA: NEW ARID CROPS

343

(1982) did extensive studies on the effects of various fertilizer formulations applied to the rooting media of jojoba and found that rooting-stage fertilization can be beneficial. High rates of Osmocote (5.93 kg M-3) and Peters (300 ppm 20-20-20) increased root weights over the controls in spring. Osmocote-treated cuttings outgrew controls in number of nodes, fresh weight, and succulence. Mineral uptake, especially of N, P, and K, was greatly increased at lining out by use of some fertilizer formulations, particularly Osmocote and Peter’s soluble. Fertilization during rooting did not enhance apparent photosynthesis or dark respiration consistently. Apparent photosynthesis declined steadily after the insertion of cuttings, but AP and root respiration rose in direct proportion to root growth as rooting advanced. Jojoba cuttings subjected to frequent misting over a period of 10-12 weeks leave the rooting bench with markedly reduced tissue contents of N, P, and K, unless these nutrients are supplied during rooting. Early growth, once rooting has taken place, is directly related to the leaf contents of these nutrients, especially N and K. Some micronutrients, such as Zn, Mn, and B, are also related to early growth of rooted cuttings, especially during the spring (Feldman, 1982). E. DISEASES AND PESTS

Wild jojoba plants have not been studied carefully to determine the extent to which diseases cause death and yield reductions. However, during the domestication process and the establishment of commercial plantings, several potentially serious jojoba diseases have been identified (Stanghellini, 1977, Alcorn and Young, 1978). These problems have occurred at the seedling stage, when cuttings are rooting under mist, and also after plants reach the seed-production stage. Bonar (1942) associated Strumella sirnmondsiae with a jojoba leaf spot but did not confirm its pathogenicity. Phytophthora parasitica, Pythium aphanidermaturn, Rhizoctonia solani, and a species Fusarium were isolated from decayed roots of 2-month-old jojoba seedlings grown in unsterilized soil (Stanghellini, 1977). Greenhouse studies found that 29 of 72 2- to %month old jojoba plants were inoculated with Phymatotrichum omnivorum when infected sorghum seeds were buried in the soil near the roots. They either showed symptoms or were dead after 7 weeks. Jojoba seedlings inoculated with Verticillium dahliae became infected as early as 3 weeks after inoculation (Alcorn and Young, 1978). Approximately 9800 rooted cuttings were transplanted into a field near Bakersfield, California, during the summer of 1979. This field had been planted with cotton for 2 years and then with barley, which was plowed under before harvest. In February, 1981, several plants having symptoms of verticillium wilt were found. Verficilliurndahliae was isolated from two plants and pathogenicity was proven. In April, 1981, a survey indicated that 49 (0.5%)of the 9800 plants

344

LEMOYNE HOGAN AND WILLIAM P. BEMIS

had symptoms of verticillium wilt. Six months later, in October, 1981, 1% of the plants showed symptoms. Some plants observed with symptoms in April, 1981, recovered, indicating that they may be tolerant. Surveys will be repeated twice each year for several years to determine the periodic influence of verticillium wilt and to determine its influence on the longevity of infected plants. Plants that remain tolerant and have other desirable characteristics will be used as a source of cuttings to determine whether tolerance persists in these materials (Orum et al., 1983). Severe defoliation of some jojoba stem cuttings has occurred during mist propagation. A species Alternaria was repeatedly isolated from leaf petioles and near nodes. Coniothryrium sp. has been identified with cuttings obtained from wild plants (Alcorn and Young, 1978). These foliar diseases usually are not a problem with cuttings obtained from cultivated plants. Replant soil treatments should be used for all nursery-grown seedlings and for growing rooted cuttings to ensure against damage caused by Phytophthora parasitica and Pythium aphanidermatum. As was the case with diseases, more attention has been given to insect and other pest problems of jojoba since it has been placed in cultivation. A moth whose larvae chew out the young, developing ovules and surrounding tissue destroyed an estimated 75-80% of the fruit set in Pinal County, Arizona, in 1957 (Gentry, 1958). This pest seemed to be confined to elevations above 1600 m. Pinto and Frommer (1978) identified 221 species representing 11 orders associated with wild jojoba in California. Twenty-five of these species were actually identified as feeding on jojoba. A leaf hopper, Homulodisca liturata, has been found breeding and feeding in large numbers on 3-year-old nursery plants in Tucson, Arizona. Thrips (Frankliniella spp .) and cigarrinha verde (Emphuasco spp.) attacked foliage at Punta del Agua, Argentina, during the humid season. In other locations in Argentina, locusts (Sistocerca cancelata) and tucuras (Dichroplus spp. and Rhamantocerus sp.) also caused damage to young plants during January, February, and March (Ayerza, 1983). Aphids attack new growth in commercial plantings in Arizona but seldom require control measures. The red spider has attacked plants in the University of Arizona germ plasm nursery at Mesa, Arizona and has been a severe problem for certain plants during the last three summers. Adjacent plants are seldom damaged, suggesting that certain selections may be resistant to red spider. Rabbits are a persistent problem in very young plantings, especially where raw desert land has been cleared and planted to jojoba in Arizona and California. In such areas it is recommended that protection be provided at least for the first year after planting, including rabbit-proof fences, trapping, and hunting. Ants of various species, especially leaf-cutters, can strip seedlings or cuttings within a few hours, and moles have caused considerable damage in some plantings. Deer and

BUFFALO GOURD AND JOJOBA NEW ARID CROPS

345

other large game animals can be a problem where other forage is limited or where animal populations are large. F. HARVESTING

For jojoba to be economically produced in the United States and other areas with high labor costs, mechanical harvesting is a necessity. As in other seedling jojoba characteristics, large variations are found in seed maturation dates and dehiscence. In Arizona most wild jojoba seed ripens during a 3-month-period from the middle of June to the middle of September. Some plants ripen seed early, others late; some plants produce seed that dehisces easily and others produce seed that clings tightly to the plant after maturity. It is extremely important to select plants that mature seed uniformly with the same pattern of seed drop to facilitate mechanical harvesting. Several research groups and commercial growers are developing mechanical harvesters (Fisher, 1983). These include an over-the-row , single-row harvester, originally developed for blueberries by BE1 Company of South Haven, Michigan, which sells for approximately $85,000. It is believed that jojoba can be mechanically harvested for $0.25 per pound of seed, as compared to $2.50 per pound by hand harvesting. G. YIELDS

Early yield data from research plots and commercial plantings have emphasized the yield variability found in plants grown from seed. Yield data from mature, cultivated jojoba plants are still scarce, but yields from the 1982 season indicate that high-yielding plants can be found. Some of the most interesting data come from the 17-ha germ plasm nursery of the University of Arizona located about 38 km south of Bakersfield, California. This planting was established in the summer of 1979 from rooted cuttings from selected mother plants. Individual plant seed harvests began in the summer of 1981 from 2.5-year-old plants. One clone consisting of 25 plants produced an average yield at 3.5 years of 759 g each; yields from individuals in this clone ranged from 307 to 1238 g. Uniformity should improve if the type and size of the original cutting as well as the quality of the cutting root system are standardized (Palzkill and Hogan, 1983). The female 7-year-old plants at the University of Arizona Mesa Experiment Station produced an average yield of 691 g in 1982; the highest producing individual plant produced 2070 g. Some plants have a tendency for annual bearing and others tend toward biannual bearing (Palzkill and Hogan, 1983). At 13 years, 13 Vista plants exceeded lo00 g each, 3 exceeded 2000 g each, and 61 produced less than lo00 g each. Maximum yields obtained at Vista from a 6-

346

LEMOYNE HOGAN AND WILLIAM P. BEMIS

year-old plant were 1760 g, and a 12-year-old plant produced 2381 g (Yermanos 1968). Nanetti (1983) reported the following yields from his direct-seeded platings near Hermosillo, Mexico: in 1980 at 3 years of age, 5 kg/ha; at 4 years, 60 kglha; and at 5 years, 270 kglha were harvested. el al.,

H.

PRODUCT USES

Jojoba has potential in several areas: cosmetics, lubrication, polishing waxes, and pharmaceuticals. Sperm whale oil has been widely used for many years in many industrial applications. Because sperm whale oil can no longer be imported into the United States and several' other industrialized countries, substitutes are needed. As the composition and physical properties of jojoba wax closely resemble those of sperm whale oil, it has been suggested as a substitute (National Academy of Sciences, 1977). Sulfurized jojoba and sulfurized sperm oils are essentially equivalent in improving the load-carrying capacity of both naphthenic and bright-stock-base oils under extreme pressure conditions. Jojoba wax can become a source of straight-chain, monounsaturated alcohols and acids, which can be used as intermediates for the preparation of other compounds, disinfectants, surfactants, detergents, lubricants, driers, emulsifiers, resins, protective coatings, corrosion inhibitors, and the bases for creams, ointments, and many other products. Jojoba wax also can be hydrogenated to produce a hard white wax, which can be used for floor finishes and for furniture, shoe, and automobile polishes. At the present time, most of the available jojoba wax is being used in the cosmetic industry. It has found worldwide acceptance by the general public in shampoos, conditioners, and lotions (Taguchi, 1976). The unpredictable availability has been a major limiting factor in the use of jojoba wax. It has limited use in the lubrication industry at the present time because of cost. Its value in the lubrication field has been well documented, but only when the cost of jojoba wax is competitive with other sperm oil substitutes will it reach its full potential.

IV. CONCLUSION Relatively few major crops dominate world agriculture. Most of these crops have been cultivated by man for extremely long periods of time and are best adapted to regions of adequate or high rainfall and fertile soils. It has become extremely difficult for a new crop to become widely accepted that it might contribute to the economy of the world. It must possess unique or valuable qualities to reach a significant level of importance.

BUFFALO GOURD AND JOJOBA NEW ARID CROPS

341

Most efforts to develop new crops have been directed toward those adapted to the better agricultural areas of the world. Little effort has been directed toward the development of new crops for the arid regions, although a significant part of the world is arid and many millions of people live in such areas. Many of these areas, such as southern Arizona, have highly developed agriculture but must rely on pumped groundwater for irrigation. Many of these areas are faced with dropping water tables and increasing energy costs for pumping water, and the production of many conventional crops is no longer economical. Large areas of southern Arizona, blessed with abundant sunlight and fertile soils, are facing this problem at the present time. The University of Arizona, College of Agriculture, as well as several other institutions, has recognized this problem and developed research programs to study new crops for arid regions that require less water than the more traditional crops. Two arid land crops which scientists at the University of Arizona are studying are the buffalo gourd and jojoba. These plants have different attributes and require different cultural techniques, but they both appear to offer considerable potential for becoming important new crops for warm, arid regions of the world. Both also require much additional research, but great strides have been made toward making them valuable additions to the list of important cultivated crops.

REFERENCES Abramvitch, R. 1977. Res. and Dev. Auth. Sci. Activ. Ben-Gunon Univ. of the Negev, annud report. Adams, J. A., Bingham, F. T., and Yermanos, D. M. 1976. Proc. Inr. Conf. Jojobu 2nd. Al-Ani, H. A., Strain, B. R., and Mooney, H. A. 1972. J. Ecol. 60,41-57. Alcom, S. M., and Young, D. 1978. Proc. Int. Conf. Jojobu 3rd. pp. 13-17. Ayemi, R. 1983. Proc. Int. Con$ Jojobu 5th (in press). Bemis, W. P., Curtis, L. C., Weber, C. W., Berry, J. W., and Nelson, J. M. 1975. U.S.Ag. Int. Dev. Tech. Bull. No. 15. Bemis, W. P., Berry, J. W., and Weber, C. W. 1977. Proc. Conf. Non-Conventional Proteins Foods Univ. Wisconsin, pp. 77-82. Bemis, W. P., Berry, J. W., Weber, C. W., and Whitaker, T. W. 1978a. HorrScience 13,235-240. Bemis, W. P., Curtis, L. C., Weber, C. W., and Berry, J. W. 1978b. Econ. Bot. 32, 87-95. Bemis, W. P., Berry, J. W., and Weber, C. W. 1979. AAAS Sel. Symp. No. 38, 65-87. Benzione, A., Migrahi, T., and Nerd, A. 1980. Proc. Inr. Con$ Jojoba 4rh, pp. 162-171. Berry, J. W., Bemis, W. P., Weber, C. W., and Philips, T. 1975. J. Agric. Food Chem. 23, 825-826. Bimbaum, E. 1980. In Proc. Int. Conf. Jojobu 4rh, pp. 158-171. Bohrer, V. L. 1975. “Symposium on Threatened and Endangered Plants.” Soc. Range Management, Albuquerque, New Mexico. Bolley, D. S., McCormack, R. H., and Curtis, L. C. 1950. J . Am. Oil Chem. SOC. 29, 571-574. Bonar, L. 1942. Mycologiu 34, 180-192. Buchmann, S. L., O’Rourke, M. K., and Shipman, C. W. 1982. Proc. Inr. Conf. Jojoba 5th.

348

LEMOYNE HOGAN AND WILLIAM P. BEMIS

Bunce, J. A., and Miller, L. A. 1976. Can. J . Bot. 54,2457-2464. Cardran, P. 1980. M.S.Thesis, Univ. of Arizona, Tucson. Collatz, G. J. 1977. Plant0 134, 127-132. Curtis, L. C. 1946. Chemurg. Dig. 5,221-224. Curtis, L. C., and Gomez, C. H. 1974. Bol. Tec. No. 4, CNIZA, Saltillo, Mexico. Curtis, L. C., and Rebeiz, N. 1974. Ford Found. Prog. Rep., Nos. I-IV. Cutler, H. C., and Whitaker, T.W. 1961. Am. Antiq. 26, 469-485. Dossey, B. F., Bemis, W.P., and Scheerens, I. C. 1981. J . Hered. 72, 355-356. Ehrler, W. L., and Fink, D. H. 1978. Proc. Int. Conf.Jojoba 3rd. pp. 361-373. Feldman, W. R. 1982. Ph.D. Dissertation, Univ. of Arizona, Tucson. Ferguson, C. W. 1977. Jojoba Happenings 19, 16. Fisher, G. L. 1983. Proc. Int. Conf. Jojoba 5th. Forti, M. 1973. Proc. Int. COP$ Jojoba Ist, pp. 13-26. Gentry, H. S. 1958. Econ. Bot. 12, 261-295. Glat, D. 1979. B.S. Honors Thesis, Univ. of Arizona, Tucson. Haase, E. F. 1976. Proc. Int. Conf. Jojoba2nd, pp. 39-47. Haldennan, A. D. 1983. Proc. Int. Con$ Jojoba 5th. Harlan, J . R. 1976. Sci. Am. 235, 88-97. Hogan, L. 1979. AAAS Sel. Symp. No. 38, 177-205. Hogan, L., and Maisari, A. A. 1976. Proc. Int. Conf. Jojoba 2nd. Hogan,L., and Palzkill, D. A. 1983. Proc. Int. Conf. Jojoba 5th. Hogan, L., Lee,C. W.,Palzkill, D. A., and Feldman, W. R. 1978. Proc. Inr. Conf.Jojoba 3rd. pp. 1-4. Katoh, M., and Kunimoto, T. 1983. Proc. Int. Conf. Jojobu 5th. Lee, C. W., Anderson, J. O., Palzkill, D. A., and Hogan, L. 1980. Proc. Int. Conf.Jojobu 4th. pp. 347-351. Libby, H. 1980. Proc. Int. Conf. Jojoba 4th. pp. 266-287. Loomis, R. S. 1976. Sci. Am. W5, 99-105. Low, C. B., and Hackett, W.B. 1981. CalifAgric. Mar.-Apr., pp. 12-13. Maisari, A. A. 1966. M.S. Thesis, Univ. of Arizona, Tucson. McKeon, E. 1983. Proc. Int. Conf.Jojoba 5th. Meneley, T. J. 1975. M.S.Thesis, Univ. of Arizona, Tucson. Mmv, N. T. 1952. Econ. Bot. 6, 410-470. Nabhan, G. 1980. Rep. Vniv. Ariz. Dep. Plunt Sci. Nanetti, M. L. 1983. Proc. Int. Conf. Jojoh 5th. National Academy of Sciences/National Research Council. 1977. Orum, T. V., Alcom, S. M., and Plazkill, D. A. 1983. Proc. fnt. Conf. Jojoba 5th. Pawrill, D. A., and Hogan, L. 1983. Proc. Int. Conf. Jojobu 5th. Paur, S. 1952. N. M. Press Bull. No. 1065. Pinto, J. D., and Frommer, S. 1. (1978). Proc. Int. Conf. Jojoba 3rd. pp. 19-24. Rasoolzadegan, Y. 1980. W.D. Dissertation, Univ. of Arizona, Tucson. Reddy, S. J. 1980. M.S. Thesis, Univ. of Arizona, Tucson. Salaman, R. N. 1949. University Press, Cambridge Chap. 16. Scheerens, J. C., Bemis, W.P., Dreher, M. L., and Berry, J. W.1978. J . Am. Oil Chem. SOC.55, 523-525. Shahani, H., Dollear, R. G., Markley, D. K., and Quimby, J. 1951. J. Am. Oil Chem. SOC. 28, 90-95. Sherbrooke, W. C. 1978. Pac. Discoverer 31, 22. Sherman, R. A., and Ray, D. T. 1978. Jojoba Happenings 22, 2.

BUFFALO GOURD AND IOJOBA: NEW ARID CROPS

349

Stanghellini, M. E. 1977. Jojobu Happenings 20, 4. Taguchi, M. 1976. Proc. Inr. Conf. Jojobu 2nd. pp. 149-170. Theisen, A. A., Knox, E. G., and Mann, F. L. 1978. “Report Prepared for National Science Foundation by Soil and Land Use Technology Inc.” US Govt. Printing Office, Washington, D.C. (Stock NO. 038-000-00389-5). Thompson, S. A., Weber, C. W., Berry, J. W., and Bemis, W. P. 1978. Nurr. Rep. Inr. 18, 515-519. Vasconcellos, I . A , , Bemis, W. P., Berry, J. W., and Weber, C. W. 1981. AOCS Monogr. No. 9, 55-68. Waymack, L. B., Weber, C. W., and Scheerens, J. C. 1979. “Arizona Cattle Feeder’s Day.” Whitaker, T. W., and Bemis, W. P. 1975. Bull. Torrey Bot. Club 102, 363-367. Whittaker, C. A. 1983. Proc. Inr. Conf. Jojobu 5th. Wood, I. W., and Jones, H. A. 1943. J . Am. Chem. SOC.65, 1783. Woodhouse, R. M. 1978. Proc. Int. Conf. Jojobu 3rd. pp. 345-360. Wortman, S. 1976. Sci. Am. 235, 31-35. Yermanos, D. M. 1974. Econ. Eor. 28, 160-174. Yermanos, D. M. 1978. Mimeo Publ. Dep. Plant Sci., Univ of Calif., Riverside. Yermanos, D. M., Kadish, A., McKell, C. M., and Goodin, J. R. 1968. Culq. Agric. 22, 3.

This Page Intentionally Left Blank

ADVANCES IN AGRONOMY, VOL. 36

PROTEIN TRANSFORMATION IN SOIL’ Michael J. Loll and Jean-Marc Bollag Laboratory of Soil Microbiology Department of Agronomy The Pennsylvania State University University Park, Pennsylvania

I. Introduction . .

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

11. Protein Sources . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 111. Proteolytic Microorganisms. . . . . . . . . . . . . . . . . . . . . . . . . . . .

A. Habitats .............................................. ............................ B. p H . . . . . . . . . . . .. . . . C. Soil Atmosphere. . . .................................. D. Temperature ............................ ................ E. Substrate.. . . . . . . . ........................ F. Salt Concentration ................................................ G. Succession ....................................................... IV. Characteristics of Proteolytic Enzymes in Soils.. . . . . . . . . . . . . . . . . . V. Environmental Factors Affecting F’roteolysis . .

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

..................................... B. Moisture . . . . . . . . . . . ................... C. Redox Conditions.. . . . . . . . . . . . . . . . . . . . . D. Soil Properties.......................................... E. Naturally Occurring Organic Compounds.. . . . . . . . . . . . . . . . . . . . . . . . . . . . F. Rhizosphere ............................. ................ G . Enzyme Activities. ..................................... . . . . . . . . . . . . . . . . . . . .. . . . . . . ... H. Agricultural Practic I. Pesticides.. . . . ... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. .. . VI. Transformation and Binding of Protein in Soil . . . .......................

.

.

C.

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

Rhizosphere Adsorption

formation. . . . . . . . . . . . . . .

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

352 352 354 355 359 359 360 360 360 360 36 1 364 364 365 365 366 367 368 368 368 369 370 372 374 376 376 377

‘Paper No. 6444 of the journal series of the Agricultural Experiment Station, University Park, Pennsylvania. 35 1

Copyright 8 by Academic Press. Inc. All rights of repduction in any form r e w e d . ISBN 0-124W0736-3

352

MICHAEL J. LOLL AND JEAN-MARC BOLLAG

1. INTRODUCTION Most of the nitrogen present in unfertilized soils is organic in nature. This organic nitrogen represents an important nutrient reservoir, and a large part of it appears to be derived from protein. In the heavily amended soils of industrialized countries, protein plays only a secondary role as a nitrogen source, but it may be of particular importance in some areas of the Third World where plant residues, household wastes, and manure are often the only fertilizers available (Ruthenberg, 1976). Sulfur, as well as nitrogen, may be added to soil through the decomposition of methionine and cysteine (Banwart and Bremner, 1976). Some research also suggests that amino compounds contribute to the structure of soil by their incorporation into humic acid (Swaby and Ladd, 1962; Ladd and Butler, 1966, 1975; Kononova, 1966; Felbeck, 1971; Raig ef al., 1975), from which they may be slowly released by enzymes (Ladd and Brisbane, 1967; Sowden, 1970). After transformation they may be used as plant nutrients. The breakdown of protein is also crucial for composting, the development of soils, the digestion of sewage sludge, and the biodegradation of solid waste. Protein degradation apparently is carried out mostly by microorganisms in conjunction with mesofauna such as earthworms and insect larvae (Parsons and Tinsley, 1975). The mineralization of protein may have a considerable effect on soil fertility, and soil scientists have been studying microbial proteolysis in situ and in vitro in order to understand the genesis of available soil nitrogen. In this article, we have tried to provide an evaluation of the work done in this area and to point out its significance in the study of soil dynamics. We have also discussed the transformations proteins undergo in the presence of biotic and abiotic components and how these reactions may influence protein decomposition and nitrogen availability. II. PROTEIN SOURCES

Proteins are present in all living things and are even found in the abiotic components of earth and water (Simonart et al., 1967; Lytle and Perdue, 1981). Their functions are manifold, and they exist in a variety of forms, often conjugated with sugars, nucleic acids, metals, and lipids (Boulter and Derbyshire, 1977; Capaldi, 1977; Neuberger, 1978). Proteins catalyze reactions as enzymes, act as the structural components of cells, serve as storage products, and are involved in reactions and responses between cells. They carry oxygen and transport molecules across cell membranes (Knowles and Gutfreund, 1974; Boulter and Derbyshire, 1977; Miller, 1978; Reid, 1978; Rupley, 1978; Warren, 1978).

PROTEIN TRANSFORMATION IN SOIL

353

Because proteins are required for so many physiological processes, they make up a significant proportion of the dry weight of living organisms. At the death and decomposition of microorganisms, plant and animal proteins and peptides provide the soil with significant amounts of nitrogen. Therefore, plants are major nitrogen suppliers (Parsons and Tinsley, 1975). Estimates of the average protein content of plants vary, the lower figures being in the 1-15% range (Buckrnan and Brady, 1966) whereas later work (Gray and Biddlestone, 1974) indicates that 5-40% of plant dry weight is proteinaceous. The amount of amino nitrogen a plant contains depends on its species and stage of development. Legumes have the most protein, about 20-30%, and soybean percentages are sometimes as high as 38%. Cereals and most vegetables are 10 and 4% protein, respectively (Boulter and Derbyshire, 1977). Forest leaf litters may consist of 2-15% crude protein (Gray and Biddlestone, 1974). Many seeds are nearly all protein, and mature leaf and stem tissue is 60%peptide and protein matter (Parsons and Tinsley, 1975). With water, proteins comprise most of the protoplasm of higher plants and participate in the synthesis of cellular components. They are not found in the cell walls of plants, and they appear to be primarily functional (Black, 1968) rather than structural. Animals are next in importance in the contribution of protein to the soil nitrogen pool. It has been estimated that 80% of muscle tissue is made of protein and nucleic acids (Gray and Williams, 1971). Although they contain more protein than plants, animals ordinarily represent a smaller proportion of the biomass of most habitats. On grazing land the excreta of sheep and cattle affect nitrogen content over long periods (Parson and Tinsley, 1975); protein accounts for 5-30% of the weight of manure (Gray and Biddlestone, 1974). In areas with large bird populations, keratin from feathers may furnish soil with a-amino nitrogen. Insects and arthropods, whose exoskeletons are made of chitin-protein and carbohydrate-protein complexes (Hunt, 1970), have also been implicated as potential nitrogen sources (Parsons and Tinsley, 1975). Bacteria, actinomycetes, fungi, and algae account for the rest of the protein found in soil. The quantities of protein that these organisms deposit in the earth are considered negligible by some authors. Using data from Stockli (1946), Parsons and Tinsley (1975) point out that of the 2 kg/m2 of microbial biomass in the Swiss meadow, only 1% is nitrogen. Assuming that protein is approximately 15% nitrogen, the amount of protein in the meadow soil is only 134 g/m2. However, biomass determinations are subject to various errors, and other authors have given much higher values for the amount of microorganisms in soil. Estimates have ranged from 300 to 6400 pounds of bacterial tissue per acre-furrow slice (Clark, 1967), and radiolabeled bacterial amino acids have been isolated from soil by Wagner and Mutatker (1968). Therefore it is possible that microbes may account for a large part of the soil nitrogen.

354

MICHAEL J. LOLL AND JEAN-MARC BOLLAG

I11. PROTEOLYTIC MICROORGANISMS Soil is a “sink” in which proteins from plants and animals accumulate, and as such it is a site of intense protein and peptide hydrolysis. This accumulation process varies with the season and ecosystem and is determined by soil characteristics, the kind of vegetation covering the soil, and the nature of the soil microbial community. Protein decomposition is carried out by the soil microflora, which can utilize the resulting amino acids as carbon and nitrogen sources. Indeed, large numbers of bacteria, actinomycetes, and fungi are involved in the biodegradation, and as many as 105-107proteolytic microbes per gram of soil have been isolated from surface horizons (Alexander, 1977). Similar concentrations have been observed in sewage waste (Hobson, 1973). Protein decomposers may comprise 22-89% of the total soil population (Hankin and Hill, 1978)and are found in a wide range of environments. Rice paddies (Kobayashi et al., 1967; Ishizawa er al., 1969),deserts (O’Brien, 1978),polders reclaimed from the sea (van Schreven and Harmsen, 1968),tundras (Dunican and Rosswall, 1974),sand dunes, salt marshes (Pugh and Mathison, 1962),meadows (Chmel and Vlacilikova, 1975),field soils (Lajudie and Chalvignac, 1950),estuaries (Sizemore and Stevenson, 1974), sewage sludge (Hobson et al., 1974;Cox, 1978) and the rumens of sheep and cattle (Blackburn and Hobson, 1962)all support proteolytic populations of various sorts. Proteases and peptidases, synthesized by soil microorganisms are the catalysts responsible for breaking down proteins. Many of these enzymes are exocellular, as a large number of native proteins are too large to be absorbed by living cells. Proteases released from cells fragment protein into smaller, membrane-permeable peptides and amino acids which microbes can metabolize. Degradation results from two modes of hydrolysis: (1) attack on the terminal amino acid of the peptide chain, which is by exopeptidases; and (2) attack on nonterminal peptide bonds by endopeptidases (Alexander, 1977). The amino acids can be metabolized to ammonia and carbon dioxide, and five different reactions have been identified. A. Hydrolytic deamination RCHzNHzCOOH + H20 + RCHOHCOOH + N H 3 RCO + HCOOH + N H 3 RCHzOH + COz + NH3

B . Reductive deamination RCHNHZCOOH + 2 H --* RCHZCOOH + NHs

C. Oxidative deamination RCHNHzCOOH

+ 4 0 2 + RCOCOOH + NHs

PROTEIN TRANSFORMATION IN SOIL

355

D. Ammonia removal RCHNHzCOOH + RHC = CHCOOH + NH3

E. Decarboxylation RCHNHzCOOH + RCHzNHz

+ COz

Intracellular enzymes from roots and dead plants and animals could also contribute to protein hydrolysis in soils. In fact, Ladd (1972) produced soil extracts having characteristics akin to plant and animal proteases, and proteolytic activity has been shown to increase when soil is incubated with plant residues (Ilyaletdinov et al., 1972). However, several investigators have been unable to detect proteases being exuded from plant roots (Estermann and McLaren, 1961; Chang and Bandurski, 1964; Vagnerova and Macura, 1974a). Therefore it is generally assumed that the living microflora are responsible for most of the degradative potential of a soil. Enzymologists have classified proteases, microbial and otherwise, into four basic groups, as shown in Table I. The substrate specificities vary, but in general tend to be rather wide. Proteases that act on high molecular weight proteins are generally exocellular, whereas intracellular hydrolases act on low molecular weight proteins (Ladd and Butler, 1975). In some fungi and bacteria proteases are inducible, and although the pure culture studies done on the subject may not present an accurate model of what happens in soil, they give us some idea of the potential mechanisms that may be at work in the environment. Gill and Modi (1981) induced the production of exocellular proteases in Aspergillus nidulans by supplementing media with cooked egg white. Proteolytic activity in a species of Serraria increased when it was grown with bovine serum albumin, a-lactalbumin, or peptone (Murakami et al., 1969); peptides have been used by Japanese scientists to induce proteases in thermophilic species of Streptomyces (Mizusawa et al., 1966). Thousands of species of soil microorganisms produce proteases and only a few of the more prominent genera are listed in Table 11. Many of these microbes are not found exclusively in soils, but are also found in fresh water and marine environments, feces, wounds, and the gastrointestinal tracts of humans and animals. Bacteria, actinomycetes, and fungi can adapt to various ecosystems, and so-called indigenous soil populations may arise from a variety of sources. The Occurrence of a proteolytic microflora depends on the following environmental conditions. A. HABITATS

Agricultural soils, such as those of orchards, pastures (Hankin et al., 1974), and plantations (Visser and Banage, 1973), often contain more protein decom-

Table I CharacteristirS and Clapsitlcrrtion of Proteaseso

Enzyme

pH optima

Molecular weight

Metal requirement

Esterase activity

-

Limited

Acid proteases Thiol proteases

4 ~ 8

35,000 20,000-50,000

Metalloproteases

7-8

35,Ooo-45,Ooo

1-5

-

Zn, Mg,Co,

Inhibitors

-

Diazoketones Iodine, iodoacetate, organic ~ K W H202 , EDTA, 0-phenanthrolie

Extensive

DFP, sarin, PMSF, DFCC

-

Fe, Mn, Ni Alkaline or serine proteases

9-11

26,000-34,000

-

“Data from Cunningham (1%5), Keay (1971), Mntsubara and Feder (1971), Priest (1977). and Walsh (1975).

Examples Pepsin Papain, ficin, bromelin, cathepsin Carboxypeptidase A, themolysin Trypsin, chymotrypsin, thrombin, subtilisin, elastase

357

PROTEIN TRANSFORMATION IN SOIL

Table II Proteolytic Microorganisms Genus Actinomycetes Acrinomyces Micromonospora Nocardia Streptomyces

Reference

Clark and Paul (1970) Ishizawa er al. (1969) Fergus (1964), Clark and Paul (1970) Fergus (1964), Clark and Paul (1970), Eklund et al. (1971), Knosel (1974)

Thermoacrinomyces Thermonospora Bacteria Achromobacrer Arthrobacter Bacillus Bacreriodes Clostridium Corynebacrerium Fluvobacrerium Micrococcus Peptococcus Proreus Pseudomonas Sarcina Serratia Staphylococcus Sfreptococcus Fungi Alrernaria Amuuroascus Anrjriopsis Arthroderma Aspergillus Auxarthon Cephalosporium Chaeromium Chrysosporium Ctenomyces Cunninghamella Curvularia

Fergus (1964) Fergus (1964) Pochon and Chalvignac (1952), Ueda and Earle (1972) Clark and Paul (1970) Pochon and Tchan (1947), Pochon and Chalvignac (1952); Appleby (1955), Clark and Paul (1970) Siebert and Toerien (1969) Pochon and Tchan (1947), Appleby (1955), Siebert and Toerien (1%9), Cox (1978) Appleby (1955), Clark and Paul (1970) Appleby (1955), Clark and Paul (1970) Clark and Paul (1970) Siebert and Toerien (1969) Pochon and Tchan (1947), Pochon and Chalvignac (1952); Appleby (1955), Clark and Paul (1970) Pochon and Tchan (1947), Kolesnikova er al. (1972), Ueda and Earle (1972) Pochon and Tchan (1947) Clark and Paul (1970) Siebert and Toerien (1969) Clark and Paul (1970) Griffin (1%0), Prudlov et al. (1973) Bohme and Ziegler (1969) Bohme and Ziegler (1969), Chmel and Vlacilikova (1975) Griffin (1960, 1972); Bohme and Ziegler (1969); Chmel and Vlacilikova (1975) Pugh and Mathison (1962) Pugh and Mathison (1962), Fergus (1964), Clark and Paul (1970) Bohme and Ziegler (1969) Clark and Paul (1970) Griffin (1960), Fergus (1964), Ong and Gaucher (1973) Bohme and Ziegler (1969), Chmel and Vlacilikova (1975) Pugh and Mathison (1962), Bohme and Ziegler (1969), Chmel and Vlacilikova (1975) Griffin (1960) ~~

~

(Continued)

MICHAEL J. LOLL AND JEAN-MARC B O U A G

358

Table II Continued Genus

~

Diheterospora Epicoccum Fusarium Ganoderma Gliocladium Gymnoascus Helminthosporium Humicola Keratinomyces Lentinus Malbranchea Martierella Microsporon Mucor Nannizia Paecilomyces Penicillium Phoma Polyporus Polystictus Pyrenochaeta Rhizopus Stilbella Talaromyces Thielavia Trametes Trichoderma Trichophyton

Reference Griffin (1960) Chmel and Vlacilikova (1975) Griffin (1960) Griffin (1960), Rudlov et al. (1973) Das et al. (1979) Griffin (1960) Griffin (1960) Griffin (1960) Griffin (1960), Fergus (1964), Ong and Gaucher (1973) Griffin (1960) Das et al. (1979) Ong and Gaucher (1973) Griffin (1960) Bohme and Ziegler (1969) Griffin (1960), Clark and Paul (1970) Pugh and Mathison (1%2), Bohme and Ziegler (1969), Griffin (1972), Chmel and Vlacilikova (1975) Griffin (1960) Griffin (1960), Clark and Paul (1970), Ong and Gaucher (1973) Griffin (1960) Das et al. (1979) Clark and Paul (1970) Griffin (1960) Clark and Paul (1970) Fergus (1964) Fergus (1964), Ong and Gaucher (1973) Griffin (1960) Das et al. (1979) Griffin (1960), Rodriguez-Kabana et al. (1978) Griffin (1960). Bohme and Ziegler (1%9), Chmel and Vlacilikova ( 1975)

posers than many virgin soils. Swamps (Visser and Banage, 1973) and tidal marshes (Cahenzli and Staffeldt, 1976) are other sites where proteolytic microbes may be in abundance. These are areas where there is much organic matter from plant and marine animal residues, and the addition of protein to soil appears to stimulate microfloral populations (Holding ef al., 1965). The plant rhizosphere has a pronounced influence upon soil proteolysis. Plant roots exude a number of nutrients, including amino acids and peptides, which promote the proliferation of microorganisms. The rhizospheric effect varies with the particular plant. More than 40% of the bacteria isolated from the roots of wheat, beans, peas, cucumbers, and barley have been found to hydrolyze protein. The percentages for other agronomic plants have been a little less, with 18%

PROTEIN TRANSFORMATION IN SOIL

359

for corn, 31% for clover, 25% for lettuce, 34% for red pepper, and 39% for tomato roots (Vagnerova and Macura, 1974~).Katznelson and Rouatt (1957) reported that more ammonifiers are present in the rhizosphere soils of wheat, rye, barley, and oats than in nonrhizosphere soils. Conversely, a decrease in the number of protein decomposers has been observed around the roots of lodgepole pines (Dangerfield et al., 1978). Apparently, isolates from the roots of these pines are less proteolytically active than isolates from the rhizospheres of other plants. B. PH

The role of pH in regulating microbial growth is not clear, at least for proteolytic species. The commonplace is that fungi (because of their tolerance of low pH) are the major users of protein in acid forest soils. Bacteria and actinomycetes are thought to be more prevalent in neutral and alkaline regimes (Chalvignac, 1953; Alexander, 1977), but this is not true in all cases. Hankin and Hill (1978) were unable to correlate bacterial numbers with pH. Many keratinophilic fungi prefer weakly acid to weakly alkaline soils (Bohme and Ziegler, 1969), although there are species which are prevalent in acid bogs (Griffin, 1972). Holding er al. (1965) observed bacterial proliferation when they added nutrients to soils with pH values as low as 3.7. They suggested that most soils, given the proper conditions for growth and small fungal populations, will support bacteria regardless of pH. For any broad class of organisms there are wide ranges of tolerance, and it is not astonishing to see differences in pH sensitivity from one species to another. C. Son, ATMOSPHERE

Whereas early work (Chalvignac, 1953) emphasized the importance of aerobes in the mineralization of protein, it is now known that both aerobic and anaerobic microorganisms are involved in the process. Aerobic metabolism is predominant at the soil surface and in the litter layers; anaerobic degradation occurs in deep horizons and waterlogged soils. Proteolytic anaerobes are common in sewage sludge digesters (Siebart and Toerien, 1969; Hobson er al., 1974), and some of these come from soil (Cox, 1978). Aerobic protein decomposers appear to be more numerous in the environment and proteolysis may be inhibited without oxygen (Sizemore and Stevenson, 1974). Keratinophilic fungi are concentrated at the soil surface and become fewer down the soil profile (Chmel and Vlacilikova, 1975). Using samples taken from a beach, Sizemore and Stevenson (1974) isolated a number of saprophytes, 75% of which were proteolytic when grown under oxygen.

360

MICHAEL J. LOLL AND JEAN-MARC BOUAG

Volatile substances produced from decaying plants can inhibit the growth of pmteindegrading soil bacteria. Lucerne hay, which releases acetaldehyde as it decomposes, reduces the number of protein hydrolyzers when incubated with soil for long periods of time (van Schreven, 1972). D. TEMPERATURE

In cold soils, psychrophilic bacteria seem to have a significant role in the biodegradation of protein (Stefaniak, 1972), but the actinomycete population decreases with decreasing temperature. Low temperatures inhibit, but do not stop, the growth of proteolytic bacteria (Stefaniak, 1968). Thermophilic fungi like Penicillium duponti and Malbranchea pulchella var. sulfurea are capable of pmtease synthesis (Ong and Gaucher, 1973) and are probably active during the summer and in compost heaps.

E. SUEISTRATE Different types of proteins have varying effects on the composition of the microflora in soil, and a particular substrate may favor the survival of a particular type of organism. Putyatina (1966) found that vegetative albumin, yeast protein, and casein promoted the development of mycobacteria; spore-forming bacteria developed when soil was amended with animal albumin and glutenin. F. SALTCONCENTRATION

Proteolytic bacteria from estuaries have been inhibited by the presence of salt (Stefani and Sequi, 1978). Mineral fertilizers could increase the salt content of soil if used improperly, with a consequent reduction of the saprophytic community. G. SUCCESSION

Ecological succession has been observed on protein substrates added to soils. Pochon and Chalvignac (1952) incubated bits of liver and kidney with soil and then examined the populations that developed. Their results showed a gradual change with time from nonsporulating gram-positive bacteria to sporulating gram-negative ones. At the end of 18 days of incubation, autolysis took place as the last of the added protein was metabolized. As they isolated few actinomycetes and no fungi, they assumed that these organisms were not involved in

PROTEIN TRANSFORMATION IN SOIL

36 1

proteolysis. However, this may have been because of the inadequacies of their media for selecting these groups of the microbial community. Okafor (1966) has proposed another scheme of succession. When proteinaceous insect wings were added to soil, proteolytic fungi were among the first colonizers. The subsequent appearance of bacteria on the wings was related to the utilization of substrates other than protein. The competitive abilities of the colonizing organisms are also important, and in the breakdown of sterile hair the less specialized fungi develop first. These are often cellulolytic organisms, such as species of Humicola or Chuetomium, which are capable of using polysaccharides. Sometime afterward keratinophilic fungi, which are more selective in their substrate utilization, begin to grow (Griffin, 1960). The same study mentions that the previous history of the soil under investigation is fundamental in determining the course of colonization. Soils having large inputs of keratin and manure from birds tend to be good sources of keratinophilic microbes. The isolation and identification of proteolytic microbes is necessary for understanding the degradation of protein in soil. However, isolations are not wholly representative of the active population and sometimes cells ordinarily dormant in soils are activated when introduced into nutrient-rich media. The metabolic properties of a cell may change when transferred from soil to media, and certain hydrolytic activities occurring in pure cultures may be absent under natural conditions. Enzyme synthesis to some extent is a function of the age of a cell or mycelium, and large numbers of bacteria do not mean a high activity. Microbiologists have sought to minimize interferences of this type by studying the enzymatic activity of soil as a whole. This approach, based on extracting and assaying soil proteases in order to measure the proteolytic potential, is thought to serve as an assay for protein degradation in soil.

IV. CHARACTERISTICS OF PROTEOLYTIC ENZYMES IN SOILS

Several investigators have tried to characterize proteolytically active extracts from soil, but such analyses have been complicated by the association of proteases with soil constituents. These complexes are thought to prolong the survival of enzymes in the environment, and in some cases the protein and soil components appear to be bound together quite strongly. Complexation reactions make the purification and characterization of soil enzymes so difficult that no one has yet made a complete purification of a protein from soil (Tinsley and Zin, 1954; Jenkinson and Tinsley, 1959; Simonart et al., 1967; Biederbeck and Paul,

362

MICHAEL J. LOLL AND JEAN-MARC BOLLAG

1973; Mayaudon and Sarkar, 1974). Analysis is also hindered by the fact that soil proteases originate from a variety of sources. Proteases come from plants, animals, and microbes, and represent a heterogeneous mixture of enzymes having different molecular weights, structures, substrate specificities, etc. The type of protease extracted depends on the proteases’ stability, their association with humic material, and their chemical nature, and on the kind of soil in which they are found and the extractant used. A particular extractant may remove several proteases from a sample simultaneously. In spite of these problems, however, work in this area has given us some insight into the reactivity and composition of “humo-enzymes” and how they act upon substrates in the soil. Determining the exact composition of soil proteases is a formidable task. They are probably bound in some way to organic and inorganic substances from humus, clays, and minerals. In fact, no complete analysis has been done on proteolytic soil extracts; they do, however, appear to contain significant amounts of humus and carbohydrates. Sarkar er al. (1980) speculate that a number of proteinases are glycoproteins in soil and that carbohydrates stabilize soil enzymes. Humic acids also have a protective effect on proteases. Pronase, when complexed with humic acid analogs prepared from p-benzoquinone, had higher temperaturn and pH optima for activity and was more thennostable, although less active (Rowel1 er al.. 1973), than its native form. Nothing is known about the cofactor requirements or structures of the active components of soil extracts. The activities of soil proteases are greatly affected by changes in temperature and pH. Working with Tris-borate extracts of some Australian soils, Ladd (1972) found that the optimal pH for protease activity was between 7.0 and 8.1 when 2-phenylalanyl leucine was used as the substrate. This range may include the optima of neutral and alkaline soil proteases, such as those described by Ambroz (1970) in central European chernozems and rendzinas. He observed maximal activity for the neutral proteases at a pH of 6.5, whereas the alkaline proteases showed their highest activities at 8.5. A multienzyme complex extracted from soil with EDTA also had its maximum caseiolytic activity at pH 8.5. Proteinase-like extracts are heat labile, and Mayaudon and co-workers (1975) reported a complete loss of activity in one extract when it was heated to 75°C. From 20 or 30”C, activity increases with temperature and reaches a peak around 50-60°C, after which it drops off rapidly (Ladd, 1972). The activation energy for one proteolytic extract has been estimated to be 20 kcal (Mayaudon et al., 1975). Proteolytic soil extracts appear to be relatively stable to a number of environmental stresses. When lyophilized, they can be stored at 1-25°C for prolonged periods of time without denaturing. Incubation at higher temperatures, however, will result in inactivation, and as much as 60% of the activity may be destroyed over a 24-hr period when the extracts are kept at 50°C. Proteolytic extracts can

PROTEIN TRANSFORMATION IN SOIL

363

undergo freeze-drying with no loss of activity (Ladd, 1972), although air-drying soils prior to extraction will reduce their effectiveness. Proteases are more sensitive to air-drying than most soil enzymes (Speir and Ross, 1981). It is not clear what the effects of y radiation are on soil protease stability. McLaren et al. (1957) were able to eliminate proteolytic activity by irradiating soil samples, but Ladd (1972) found only a 33% reduction in the activity of his extracts after irradiating them with a larger dosage than McLaren used. Voets et al. (1963, on the other hand, saw no differences in proteolytic activity between irradiated and untreated soils. As the molecular weight of an enzyme is related to the amount of radiation needed to deactivate it (Dertinger and Jung, 1970), these contradictory results imply that proteases of various types and molecular weights are present in different soils. An alternative explanation is that soils may differ in their abilities to protect enzymes from y rays. Ladd (1972) found that extracted soil proteases were also resistant to attack by the pure proteases thermolysin and subtilisin, whereas Mayaudon et al. (1975) could effect a partial degradation of a proteolytic extract with commercial pronase. This discrepancy may be caused by differences in the soils used or in the preparation of the extracts; at present the cause(s) responsible has not been discovered. There is little information about the specificities of soil proteases for proteins and peptides. It is known that casein and gelatin are readily decomposed in soil but few studies have examined other substrates. According to Ladd and Butler (1972), the proteinases of several Australian soils had very high rates of hydrolysis for dipeptides with hydrophobic side groups. Hemoglobin, casein, and peptides with hydrophilic side chains were hydrolyzed much more slowly, and the degradation of the polar benzoyl arginine amide varied in different soils. This preferential attack on substrates with hydrophobic moieties is characteristic of leucine aminopeptidase, chymotrypsin, and carboxypeptidase. In order to characterize proteolytic extracts, scientists have studied their behavior in the presence of inhibitors. Ladd (1972) used EDTA, o-phenanthroline, P-phenyl propionate, mercuric chloride, and dimethyl sulfoxide to inhibit the activity of his extracts. The first three inhibitors are all metal chelators which work against carboxypeptidase, an animal proteinase. Dimethyl sulfoxide is known to attack plant proteases such as phaseolain. Therefore Ladd’s extract may contain enzymes from several sources. Another type of proteolytic extract is inhibited by diisopropylphosphorofluoridate (Mayaudon et al., 1975) but is unaffected by chemicals which react with the sulfhydryl group of proteins. This sort of behavior is typical of alkaline serine proteases. Conclusions about the origin and nature of proteases from inhibition kinetics and substrate specificities may be premature because little is known about the effect of soil materials on enzyme structure and reactivity or about the respective contributions of plants, animals, and microorganisms to the enzymatic activity of a given soil. Enzymes in soils

364

MICHAEL J. LOLL AND JEAN-MARC BOLLAG

are not the same as purified enzymes extracted from living organisms, and the characteristics of soil proteases must be examined in this context rather than in comparison with pure proteases.

V. ENVIRONMENTAL FACTORS AFFECTING PROTEOLYSIS Protein decompositionin soil is governed by the peculiar nature of the soil, the indigenous microflora, and a range of environmental factors. Many farming practices also have direct and indirect effects on soil proteolysis. In order to determine the importance of protein degradation for soil fertility we must know how this process is delimited by environmental conditions and agronomic methods. A. TEMPERAW

Proteolysis in soils has been observed over a wide range of temperatures (10-60°C) (O’Brien, 1978). In general, the decomposition of proteinaceous substances seems to increase with temperature, although Khaziyev (1977) reported a negative correlation between temperature and protease activity. He relates this unusual finding to changes in the soil structure caused by freezing and hypothesizes that at low temperatures microaggregates are broken, releasing enzymes which cannot be extracted during the summer. Others have found that proteases in soil are not as active at cold temperatures (Kuprevich and Shcherbakova, 1971). Temperature reduction decreased nitrogen mineralization in soil samples that underwent a simulated fall season (Campbell et al., 1971). The soils were cooled from 14 to 3°C in a temperature cycle which mimicked the diurnal temperature variations occurring in Saskatchewan in the fall. Samples treated in this fashion lost some of their ability to convert organic nitrogen to inorganic forms. According to Cahenzli and Staffeldt (1976), animal muscle decays faster in soil at 40 than at 5OC, and O’Brien (1978) observed that proteolysis in a desert soil was three times higher at 37 than at 20°C. The disappearanceof protein zein buried in a tussock grassland was positively correlated with the number of degree days greater than 0°C (Ross and Cairns, 1978). The thermal stimulation of protein degradation may be the result of several factors. Metabolism, growth, and enzyme synthesis in many species of microorganisms speed up with a rise in temperature. A cell may produce more protease at an elevated temperature, and if warmth encourages cell division there could be

PROTEIN TRANSFORMATION IN SOIL

365

proliferation of the more proteolytically active microorganisms, assuming that the proper nutrients are available. Nonproteolytic organisms may be affected by temperature in an analogous manner. Although it is possible that they could compete with protein decomposers, it may be that they also act to break down materials to which proteins are bound, freeing substrate for the proteolytic microbes. Finally, heating could increase the rates of enzymatic and chemical reactions involved in the decay process. There is, no doubt, an upper limit on stimulatory phenomena that is determined by the heat tolerance of the soil population and the inactivation temperature of the proteases. As temperatures in excess of 60°C are required to denature soil proteinases (Mayaudon et al., 1975; Ladd and Butler, 1972), frequent thermal inactivation in nondesert habitats is improbable. Temperature regulates decomposition indirectly through its effects on oxygen tension and moisture. In addition, a number of biological and chemical processes occur coincidently with, and not always because of, the onset of warmer summer temperatures, and these processes have their effects on the degradative capabilities of a soil. B. MOISTURE

Moisture is essential for the mineralization of proteins. Plants and microbes need water for growth and metabolism. Proteases, in order to function, must be in aqueous solution or, if bound to a solid surface, at the interface of an aqueous solution. Complete or partial dissolution of a protein substrate helps in its hydrolysis as more of its surface area becomes accessible to proteolytic enzymes. It is not surprising, then, that there is a strong correlation between protein hydrolysis and soil moisture (Kuprevich and Shcherbakova, 1971; Cahenzli and Staffeldt, 1976; Khaziyev, 1977; Klein, 1977). In areas having high temperatures and little rainfall, conditions typical of deserts and prairie summers, proteolysis is severely restricted (Klein, 1977; O’Brien, 1978). C . REDOX CONDITIONS

Most of the current data suggest that the breakdown of protein is facilitated by the presence of oxygen, although hydrolysis probably goes on at reduced rates under anaerobic conditions. The proteolytic activity of aquatic organisms has been shown to decline with decreasing oxygen tension (Sugahara et al., 1974). Ross and Cairns (1978) found that protein zein decomposed much faster at a depth of 5 cm than at 15 cm in New Zealand soil, and that the proteolytic capability of mineral soils in Karelia decreased with depth in the soil profile. Peat bog soils from the same part of the Soviet Union had hydrolytic activities

366

MICHAEL J. LOLL AND JEAN-MARC BOLLAG

throughout their profiles (Katznel’son and Ershov, 1958). This would indicate the presence of anaerobes in the lower horizons or the permeation of peat soils by oxygen. D. SOILPROPERTIES

The type, characteristics, and composition of the soil all influence the behavior of a soil enzyme. Peat bog soils, for instance, have higher proteolytic activities than meadow soils, sod podzols, and iron podzols (Katznel’son and Ershov, 1958; Kuprevich and Shcherbakova, 1971). The large amount of organic matter in peat probably acts as a favorable nutrient source for the development of protein hydrolyzers and stabilizes exocellular proteases. Organic matter content and protease action are strongly correlated (Kuprevich and Shcherbakova, 197 1; Ladd and Butler, 1972; Mayaudon and Sarkar, 1974). Proteases are also quite active in agrillaceous soils (Katznel’son and Ershov, 1958). Soil reaction is an important consideration in studying proteolysis. Chunderova (1970) was able to raise the relatively low proteolytic activity of acid soils (pH 4.2-4.8) by liming them. Activity peaked at a pH of 6.3 and did not change with further increases in pH. In soils infested with Trichoderma viride, fungal proteinases are functional only at pH values between 5.5 and 6.5; pH 6.0 is optimum for substrate hydrolysis (Rodriguez-Kabana et al., 1978). The pH has an effect on the kind as well as the activity of soil proteases. Alkaline proteases are uncommon or undetectable in acid soils (Ambroz, 1970). Ladd and Butler (1972) could not correlate proteolysis with soil pH, but they could link it to other soil properties. The activity of a peptidase specific for Z phenylalanylleucine was related to cation-exchange capacity (CEC), clay content, and surface area, whereas that of a protease specific for benzoylargininamide was correlated with soil surface area, clay content, CEC, total nitrogen, and organic matter content. Mayaudon et al. (1975) reported a positive correlation between organic matter content and the proteolytic capacity of a soil extract, and a negative correlation between activity and clay content. These results are interesting because they indicate that adsorption phenomena between proteases and organic and mineral substances may take place in soil. Ambroz (1966) has briefly discussed the affinities of neutral and alkaline proteases for different clays, pointing out that protein adsorption is more pronounced with montmorillonite and illite, which have high CECs, than with kaolinite, which has a lower CEC. Reactions with the inorganic fraction of soil decrease, but do not eliminate, the ability of proteases to attack proteins (GHith and Thomas, 1979). Protein adsorbed into the clay matrix is better protected from decomposition than surface-adsorbed protein. It is possible, however, that some smaller proteases can penetrate the clay layers and hydrolyze proteins and peptides. Allophane at a pH of 5 is a strong inhibitor of pronase, as are montmorillonite and halloysite to

PROTEIN TRANSFORMATION IN SOIL

367

lesser degrees at the same pH (Aomine and Kobayashi, 1964). Inhibition occurs even when montmorillonite is partially saturated with protein substrate (Griffith and Thomas, 1979). Reduction in activity is probably caused by conformational changes in the enzyme when it binds to clay. In some cases, soil minerals promote proteolysis. When lysozyme was adsorbed to bentonite and incubated with pure cultures of Pseudomonas or Fluvobacterium species, the bacteria decomposed the complexed protein faster than the uncomplexed protein. Clay apparently can concentrate protein on its surface, whereas in solution the substrate is too dilute for rapid hydrolysis (Estermann and McLaren, 1959). Soil mineral matter can, on occasion, provide sites for enzyme-substrate interactions (McLaren and Estermann, 1956). E. NATURALLY OCCURRING ORGANIC COMPOUNDS

Polyphenolics, specifically plant tannins and humic acids, are the most studied organic compounds in terms of their reactions with proteins and enzymes in soil. Tannins have long been known to lessen enzymatic activity (Goldstein and Swain, 1965) and exert a depressive effect on proteolysis (Basaraba and Starkey, 1966), either changing the shape of the protease or protecting the substrate from hydrolysis. Dihydroxphenyalanine (DOPA) melanin, found in fungi, inhibits protease (Kuo and Alexander, 1967), and humic and fulvic acids may increase or decrease the potency of proteolytic enzymes (Ladd and Butler, 1969b). The particular effect depends on the protease in question and the molecular weight (Butler and Ladd, 1971) and carboxyl content (Butler and Ladd, 1969) of the humic acid. Pronase, trypsin, carboxypeptidase A, and subtilopeptidase A have been inhibited by soil humic acids, and this inhibition varied for some of the proteinases depending on the substrate used (Ladd and Butler, 1969b). The action of Pronase on peptides was more sensitive to humic acid than was its action on protein (Ladd and Butler, 1969a). Papain, ficin, and thermolysin have been stimulated by humic acids, and phaseolain and chymotrypsin were unaffected by similar treatments (Ladd and Butler, 1969b). Butler and Ladd (1971) theorize that humic substances, especially those with large molecular weights (greater than 30,000), are rigid, and when they bind to proteases they distort the enzymes’ structures, thereby changing their reactivities. The binding sites involve the carboxyl groups of the humic acids (Butler and Ladd, 1969). Plants and animals synthesize their own protease inhibitors for physiological regulation. Little is known about the survival of these compounds in soil, but it is possible that they could affect proteolysis. Natural inhibitors are common in mammals, potatoes, beets, and soybeans (Vogel et al., 1968). Proteases are subject to autodigestion and may act as their own substrates (Cunningham, 1965). It is not known whether reaction with soil components

368

MICHAEL J. Lou AND JEAN-MARC BOLLAG

prevents this. Calcium stabilizes some proteolytic enzymes (Cunningham, 1965) and may protect them from self-digestion in soils if that element is prevalent.

Plant roots exude substances that can be used by microorganisms for growth. As a result, protease activity is higher in the rhizosphere than in the soil outside of this area, and even more activity is evident on the rhizoplane (Vagnerova and Macura, 1974~).Conditions at the root surface are optimal for nutrient assimilation by microbes, but this is not the only reason for the superior protease activity of the rhizoplane. Roots, like clays and humic acids, have exchange capacities and can bind enzymes, reacting with them electrostatically or actually taking them up into the root’s free space (McLaren et al., 1960). Sorption reduces the efficacy of the proteases (Vagnerova and Macura, 1974b), but the amount of bound protease can be so great that the entire activity of the root surface is well above that of the surrounding soil. The sorption of proteinases by roots could be another way in which enzymes are preserved in soil (Vagnerova and Macura, 1974~). G . ENZYMEACTIVITIES

The physiological processes of plants and nonproteolytic bacteria and fungi may coincide with the breakdown of protein. There are strong relationships between protease activity and the activities of sulfatase (Ladd and Butler, 1972), DOPA oxidase (Mayaudon et al., 1975), invertase, amylase, and catalase (Ambroz, 1970), implying that protein decomposers have a broad range of metabolic abilities or that environments that favor proteolytic organisms also favor organisms which synthesize these other enzymes.H. AGRICULTURAL PRACTICES

Farming produces fundamental alterations in the biochemistry of soils, and planting methods, fertilization, and herbicide application can effect drastic changes in protease activity. Plowing mineral soils can improve their proteolytic activity (Romeiko, 1969), and in some cases cultivation decreases the nitrogen content of a soil (Keeney and Bremner, 1964). Ammonium sulfate and potassium nitrate fertilization may be beneficial for protease synthesis (Bei-Bienko, 1970). Blagoveshchenskaya and Danchenko (1974) saw higher enzymatic activity in crop rotations than in a corn monoculture. When they added nitrogen, phosphorus, and potassium fertilizers to a soil after manuring, protease activity

PROTEIN TRANSFORMATION IN SOIL

369

increased as compared with soils amended with only mineral fertilizers. Rates of application of mineral nitrogen over manure had little influence on activity in the monoculture but increased it by 20% in the rotation. Amending a Belgian soil with manure diminished the numbers of proteolytic microbes but enhanced proteinase activity (Batistic and Mayaudon, 1978). Sewage sludge, in spite of its heavy metal content, similarly can aid protein decomposition (Varanka et al., 1976). Manures and sludges nurture soil microflora, add their own active microbial populations to soil, and may contain extracellular proteases stabilized by organic matter (Varanka et al., 1976). I. PESTICIDES

Pesticides often affect helpful as well as harmful organisms in the area of application. For soil populations, these xenobiotics are beneficial or detrimental depending on the concentration and type of pesticide, the persistence of the compound in soil, and the composition and sensitivity of the microbial community. Eptam (S-ethyl-N,N-dipropylthiocarbamate),for example, stimulates soil proteolysis at low concentrations but is inhibitory at higher levels (Cullimore and Ball, 1978). The same study showed that bromacil [5-bromo-6-methyl-3 (1methylpropy1)-2,4-(lH,3H)-pyrimidinedione] fosters protein decomposition whereas MCPA (2-methyl-4-chlorophenoxyacetic acid), diquat (1,l ’-ethylene-2,2’-dipyridylium dibromide), paraquat (1, l ’-dimethyl-4,4’-bipyridylium ion), dicamba (3,6-dichloro-o-anisic acid), 2,4-D (2,4-dichlorophenoxyacetic acid), 2,4,5-T (2,4,5-trichlorophenoxyacetic acid), and monuron [3-@-chloropheny1)- 1,1-dimethylurea] all reduce the hydrolysis of proteins. Dinoseb (2,4dinitro-6-sec-butyphenol)decreases the number of protein decomposers in soil (Wainwright and Pugh, 1973). In Swedish soils, 2,4,5-T slightly promoted the growth of proteolytic microorganisms (Torstensson, 1974), as did bentazon [3isopropyl- lH-2,1,3-benzothiadiazin-4-(3H)-one-2,2-dioxide]at a concentration of 1 ppm (Torstensson, 1975). The fungicides thiram (tetramethylthioperoxydicarbonic diamide), verdasan (phenylmercuric acetate), and captan {N-[(trichloromethyl)thio]-4-cyclohexene-1,2-dicarboximide} in high concentrations caused increases in soil ammonification (Wainwright and Pugh, 1973), indicating that ammonifying or proteolytic microbes (or both) were stimulated. Malathion (0,0-dimethyl-S-( l ,2-dicarbethoxyethyl)dithiophosphateis not harmful to species of Bacillus and Pseudomoms (Stanlake and Clark, 1975), two proteolytic genera often found in soil. Proteinase activity is relatively unaffected by the application of atrazine (2-chloro-4-ethylamino-6-isopropylamino-s-triazene) (Ghinea, 1964). The clay and sand contents of two Belgian soils appeared to govern the growth of les germes protiolytiques in the presence of three phenylcarbamates (Bellinck and Mayaudon, 1978). The pesticides encouraged ini-

370

MICHAEL J. LOLL AND JEAN-MARC BOLLAG

tial growth in a clay soil, but bacterial development eventually peaked and declined sharply over time. This pattern was also evident in a sandy soil to a lesser degree, except that monophenylcarbamateproduced an initial decrease in the number of protein hydrolyzers. This inhibition was followed by an increase in growth which gradually leveled off. The growth phase and decline was much shorter lived in the sandy soil than in the clay soil. Bellinck and Mayaudon (1978) see this as the result of herbicide absorption by soil colloids with slow herbicide release for microbial nutrition. Pesticides occasionally exhibit synergistic phenomena with fertilizers in altering enzymatic operations. Following an initial surge in protease synthesis, soil treated with both urea and paraquat in the laboratory showed a 42% loss of activity after 1 month (Murakami et al., 1969). The addition of dalapon and urea to samples caused an immediate reduction in activity of 11%. Plots in the field lost 41-56% of their prior activity with dalapon-urea combinations and 46% activity with the addition of urea and paraquat (Namdeo and Dube, 1973b). Soil peptidases and proteases are not equally susceptible to fumigation. Peptidases capable of hydrolyzing Z-phenylalanyl leucine and benzoylargininamide were not affected by methyl bromide, whereas caseinases from the same sterilized soil were markedly inhibited by that compound. The activities of the enzymes were partially suppressed by chloropicrin and then rebounded, exceeding the activities of untreated controls (Ladd and Butler, 1966). It has been proposed that the deactivation of caseinase comes from its binding to clay and the destruction of unbound caseinase. The subsequent recovery of activity resulted from a microbial flush, which happens when surviving organisms begin to develop on the lysed cell material produced by pesticide treatment. In fact, fumigation probably releases relatively large amounts of protein from the microbial biomass, and the dead cells then provide carbon and nitrogen for the few unharmed cells. This conclusion was supported by a study done by Wainwright and Pugh (1975), who extracted more free amino acids from soil after, than before, fumigation.

VI. TRANSFORMATION AND BINDING OF PROTEIN IN SOIL Protein undergoes different transformations in the soil environment (Fig. 1). Quantitative studies are practically nonexistent, but a large proportion of the protein introduced into soil is broken down into peptides, which in turn are hydrolyzed to amino acids. The amino acids are metabolized by the soil microflora or react with soil constituents. Surviving proteins may react with soil minerals, lignins, tannins, melanins, and humic acids. The aggregates that proteins form with these materials protect peptides from microbial action and weath-

37 1

PROTEIN TRANSFORMATION IN SOIL MICROORGANISMS PLANTS ANIMALS

PLANT ROOTS

1 -

HUMIC SUBSTANCES

PROTEINS

t

CLAYS

PEPTIDES

AMINO ACIDS

NH3

FIG. 1. Protein transformation in soil.

ering (Haworth, 1971). Complexed protein is still susceptible to some biodegradation, but decays at a much slower rate than does uncomplexed protein (Estermann et al., 1959). The evidence for clay and humoprotein complexes is indirect, and there is still some controversy about whether or not intact protein can exist in soil over long periods of time. Infrared spectra of some humic acids and straw extracts show what appear to be protein amide I and amide I1 peaks which disappear after treatment with 6 N HCI (Goulden and Jenkinson, 1959; Stevenson and Goh, 1971; Boyd et al., 1980). Acid hydrolysis of soil releases amino acids, which comprise 20-50% of the total nitrogen content (Bremner, 1949, 1955, 1965; Cheng and van Hove, 1964; Piper and Posner, 1968). Humic acids contain up to 15% a-amino nitrogen (Piper and Posner, 1968). Some nonprotein amino acids synthesized by bacteria have been occasionally

312

MICHAEL J. LOLL AND JEAN-MARC BOLLAG

detected in soil, but these compounds do not seem to be a large proportion of the total amino acid content and may even be artifacts in some instances (Stevenson, 1956; Bremner, 1965). The presence of peptides in humic acid was indicated by Piper and Posner (1968) and by Sowden (1966). Peptide-like compounds from soil have been found to react with phenylisothiocyanate and fluorodinitrobenzene, two compounds that characteristically react with the terminal groups of proteins (Sowden, 1966). Because of the heterogeneity of soil material, conclusions about such findings can only be speculative, as it is not known how soil constituents may interfere with reactants for proteins. Classical tests for protein content (i.e., the Lowry and Biuret methods) have given low values for the peptide composition of humic acids, but soil polymers may serve to mask protein from detection by conventional means.(Ladd and Butler, 1966). The vast amount of research performed in soil enzymology also lends support to the notion that proteins exist in a stabilized form in soils, but clear data are lacking.

A. CLAY-PROTEINCOMPLEXES

Because of their cation-exchange capacity and relatively large surface area, clays act as absorbents of positively charged organic material. Association with clay confers upon proteins some resistance to degradation (Ensminger and Gieseking, 1942; Pinck and Allison, 1951; Pinck et ul., 1954; McLaren, 1954b; Lynch and Cotnoir, 1956). The effect of clay is attributed to its inhibition of proteolytic enzymes, its shielding of protein from enzymatic attack, or a combination of the two processes. In general, bentonite-protein complexes are more resistant to hydrolysis than are kaolinite- and illite-protein complexes (Lynch and Cotnoir, 1956; Estermann et al., 1959). Bentonite has a higher cationexchange capacity than kaolinite and illite, and more protein is adsorbed to montmorillonitic materials. X-Ray diffraction shows that proteins can form interlamellar monolayers in montmorillonite and that increasing the amount of protein added to clay suspensions results in the formation of bilayers (Talibud e n , 1950). Protein is best shielded by a clay lattice when it forms a single layer on the clay surface, a monolayer produced when the weight of the protein in the complex is roughly 8% of that of the montmorillonite. The upper portions of protein multilayers are more susceptible to degradation by proteolytic enzymes (Pinck et al., 1954). As excess protein in the clay structure is hydrolyzed, the layers contract and may help to protect the thin coating of adsorbed protein (Estermann et al., 1959). Aggregates of lysozyme and bentonite or kaolinite are less vulnerable to decomposition when dried and then rewetted (Estermann et ul., 1959); this treatment may wash superfluous protein out of the matrix. The conformation of the protein in a clay complex either speeds up or slows down its hydrolysis.

PROTEIN TRANSFORMATION IN SOIL

373

Denatured lysozyme is broken down more quickly in combination with kaolinite than is the native globular form in a free or complexed state (McLaren, 1954b). Inorganic compounds other than clay may prolong the survival of proteins in soil. Free iron and aluminum oxides can be chelated by organic compounds (Greenland, 1965) and seem to prevent decay, perhaps by changing the shape of the organic substrate and making it less suitable for enzymatic attack. The amino groups of humus seem to be involved in chelating metal ions, forming complexes that help to stabilize clays (Brydon and Sowden, 1959). Montmorillonite and other soil minerals inhibit biodegradation but do not stop it entirely. Species of Pseudomonas, Flavobacterium, and Bacillus exude proteases that can penetrate clay layers (Estermann and McLaren, 1959; Estermann et al., 1959; Marshman and Marshall, 1981). McLaren and Estermann (1956) theorized that proteases may be adsorbed to kaolinite and can move over it, in some fashion, to react with substrates. Marshman and Marshall (1981) investigated the growth of protein-decomposing bacteria on clay-adsorbed proteins, and from their data proposed a model based on four assumptions: 1. Protein binds to clay matrices at sites that are available to proteolytic organisms as well as to sites that are not available to them 2. Protein binds strongly to both sites and is not readily released from them 3. Protein prefers to react with unavailable sites 4. The amount of protein bound to the two types of sites is determined by the nature of the proteases present. The problem of inhibition has led soil scientists to examine the factors which affect the bonding of a protein or proteinase to clay. Proteins bind to clays very quickly, and 90% of the protein added to mineral systems may be adsorbed in only 3 minutes (McLaren, 1954b; Armstrong and Chesters, 1964). Reaction sites on clays include not only the interlamellar surfaces but also the edges of the platelets (Harter and Stotzky, 1973), depending on the peptide type and concentration. Adsorption may be in mono- or multilayers (Talibudeen, 1950). The primary mechanism of binding is ionic and results from the cation-exchange capacities of soils (Ensminger, 1942; Ensminger and Gieseking, 1939, 1941). Amino, imidazole, and protonated carboxyl groups on proteins compete with inorganic ions for sites on the CECs of clays (Ensminger, 1942; Armstrong and Chesters, 1964; Schnitzer et al., 1980; Stefani and Sequi, 1978). Hydrogen bonding may also be involved in the formation of protein-mineral aggregates (McLaren, 1954a; McLaren e? al., 1958; Albert and Harter, 1973). It is possible that the stresses induced by adsorption denature proteins as they attach to the clay surface (Talibudeen, 1950), but this is a matter of some contention. Maximal adsorption is sometimes observed when the pH of the clay suspension is equal to the isoelectric pH of the protein (McLaren, 1954a; Armstrong and Chesters, 1964; Albert and Harter, 1973). At pH values above the isoelectric

374

MICHAEL J. LOLL AND JEAN-MARC BOLLAG

pH, the protein is negatively charged and repulsed from the mineral matrix. At pH values below the isoelectric value, the protein is positively charged, and although a greater affinity for the clay would be expected, this is not necessarily the case. In fact, adsorption decreases with lower pH, a phenomenon variously ascribed to competition between protein molecules and hydrogen ions for exchange sites (McLaren et al., 1958) and charge density effects. As the pH drops, the number of positive functional groups on a peptide increases and fewer molecules are needed for satisfying the clay’s negative charges (Armstrongand Chesten, 1964). Protein adsorption to kaolinite, according to Albert and Harter (1973), is not dependent on pH, and bonding in this instance must be nonionic. Harter and Stotzky (1971) criticized the importance of the influence of pH in the formation of protein-mineral complexes. First, they pointed out that even at pH values above the iwelectric pH of a peptide there are still some positively charged groups on a protein that are available for reaction with clays. Second, they mentioned that the pH at the surface of a clay is not always the same as that of the overall clay suspension, and the surface pH may be more favorable for binding. More important factors governing adsorption appear to be protein molecular weight and the ion population on the clay surface. Proteins with higher molecular weights are more likely to be adsorbed than those with low molecular weights and evidently have a certain competitive advantage. The number of positively charged functional moieties has a certain secondary influence, explaining discrepancies in the adsorption of materials of different weight. Catalase, for example, has a molecular weight more than twice that of casein (25 1,OOO versus 121,0oO), yet the difference in their respective affinities for clay is less than a factor of two. Casein probably has more positive charges at low pH than does catalase, and proportionately less casein is required for saturation (Harter and Stotzky, 1971). Lysozyme adsorption has been related to the valence or ionic potential of cations on the cation-exchange capacity and decreased with an increase in the size of the ionic radius of the saturant. Ovalbumin adsorption on the same clay was affected more by the pH of the suspension. Protein adsorption was found to be highest for clays homoionic to hydrogen and decreased in the order H > Na > Ca > A1 > La > Th. Proteins seem to compete better with monovalent ions for clay exchange sites as divalent and trivalent cations are bound more strongly (Harter and Stotzky, 1971). Only hydrogen ions behave differently. The presence of salts in clay suspensions discourages peptide adsorption and can desorb bound organic matter (Harter and Stotzky, 1971). B . POLWHFNOLIC-PROTEIN COMPLEXES

Lignins, tannins, and melanins are common plant and fungal metabolites found in leaf litter, crop residues, and soil and are the major components of

PROTEIN TRANSFORMATION IN SOIL

375

humic acids. Phenolic polymers such as these adsorb and react with proteins, stabilizing them against enzymatic digestion. As with clays, stabilization is the probable consequence of protease inhibition and protection of the substrate from attack. In 1932 Waksman and Iyer discovered the inhibitory effect of lignin upon proteolysis and proposed that most of the organic nitrogen of soil was present in a lignoprotein “humic nucleus.” Since that time tannins (Davies et al., 1964; Basaraba and Starkey, 1966; Benoit et al., 1968; Lewis and Starkey, 1968), melanin (Kuo and Alexander, 1967), humic acids (Lynch and Lynch, 1958), and humic acid analogs (Verma ef al., 1975; Verma and Martin, 1976; Martin et al., 1978; Martin and Haider, 1979) have been proven to retard protein breakdown. It is now clear that the nature of soil nitrogen is quite complicated. This complexity is compounded by the variety of interactions phenols and proteins undergo. There are several mechanisms by which protein-phenol complexes are formed, and the Occurrence of any one is regulated by certain environmental conditions. Hydrogen bonding takes place between proteins and phenols and is probably a rather common event in litters and soils. The bond which forms between substituted amides and phenolic hydroxyl groups is very strong and no doubt is important in the production of resistant complexes. Strong hydrogen bonds also occur with carboxyl groups (Loomis and Battaile, 1966; Ladd and Butler, 1975). Hydrogen bonding would explain the widespread observation that tannin-protein complexes synthesized at low pH are more resistant to decomposition than complexes produced at high pH (Davies et al., 1964; Basaraba and Starkey, 1966; Benoit et al., 1968). It has been noted that hydrolyzable tannins do not provide protein with as much protection as condensed, more aromatic ones do, and this may be because of intermolecular hydrogen bonding within hydrolyzable tannins. Tannins of this kind would be less likely to bind to proteins (Ladd and Butler, 1975). Acid forest soils with relatively large amounts of condensed tannins would probably be the best areas for the formation of stabilized proteins (Basaraba and Starkey, 1966). Covalent bonding is another widespread mode of reaction in soil and is responsible in part for the origin of humic and fulvic acids. Proteins, peptides, amines, and amino acids react with phenols, giving rise to polymeric products. These oxidations can be strictly chemical or mediated by phenolase enzymes (Loomis and Battaile, 1966; Taylor and Battersby, 1967; Ladd and Butler, 1975). Catechol forms a p-aminohydroquinone with primary amines in the presence of a chemical oxidant (Mason, 1955). Theis (1945) has hypothesized that proteins react with quinones via meta linkages between the ring and the free amino groups of the peptide backbone or of lysine. Glycine does not combine with phenols as quickly as glycylglycine does, and the tripeptide reacts fastest of all, in what is referred to as the peptide effect (Mason, 1955). The bond between the aromatic ring and the protein amino group is thought to be the most impervious to enzymatic action (Ladd and Butler, 1966; Robert-Gero et al., 1967), although a species

376

MICHAEL J. LOLL AND JEAN-MARC BOLLAG

of Achromobucter has been isolated that can utilize N-(0-carboxypheny1)glycine. The same species, however, was unable to metabolize N-phenylglycine, N-knitrophenyl)glycine, or N-@-hydroxypheny1)glycine. Clearly the susceptibility to degradation is a function of other moieties (Ladd, 1964). Covalent reactions may also occur between free amino groups and humic acid carboxyls and other phenolic substituents. Ionic interactions between organic compounds occur in soil between basic and acidic functional groups. Associations of this kind are highly pH dependent and are similar to mineral cation exchange. Ladd and Butler (1971) and Butler and Ladd (1969) implied that the amino groups of enzymes react ionically with humic acid carboxyls. The activity of the enzyme can be greatly altered by this complexation. C. RHIZQSPHFXEADSORITION

Finally, roots are capable of adsorbing proteinaceous matter. Some proteins can penetrate the outer spaces of barley roots, which have a negative charge (McLaren et al., 1960). Whether plants can actually utilize native proteins is unknown, but roots may be another “shelter” against enzymatic hydrolysis. Covalent and coulombic processes are always at work in soil and it would not be surprising if a protein were bound to organic and inorganic matter by two or three mechanisms at the same time. The heterogeneity of bonding is one reason that proteins are stabilized in soil. A battery of enzymes would be needed to destroy the various linkages between peptides and soil components, and this is why protein or portions of protein can remain intact in soil.

VII. ECOLOGICAL AND AGRONOMIC IMPORTANCE OF PROTEIN TRANSFORMATION

Protein transformations in soil have a considerable influence on soil ecology, agriculture, and public health. From an ecological point of view, protein is important, as discussed previously, in its contribution to soil structure and through its reactions with clays and naturally occurring phenols during humus formation. Apparently a number of extracellular microbial enzymes are stabilized by their complexation with soil components and can survive conditions inimical to intact cells (Ladd and Butler, 1975). Some of these enzymes act for long periods in degrading the basic constituents of plant and animal residues. Others such as ureases, amidases, and phenoloxidases transform fertilizers and alter pesticides and their intermediates and thus have agricultural significance.

PROTEIN TRANSFORMATION IN SOIL

377

Protein mineralization affects public health because of its role in the biodegradation of industrial and sewage wastes (Gray and Biddlestone, 1974). In addition, the tendency of protein to bind with soil substituents could pose problems for sanitation. It is feared that the protein coats of enteric viruses are sorbed to clays in sludge-amended soils. The sorbed viruses may later be released (especially if there is an increase in the ionic strength of the soil solution) and could contaminate the soil, crops, and groundwater (Harter, 1975; Seidler et al., 1980). The most important function of protein in soil is as a nitrogen source. In the years to come more emphasis will probably be placed on protein function in this respect. With increasing petroleum prices, crop fertilization becomes a more expensive proposition; to offset cost increases, it is probable that farmers will use more organic fertilizer and will apply agricultural practices that enhance the availability of nitrogen from protein. For instance, pesticide treatments and tillage could be changed so as to facilitate maximum proteolysis by the soil microflora. This might include the substitution of pesticides which do not interfere in protein metabolism. No-till methods may not be best for the degradation of proteins and peptides, as proteolysis is an aerobic process. Plowing helps to increase soil aeration, which is important for the incorporation and transformation of organic fertilizer in soil. In some cases liming acid soils may improve the degradation of proteinaceous wastes. Protein transformation affects many agricultural and biological processes. Therefore, more research on proteolysis and protein transformation in soil is needed. A better understanding of these problems could help agronomists to improve and maintain soil fertility through the use of proteinaceous wastes as supplemental fertilizers. ACKNOWLEDGMENT The authors thank Drs. Jon K. Hall and Les E. Lanyon for their assistance and helpful comments.

REFERENCES Albert, J. T., and Harter, R. D. 1973. Soil Sci. 115, 130-136. Alexander, M. 1977. “Introduction to Soil Microbiology,” 2nd ed. Wiley, New York. Ambroz, Z. 1966. Sb. Vys. Sk. Zemed. Brne Rada Al, 57-62. Ambroz, 2. 1970. Zentralbl. Bakteriol. Parasitende. Abt. 2 125, 433-437. Aomine, S.,and Kobayashi, Y. 1964. Soil Sci. P h Nutr. (Tokyo)10, 28-32. Appleby, J. C. 1955. J . Gen. Microbiol. 12, 526-533. Armstrong, D. E., and Chesters, G. 1964. Soil Sci. 98, 39-52. Banwart, W. L., and Bremner, J. M. 1976. Soil Biol. Biochem. 8, 439-443. Basaraba, J., and Starkey, R. L. 1966. Soil Sci. 101, 17-23.

378

MICHAEL J. LOLL AND JEAN-MARC BOLLAG

Batistic, L., and Mayaudon, J. 1978. Rev. Ecol. Biol. Sol 14, 499-507. Bei-Bienko, N. V. 1970. Pochvovedenie 2, 87-93. Bellinck, C., and Mayaudon, J. 1978. Rev. Ecol. Biol. Sol 15, 435-444. Benoit, R. E., Starkey, R. L., and Basaraba, J. 1968. Soil Sci. 105, 153-158. Biederbeck, V. O., and Paul, E. A. 1W. Soil Sci. 115,357-366. Black, C. A. 1968. “Soil-Plant Relationships,” 2nd ed. Wiley, New York. Blackbum, T. H., and Hobson, P. N. 1962. J. Gen. Microbiol. 29, 69-81. Blagoveshchenskaya,Z. K., and Danchenko, N. A. 1974. Sov. SoilSci. (Engl. Transl.) 6,569-575. Behme, H., and Ziegler, H. 1969. Mycopathol. Mycol. Appl. 38, 247-255. Boulter, D., and Derbyshire, E. 1977. In “Plant Proteins” (G. Norton, ed.),pp. 3-24. Buttenworth, Boston. Boyd, S. A., Sommers, L. E., and Nelson, D. W. 1980. Soil Sci. Soc. Am. J . 44, 1179-1186. Bremner, J. M. 1949. J. Agric. Sci. 39, 181-193. Bremner, J. M. 1955. J. Agric. Sci. 46, 247-256. Bremner, J. M. 1965. In “Soil Nitrogen’’ (W. V. Bartholomew and F. E. Clark, eds.). Amer. Soc. Agron., Madison, Wisconsin. Brydon, J. E., and Sowden, F. J. 1959. Can.J. Soil Sci. 39, 136-143. Buckman, H. O., and Brady, N. C. 1966. “The Nature and Proprties of Soils.” MacMillan, New York. Butler, I. H. A., and Ladd, J. N. 1969. Aust. J. Soil Res. 7, 263-268. Butler, J. H. A., and Ladd, J. N. 1971. Soil Biol. Biochem. 3, 249-257. Cahenzli, M., and Staffeldt, E. E. 1976. Proc. Int. Biodegradation Symp. 3rd, pp. 523-532. Campbell, C. A., Biederbeck, V. O., and Warder, F. G. 1971. Soil Sci. SOC. Am. Proc. 35, 480-483. Capaldi, R. A. 1977. In “Membrane Proteins and Their Interactions with Lipids” (R. A. Capaldi, ed.), pp. 1-19. Dekker, New York. Chalvignac, M. A. 1953. Ann. Inst. Pusteur (Paris)84, 816-819. Chang, C. W., and Bandurski, R. S. 1964. Plant Physiol. 39, 60-64. Cheng, H. H., and van Hove, J. 1964. Pedologie 14, 8-23. Chmel, L., and Vlacilikova, A. 1975. Sabouraudia 13, 185-191. Chunderova, A. N. 1970. Sov. Soil Sci. (Engl. Transl.) 2, 308-314. Clark, F. E. 1967. In “Soil Biology” (A. Burges and F. Raw, eds.), pp. 15-49. Academic Press, New York and London. Clark, F. E., and Paul, E. A. 1970. Adv. Agron. 22, 375-435. Cox, D. J. 1978. J . Appl. Bactetiol. 45, 259-266. Cullimore, D., and Ball, L. 1978. Appl. Environ. Microbiol. 36, 959-961. Cunningham, L. 1%5. Compr. Biochem. 16, 85-188. Dangerfield, J. A., Westlake, D. W. S., andcook, F. D. 1978. Can.J . Microbiol. 24, 1520-1525. Das, A., Chatterjee, M., and Roy, A. 1979. Mycologia 71, 530-536. Davies, R. I., Coulson, C. B., and Lewis, D. A. 1964. J. Soil Sci. 15, 299-309. Dertinger, H., and Jung, H. 1970. “Molecular Radiation Biology.” Springer-Verlag, New York. Dunican, L. K., and Rosswall, T. 1974. I n “Soil Organisms and Decomposition in Tundra” (A. J. Holding et al.. eds.),pp. 79-92. IBP Tundra Biome Steering Comm., Stockholm. Eklund, E., Backman, T., and Gyllenberg, H. G. 1971. Zentralbl. Bokteriol. Parusitenkde. Abt. 2 126, 725-734. Ensminger, L. E. 1942. Soil Sci. 54, 191-197. Ensminger, L. E., and Gieseking, J. E. 1939. Soil Sci. 48, 467-473. Ensminger, L. E., and Gieseking, J. E. 1941. Soil Sci. 51, 125-132. Ensminger, L. E., and Gieseking, J. E. 1942. Soil Sci. 53, 205-209. Estemann, E. F., and McLaren, A. D. 1959. J. Soil Sci. 10,-78.

PROTEIN TRANSFORMATION IN SOIL

379

Estermann, E. F., and McLaren, A. D. 1961. Plant Soil 15, 243-260. Estermann, E. F., Peterson, G. H., and McLaren, A. D. 1959. Soil Sci. Soc.Am. Proc. 23,31-36. Felbeck, G . T. 1971. Soil Sci. 111, 42-48. Fergus, C. L. 1964. Mycologia 56, 267-284. Flaig, W., Beutelspacher, H., and Rietz, E. 1975. In “Soil Components” (J. E. Gieseking, ed.), Vol. 1, pp. 1-211. Springer-Verlag, New York. Ghinea, L. 1964. Trans. In?. Congr. Soil Sci. 8th 3, 857-869. Gill, P. P., and Modi, V. V. 1981. Foliu Microbiol. 26, 78-82. Goldstein, J. L., and Swain, T. 1965. Phytochemistry 4, 185-192. Goulden, J. D. S., and Jenkinson, D. S . 1959. J. Soil Sci. 10, 264-270. Gray, K. R., and Biddlestone, A. J. 1974. In “Biology of Plant Litter Decomposition” (C. H. Dickinson and G. J. F. Pugh, eds.), Vol. 2, pp. 743-775. Academic Press, New York. Gray, T. R. G., and Williams, S. T. 1971. “Soil Micro-Organisms.” Hafher, New York. Greenland, D. J. 1965. Soils Fen. 28, 521-532. Griffin, D. M. 1960. Trans. Br. Mycol. SOC. 43, 583-5%. Griffin, D. M. 1972. “Ecology of Soil Fungi.” Syracuse Univ. Press, Syracuse. Griffith, S. M., and Thomas R. L. 1979. Soil Sci. SOC.Am. J . 43, 1138-1140. Hankin, L., and Hill, D. E. 1978. Soil Sci. 126, 40-43. Hankin, L., Sands, D. C., and Hill, D. E. 1974. Soil Sci. 118, 38-44. Harter, R. D. 1975. Soil Sci. 120, 174-181. Harter, R. D., and Stotzky, G. 1971. Soil Sci. SOC. Am. Proc. 35, 383-389. Harter, R. D., and Stotzky, G. 1973. Soil Sci. SOC.Am. Proc. 37, 116-123. Haworth, R. D. 1971. SoilSci. 111,71-79. Hobson, P. N. 1973. Proc. Biochem. 8(1), 19-25. Hobson, P. N., Bousefield, S., and Summers, R. 1974. CRC Crit. Rev. Environ. Control 4, 131-191. Holding, A. J., Franklin, D. A., and Watling, R. 1965. J. Soil Sci. 16, 44-59. Hunt, S. 1970. “Polysaccharide-Protein Complexes in Invertebrates.” Academic Press, New York. Ilyaletdinov, A. N., Mamilov, Sh. Z., and Adiev, A. 1972. Symp. Biol. Hung. 11, 117-120. Ishizawa, S., Araragi, M., and Suzuki, T. 1969. Soil Sci. Plant Nutr. (Tokyo) 15, 104-112. Jenkinson, D. S., and Tinsley, J. 1959. J. Soil Sci. 10, 245-263. Katznelson, H., and Rouatt, J. W. 1957. Can. J . Microbiol. 3, 265-269. Katznel’son, R. S., and Ershov, V. V. 1958. Microbiology (Engl. Transl. of Mikrobiologiya, Washington, D.C.) 27, 81-88. Keay, L. 1971. Proc. Biochem. 6(8), 17-21. Keeney, D. R., and Bremner, J. M. 1964. Soil Sci. Soc.Am. Proc. 28, 653-656. Khaziyev, F. Kh. 1977. Sov. Soil Sci. (Engl. Transl.) 9, 552-563. Klein, D. A. 1977. Ecology 58, 184-190. Knosel, D. 1974. Z. Pfanzenernaehr. Bodenkd. 81, 602-605. Knowles, J. R., and Gutfreund, H. 1974. In “Chemistry of Macromolecules” (H. Gutfreund, ed.), pp. 376-397. Univ. Park Press, Baltimore, Maryland. Kobayashi, M., Takahashi, E., and Kawaguchi, K. 1967. Soil Sci. 104, 113-118. Kolesnikova, I. G., Khokhlova, Yu. M., Sergeeva, L. N., and Shabanova, E. M. 1972. Microbiology (Engl. Transl. of Mikrobiologiya, Washington, D.C.) 41, 784-788. Kononova, M. M. 1966. “Soil Organic Matter: Its Nature, Its Role in Soil Formation and in Soil Fertility,” 2nd Engl. ed. Pergamon, New York. Kuo, M. J., and Alexander, M. 1967. J. Bacteriol. p4, 624-629. Kuprevich, V. F., and Shcherbakova, T. A. 1971. “Soil Enzymes.” Indian Natl. Documentation Centre, New Delhi. Ladd, J. N. 1964. Aust. J . Biol. Sci. 17, 153-169.

380

MICHAEL J. LOLL AND JEAN-MARC BOLLAG

Ladd, J. N. 1972. Soil Biol. Biochem. 4, 227-237. Ladd, J. N., and Brisbane, P. G. 1967. Ausf. J . Soil Res. 5, 161-171. Ladd, J. N., and Butler, J. H. A. 1966. Ausf. J. Soil Res. 4, 41-54. Ladd, J. N., and Butler, J. H. A. 1%9a. Ausr. J. Soil Res. 7, 241-251. Ladd, J. N., and Butler, J. H. A. 1969b. Ausr. J. Soil Res. 7, 253-261. Ladd, 1. N., and Butler, J. H. A. 1971. Soil Biol. Biochem. 3, 157-160. Ladd, J. N., and Butler, J. H. A. 1972. Soil Biol. Biochem. 4, 19-30. Ladd, J. N., and Butler, J. H. A. 1975. In “Soil Biochemistry” (E. A. Paul and A. D. McLaren, eds.), Vol. 4, pp. 143-194. Dekker, New York. Lajudie, J., and Chalvignac, M. 1950. Ann. Insr. Pusreur (Paris) 90, 359-361. Lewis, J. A., and Starkey, R. L. 1968. Soil Sci. 106, 241-247. Loomis, W. D., and Battaile, J. 1966. Phytochemisrry 5, 423-438. Lynch, D. L., and Cotnoir, L. J., Jr. 1956. Soil Sci. Soc. Am. Proc. 20, 367-370. Lynch, D. L., and Lynch, C. C. 1958. Nurure (London) 181, 1478-1479. Lytle, C. R., and Perdue, E. M. 1981. Environ. Sci. Technol. 15, 224-228. Marshman, N. A., and Marshall, K. C. 1981. Soil Biol. Biochem. 13, 127-134. Martin, J. P., and Haider, K. 1979. Appl. Environ. Microbiol. 38, 283-289. Martin, J. P., Parsa, A. A., and Haider, K. 1978. Soil Biol. Biochem. 10, 483-486. Mason, H. S. 1955. Nature (London) 175, 771-772. Matsubara, H., and Feder, J. 1971. In “The Enzymes” (P. D. Boyer, ed.), Vol. 3, pp. 721-795. Academic Press, New York. Mayaudon, J., and Sarkar, J. M. 1974. Soil Biol. Biochem. 6, 275-285. Mayaudon, J., Batistic, L., and Sarkar, J. M. 1975. Soil Biol. Biochem. 7, 281-286. McLaren, A. D. 1954a. J. Phys. Chem. 58, 129-137. McLaren, A. D. 1954b. Soil Sci. SOC. Am. Proc. 18, 170-174. McLaren, A. D., and Estemann, E. F. 1956. Arch. Biochem. Biophys. 61, 158-173. McLaren, A. D., Reshetko, L., and Huber, W. 1957. Soil Sci, 83,497-502. McLaren, A. D., Peterson, G. H., and Barshad, I. 1958. Soil Sci. Soc.Am. Proc. 22, 239-244. McLaren, A. D., Jensen, W. A., and Jacobson, L. 1960. Plant Physwl. 35, 549-556. Miller, A. 1978. In “Characterization of Protein Conformation and Function” (F. Franks, ed.),pp. 62-69. Symposium Press, London. Mizusawa, K., Ichishima, E., and Yoshida, F. 1966. Agric. Biol. Chem. 30, 35-41. Murakami, M., Fukunaga, K., Hatsuhahi, M., and Ono, M. 1969. Biochim. Biophys. Acra 192, 378-380. Namdeo, K. N., and Dube,J. N. 1973. Soil Biol. Biochem. 5, 855-859. Neuberger, A. 1978. In “Characterization of Protein Conformation and Function” (F. Franks, ed.), pp. 144-156. Symposium Press, London. O’Brien, R. T. 1978. In “Nitrogen in Desert Ecosystems” (N. E. West and J. J. Skujins, eds.), pp. 50-59. Dowden, Hutchinson, & Ross, Stroudsburg, Pennsylvania. Okafor, N. 1966. Plant Soil 25, 211-237. Ong, P. S . , and Gaucher, G. M. 1973. Can. J. Microbiol. 19, 129-133. Parsons, J. W., and Tinsley, J. 1975. In “Soil Components” (J. E. Gieseking, ed.), pp. 263-304, Springer-Verlag, New York. Pmck, L., and Allison, F. 1951. Science (Wushingron, D.C.) 114, 130-131. Pmck, L., Dyal, R., and Allison, F. E. 1954. Soil Sci. 78, 109-118. Piper, T. J., and Posner, A. M. 1968. Soil Sci. 106, 188-192. Pochon, J., and Chalvignac, M. 1952. Ann. Imr. Pusreur (Paris) 82, 690-695. Pochon, J., and Tchan, Y. 1947. Ann. Inst. Pusfeur (Paris) 73,696-700. Priest, F. G. 1977. Bacreriol. Rev. 41, 711-753. Prudlov, B., Ushakova, V. I., and Egorov, N. S. 1973. Microbiology (Engl. Trunsl. of Mikrobiologiyu, Washington, D.C.) 42, 18 1-185.

PROTEIN TRANSFORMATION IN SOIL

38 1

Pugh, G. J. F., and Mathison, G. E. 1962. Trans. Br. Mycol. SOC. 45, 567-572. Putyatina, T. N. 1966. Sel‘khoz. Biol. 1, 770-774. Reid, K. B. M. 1978. In “Characterization of Protein Conformation and Function” (F. Franks, ed.), pp. 70-83. Symposium Press, London. Robert-Gero, M., Hardisson, C., LeBorgne, L., and Vidal, G. 1967. Ann. Insr. Pusteur (Paris)113, 903-909. Rodriguez-Kabana, R., Kelley, W. D., and Curl, E. A. 1978. Can. J. Microbiol. 24, 487-490. Romeiko, I. N. 1969. Pochvovedenie 10, 87-90. Ross, D. J., and Cairns, A. 1978. N.Z.J. Sci. 21, 481-486. Rowell, M. J., Ladd, J. N., and Paul, E. A. 1973. Soil Biol. Biochem. 5, 699-703. Rupley, J. A. 1978. In “Characterization of Protein Conformation and Function” (F. Franks, ed.), pp. 54-61. Symposium Press, London. Ruthenberg, H. 1976. “Farming Systems in the Tropics.” Oxford Univ. Press (Clarendon), Oxford. Sarkar, J. M., Batistic, L., and Mayaudon, J. 1980. Soil Biol. Biochem. 12, 325-328. Schnitzer, M., Kodama, H., and Ivarson, K. C. 1980. Z. Pflanzenernuehr. Bodenkd. 143,334-343. Seidler, R. J., Allen, D. A,, Colwell, R.R., Joseph, S. W., and Daily, 0. P. 1980. Appl. Environ. Microbiol. 40,715-720. Siebert, M. L., and Toerien, D. F. 1969. Water Res. 3, 241-245. Simonart, P., Batistic, L., and Mayaudon, J. 1967. Plant Soil 27, 153-161. Sizemore, R. K., and Stevenson, L. H. 1974. Life Sci. 15, 1425-1432. Sowden, F. J. 1966. Soil Sci. 102, 264-271. Sowden, F. J. 1970. Can. J . Soil Sci. 50, 233-241. Speir, T. W., and Ross, D. J. 1981. Soil Biol. Biochem. 13, 225-229. Stanlake, G. J., and Clark, J. B. 1975. Appl. Environ. Microbiol. 30, 335-336. Stefani, A., and Sequi, P. 1978. Z. Pflanzenernuehr. Bodenkd. 141, 3-10. Stefaniak, 0. 1968. Plant Soil 29, 193-204. Stefaniak, 0. 1972. Symp. Biol. Hung. 11, 121-124. Stevenson, F. J. 1956. Soil Sci. SOC. Am. Proc. 20, 201-204. Stevenson, F. J., and Goh, K. M. 1971. Geochim. Cosmochim. Acra 35, 471-483. St&kli, A. 1946. Schweiz. Z. Forstwes. 97, 356-378. Sugahara, I., Sugiyama, M., and Kawai, A. 1974. In “Effect of the Ocean Environment on Microbial Activities” (R. R. Colwell and R. Y. Morita, eds.), pp. 327-340. Univ. Park Press, Baltimore, Maryland. Swaby, R. J., and Ladd, J. N. 1962. Trans. Inr. Soil Sci. SOC., Commissions IV & V, pp. 197-202. Talibudeen, 0. 1950. Nature (London) 166, 236. Taylor, W. I., and Battersby, A. R. (eds.) 1967. “Oxidative Coupling of Phenols.” Dekker, New York. Theis, E. R. 1945. J . Biol. Chem. 157, 23-33. Tinsley, J., and Zin, M. K. 1954. Congr. Int. Sci. Sol 5th 2, 342-347. Torstensson, L. 1974. Swed. J . Agric. Res. 4, 151-160. Torstensson, L. 1975. Swed. J . Agric. Res. 5, 177-183. Ueda, S., and Earle, R. C. 1972. J. Gen. Appl. Microbiol. 18, 239-248. Vagnerova, K., and Macura, J. 1974a. Folia Microbiol. 19, 322-328. Vagnerova, K., and Macura, J. 1974b. Folia Microbiol. 19, 329-339. Vagnerova, K., and Macura, J. 1974c. Folia Microbiol. 19, 525-535. van Schreven, D. A. 1972. Plant Soil. 36, 561-569. van Schreven, D. A., and Harmsen, G. W. 1968. In “The Ecology of Soil Bacteria” (T. R. G. Gray and D. Parkinson, eds.), pp. 474-499. Univ. of Toronto Press, Toronto. Varanka, M. W., Zablocki, Z. M., and Hinesly, T. D. 1976. J. Water Pollur. Conrrol Fed. 48, 1728- 1740. Venna, L., and Martin, J. P. 1976. Soil Biol. Biochem. 8, 85-90.

382

MICHAEL J. LOLL AND JEAN-MARC BOLLAG

Verma, L., Martin,J. P., and Haider, K . 1975. Soil Sci. Soc. Am. Proc. 39, 279-284. Visser, S. A., and Banage, W. B. 1973. Rev. Ecol. Biol. Sol 10, 55-70. Voets, J. P., Dedeken, M., and Bessems, E. 1%5. Naturwissemchajlen 52, 476. Vogel, R., Trautschold, I., and Werle, E. 1%8. “Natural Proteinase Inhibitors.” Academic Press, New York and London. Wagner, G. H., and Mutatker, U. K. 1%8. Soil Sci. SOC.Am. Proc. 32, 683-686. Wainwright, M., and Pugh, G. J. F. 1973. Soil Biol. Biochem. 5, 577-584. Wainwright, M., and Pugh, G. J. F. 1975. Soil Biol. Biochem. 7 , 1-4. Waksman, S. A., and Iyer, K. R. N. 1932. Soil Sci. 34,43-69. Walsh, K. A. 1975. I n “Roteases and Biological Control” (E. Reich, D. B. Rifkin, and E. Shaw, eds.), pp 1-11. Cold Spring Harbor Lab., Cold Spring Harbor, New York. Warren, G. B. 1978. I n “Characterization of Protein Conformation and Function” (F. Franks, ed.), pp. 94-1 14. Symposium Press, London.

ADVANCES IN AGRONOMY, VOL. 36

APPLICATIONS OF INDUCED AND SPONTANEOUS MUTATION IN RICE BREEDING AND GENETICS J. Neil Rutger U.S. Department of Agriculture, Agricultural Research Service Agronomy and Range Science Department University of California at Davis Davis, California

I. Introduction ................................ ............... ..... U. Breding Applications of Semidwarf Mutants ..... A. In California .... .................................. B. InJapan ......................... ................. C. Other Localities ...............................

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

111. Breeding Applications of Early Maturity Mutants ............ IV. Breeding Applications of Other Types of Mutants ...........................

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

B.

Disease Resistance . .

A.

Genetic Male Steriles.. . . . . . . . . . . Marker Genes.. ...........................

B.

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

383 385 385 392 393 395 396 399 399 400 40 1 402 404 404 405 407 408 410

I. INTRODUCTION In the last decade, induced mutation has assumed its correct role as one of the several tools available to progressive plant breeders. A striking example of this role has been in rice breeding in California, where induced and spontaneous mutants have not only been released directly as improved cultivars but more importantly have been used as donor parents in standard cross-breeding procedures. Since 1976, two induced semidwarf mutants have been released directly 383

Copyright 8 by Academic Press, Inc. All rights of reproduction in any form reserved. ISBN 0-12000736-3

384

J.

NEIL RUTGER

as rice cultivars (Rutger et al., 1977; Camahan et al., 1981b); six more semidwarf cultivars have been developed by crossing with an induced semidwarf mutant or its derivatives (Carnahan et al., 1982; Rutger, 1982a). An induced glutinous endosperm mutant was released directly as an improved cultivar during the same period (Carnahan et al., 1979) and a spontaneous mutation for early maturity was released previously (Camahan et al., 1975). Both of these cultivars have also served as the donor parent of subsequent releases (Carnahan et al., 1981a; Johnson et al., 1980, 1981). Induced semidwarf mutants also are being used widely in breeding programs in Japan (Sato, 1980). The increase in use of mutants in rice breeding worldwide is further documented by Kawai (1982), who listed 45 rice cultivars derived either directly from induced mutation or from crossbreeding with mutants; this may be compared to a similar listing of 30 cultivars by Mikaelsen (1980) just 3 years earlier, and of 13 cultivars by Sigurbjomsson and Micke (1974) some 6 years before that. Until the present time, the most useful types of mutants in rice have been characters that were simply inherited and usually controlled by single recessive genes: semidwarfism, early maturity, waxy endosperm, genetic male sterility, and various phenotypic markers such as hull color. Such mutants have been used in three general modes: for direct release as improved cultivars, for donor gene sources in standard cross-breeding or hybridization programs, and for developing near-isogenic comparisons for testing agronomic and physiologic hypotheses. The second category, that is, as parents in hybridization programs, is becoming the most important practical application of mutants. Thus, of the rice cultivars released since 1975, 16 came from hybridization with an induced mutant donor source and 9 were direct releases of induced mutants, as compared to 4 and 16 from hybridization and direct releases, respectively, of those released before 1975 (Kawai, 1982). Mutants are very useful in situations where only one or two simple changes in well-adapted local cultivars are needed, especially when the local cultivar carries gene complexes adapted to modem agriculture (Micke, 1979). Such complexes may include cold tolerance, grain quality, insect or pest resistance, and tolerance to environmental stress. Under these circumstances it may prove easier and faster to improve the local cultivar by inducing a needed mutation than by hybridizing it with an unadapted donor from world collections, because in the latter case the desired complex of characters may be difficult and more time-consuming to recover (Rutger and Lehman, 1977). As Nilan et al. (1977) have pointed out, induced mutation is best considered as a supplement to natural genetic variability. The challenge to plant breeders is to be alert for situations in which this supplementation can be used efficiently. Some of the more successful applications of induced and spontaneous mutation in rice improvement are reviewed in this article.

INDUCED MUTATIONS AND GENETICS OF RICE

385

II. BREEDING APPLICATIONS OF SEMIDWARF MUTANTS A. IN CALIFORNIA Induced mutants have played a significant role in rice improvement in California, both in direct releases as cultivars and as parents in cross-breeding programs (Fig. 1). The first semidwarf rice cultivar in California, Calrose 76, was released in 1976; it originated as a single gene semidwarf mutant from the very welladapted tall cultivar Calrose (Rutger et al., 1977). Genetic studies showed that Calrose 76 possessed a single recessive gene for semidwarfism, designated sd, (Rutger et al., 1976; Foster and Rutger, 1978). The sd, gene reduces plant height about 25% through approximately proportional reductions in lengths of the top five internodes; panicle length remains essentially unchanged (Rutger ef al., 1976). Almost from the beginning, it was apparent that the semidwarf mutant would be useful in breeding programs (Rutger and Peterson, 1976). In practice Calrose 76 has been more important as an adapted semidwarf donor than as a cultivar per se. It has been the source of semidwarfism either directly or indirectly for five additional cultivars; M7, released in 1977 (Carnahan et al., 1978a); M-101, released in 1979 (Rutger et al., 1979b); S-201 and M-301, both released in 1980 (Carnahan et al., 1980; Johnson ef al., 1980); and M-302, released in 1981 (Johnson et al., 1981). In addition, the semidwarf mutant D51, which was selected from the same irradiated population as Calrose 76 and has the same semidwarfing gene, is the semidwarf donor for another cultivar, Calmochi-202 (Carnahan et al., 1981a) (Table I). Because Calrose 76 and D51 are believed to have arisen from the same mutational event (Rutger et al., 1976), they are commonly referred to as the Calrose 76 sd, source. In 1981, the seven cultivars with the Calrose 76 source of semidwarfism were grown on an estimated 54% of the 245,000 ha of rice in California (Brandon et al., 1981). These cultivars were developed to take advantage of semidwarfism in combination with different maturity classes, grain types, grain quality, and leaf and hull pubescence or glabrousness. Another directly released semidwarf mutant, M-401 (Carnahan et d., 1981b), has been shown to carry the same sd, allele as Calrose 76 (Rutger, 1982b). The accelerated development of semidwarf rice cultivars in California has resulted from the cooperative efforts of three agencies: U.S. Department of Agriculture- Agricultural Research Service and the University of California, both at Davis, California, and the California Co-operative Rice Research Foundation, Inc., located at Biggs, California. The principal activities of the Davisbased agencies are in basic research on breeding methods and germplasm development; the Biggs agency is engaged primarily in cultivar development. The

-

Early Watar ibunr

Ch

1920

1948

6 Calrarr

Smooth No.4

I

C 6 Smoorh

QI

1968 Kokuhorosr

I

1976

1982

FIG.1. Ancestry of publicly developed short- and medium-grain rice cultivars in California. Semidwarf cultivars are in double boxes, tall cultivars in single boxes. Introductions, breeding lines, and proprietary cultivars are underlined. Female parent is on the left, except for three cases in which it is designated by 0 . The cross that produced S-201 was actually between a sister line of S6 and the induced mutant D51, a mutant which has the same semidwarfing gene as Calrose 76.

Table I Semidwarf Public Cultivars and Released Germplasm Lines Derived from Induced Mutation or Cross-Breeding with Induced Mutants in California Cultivar or germplasm line

W

4 00

Cultivars Calrose 76 M7 M-101 M-301 s-201 M-302 Calmochi-202 M-401 Germplasm lines CI 11033 (D66) CI 11034 (D24) CI 11035 (D38) CI 11036 (DDI) CI 11045 (S-6190-96) CI 11046 (S-6190-110) CI 11047 (S-8158-39) CI 11048 (WC 1403-12) CI 11049 (72/16439) CI 11050 (S-6193-3) 'sd? # sd,

Semidwarf allele(s)

Year of release

Origin

1976 1977 1979 1980 1980 1981 1981 1981

y-Ray induced mutant of Calrose Calrose 76/CS-M3 CS-M3/Calrose 76//D31 Calrose 76/CS-M3//M5 Calrose 76/CS-M3//S6 Calrose 76/CS-M3//M5 Colusa/CS-M3//D51/3/Calmochi-201 y-Ray induced mutant of Terso

Rutger et al. (1977) Camahan et al. (1978a) Rutger et al. (1979b) Johnson et al. (1980) Camahan et al. (1980) Johnson et al. (1981) Camahan et al. (1981a) Camahan et a[. (1981b)

1977 1977 1977 1977 1981 1981 1981 1981 1981 1981

y-Ray induced mutant of Calrose y-Ray induced mutant of Calrose y-Ray induced mutant of Colusa Calrose 76/CI 11033 y-Ray induced mutant of M5 y-Ray induced mutant of M5 y-Ray induced mutant of Maxwell y-Ray induced mutant of WC 1403 y-Ray induced mutant of Calrose y-Ray induced mutant of Tsuru Mai

Rutger et al. Rutger ef al. Rutger ef al. Rutger et al. Rutger er al. Rutger et al. Rutger ef al. Rutger er al. Rutger et al. Rutger et al.

Reference

(1979a) (1979a) (1979a) (1979a) (1982a) (1982a) (1982a) (1982a) (1982a) (1982a)

means that this genotype has a semidwarfing gene nonallelic to sd, but that its relationship to other semidwarfing loci is unknown.

388

J. NEIL RUTGER

Davis group investigated a breeding method, induced mutation, and took the lead in developing two semidwarf cultivars, Calrose 76 and M-101, and several germplasm lines (Rutger et al., 1979a, 1982a). The Biggs group has developed numerous additional cultivars, both tall and semidwarf, as well as germplasm lines (Carnahan ef al., 1982). Height of the California semidwarfs is generally about 90 cm, compared to 120-130 cm for the previous tall cultivars. The semidwarfs yield more because of their increased responsiveness to nitrogen fertilizer (Fig. 2). In addition to showing greater lodging resistance at the higher nitrogen levels, the semidwarfs show increased numbers of panicles per square meter (Dat et al., 1978). Because rice yields in California were at a relatively high base level, semidwarf cultivars did not produce the dramatic doubling of yields that frequently occurred in the Green Revolution cultivars in Asia. Nevertheless, the cumulative evidence indicates that the semidwarfing gene increases rice yields 15%, and when the semidwarf cultivars are used with intensified cultural practices, farm yields increase about 25%. In 1981, approximately 95% of California’s 245,000 ha of rice were sown to semidwarfs [54% to cultivars carrying the Calrose 76induced mutant source and about 35% to the semidwarf cultivar M9, which derives its semidwarfing gene from IR8 (Carnahan et al., 1978b)I. The combination of favorable growing conditions, intensive cultural practices, and high-

9 9

8

- 70001

67

101

135

168

202

Nitrogen (kg/ha)

FIG. 2. Response of the semidwarf cultivars Calrose 76 and M7 and the tall cultivar CS-M3to nitrogen fertilization. Adapted from Brandon er al. (1981).

INDUCED MUTATIONS AND GENETICS OF RICE

389

yielding semidwarf cultivars resulted in record California average yield of 8.07 metric tons/ha in 1981. Genetic studies have shown that the induced mutant semidwarfing gene sd, is allelic to the major semidwarfing gene in Deo-Geo-Woo-Gen (DGWG) and the widely grown Green Revolution cultivars derived from DGWG (Foster and Rutger, 1978; Mackill and Rutger, 1979). Because the three widely used indica semidwarf cultivars DGWG, I-geo-tze, and TNl all carry the same recessive semidwarfing gene (IRRI,1966), the simplified designation DGWG will be used for the indica semidwarf gene source in this article. In F, generations of crosses between sd, and the DGWG types, no truly tall recombinants have been recovered, although considerable variation exists in height of the F, semidwarfs (Foster and Rutger, 1978; Mackill and Rutger, 1979). In crosses with tall cultivars, the recessive sd, mutant gene shows rather discrete semidwarf versus tall segregation. On the other hand, the DGWG semidwarfing source shows more continuous variability, which is usually attributed to the presence of a single major gene plus minor gene modifiers (Aquino and Jennings, 1966). Thus, although it is possible to obtain a broad range of semidwarf heights in crosses of tall lines with DGWG, a narrower range of semidwarf heights is recovered in crosses of tall lines with the induced sd, source. The major gene plus modifier nature of the DGWG source also would explain the variation in semidwarf height that is seen in crosses between the induced mutant sd, and DGWG sources. The sd, allele appears to be a recumng mutation, as it has been induced independently in three California cultivars: Calrose, to produce Calrose 76 (Rutger, 1982a); Colusa, to produce a germplasm line (Rutger, 1982a); and Terso, to produce M-401 (Carnahan et al., 1981b). In addition to the 8 semidwarf cultivars which have been developed from use of induced mutants, 10 semidwarf germplasm lines have been released from the California programs (Table 1). Allelism tests have shown that at least three independent, recessively inherited semidwarf genes were induced in the tall cultivar Calrose: the sd, locus present in Calrose 76 and its six crossbred derivative cultivars, the sd, locus in CI 11033, and the sd, locus in CI 11034 (Foster and Rutger, 1978; Mackill and Rutger, 1979). Typically, F,s among the nonallelic semidwarfs are tall, and 9 tall : 6 semidwarf : 1 doubledwarf ratios are observed in the F,. However, neither the sd, nor the sd4 source has been as agronomically useful as the sd, source. The sd, source reduces height only 15 cm and thus is still somewhat lodging susceptible at high fertility levels (sd, reduces height about 30 cm, a more desirable reduction). The sd, source also reduces height only 15 cm and has an additional pleiotropic effect for a 20% reduction in seed size. The inheritance of height in three other induced semidwarfs of Calrose was investigated, but these three (D32, D23, and D25) were not released because genetic studies showed that D32 was allelic to the Calrose 76 sd, source and to DGWG and that D23 and D25 were allelic to the CI 11034

J. NEIL RUTGER

390

CALROSE 76 032 DG WG

FIG. 3. Allelic relationships of induced semidwarf mutants from Calrose and Deo-geo-woo-gen (DGWG). Genotypes at the same comer of the triangle are allelic; those at different corners are nonallelic. Adapted from Mackill and Rutger (1979).

sd, source (Fig. 3). Another induced semidwarf, CI 11035, from the cultivar Colusa, also was found to be allelic to the sd, source. The released doubledwarf carrying sd, sd, (CI 11036) and an unreleased doubledwarf carrying sd, + sd, have been too short (about 65-75 cm) and not as productive agronomically as the sd, semidwarf (Rutger, 1982b). The third doubledwarf, sd, + sd4, was created, but as its height was about the same as the sd, semidwarf(85-95 cm), it was not pursued. After the three independent sernidwarfiig alleles, sd,, sd,, and sd, were identified, subsequent genetic studies concentrated only on determining if new mutants were allelic to sd,. Two semidwarf mutants induced by Carnahan and co-workers from the tall cultivar M5 (CI 11045 and CI 11046) were thus found to be nonallelic to sd, (Table I). Both are about 30 cm shorter than their parent and, except for a tendency to show discolored hulls at harvest, both are phenotypically identical to the sd, source (Rutger et al., 1982a). Although neither of the M5 “raw” semidwarfs has been more productive than its tall parent (and thus by inference each is less productive than the sd, source), neither has been evaluated for yield potential after crossing to other genotypes. Gale el al. (1982) have stressed the importance of influence of background genotypes on newly induced mutants. They showed data on a wheat case in which the plot yield of an EMS-induced semidwarf mutant was increased from 14 to 38% of its tall parent by backcrossing to its parent. This “cleaning up” process apparently helped eliminate other mutated (deleterious) genes in the original line. The cleaning up process probably would not have such dramatic beneficial effects on the M5 semidwarfs, as they yielded 95% of their tall parent (Rutger et al., 1982a), but some increases might result from putting these genes

+

INDUCED MUTATIONS AND GENETICS OF RICE

39 1

in different backgrounds. Should the sd, source ever prove genetically vulnerable, the M5 semidwarfs will certainly receive greater attention. An induced semidwarf mutant, CI 11047, was also selected from the tall, early-maturing cultivar Maxwell (Table I). The height of CI 11047 is similar to the sd, sources, but its inheritance has not been studied. Yield was similar to the tall check cultivar Earlirose, but both the semidwarf mutant and the tall check showed considerable lodging (Rutger et al., 1982a). Carnahan and co-workers also induced a semidwarf mutant, CI 11048 (Table I), in the tall, weak-strawed, rice water weevil-tolerant line, WC-1403. The mutant is about 110 cm tall, which is about 20 cm shorter than its parent (Rutger et al., 1982a), and retains its parental level of water-weevil tolerance (Johnson and Carnahan, 1982). Inheritance of its short stature has not been studied. Narrow-leaf semidwarf mutants have been induced in two different cultivars (Table I). These two mutants, CI 11049 and CI 11050, have been of interest because they produce more than three-fourths of normal semidwarf yields with only half as much leaf blade area (Rutger et al., 1982a; Lafitte, 1982). Thus, these lines apparently trap sunlight energy more efficiently. It is postulated that their reduced yield may result from being too short, as the narrow-leaf semidwarfs are 10-15 cm shorter than sd, lines. These narrow-leaf mutants may be of interest in physiological studies on drought tolerance, because many ecotypes of grasses from semiarid and low-rainfall areas have narrower leaves than those from humid and higher rainfall areas. Similarly, it would be useful to determine whether such mutants are beneficial in reducing respiration losses in rice-growing climates where high night temperatures prevail. One narrow-leaf semidwarf mutant, CI 11049, was shown to be nonallelic to the sd, source; inheritance of height in CI 11050 has not been studied. The semidwarfhg gene in CI 11049 has a pleiotropic effect for short, narrow leaves and reduced seed size. Attempts are being made to obtain narrow-leaf recombinants with 85-95 cm height that might raise yields to a competitive level. Another semidwarf mutant selected in California, “Short Labelle,” represents a semidwarf in a long-grain background. Short Labelle equalled the yield of its tall parent in one test in Arkansas (McKenzie et al., 1982), but when averaged over four tests it was noticeably lower yielding than Labelle. Attempts to “clean up” Short Labelle by hybridizing it with its parent and other adapted long grain cultivars are also being made. Three reasons that induced mutation has been so useful in California are:

1. The objectives were clearly specified and limited to characters under simple genetic control, primarily semidwarfism, early maturity, and waxy endosperm. 2. Once found, the desired characters were quickly incorporated into standard

392

I. NEIL RUTGER

hybridization programs, frequently in a stepwise fashion (Rutger and Peterson, 1976). 3. The rate of incorporation was speeded by the close relationships of the California cultivars. Except for some Japanese and Chinese cultivars, most introduced lines are not well adapted to the long days and cool nights of California’s climate. Thus, it has been easier in breeding programs to use induced mutants than introduced sources, although both have been successful. B. INJAPAN

An outstanding example of a useful semidwarf mutant in Japan has been the cultivar Reimei, which was induced by y irradiation of seeds of the tall cultivar Fujiminori (Futsuhara, 1968). The 15-cm height reduction of Remei was controlled by a single nondominant gene. Reimei has been used extensively in crossbreeding. Sat0 (1980) reported that Reimei has contributed short and stiff culm to 5 cultivars through crossbreeding, that it was a parent in the pedigree of 40 “candidate” cultivars, and that it is in the pedigree of 19 additional candidate strains. Another short culm mutant, Fukei 71, with a 30-cm height reduction, also was induced from Fujiminori (Futsuhara, 1968). Fukei 71 is in the ancestry of 2 additional cultivars in Japan (Sato, 1980). Short-culm mutants also were induced in the popular cultivar Koshihikari (Samoto and Kanai, 1975). Koshihikari mutants have been used to breed 2 cultivars (Sato, 1980). Kawai (1982) listed 2 more cultivars that had been bred in Japan by using induced mutants in crossbreeding, making a total of at least 11 cultivars developed in this fashion. In addition to Reimei, two other mutants have been released directly as improved cultivars in Japan: Miyuki-Mochi, a waxy mutant, and Miyamanishiku, a large-grain mutant (Toda, 1979). Ikehashi and Kikuchi (1982) studied the allelism of Reimei and the DGWG source and suggested that the semidwarfism gene in Reimei is the same as the one from DGWG. They also studied the allelism of the native semidwarf Japanese cultivar Jikkoku and DGWG, after making four backcrosses of the respective gene sources into the cultivar Norin 29 to minimize confusing background effects from the usual wide segregation occurring in indica-japonica hybrids. Jikkoku has been used widely in southwest Japan as a semidwarf donor. From crosses between the two semidwarf sources, they concluded that DGWG and Jikkoku have an identical semidwarf gene. Okuno and Kawai (1977) studied many short-culm mutants in the rice cultivars Norin 8 and Norin 22. Undesirable changes in other agronomic characters were often associated with the short mutants. Induced mutants could be

INDUCED MUTATIONS AND GENETICS OF RICE

393

classified into three types; “upper shortening”, “lower shortening”, and “normal” internodes. Genetic studies of six short-culm mutants showed that culm length was controlled by single recessive genes at different loci, although the possibility of simultaneously induced mutant genes at closely linked loci could not be completely eliminated (Okuno and Kawai, 1977). They noted that the most desirable mutants would be those that showed minimal undesirable changes in other characters. Yamaguchi et al. (1981) described an erect, narrow, thick-leaf dwarf rice mutant induced in the cultivar Hatsunishiki. Narrow leaf was controlled by a single recessive gene. A semidwarf recombinant, G31, derived by crossing the dwarf mutant to another cultivar, also had erect, narrow, thick leaves. G31 had net photosynthetic and transpiration rates per unit leaf area about 1.5 times that of the best check cultivar. In dense broadcast plantings, G31 yielded more rhan did two check cultivars. The yield increase was attributed to less mutual shading of leaves. The leaf characteristics of the mutant described by Yamaguichi et al. (1981) appear to be similar to the two narrow leaf mutants (C1 11049 and CI 11050) found in California (see p. 391, this article; Rutger et al., 1982a). C . OTHERLOCALITIES

Numerous other induced dwarf and semidwarf mutants of rice have been reported in the literature (Hajra et al., 1980). Notable examples among indica cultivars include the mutants Shuang-chiang 30-21, Keh-tze 20-74, and I-kungbau 4-2 in Taiwan (Hu, 1973); a short stature mutant from H4 in Sri Lanka (Gunawardena et al., 1971); and the semidwarf mutant Jagannath in India (Mohanty and Das, 1979). Examples among japonica cultivars include the semidwarf mutant Milyang 10 in Korea, which was reported to be controlled by a single recessive gene (Ree,1973); two semidwarfing mutants from the cultivars Chianung 242 and Tainan 5 in Taiwan, which were reported to be nonallelic to each other (Woo et al., 1974); and six short mutants from the cultivar Tainung 61 in Taiwan (Buu and Huang, 1975). In the first two reports on japonica mutants, no mention was made of allelism tests to DGWG, but in the last report one mutant was found to be allelic to the semidwarf gene in TN1. Until recently, most workers have concentrated on making direct releases of the mutants and thus failed to realize the major benefits of mutants as donor parents in crossbreeding programs. Also, the DGWG semidwarf source was coming into widespread use in breeding programs, and because the gene most frequently induced was the same as in DGWG, many breeders thought that induced mutants would have minimal benefits. However, as the role of induced mutation in supplementing natural genetic variability became better understood

394

I. NEIL RUTGER

(Nilan et al., 1977), several researchers have pursued programs which show promise either for inducing nonallelic semidwarfs or for inducing mutants in specialized cultivars. Thus, Reddy and Padma (1976) induced several semidwarf and dwarf mutants in the tall cultivar Tellakattera in India. Four dwarf and one sernidwarf mutants were nonallelic to each other and to the DGWG semidwarf source (Padma and Reddy, 1977). Jacquot (1978) reported that short-strawed mutants were induced by irradiation of the line 63-83 (IRAT2) from Senegal. The mutants had increased lodging resistance and retained other desirable attributes for upland rice. Three cultivars resulted from irradiation of 63-83: IRAT 13, IRAT 78, and IRAT 79, IRAT 13 has been widely grown in West Africa. It is of medium height, 110 cm, and carries a recessive gene for reduced height, although it is not considered a dwarf (Jacquot, 1978). It has also been used as a parent in breeding programs. Another mutant from 63-83, mutant 312A, has a recessive gene for semidwarfism that is different from the gene in TNl. Saini et al. (1977) induced a semidwarf mutant, PAU Mutant Basmati 370, in the tall cultivar Basmati 370 and reported that it yielded 58% more than its parent in India. Awan et al. (1982) induced several semidwarf mutants in Basmati 370 in Pakistan and found considerable yield advantages for the mutants. The Basmati rice case appears to be a classic situation in which semidwarf mutants would be useful: Basmati rices are characterized by a scent, and fine grains that elongate greatly during cooking; thus they are highly valued by rice consumers in several Middle East countries. Unfortunately, the original Basmati 370 cultivar grows very tall (150-180 cm), and is very susceptible to lodging, hence low yielding. High-yielding semidwarf Basmati cultivars developed by crossing to DGWG or its derivatives apparently have had limited acceptance. Yield has definitely been increased, but some consumers report that the odor and cooking characteristics do not match the original distinctive Basmati type. Thus, it would seem that the Basmati rices would be an ideal situation for inducing semidwarf mutants; this should provide a means of changing a single character (height) without disrupting the distinctive complex of Basmati adaptation and quality characters. However, it was reported in the Pakistan study that even the induced Basmati semidwarf mutants do not have quality characters equivalent to the original parent (Awan et al., 1982). Backcrosses to the tall parent are being made in attempts to “clean up” the quality deficiencies in the semidwarf mutants. Mahadevappa et al. (1981) undertook studies to induce early maturing, shortstature mutants in eight cultivars adapted to tidal swamps in Indonesia, because the introduction of modern semidwarf cultivars had not been successful in areas with tidal swamps and adverse soil conditions. Mahadevappa et al. (1981) used the mutagen ethyleneimine in attempts to improve these native cultivars and the mutants thus induced are being tested in the tidal swamps.

INDUCED MUTATIONS AND GENETICS OF RICE

395

Similarly, researchers at CIAT in Colombia, noting that broad-spectrum resistance to blast had been found only in tall, lodging-susceptible rice lines, decided to induce dwarf mutants in these lines. It was anticipated that the dwarf mutants would be more suitable as recurrent donor parents in crossbreeding (CIAT, 1980). Several mutants were induced and now are under evaluation (CIAT, 1981). Buddenhagen embarked on a similar induced mutation project at IITA in Nigeria in order to obtain shorter versions of local land races that have horizontal resistance to blast and adaptation to local problems. Several semidwarf mutants have been induced and are being used in breeding (I. W. Buddenhagen, personal communication, 1982).

D. RECURRENTMUTATIONFOR SEMIDWARFISM

The cumulative evidence indicates that the major semidwarfing gene at the locus of the DGWG semidwarfing gene is a recurring mutant. First, Hu (1973) noted that DGWG itself may be a spontaneous semidwarf mutant of the tall cultivar Woo-Gen (Hu reported that Deo-Geo-Woo-Gen in Chinese means “short leg” Woo-Gen). Second, Hu (1973) reported that the semidwarf gene of induced mutants of indica cultivars in Taiwan were at the same locus as the DGWG source. Third, Singh et al. (1979) studied 12 representative spontaneous and induced dwarf lines and found that 11 were allelic to the DGWG source, which further indicates that this locus is subject to recurring mutation. Fourth, semidwarfing mutants at the sd, locus (and by direct or indirect tests, at the DGWG semidwarf locus) have been induced independently in three separate cultivars in California: Calrose, Colusa, and Terso (Rutger, 1982b). Fifth, both the dwarfing genes of the induced mutant Reimei and the native semidwarf Jikkoku in Japan seem to be allelic to the DGWG semidwarf gene (Ikehashi and Kikuchi, 1982). Sixth, IRRI (1980) reported that about 70 semidwarf sources have been tested against the sd, gene from DGWG. More than 40 sources were nonallelic to the sd, gene; others were identical or belonged to the same compound locus. Interestingly, in the IRRI study Reimei was reported to have a nonallelic gene for short stature. The importance of the DGWG semidwarf locus in world rice production is well known. Hargrove ef al. (1979) noted that all named cultivars from IRRI (except IR5) and virtually all semidwarfs developed in national rice breeding programs in Asia derive their semidwarfing gene from DGWG. Widespread usage of a single semidwarfing gene increases the potential genetic vulnerability of a crop, although no evidence has yet accumulated of associated detrimental effects of the DGWG source. However, it would be desirable to have backup semidwarfing genes that are nonallelic to the DGWG source, preferably on a

396

J. NEIL, RUTGER

different chromosome or at least at a locus not closely linked to the DGWG locus. The DGWG semidwarfing gene is in linkage group three (Suh and Heu, 1978). In California, none of the induced semidwarf sources nonallelic to sd, (and, by

inference, to the DGWG source) has become important in breeding. Whether this results from undesirable associated or pleiotropic effects of the nonallelic sources or only from lack of usage in breeding programs is not clear at the present time. In Japan, the two nonallelic (to DGWG) cultivars Hokuriku 100 and Kochihibiki were noted for their high yield (Ikehashi and Kikuchi, 1982), thus useful nonallelic semidwarfs apparently can be found. The general usefulness of the sd, locus, as opposed to other semidwarfing loci, sometimes suggests that there is some basis, as yet unknown, for productivity with this particular locus. Following reports by wheat researchers that the semidwarfness of the widely used Norin 10 genes in wheat are invariably associated with nonresponsiveness to gibberellic acid (GA,) (Gale and Gregory, 1977), while the less used induced semidwarf mutants in wheat are responsive to GA,, rice workers have been interested in possible analogies in their crop. The DGWG semidwarf locus in indica backgrounds appears to be GA,-responsive (Harada and Vergara, 1971). The induced mutant sd, in japonica backgrounds appear to be no more GA,-responsive than are tall japonica cultivars, but this may be compounded by the fact that the tall japonicas themselves are relatively GA,responsive (T. R. LaVelle and J. N. Rutger, unpublished). Further studies are needed to clarify whether GA, response in rice semidwarfs has any economic significance.

111. BREEDING APPLICATIONS OF EARLY MATURITY MUTANTS Because early maturing mutants are among the types of mutants most easily identified (Micke, 1979), it is not surprising that many have been found in rice. Good examples are the two widely grown rice cultivars in California which were direct releases of spontaneous mutations for early maturity. The first, Earlirose, was a proprietary cultivar which originated as an early maturing selection from Calrose (Davis, 1965). Earlirose is 7-25 days earlier than Calrose but is similar in height and other characteristics. In the early 1970s, Earlirose was grown on 15-20% of the California rice area. The second cultivar, M5,originated from two early maturing, pure-line mutations from CS-M3 (Carnahan ef al., 1975) (Table II). M5 is 10-12 days earlier than CS-M3 but is similar to its parent in height, yield, and other characteristics. In the late 1970% M5 was grown on as

397

INDUCED MUTATIONS AND GENETICS OF RICE

Table II Early Maturing Public Cultivar and Released Germplasm Lines Derived from Spontaneous or Induced Mutation in California Cultivar or germplasm line Cultivar M5 Germplasm lines CI 11037 (D18) CI 11038 (D31) CI 11051 (ED7) CI 11052 (S-6190-57) CI 11053 (S-8157-82) CI 11054 (S-6189-21)

Year of release

Origin

Reference

1975

Spontaneous mutant from CS-M3

Camahan et al. (1975)

1977 1977 1981 1981 1981 1981

y-Ray induced mutant of Calrose y-Ray induced mutant of Calrose Spontaneous mutant of Calrose 76 y-Ray induced mutant of M5 y-Ray induced mutant of S6 y-Ray induced mutant of Terso

Rutger et al. (1979a) Rutger et al. (1979a) Rutger et a[. (1982a) Rutger et al. (1982a) Rutger et al. (1982a) Rutger et al. (1982a)

much as 30-40% of the California rice area. The early maturity characteristic of M5 has also been recombined with the induced mutant semidwarfhg gene sd, to breed two semidwarf cultivars with early maturity, M-301 and M-302 (Johnson et al., 1980, 1981) (Fig. 1). An induced mutant for very early maturity, germplasm line CI 11038 (D31 in Fig. l), was the source of the very early maturity of the semidwarf cultivar M-101 (Rutger et al., 1979a,b) (Fig. 1). CI 11038 and M-101 are about 20 days earlier than the original Calrose parent, so that M-101 is especially well suited for production in short season areas of California, for late planting in doublecropping situations following barley or wheat, or for delayed plantings resulting from heavy winter rains as occurred in the 1982 season. Several other induced mutants for early maturity have been released as germplasm lines in California. These lines, which otherwise possess the grain cooking quality and cold tolerance of their parent cultivars, are expected to be useful germplasm sources for breeding additional cultivars. The early maturing mutant CI 11037 came from the same 25 kR-treated seed lot that produced Calrose 76 and CI 11038 (Rutger et al., 1979a) (Table 11). CI 11037 is about 15 days earlier than Calrose but its height and other characteristics are similar. CI 11037 carries a single, weakly dominant gene for early maturity. Its early maturity was independent of the semidwarf gene in Calrose 76 and the glabrous hull gene in CS-M3 (McKenzie et al., 1978). A spontaneous mutant for early maturity, CI 11051 (ED7), was also released as a germplasm line (Table 11). CI 11051 was found in a seed-increase field of the semidwarf cultivar Calrose 76; it is 15-20 days earlier than its parent and similarly carries the semidwarfing gene sd,. Except for its earlier maturity, the

398

I. NEIL RUTGER

phenotype of CI 11051 is nearly identical to that of Calrose 76. Studies on the inheritance of the early maturity of CI 11051 suggest that its early maturity was the result of a single nondominant gene (Rutger et al., 1982a). CI 11051 would have been considered for direct release as a cultivar had its performance and adaptation not been identical to those of the cultivar M-101. Furthermore, M-101 has the additional merit of having glabrous hulls, which are generally preferred by U.S. rice growers. Three other early maturing mutants were selected following ~ C irradiation O of three medium-to-late maturing cultivars (Table II). CI 11052 is about 1 week earlier than M5 but is otherwise similar to its parents; CI 11053 also is about 1 week earlier than S6 and otherwise similar to its parent; CI 11054 is about 3 weeks earlier than Terso and is otherwise similar to its parent. Inheritance of early maturity of CI 11052, CI 11053, and CI 11054 has not been studied (Rutger et al., 1982a). Other notable examples of early maturity mutants that have been directly released as cultivars are Nucleoryza in Hungary (Mikaelsen et al., 1971) and Kashmir Basmati in Pakistan (Awan and Cheema, 1976). Nucleorzya is 3 weeks earlier than its parent and is thus better adapted to short-season climates. Similarly, Kashmir Basmati is about 20 days earlier and can be grown at higher elevations than its parent. Two mutants released in Bangladesh, IRATOM 24 and IRATOM 38, required 23 and 36 days less, respectively, from seeding to maturity than did the original cultivar IR8 (Miah et al., 1981). Four additional mutants were found to be 23-41 days earlier than IR8. One, Mut 1-2, yielded 2884 kg/ha compared to 1825 kg/ha for the best check cultivar in the experiment, BR-3. In the People’s Republic of China, an early maturing rice cultivar reported to be derived from radiation breeding, Yuanfeng Early, is 40 days earlier than its parent cultivar, Kazi No. 6. Yuanfeng Early is grown on more than 600,OOO hectares in the lower Yangtze River region ( M A , 1982), which undoubtedly signifies that it is the world’s most widely grown mutant cultivar. Notable efforts at inducing large numbers of early maturing mutants in rice have been reprted by Kawai and Sat0 (1969) and Ismachin and Mikaelsen (1976). Kawai and Sat0 (1969) induced 59 significantly earlier heading mutants in the cultivar Norin 8. These mutants ranged from 1.3 to 18.4 days earlier than their parents, and grain yields ranged from 53 to 104% of the parent. Changes in other characters occurred in most mutants. The frequency of mutants that were at least 3 days earlier and yielded 95% of the parent was less than one per lo00 M, strains. Ismachin and Michaelsen (1976) induced a large number of mutants which matured in 110-120 days as compared to 140-150 days for the parent cultivar Pelita I. In yield trials, 11 mutants equaled or exceeded the parent. In addition to consideration for direct release, the mutants were put into a crossbreeding program.

INDUCED MUTATIONS AND GENETICS OF RICE

399

VI. BREEDING APPLICATIONS OF OTHER TYPES OF MUTANTS A. WAXYENDOSPERM An induced mutant for waxy (wx) endosperm was directly released as the cultivar Calmochi-201 in California (Camahan et af., 1979). Calmochi-201 was one of three waxy mutants found among 2000 X, generation panicles following W o irradiation of seeds of the widely grown nonwaxy tall cultivar S6. Calmochi-201 closely resembles its parent, but Calmochi-201 has waxy endosperm, about 11% reduction in kernel weight, and 6% reduction in yield (Carnahan et af., 1979). Almost immediately, Calmochi-201 was succeeded by Calmochi-202, a semidwarf waxy recombinant from a cross between a line carrying the sd, gene and Calmochi-201 (Fig. 1). At the time of its release, Calmochi-202 yielded 17% more than either Calmochi-201 or S6. Both Calmochi-201 and Calmochi-202 are considered unsatisfactory for making mochi cakes but are acceptable for other glutinous rice markets (Carnahan et af., 1981a). An induced mutant for waxy endosperm also was released directly as the cultivar Miyuki-Mochi in Japan (Toda, 1979). Miyuki-Mochi originated from one of two panicles having waxy grains in the MI generation, following y irradiation of seeds of the nonwaxy cultivar Toyonishiki. Miyuki-Mochi closely resembles its parent, but Miyulu-Mochi has waxy endosperm, about 9% reduction in kernel weight, and 8% reduction in yield. When compared to the standard local waxy cultivar Shinano-Mochi No. 3, the mutant cultivar had a yield advantage of 15%. Toda (1979) also induced waxy mutants in seven additional nonwaxy cultivars to study the action of the waxy gene. Kernel weights were reduced 3-16%, and grain yields ranged from 2% increase to 23% decrease. Khambanonda et al. (1982) reported that three cultivars have been produced by direct release of induced mutants in Thailand, of which two were induced mutants for waxy endosperm. RD6 arose from irradiation of Khao Dawk Mali 105, and RDlO arose from fast neutron treatment of the semidwarfcultivar RDl. A major significance of induced mutation for waxy endosperm is that this technique permits the almost instantaneous development of high-yielding waxy cultivars. In the United States, for example, waxy rice represents a very small fraction (perhaps 1%) of the rice market. Consequently, it was not economically feasible to spend much effort on breeding improved waxy cultivars, and most existing waxy lines were unadapted or otherwise low in yield. In Japan, Toda (1979) also noted that breeding progress of waxy rice in Japan is slow because of limited interest by breeders. Instead of spending years putting the waxy gene into high-yielding backgrounds through hybridization, the work by Carnahan et af.

400

J. NEIL RUTGER

(1979) and Toda (1979) demonstrated that it is possible to take a good local cultivar and convert it to waxy endosperm by irradiation. In California, this approach followed by hybridization with a semidwarf source led to the rapid development of a very high-yielding semidwarf waxy cultivar, Calmochi-202. B . DISEASE RESISTANCE

Induction of disease resistance would be a desirable goal for many breeding programs, especially where cultivars otherwise satisfactory are being grown. Induced mutants for disease resistance in rice in several countries have been reported for blast (Pyricularia oryzae) and bacterial leaf blight (Xanthomonas oryzae) (Mikaelson, 1980). Many of these reports have been presented in IAEA publications (1974, 1977) and are also reviewed by Hajra et al. (1980). However, there seem to be no examples in rice of the widespread use of induced mutants for disease resistance. Some of the problems associated with use of induced mutants for disease resistance are evident from the very thorough series of investigations conducted by Kawai (1974) and Tanaka et al. (1978) on induction of blast-resistant mutants. Frequency of mutants was very low; most mutants did not show greatly improved resistance and had negative changes in other agronomic traits (Kawai, 1974). Few workers seem to have taken precautions to eliminate pollen contamination in the mutated populations. Because spontaneous or “field” hybrids resulting from fertilization of partially sterile M,plants by foreign pollen have been noted by several workers (Caldecott et al., 1959; Konzak, 1959; Simons et al., 1962) as the probable source of disease-resistant plants in irradiated populations of oats, similar outcrossing probably occurs in mutated populations of rice. Kawai (1974) and Tanaka et al. (1978) prevented outcrossing in the M, generation by bagging panicles but did not protect plants in subsequent generations. Outcrossing apparently occurred in the later generations, as resistant plants were found with disease reactions similar to a blast-differential check cultivar growing in the same nursery. The variants also showed morphological characters different from the original parent cultivar. Unless the possibility of foreign pollen is eliminated, a researcher can never be completely certain that resistance arose from induced mutation. Although from a breeding standpoint it might be argued that the source of resistance, whether from natural crossing or induced mutation, does not matter, it is important to know whether or not mutations actually are being induced. If not, then the considerable efforts in attempting to induce mutants can be avoided. Another problem is that recessive mutants occur at higher frequency than dominant mutants. Because most disease-resistant genes in plants are dominant,

INDUCED MUTATIONS AND GENETICS OF RICE

40 1

the probability of obtaining useful mutants is further reduced. Tanaka et al. (1978), in studies which were carefully controlled to eliminate obvious field hybrids observed recessive blast-resistant mutation frequencies of 5 per 4575 and 4 per 5851 M, or M, lines in y ray and chemical mutagen treatments, respectively. Dominant mutation frequency was even lower, only 3 per 60,101 M, plants. Finally, Tanaka et al. (1978) also studied the frequency of virulent mutations in the blast fungus and found that mutants for increased pathogenicity occurred at a frequency of 0.5%, approximately five times as frequently as resistant recessive mutants occurred in the host. Therefore, they concluded that induced mutants for resistance in the host would soon succumb to blast mutants with increased pathogenicity. C. GRAINSHAPE Grain-shape mutants, which have been reported by several authors, generally comprise two types: more slender than the parent, or shorter than the parent. Among indica cultivars the emphasis is usually on obtaining slender grains, although Reddy and Reddy (1973) obtained both types in populations of the cultivar IR8 which had been treated with chemical mutagens. The slender-grain types had less white belly than the parent, a generally more desirable appearance, and increased whole-grain recovery after milling. The short-grain types resembled the seeds of japonica cultivars. Bhivare and Das (1980) also obtained a fine-grain (slender) mutant of IR8; Borah and Goswami (1981) induced finegrain mutants of the cultivar Pusa-33; Kaul and Kumar (1981) induced fine-grain mutants in the cultivars Jhona 349 and IR8, in addition to high-yielding mutants of Basmati 370; Reddy (1979) induced short-grain mutants in the long-grain cultivar Sona and proposed that the short-grain types would have higher wholegrain milling yields; and Hajra and Halder (1980) induced a small-seeded, highyielding mutant from the cultivar Pankaj. Grain-shape mutants in japonica cultivars have been reported by Kawai (1968) and Hu and Alionte (1978). Kawai (1968) conducted genetic studies on five short-grain mutants of the cultivar Norin 8. Each of the five mutants was controlled by a single gene, involving at least four different loci. Short grain length was partially recessive to normal grain length. The mutants showed pleiotropic effects for other agronomic traits, including compact erect panicles. Hu and Alionte (1978) induced both short- and long-grain variants in California mediumgrain cultivars. The short-grain types were more frequent than long-grain types. Implicit in the induced mutant studies on grain shape is the assumption that grain quality and acceptability depends primarily on shape and appearance. Although shape and physical appearance of grain are definitely first-order factors

402

J. NEIL RUTGER

of quality in rice, chemical composition, especially amylose content, and cooking characteristics are also very important. Because it is unlikely that changing the shape of a grain by induced mutation will also change chemical Composition, it is important to remember that grain shape mutants may not be acceptable if the problems with the original cultivar also involved chemical composition. H. L. Carnahan (personal communication, 1982)has purused a useful application of induced mutation for short grain shape in California medium-grain cultivars. The cultivar Calrose, developed by backcrossing medium-grain shape into the short-grain Caloro background (Fig. l),is more resistant to cold-induced sterility than Caloro. Despite much breeding effort, the cold tolerance of Calrose, Calrose 76,or M 7 (all medium-grain cultivars) has not been recovered in short-grain derivatives. Therefore, Carnahan reasoned that genes for cold tolerance are closely linked with the gene(s) for grain shape. With this in mind, he irradiated the medium-grain cultivar M 7 with the objective of obtaining a shortgrain type in the cold-tolerant M 7 background. To quote H. L. Carnahan (personal communication, 1982), “Our derived short grain mutant seems to have fulfilled this objective. A . . . seed increase was grown this year in anticipation of possibly releasing it . . . in 1984.” In this case, there should be no problem with chemical composition; the California short- and medium-grain cultivars have essentially the same chemical composition and cooking characteristics.

In the last 15 years, there has been considerable interest in breeding for increased protein content in rice, both through induced mutation and through conventional techniques. However, no high-protein cultivars have been released. Several optimistic reports on raising the protein content of rice through induced mutation have been summarized in IAEA publications ( M A , 1973, 1978, 1979). The protein content of many mutants was higher than the parent cultivars, but in most cases the increased protein content was associated with malformed or sterile plants or with other factors which resulted in decreased grain yield. Probably the most thorough series of studies on increasing protein content through induced mutation was conducted by Tanaka and Hiraiwa (1978), who began selection for high-protein mutants in the cultivar Nihonbare in 1968. From a population of 6600 M, plants, they selected 201 high-protein variants without regard to other changes in the selected plants. Because most M, lines were lower yielding than the parent cultivar, the population was reduced to 1 1 high-protein M, limes (families) that were similar phenotypically and in yield to the parent. These were reduced to 7 nonsegregating families in the M, generation, which were then yield tested for 3 years. Two limes were identified that had yields similar to the parent but with higher protein productivity. One line pro-

INDUCED MUTATIONS AND GENETICS OF RICE

403

duced 11% and the other 14% more protein per hectare than the parent. The increased protein productivity resulted from increases in both grain yield and protein content; protein contents of the two mutants were 8.20 and 8.26%, compared to 7.77% fwthe parent. One mutant when studied in detail was found to absorb more nitrogen from the soil than the original cultivar (Tanka, 1978). However, further use of these high-protein mutants apparently has not been reported. Schaeffer and Sharpe (1982) used tissue culture techniques to select for increased lysine and protein in rice. Anther-derived tissues were screened for resistance to a lysine analog, S-(2-aminoethyl)-~-cysteine.Plants were regenerated from resistant tissues and carried through five generations of selfing. Reduced seed fertility was evident in the early generations, thus selection for both high protein and high seed set was practiced in the later generations. After the fifth selfing generation, there was a 10% increase in protein and a 3-4% increase in per cent lysine over the control. The grain of the 8 plants of the fifth generation line, MA-99, averaged 9.37% protein and 3.93% lysine (as percentage of total amino acids) compared to 8.53% protein and 3.79% lysine for the 13 control plants (see Schaefer and Sharp, 1982; Table 4). Problems with increasing protein through breeding have not been limited to studies with induced mutation or tissue culture screening. Large environmental and genotype X environment effects on protein were noted by HilleRisLambers et al. (1973), who attempted to select for increased protein by conventional crossbreeding. Low heritabilities and negative correlations between yield and protein were common (Rutger and Qualset, 1976). Since then, breeding efforts for increased protein content of rice in California have been dropped (Carnahan et al., 1982). The most extensive studies on breeding for increased protein have been conducted at IRRI, as summarized by Coffman and Juliano (1982). The best example of progress in breeding for increased protein was the line IR2153-338-3 which, in 1976, after six seasons of testing, was shown to yield the same as IR8 and to have one percentage point higher protein. However, by that time the best yielding cultivar was IR26 rather than IR8. IR2153-338-3 was still higher in protein than IR26, but it was not comparable in yield. In 1977, IR2153-338-3 was devastated by a new disease, ragged stunt virus, which greatly reduced its usefulness. Coffman and Juliano (1982) noted that this is illustrative of a major problem in developing high-protein rice cultivars. Extended testing is needed to confirm that a line is truly high protein, and in the meantime the line may have either been surpassed by a new standard cultivar or become susceptible to new pests or diseases. Given the quantitative nature of protein inheritance, and the problems of separating out environmental effects, it seems improbable that induced mutation will be useful in developing high-protein, high-yielding rice cultivars. Nor does

404

J. NEIL RUTGER

induced mutation appear suitable for increasing lysine content. The high lysine mutants in maize, sorghum, and barley resulted from large reductions in the lysine-poor prolamine fraction of protein; rice already has low prolamine fractions when comparable to the high-lysine mutants in other cereals (Rutger and Qualset, 1976).

V. GENETIC APPLICATIONS OF MUTANTS

A. GENETICMALESTERILES A current application of induced mutation is the induction of genetic male sterility for use in facilitating crossing in population improvement schemes. Simply inherited genetic male steriles of spontaneous origin have been reported in rice for over 50 years (the literature to 1976 was reviewed by Trees and Rutger, 1978), but most of the early steriles showed little outcrossing. Trees and Rutger (1978), for example, observed four single-gene sterile sources of spontaneous origin, but outcrossing on most sterile plants was zero, with occasional plants having as much as 5% seed set. However, in the last few years scientists in Japan, the Philippines, and the United States have induced genetic male-sterile mutants that show moderate levels of seed set from open pollination. Fujimaki er al. (1977) induced genetic male sterile mutants in the japonica rice cultivar Nihonmasari, which showed 1.7-8.3% seed set under open pollination. Hiraiwa and Tanaka (1980) induced additional male sterile mutants in the cultivar Nihonmasari, some of which were at different loci from those induced by Fujimaki et al. (1977). Seed set under open pollination of the mutants induced by Hiraiwa and Tanaka (1980) ranged from 1 to 12%. KO and Yamagata (1980), using three different japonica cultivars, induced genetic male steriles at six different loci. Under open pollination the induced male-sterile mutants in the M, generation showed 2.5-13.5% seed set, but when surrounded by normal plants seed set was 10-30%. Singh and Ikehashi (1981) induced monogenic male steriles in the indica cultivar IR36. In F2populations resulting from crossing four male steriles back to the IR36 parent, open-pollinated seed set on sterile plants ranged from 14.8 to 33.6%. Rutger et al. (1982b) induced 11 recessive genetic male-sterile lines in the semidwarf japonica cultivar M-101 by y irradiation. Seed set on the male steriles from open pollination by sib plants was 1-40% in 1980. In 1981, open-pollination seed set was lower, 0-29%, but evidence of considerable crossing was confirmed by the use of pollinators with a dominant marker gene. These 11 lines represent four different pollen types; stainable pollen abortion type, partial pollen abortion type, complete pollen abortion type, and no pollen type; however, their allelic relationships have not yet been determined.

INDUCED MUTATIONS AND GENETICS OF RICE

405

B. MARKERGENES

All of the new mutant genes-for semidwarfism, early maturity, waxy endosperm, disease resistance, male sterility, etc.-are useful additions to the array of marker genes in rice. These genes can be used to define linkage groups more fully in this important crop. The additional induced mutant genes that may not have immediate breeding or genetic applications are also useful as marker genes. These latter genes include such as the three hull color mutants that have been found in California in recent years. One is a “yellow panicle” mutant found by the author after 6oCo irradiation of the semidwarf cultivar M-101, which is characterized by light yellow-green leaves that are especially noticeable during the seedling stage, by light yellowgreen panicles persisting until about midway between heading and maturity, and by maturity approximately 7 days earlier than the parent cultivar. The yellow panicle mutant appears to be controlled as a single recessive gene with a pleiotropic effect for early maturity (Azzini, 1983). Yield potential has not been measured, but the mutant appears generally productive. “Pale green hull” mutants of the cultivars M-101 and M-201 were found independently by the author and C. W. Johnson, respectively, in 1980. The mutants are characterized by pale green hulls that are noticeable from heading until about midway between heading and maturity. The pale green color apparently results from less chlorophyll in the glumes. The mutant found by the author appears to be recessive, because F, plants from a cross between it and the normal color germplasm line CI 11051 had normal color glumes. Yield potential has not been measured but the mutant appears to be as productive as its parent, M-101. “Gold hull” mutants of the cultivar M-101 also were found independently by the author and by H:L. Carnahan in 1980. The mutants are characterized by golden glume colors at maturity. These are assumed to be the same as the recessive gold hull gene (gh) that has been used widely in breeding programs in the southem United States. Although it might be anticipated that the gold hull M-101 lines arose from outcrossing with southern cultivars in the breeding nurseries, the otherwise normal M-101-like nature of the gold hull lines indicates that they arose from mutations. Yield of the gold hull mutants has not been measured, but they also appear to be as productive as the M-101 parent. Long-culm or tall mutants also may be included in the lists of useful genetic markers. Okuno and Kawai (1977, 1978a) noted that long-culm mutants are much rarer than short-culm mutants, but they found eight induced long-culm mutants in the cultivar Norin 8. In subsequent genetic studies on five of the longculm mutants, Okuno and Kawai (1978b) found that long culm in one mutant, LM- 1, was controlled by a single recessive gene; however, in another mutant, LM-3, long culm was controlled by a single dominant gene. Single gene control was not indicated in the other three mutants.

406

J. NEIL RUTGER

Hajra et al. (1982) recently reported a recessively inherited, extremely tall mutant in the tall cultivar Latisail. Earlier, the same group of researchers had reported a dominant dwarf in the cultivar IR8 (Mallick et al., 1980). In addition to uses as marker genes, these recessive tall or dominant dwarf mutants may have applications in producing hybrid rice seed; Rutger and Camahan (1981) reported the discovery of a putative natural mutation for recessive tall height. They proposed the use of this recessively inherited tall plant type as a genetic tool for facilitating hybrid seed production. The tall plant should have a pollen distribution advantage over a semidwarf parent when interplanted with semidwarf maternal plants of a commercial hybrid. Following pollination, the tall males can be removed from the seed field. Because tall height is recessively inherited, the resulting hybrids are short as in the female parent, which is desirable in situations in which short plants are known to be more productive than tall plants. The recessive tall reported by Rutger and Carnahan (1981) was designated eui for the elongated uppermost internode that is responsible for the increase in height. The eui character originated in an F3 population of a cross between a semidwarf and a tall cultivar, presumably by spontaneous mutation. Other marker genes include an endosperm mutant induced in California, which resembles waxy mutants phenotypically. However, thus mutant, termed “opaque,” apparently has near-normal amylose content (10-15%) compared to the <0.1% amylose of Calmochi-202. The opaque mutant is inherited as a simple recessive character (Azzini, 1983) and may be the same as the “crumbly” mutant in rice. Seeds on recessive homozygotes show a xenia effect when pollinated by normal plants, making the opaque endosperm mutant useful as a marker gene in outcrossing studies. The possible allelic nature of the opaque gene and the waxy gene has not been clarified. Satoh and Omura (1981) induced several different kinds of endosperm mutants in rice in order to supply materials for improving seed quality. As a rationale for their work, they noted that several kinds of endosperm mutants have been found and used in genetic studies in maize but that almost the only known endosperm mutation in rice was the waxy character. They found endosperm mutants which they labelled as white core, floury,waxy, dull, wrinkled, or immature grain. They also found a giant embryo mutant. Each of the induced mutants, except for one of the flourymutants, was controlled by a single recessive gene. They found two other endosperm mutants, labeled sugary and shrunken, for which no genetic information was given. Biochemical and other studies are continuing to determine if there are uses for these mutants analogous to their counterparts in maize (Satoh and Omura, 1981). Regardless of the possible breeding uses of these mutants, they are valuable additions to the stocks of marker genes. Other morphological mutants, such as chlorophyll deficiencies, round grain shape, extreme dwarfs, etc., are observed frequently in irradiated populations. However, as most mutants seem to be repetitions of known marker genes, they

407

INDUCED MUTATIONS AND GENETICS OF RICE

are not expected to be of much value unless they are in a specific background that is needed for genetic studies.

c. As RESEARCHTOOLSIN AGRONOMIC AND PHYSIOLOGICAL STUDIES

A recent IAEA symposium on “Induced Mutations-A Tool in Plant Breeding” highlighted the uses of mutants for developing better understanding of plant growth and functions. In this symposium, Vose (1981) observed that induced mutants are useful in studies of physiology because the mutants may differ in only one major physiological character. He also noted that mutants can provide previously unknown variation for physiological characters, such as the nitrate reductase-deficient mutants reported in the same symposium (Feenstra ef al., 1981; Nilan et al., 1981). In rice, Rutger and Peterson (1981) described the use of mutants in a three-phase program to develop higher yielding plant models (Fig. 4). In Phase 1, induced mutant genes were used with existing genes to develop near-isogenic comparisons quickly. In Phase 2, the near-isogenics were used to test agronomic and physiologic hypotheses about the bases for increased yield. In

””’

Develop near-lsogenlcs for testlng agronomic and physlol oglc hypotheses

Design higher ylel ding plants ; release gemplasm

z

APPLICATIONS

1

AGRONOMY

and

PHYSIOLOGY Use near-lsogenlcs to &tennine yieldlinltlng factors

FIG. 4. Three-phase program to develop higher yielding models of the rice plant. From Rutger and Peterson (1981).

408

J.

NEIL RUTGER

Phase 3, the information developed from Phase 2 was used to design more productive future plants. A critical assumption was that induced mutants for single gene characters such as height and early maturity were isogenic to the parent cultivar. Because there is considerable evidence that mutants often have associated or pleiotropic effects on other characters (Okuno and Kawai, 1977; Gale et al., 1982), it is usually considered desirable to backcross the “raw” mutant to its parent so that it may be “cleaned up” and thus more truly isogenic to its parent. In the California work, there have been minimal side effects associated with the useful mutants, perhaps because emphasis was placed on selecting plants normal for attributes other than the mutant characteristics. Thus the need for clean up of mutants has not seemed critical in most cases investigated. The mutants and various recombinants involving mutants enabled Rutger and Peterson (1981) to demonstrate that critical elements of high yield included semidwarfism, seedling vigor, resistance to cold-induced sterility, and early maturity. In addition, they developed evidence that sink size is a limiting factor for yield. They proposed that future work involve use of mutants and their recombinants to manipulate yield components in source-sink and partitioning studies. Examples of mutants that may be useful in such studies include the narrow-leaf types found in both California and Japan (Rutger et al., 1982a; Yamaguchi et al., 1981), which have higher net photosynthesis per unit leaf area. Satoh and Omura (198 1) also noted that their various endosperm mutants may be useful in advancing knowledge of plant development, as each mutant affected certain grain components. Induced mutants in both the rice host and in the blast pathogen also have been helpful in understanding host-pathogen interactions and in demonstrating the reasons that host resistance soon breaks down (Tanaka et al., 1978). VI. FUTURE USES OF MUTATION IN RICE IMPROVEMENT The best examples of useful mutants in rice have occurred in situations where only one or two simply inherited changes, such as short stature, early maturity, waxy endosperm, short grain shape, etc., have been needed in locally adapted cultivars. Most of these mutants have been recessively inherited. It follows that emphasis in induced mutation for future work should be placed in similar situations. It is further evident that the benefits of mutants in breeding will be optimized when the mutants are utilized in standard crossbreeding programs. Other guidelines include using mutants to supplement natural genetic variability. Thus, if a desired gene is known in world collections, the breeder will wish to

INDUCED MUTATIONS AND GENETICS OF RICE

409

consider the relative merits of induced mutation versus crossing. Induced mutation may be a means of quickly obtaining a needed gene in an adapted background, but the additional genetic diversity introduced during hybridization with world collection sources may give greater long-term returns. Induced mutation research relies heavily on availability or development of effective screening methods. Most cases to date have been based on visual screening techniques for semidwarfkm, early maturity, etc. Much current work on tissue culture is directed at using physiological or biochemical screening methods for identifying spontaneous or induced mutants. Tissue culture offers the potential for screening extremely large populations, but this is counterbalanced by the problems of regenerating plants and determining if desired characteristics expressed in culture also are expressed at the whole plant level. It is important to remember that some of the physiological screening techniques currently in use in tissue culture, for herbicide tolerance, salt tolerance, etc., can also be used on whole-plant populations. Tseng and Seaman (1982), for example, were able to select for increased tolerance in rice breeding lines to the herbicide thiolcarbamate by a seedling screening test. Similarly, Bright er al. (1982) selected barley mutants with an altered aspartate kinase enzyme by screening embryos for growth on a medium containing lysine plus threonine. Although whole-plant screening may require more space than tissue culture, the benefits of having a plant in hand at the end of the experiment can outweigh the disadvantages. A particularly powerful tool for screening for mutants would result if genetic systems were available for producing easily identified haploid seeds. Recessive mutants could then be identified in the M, generation. Genetic techniques for haploid seed production are available in maize (Sarkar, 1974), but so far similar techniques have not been reported for rice. Some specific examples of needed diversity in rice that might be obtained through induced mutation include genes for naked or free-threshing grain, that is, easy separation of the lemma and palea from the caryopsis as in wheat; longer floret opening time; cytoplasmic male sterility; and herbicide resistance. According to Vavilov’s proposals (1951) on homologous series and parallel variation, the naked gene should occur in rice, because it is present in the other cereals. However, free-threshing, normal kernel shape types have not been found in rice, probably because the hulls provide a strong selective advantage for survival of the kernel in aquatic environments. With the advent of peroxide seed coatings, the needs for a protective hull are diminished, and a naked rice kernel has a better survival chance. Longer floret opening time would be useful in hybrid rice seed production, as more time would be available for pollen transfer; rice florets remain open only for an hour or less, whereas wheat florets remain open for several hours. A recent report on another member of the grass family, pearl millet, indicates that the induction of cytoplasmic male sterility is possible (Bur-

410

J. NEIL RUTGER

ton and Hanna, 1982), and it would seem worthwhile to make similar attempts in rice. Cytoplasmic male sterility is available in several background genotypes in rice, but it would be helpful to have more sources. Finally, resistance in crop plants to grass-killing herbicides would be most useful in rice culture as an aid to grassy weed control. Sufficient examples of herbicide resistance have been reported in other crops that additional’searchesin rice seem worthwhile. ACKNOWLEDGMENTS Appreciation is expressed to Dr. H. L. Carnahan of the California Co-operative Rice Research Foundation, Inc., for his reviews and suggestions for improving the manuscript.

REFERENCES Aquino, R. C., and Jennings, P. R. 1966. Crop Sci. 6, 551-554. Awan, M.A., and Cheema, A. A. 1976. Mu#ut. Breed. Newsl. 7,4-5. Awan, M. A., Cheema, A. A., and Tahir, G. R. 1982. Paper presented at the 2nd Res. Coordination Meet. FAOIIAEA Pmg. Evaluation of Semi-Dwarf Cereal Mutants for Cross Breeding, Davis, California, 30 August-3 September 1982. &hi, L. E. 1983. Unpublished data. Bhivare, L. N., and Das, P. K . 1980. J . Nucl. Agric. Biol. 9, 106-107. Borah, S. P., and Goswami, B. C. 1981. J . Nucl. Agric. Biol. 10,6-8. Brandon, D. M., Carnahan, H. L., Rutger, J. N., Tseng, S. T., Johnson, C. W., Williams, J. F., Wick, C. M., Canevari, W. M., Scardaci, S. C., and Hill, J. E. 1981. “California Rice Varieties: Description, Performance and Management.” Univ. o;California, Davis (SpecF’ubl. No. 3271, Div. Agric. Sci.). Bright, S. W. J., Miflin, B. J., and Ropes, S . E, 1981. Biochem. Gener. 20, 229-243. Burton, G. W., and W. W . Hanna. 1981. Crop Sci. 22, 651-652. Buu, J. H., and Huang, C. S. 1975. Chung-huaNung Hsuch Hui Po0 (J. Agric. Assoc. China) 92, 24-30. Caldecon, R. S., Stevens, H., and Roberts, B. 1. 1959. Agron. J . 51,401-403. Camahan, H. L., Mastenbroek, J. H., Tseng, S. T., and Johnson, C. W. 1975. Crop Sci. 15,887. Camahan, H. L., Johnson, C. W., and Tseng, S . T. 1978a. Crop Sci. 18, 356-357. Camahan, H. L., Johnson, C. W., Tseng, S. T., and Mastenbroek, J. H. 1978b. Crop Sci. 18, 357-358. Carnahan, H. L., Johnson, C. W., Tseng, S. T., and Brandon, D. M. 1979. Crop Sci. 19, 746. Carnahan, H. L., Johnson, C. W., Tseng, S. T., and Brandon, D. M. 1980. Crop Sci. 20, 551. Camahan, H. L., Johnson, C. W., Tseng, S. T., and Rutger, J. N. 1981a. Crop Sci. 21,985-986. Camahan, H. L., Johnson, C. W., Tseng, S. T., and Brandon, D. M. 1981b. Crop Sci. 21, 986-987. Carnahan, H. L., Johnson, C. W., Tseng, S. T.,and Oster, J. H. 1982. Proc. Rice Technul. Work. Group 19th. pp. 19-20. CIAT. 1980 Report, Cali, Colombia, pp. 59-60. CIAT. 1981 Report, Cali, Colombia, p. 52. Coffman, W. R., and Juliano, B. 0. 1983. In ‘‘Nutritional Quality of Cereal Grains: Genetic and Agronomic Improvements” (K.J. FEY, ed.). Amer. Soc. Agron., Madison, Wisconsin. (In press.)

INDUCED MUTATIONS AND GENETICS OF RICE

41 1

Dat, T. V., Peterson, M. L., and Rutger, J. N. 1978. Crop Sci. 18, 1-4. Davis, L. L. 1965. Certification Application for Earlirose, Calif. Crop Improvement Assoc. Feenstra, W. J., Jacobsen, E., and deviser, A. J. C. 1981. Inr. Symp. Induced Mutations. (pp. 321-332). IAEA, Vienna. Foster, K. W., and Rutger, J. N. 1978. Generics 88 , 559-574. Fujimaki, H., Hiraiwa, S., and Kushibuchi, K. 1977. IkushugukuZusshi (Jpn. J. Breed.) 27,70-77. Futsuhara, Y. 1968. Gamma Field Symp. 7, 87-109. Gale, M. D., and Gregory, R. S . 1977. Euphyticu 26, 733-738. Gale, M. D., Law, C. N., Marshall, G. A., Snape, J. W., and Worland, A. J. 1982. In “SemiDwarf Cereal Mutants and Their Use in Cross-Breeding,” pp. 7-23. IAEA, Vienna (IAEATEC DOC-268). Gunawardena, S . D. I. E., Navaratne, S . K., and Ganashan, P. 1971. In “Rice Breeding with Induced Mutations 111,” pp. 29-33. IAEA, Vienna. Hajra, N. G., and Halder, S . 1980. Genet. Agr. 35, 327-338. Hajra, N. G., Bairagi, P., and Dasgupta, P. 1980. SubruoJ. 12, 125-138. Hajra, N. G., Mallick, E. H., and Bairagi, P. 1982. Actu Agron. Acnd. Sci. Hung. 31, 35-41. Harada, J., and Vergara, B. S . 1971. Crop Sci. 11, 373-374. Hargrove, T. R., Coffman, W. R., and Cabanilla, V. L. 1979. IRRf Res. Paper Ser.. No. 23. HilleRisLambers, D., Rutger, J. N., Qualset, C. O . , and Wiser, W. J. 1973. Euphyricu 22, 264-273. Hiraiwa, S . , and Tanaka, S . 1980. Gamma Field Symp. 19, 103-115. Hu, C. H. 1973. Euphyticu 22, 562-574. Hu, C. H., and Alionte, G. 1978. Proc. Rice Technol. Work. Group 17th. pp. 14-15. IAEA. 1973. “Nuclear Techniques for Seed Protein Improvement.” IAEA, Vienna. IAEA. 1974. “Induced Mutations for Disease Resistance in Crop Plants.” IAEA, Vienna. IAEA. 1977. “Induced Mutations against Plant Diseases.” IAEA, Vienna. IAEA. 1978. “Seed Protein Improvement by Nuclear Techniques.” IAEA, Vienna. IAEA. 1979. “Seed Protein Improvement in Cereals and Grain Legumes.” IAEA, Vienna. IAEA. 1982. Mutar. Breed. Newsl., No. 19. Ikehashi, H., and Kikuchi, F. 1982. JARQ 15, 231-235. IRFU. 1966 Annu. Rep., Los Banos, Philippines. IRRI. 1980 Annu. Rep., Los Banos, Philippines. Ismachin, M., and Mikaelsen, K. 1976. In “Induced Mutations in Cross-Breeding,” pp. 119-121. IAEA. Vienna. Jacquot, M. 1978. In “Rice in Africa” (I. W. Buddenhagen and G. J. Persley, eds.), pp. 117-129. Academic Press, New York. Johnson, C. W., Carnahan, H. L. 1982. Proc. Rice Technol. Work. Group 19th p. 45. Johnson, C. W., Carnahan, H. L., Tseng, S. T., and Brandon, D. M. 1980. Crop Sci. 20, 551. Johnson, C. W., Carnahan, H. L., Tseng, S. T., and Hill, J. E. 1981. Crop Sci. 21, 986. Kaul, M. L. H., and Kumar, V. 1981. Int. Rice Res. Newsl. 6 , 3-4. Kawai, T. 1968. In “Mutations in Plant Breeding 11,” pp. 161-192. IAEA, Vienna. Kawai, T. 1974. In “Induced Mutations for Disease Resistance in Crop Plants,” p. 153. IAEA, Vienna. Kawai, T. 1982. Paper presented at the 2nd Res. Coordination Meet. FAO/IAEA Prog. Evaluation of Semi-Dwarf Cereal Mutants for Cross Breeding, Davis, California, 30 August-3 September 1982. Kawai, T., and Sato, H. 1969. Nogyo Gijutsu Kenkyusho Hokoku D (Bull. Nutl. Inst. Agric. Sci. Ser. D ) No. 20, 1-33. Khambanonda, P., Dookamana, P., and Sarigabutr, A. 1982. Paper presented at the 2nd Res. Coordination Meet. FAOlIAEA h g . Evaluation of Semi-Dwarf Cereal Mutants for Cross Breeding, Davis, California, 30 August-3 September 1982.

412

J. NEIL RUTGER

KO, T., and Yamagata, H. 1980. Ikushugaku Zusshi (Jpn. J . Breed.) 30, 367-374. Konzak, C. F. 1959. Agron J . 51, 518-520. Mine, H. R. 1982. M.S. Thesis, Univ. of California, Davis. Mackill, D. J., and Rutger, J. N. 1979. J . Hered. 70, 335-341. Mahadevappa, M., Ikehashi, H., Noorsyamsi, H., and Coffman, W. R. 1981. IRRIRes. Paper Ser., No. 57. Mallick, E. H., Hajra, N. G., and Bairagi, P. 1980. Riso 28, 3-7. McKenzie, K. S., Board, J. E., Foster, K. W., and Rutger, J. N. 1978. Sabrao J . 10, 96-102. McKenzie, K. S., Lee, F. N., and Wells, B. R. 1982. Ark. Farm Res. 31, 3. Miah, A. J . , Mansur, M. A., and Uddin, M. J. 1981. Indian J . Agric. Sci. 51, 145-146. Micke, A. 1979. Gamma Field Symp. 18, 1-23. Mikaelsen, K. 1980. In “Innovative Approaches to Rice Breeding,” pp. 67-79. IRRI, Los Banos, Philippines. Mikaelsen, K., Saja, J., and Simon, J. 1971. In “Rice Breeding with Induced Mutations III,” pp. 97-101. IAEA, Vienna. Mohanty, H. K., and Das, S. R. 1979. Proc. Symp. Role Induced Mutat. Crop Imrovement, pp. 55-64. Nilan, R. A., Kleinhofs, A., and Konzak, C. F. 1977. Ann. N. Y. Acad. Sci. 287, 367-384. Nilan, R. A., Kleinhofs, A., and Wamer, R. L. 1981. Int. Symp. Induced Mutations. pp. 183-200 IAEA, Vienna. Okuno, K., and Kawai, T. 1977. Gamma Field Symp. 16, 39-62. Okuno, K., and Kawai, T. 1978a. Ikushugaku-Zasshi (Jpn. J . Breed.) 28, 243-250. Okuno, K., and Kawai, T. 1978b. Ikushugaku Zasski (Jpn. J . Breed.) 28, 336-342. Padma, A., and Reddy, G. M. 1977. Crop Sci. 17,860-863. Reddy, G. M., and Padma, A. 1976. Theor. Appl. Genet. 47, 115-118. Reddy, G. M., and Reddy, T. P. 1973. Radiat. Bot. 13, 181-184. Reddy, T. P. 1979. Riso 28, 9-13. Ree, J. H. 1973. Sabra0 J. 6, 83-85. Rutger, J . N. 1982a. Int. Rice Comm. Newsl. 31, 31-33. Rutger, J. N. 1982b. Paper presented at the 2nd Res. Coordination Meet. FAOlIAEA Prop. Evaluation of Semi-Dwarf Cereal Mutants for Cross Breeding, Davis, California, 30 August-3 September 1982. Rutger, J. N., and Camahan, H. L. 1981. Crop Sci. 21, 373-376. Rutger, J. N., and Lehman, W. F. 1977. Calif. Agric. 31, 29. Rutger, J . N., and Peterson, M. L. 1976. Calif. Agric. 30, 4-6. Rutger, J. N., and Peterson, M. L. 1981. Int. Symp. Induced Mutations, pp. 457-468. IAEA, Vienna. Rutger, J. N., and Qualset, C. 0. 1976. In ‘‘Opportunities to Improve Protein Quality and Quantity for Human Food,” pp. 143-158. Univ. of California, Davis (Special Publ. No. 3058). Rutger, 1. N. Peterson, M. L., Hu, C. H., and Lehman, W. F. 1976. Crop Sci. 16, 631-635. Rutger, J. N., Peterson, M. L., and Hu, C. H. 1977. Crop Sci. 17, 978. Rutger, J. N., Foster, K. W., McKenzie, K. S . , Mackill, D. J., Peterson, M. L., and Hu, C. H. 1979a. Crop Sci. 19, 229-300. Rutger, J. N., Peterson, M. L., Camahan, H. L., and Brandon, D. M. 1979b. Crop Sci. 19,929. Rutger, J. N., Camahan, H. L., and Johnson, C. W. 1982a. Crop Sci. 22, 164-165. Rutger, J. N., Mese, M. D., and Lu, Y. G. 1982b. Agron. Abstr., p. 82. Sairk S. S., Gagneja, M. R., and Brar, G. S. 1977. Sci. Cult. 43, 259. Samoto, S . , and Kanai, D. 1975. Ikushugaku Zasshi (Jpn. J . Breed.) 25, 1-7. Sarkar, K. R. 1974. In “Haploids in Higher Plants” (K.J. Kasha, ed.), pp. 33-41. Univ. of Guelph, Ontario.

INDUCED MUTATIONS AND GENETICS OF RICE

413

Sato, H. 1980.Mutat. Breed. Newsl. 15, 2-4. Satoh, H., and Omura, T. 1981. Ikushugaku zOsshi-(Jpn. J . Breed.) 31, 31-6-326. Schaeffer. G. W.. and Sharpe, F.T., Jr. 1982.Beltsville Symp. Agricult. Rex 7th, Genet. Eng. pp.

237-254. Sigurbjijmsson, B., and Micke, A. 1974.In “Polyploidy and Induced Mutation in Plant Breeding,” pp. 303-343. IAEA, Vienna (IAEA-PL-503-40). Simons, M. D., Caldecott, R. S., and Frey, K. J. 1%2. Plant Dis. Repl. 46, 88-91. Singh, R. J., and Ikehashi, H. 1981.Crop Sci. 21, 286-289. Singh, V. P., Siddiq, E. A,, and Swamhathan, M. S. 1979. Theor. Appl. Genet. 55, 169-176. Suh, H.S., and Heu, M. H. 1978. Yuk Chong Hakhoe Chi (Korean J . Breed.) 10, 1-6. Tanaka, S. 1978. In “Seed Protein Improvement by Nuclear Techniques,” pp. 199-201. IAEA, Vienna. Tanaka, S., and Hiraiwa, S. 1978. In “Seed Protein Improvement by Nuclear Techniques,” pp. 191-198. IAEA, Vienna. Tanaka, S., Kawai, T., Yamasaki, Y.,Niizeki, H., Kiyosawaa, S.,Wada, M., Moue, T.,and Sekiguchi, F. 1978. Gamma Field Symp. 17, 61-74. Toda, M. 1979.Gamma Field Symp. 18, 73-82. Trees, S. C., and Rutger, J. N. 1978.J. Hered. 69, 270-272. Tseng, S. T., and Seaman, D. E. 1982.Proc. Rice Techno!. Work. Group 19tk p. 18. Vavilov, N. I. 1951.Chron. Bot. 13, 1-364. Vose, P. B. 1981.Int. Symp. Induced Mutations. pp. 159-181. IAEA, Vienna. Woo,S. C., Wu, W. H., and Tung, I. J. 1974. Bor. Bull. Acad. Sin. 15, 54-56. Yamaguchi, H.,Watanabe, M., Sato, S., and Kanbayashi, Y. 1981.Int. Symp. Induced Murations pp. 201-211. IAEA, Vienna.

This Page Intentionally Left Blank

ADVANCES IN AGRONOMY, VOL 36

NITROGEN AVAILABILITY INDEXES FOR SUBMERGED RICE SOILS K. L. Sahrawat International Crops Research Institute for the Semi-Arid Tropics (ICRISAT) ICRISAT Patancheru P.O., Andhra Pradesh, India

I. Intmduction Factors Affecting Mineralization of Organic Nitrogen . A. Temperature B. Moisture Regime and Soil Drying. C. Soil Characteristics. D. Organic Amendments E. Land Preparation and Tillage Practices III. Biological Indexes . A. Anaerobic Incubation Methods B. Factors Affecting Results of Laboratory Incubation Tests. C. Ammonium Content in Soil Solution . D. Soil Biomass Nitrogen IV . Chemical Indexes. A. Organic Carbon and Total Nitrogen Content of Soils B. Alkaline Pennanganate Method. C. Acid Permanganate Method D. Other Chemical Methods. V. Simple Models of Nitrogen-Supplying Capacity Based on Biological and Chemical Indexes. VI . A Values. VII. Electro-Ultrafiltration. VIII. Plant Analyses . . IX. Nitrogen-Supplying Capacity and Fertilizer Recommendations X. Perspectives . References

n.

415 417 417 418 419 420 42 1 42 1 422 425 421 428 428 428 430 432 432 435 439 441 442 443 445

447

1. INTRODUCTION

Rice culture occurs on soils that differ considerably in their characteristics, including their nitrogen-supplying capacity. According to Moormann ( 1978), rice is grown on all the 10 soil orders given in the soil classification system of 415

Coonieht Q bv Academic Press. Inc. AII rights of;;pr;ductio;l in any form reserved. ISBN 0-12-000736-3

416

K. L. SAHRAWAT

Table I Major sdtl CLe4sweations According to Soil Taxonomy Used for Rice Growinga ~~

Suborders of soils used for rice culture Order

Major importance

Alfisols Aridisols Entisols Histosols

Aqalfs, Ustalfs

Inceptisols

Aquepts, Ochrepts, Troppts

Mollisols Oxisols spod0S01s Ultisols Vettisols

Aquents

-

-

Aquult, Udults

-

Local importance

Minor importance

Udalfs

Xeraffs Orthids, Argids Orthents, Psamments Hemists, Saprists, Andepts

-

Fluvents -

Aquolls

Humults Uderts Usterts

Udolls orthox, ustox

Aquods Usfults Torrerts Xererts

“From Moomann (1978).

soil taxonomy. The relative importance of the various soil suborders is shown in Table I. The mineralizable nitrogen (N) pool in soils plays a dominant role in nitrogen nutrition of wetland rice. Studies using lSN-labeled fertilizers have shown that approximately one-half to two-thirds of the total N utilized by a rice crop, even in well-fertilized rice paddies, comes from the soil-mineralizable N pool (Broadbent, 1978; IAEA, 1978; Reddy and Patrick, 1980; Koyama, 1981). The current shortage of fertilizers coupled with soaring prices resulting from energy costs involved in their manufacture warrant the most judicious and efficient use of fertilizer N, for which it is essential to know the nitrogen-supplying capacity of soils. Thus, development of laboratory indexes for predicting soil nitrogen availability to rice forms an important component of research for efficient use of fertilizer nitrogen. Numerous biological and chemical laboratory methods have been proposed for predicting soil N availability to various crops, including rice, and these have been reviewed by Bremner (1963, Gasser (1969), Robinson (1975), and Chang (1978). However, there is no comprehensive review available on the nitrogen availability indexes for submerged soils, although rice yields more than those of any other crop depend on soil nitrogen availability. This article will review the recent literature on methods proposed for assaying the nitrogen-supplyingcapacity of wetland rice soils and recommend those methods which have potential for predicting soil N availability, thus making possible the judicious and efficient

NITROGEN AVAILABILITY INDEXES FOR RICE

417

use of fertilizer nitrogen for rice production. No attempt will be made to discuss the literature covered in earlier reviews (Bremner, 1965; Patrick and Mahapatra, 1968; Gasser, 1969; Robinson, 1975; Chang, 1978; Sahrawat, 1982d). However, the mineralization process, which is basic to soil nitrogen availability to wetland rice, will be discussed in some detail.

II. FACTORS AFFECTING MINERALIZATION OF ORGANIC NITROGEN Mineralization of organic nitrogen, which does not proceed past ammonium production in wetland rice soils, is the most important biological process that is involved in the availability of soil N to rice grown under submerged conditions. The term rnineralizable N will be used to signify NH, production, because in a flooded paddy soil nitrification is at a low ebb because of the lack of oxygen. The process of mineralization in submerged soils is adequately described by Ponnamperuma (1972), Broadbent (1979), and Sahrawat (1983b), and only the literature pertinent to the factors affecting the mineralization and availability of soil N will be discussed here. +

A. TEMPERATURE

Among the several factors which affect soil N mineralization, temperature and moisture regime are perhaps the most important. Studies at the International Rice Research Institute (IRRI) in the Philippines demonstrated that the release of water-soluble NH,+ in eight soils was least at 15°C and increased progressively with increases in temperature from 15 to 45°C through 25 and 35°C increments (WU, 1967). The amount of water-soluble NH; released was twice as much at a temperature regime of 38-35°C as at 20°C in a silt loam (Cho and Ponnamperuma, 1971). In another study, it was found that ammonium production in four soils under anaerobic incubation increased with an increase in temperature from 15 to 45°C (IRRI, 1974; Table 11). Similarly, Myers (1975) reported that the ammonification of organic N in a tropical soil increased with an increase in temperature from 20 to 5O"C, was maximum at 50"C, and then decreased slightly with further increases in temperature. The ammonification rate was highest during the first week. These results are consistent with the conclusion drawn by Ponnamperuma (1972) that most of the organic N in a soil is mineralized within 2 weeks after submergence if the temperature is optimum. Numerous studies made in Japan (Kai et al., 1969; Onikura et al., 1975; Yoshino and Dei, 1977) and elsewhere (see reviews by Ponnamperuma, 1972; Broadbent, 1979) emphasize the importance of temperature to N mineralization

K.L. SAHRAWAT

418

TaMe II

JIffectof Temperature on NHJ Release in Four Soils Incubated Anaerobically for 2 Weeksa ~

NHJ released (ppm) in dry soil ("C) Soil

15

25

35

40

Keelung silt loam Casiguran clay loam Pila clay loam Luisiaua clay

65 205 125 50

140 250 150 115

165 325 195 200

200 340 240 130

"From IRRI (1974).

in submerged soils, These studies suggest that the differences in rice yields obtained in cooler and warmer regions could be explained by the effects of temperature on mineralization-immobilization and the availability of soil N during the growing season, especially in the later stages of growth (see Yanagisawa and Takahashi, 1964; Yamane, 1967; Gotoh and Onikura, 1971). B. MOISTURE RFKXME AND

SOIL

DRYING

Similarly, the moisture regime of a soil is critical for mineralization of organic N. It has been found that soil drying prior to flooding changes the pattern of soil mineral N release. The pioneering work of Shioiri and co-workers (Shioiri et al., 1941; Shioiri, 1948) in Japan showed &at soil drying enhances mineralization of soil organic N. Subsequent studies by Ventura and Watanabe (1978) and Sahrawat (1980a, 1981a) have also shown that soil drying enhances soil N mineralization and thus its availability to rice. The soil-drying effect was very marked in four permanently waterlogged Philippine Histosols. There was a virtual absence of mineralization in these permanently waterlogged soils, but airdrying the soils prior to flooding caused a surge in release of NHZ (Sahrawat, 1981a; Fig. 1). These results suggest that careful water management is necessary to avoid either nitrogen deficiency or excessive loss of organic matter accompanying rapid mineralization when the Histosols are used for wetland rice culture. Sahrawat (1981b) performed greenhouse pot and field studies to examine the effects of soil drying versus flood fallowing on the availability of soil and fertilizer N; he found that the N availability of unfertilized soils was not affected by continuous flooding or soil drylng and flooding one or more times during the growing season, as judged by crop growth and N uptake. Perhaps soil drying compensated for the small N losses that occurred because of nitrification-denitrification by enhanced mineralization.

NITROGEN AVAILABILITY INDEXES FOR RICE

50

-

0) Y

1 , A I R DRIED

4300 001

\

m E

v

Z

+L* I

Z

200

419

//'/

Y G

-

4 , AIR DRIED

3 , A I R DRIED

2 , A I R DRIED

2,3,4

0 6 0

-

n

A

"

I

2

A

w

I

I

4

6

NOT A I R DRIED

n I

a

WEEKS

FIG.1. Effect of air-drying on ammonification of soil organic N in four Philippine lowland Histosols incubated anaerobically at 30°C (Sahrawat, 198la). ~

Soil

PH ( 1 ~ 2H20)

Organic matter (%)

6.2 5.6 6.1 5.9

36.1 22.0 39.0 42.0

Total N (8) 1.48 0.56 1.20

1.40

C . SOILCHARAITERISTICS

It is also recognized that soil nitrogen availability and rice growth improve on acid soils with liming (Ponnamperuma, 1958; Borthakur and Mazumdar, 1968). These effects may perhaps be attributable at least partially to the alleviation of toxicity from elements such as iron as well as to a general improvement in the availability of other nutrients with liming (Ponnamperuma, 1958). Sahrawat (1980b) reported that ammonification of organic N proceeded well in two acid

K. L. SAHRAWAT

420

Table Ill

Correlations between Ammonium produced by Anaerobic Incubation and Organic Matter, and Total Nitrogen Content of Philippine Soils ~~~~

~

~

Correlation coefficient (r) Number of samples

NH,+released as

9 31 39 43 280 483

2.2- 12.3 NRa 1.9-10.7 2.615.0 3.5-26.0 1.8-14.2

% of total

N

Nb+ versus organic matter

Nb+ versus total N

Reference

0.897b

0.9126 NRa

Sahrawat (1982a) IRRI (1964) Sahrawat (1983b) Gaballo ( 1973) IRRI (1978) IRRI (1973)

0.816

0.916 0.866

NRa 0.726

0.946 0.85" 0.796 0.796

"Not reported. bSignificant at p = .01.

sulfate soils from the Philippines that had pH values of 3.4 and 3.7, respectively. These results suggest that ammonification in soils is perhaps adapted even in extremely acidic soil conditions, although nitrification may not operate under such conditions (Sahrawat, 1980a,b, Sahrawat 1982b). Studies made at the .IRRI with a large number of Philippine soils have shown that ammonium produced by anaerobic incubation is highly significantly correlated with organic matter and total N; total N recovered as ammonium varied from 1.8 to 26.0% (Table 111). In a detailed study of 39 diverse Philippine rice soils, Sahrawat (1983b) found that ammonification of soil organic nitrogen under waterlogged conditions was highly positively correlated with organic carbon (r = .91**) and total N (r = .94**) and highly negatively correlated with the C/N ratio (r = -.46**) of the soils. Other soil properties, such as pH, clay content, and cation-exchange capability (CEC) were not significantly correlated with ammonium production. Multiple regression analyses showed that organic matter content, measured by organic C and total N, accounted for most of the variation in mineralizable N in these soils. The soils tested differed greatly in mineralizable N contents (17-428 ppm NH,+-N), suggesting that they would need different amounts of fertilizer N to obtain a given yield level. D. ORGANIC AMENDMENTS

In addition to soil and environmental factors, the incorporation of organic residues also greatly influences the mineralization of soil N and its availability to rice (for reviews see Broadbent, 1979; Sahrawat, 1979, 198Od). Pioneering work

NITROGEN AVAILABILITY INDEXES FOR RICE

42 1

by Acharya (1935) clearly established that although the decomposition of crop residues was slower under anaerobic conditions, the net immobilization and nitrogen factors were much lower for the anaerobic decomposition than for the aerobic decomposition of the plant materials. The important factors that govern the effect of organic residues on the mineralization and availability of soil nitrogen are soil nitrogen status, nitrogen content and C/N ratio of the residues added, temperature, and moisture regime. E. LANDPREPARATIONAND TILLAGE F’RA~CES

Similarly, tillage and land preparation operations also influence soil N mineralization and its availability to rice crops (see DeDatta and Barker, 1978). Puddling, a common practice for lowland rice culture in Asia, has been found to have beneficial effects on rice growth. However, with the present state of knowledge, it is rather difficult to attribute the beneficial effects of soil puddling entirely to enhanced mineralization and availability of soil N because puddling affects rice growth in many ways other than by increasing the supply of soil N (see Sanchez, 1973a,b; DeDatta and Kerim, 1974). Japanese scientists, however, have claimed that cultivation and puddling of soils enchances mineralization of soil organic nitrogen (Harada et al., 1964; Sakanoue and Matsubara, 1967). It is evident that the time of soil sampling is most critical, particularly with regard to the season and cultural practices that are followed for land preparation for rice culture before and after taking the soil samples for assessing the Nsupplying capacity in the laboratory.

111. BIOLOGICAL INDEXES

The techniques involving estimation of mineral N formed during brief incubation periods have been suggested as the most satisfactory means of assessing the nitrogen-supplying capacities of soils. The various incubation techniques that have been used for estimation of potentially mineralizable N in soils are discussed by Bremner (1965), and most of the earlier references on the subject can be found in this review. However, it is evident that the methods have been tested more extensively for the upland soils, with less emphasis on submerged rice soils (Patrick and Mahapatra, 1968; Chang, 1978; Sahrawat, 1982d). This section discusses the incubation methods employed for measuring mineralizable N particularly for submerged soils and its availability to rice crops as evaluated by greenhouse and field studies (see Table IV).

K. L. SAHRAWAT

422

Table IV Incubation Methods Used for RedMiag Soil Nitrogen Availability to Rice in Greenhouseand Field Studies Method

Reference

Anaerobic incubation (under waterlog- Lopez and Galvez (1958); Subbiah and Bajaj (1962); ged conditions) at 30, 35, or 40°C Ponnampe~ma(1%5, 1978); Sims and Blackmon (1967); Sims et al. (1%7); Bajaj et al. (1%7); Kafor 6-14 days and in some cases waguchi and Kyuma (1969); Koyama (1971); Lin et for as much as 12 weeks al. (1973); Yoshmo and Dei (1974, 1977); Onikura et 01. (1975); Shiga and Ventura (1976); Singh and Pasricha (1977); Bajaj and Hasan (1978); P o ~ a m peruma and Sahrawat (1978); Bajaj and Singh (1980); Dolmat et al. (1980); Sahrawat (198Oc, 1982d) Aerobic incubation at 30°C for 1-2 Bajaj et al. (1967); Bajaj and Hasan (1978); Bajaj and weeks Singh (1980); Dolmat et al. (1980)

A. ANAEROBIC INCUBATION METHODS

Ponnamperuma (1965) has discussed the implications of ammonia release in wetland soils for rice culture and reported that ammonia production in submerged soils followed approximately an asymptotic course described by the following equation, which could be used for predicting NH4+ production in wetland soils: where A is the mean maximum NH4+ nitrogen concentration, Y is the actual concentration t days after flooding, and c is a parameter of the soil. A is a characteristic for a soil under a given temperature regime that is related to the organic matter content of paddy soils (Ponnamperuma, 1972). Work done at the IRRI in the Philippines has indicated that almost all the mineralizable nitrogen in a flooded soil is converted into ammonium within 2 weeks of submergence provided the temperature is favorable and the soil is neither strongly acid nor greatly deficient in available phosphorus (Ponnamperuma, 1972). Working with tropical rice soils, Lopez and Galvez (1958) and Ponnamperuma (1965) suggested that the amount of ammonia released by anaerobic incubation is a good measure of available nitrogen in flooded soils. Lopez and Galvez (1958) reported that rice grain yields correlated significantly with the mineralized N released within 2 weeks under waterlogged conditions in greenhouse experiments but not in field conditions. Subbiah and Bajaj (1962) found that in 18 Indian soils under waterlogged conditions the ammonium nitrogen released at 35°C within 1 week was a good index of nitrogen availability to rice.

423

NITROGEN AVAILABILITY INDEXES FOR RICE

It was further noted that the NH4+ released under waterlogged conditions within 1 week was significantly correlated with crop response (r = -.783**), but NH,+ released after 2 and 3 weeks was not significantly correlated with crop response. On the other hand, Waring and Bremner (1964a) suggested that the ammonium produced in soils after 2 weeks of incubation under waterlogged conditions at 30°C was a good index of N availability to upland crops. This conclusion was based on their results which showed that mineralization of organic N under aerobic conditions was highly correlated with NH, -N released under anaerobic incubation. Anaerobic incubation tests, because they are simple and involve only the measurement of NH, -N, are more adaptable than aerobic incubation tests. Similarly, studies by Japanese scientists have shown that the ammonia produced during 2 weeks of anaerobic incubation is a good index of available nitrogen in paddy soils (Kawaguchi and Kyuma, 1969; Koyama, 1971, 1981; Yoshino and Dei, 1974). Lin et al. (1973) evaluated nine nitrogen availability indexes for predicting N availability in 20 Taiwanese soils in a pot experiment, reporting that the amount of NH4+ released after 1 week of anaerobic incubation at 40°C was highly correlated with N uptake or percentage yield of rice grain. These authors proposed that ammonia production after 1 week of waterlogged incubation at 40°C can be employed for predicting the N-supplying capacity of rice soils. Similarly, in the United States, Sims et al. (1967) and Sims and Blackmon (1967) assessed the nitrogen-supplying capacity of 61 soils from Arkansas (in a pot study) and concluded that a single determination of NH4+ after 6 days of waterlogged incubation at 40°C provided one of the better indexes of N availability to rice. Soluble plus extractable NH, -N after 6 days of incubation accounted for 91% of the variation in yield on the 19 clay soils, but it accounted for only 18% variation on the 42 silt loams, which suggests that similar incubation tests may not be equally effective in predicting the nitrogen-supplying capacity of diverse types of soils. Dolmat et al. (1980) found that the amounts of ammonium released in surface soils from Louisiana during anaerobic incubation (30°C, 2 weeks) was the best of the eight nitrogen availability indexes tested for predicting the yield of rice grain in field experiments conducted at 34 locations (Table V). These results indicated the usefulness of the anaerobic incubation method for predicting the availability of soil N to rice under field conditions. Sahrawat (1980~)studied the N-supplying capacity of eight Philippine rice soils by growing six crops of rice under flooded conditions in a greenhouse experiment, and found that the dry matter yield and N uptake were highly significantly correlated with the amounts of ammonium released in these soils under waterlogged conditions within 2 weeks at 30°C. In a more detailed study with 39 diverse rice soils it was further confirmed that the mineralizable N (NH,+) released during anaerobic incubation of soils either at 30°C for 2 weeks or at 40°C for 1 week was a very good index of soil N availability to rice grown +

+

+

K. L. SAHRAWAT

424

Table V Liiear Correlation Coefficients (r) between Soil N Availability Indexes and Rough Rice Yields at 34 Lucations in Louisiana without Application of Fertilizer Na Correlation coefficient

N availability index

Total N

(r)b

0.363 0.273 0.560 0.433 0.352 0.315 0.323 0.289

Alkaline permanganate' Anaerobic incubationd Aerobic incubatione Autoclaving in 0.01 M CaCl$ Boiling in 0.01 M CaC@ Boiling in 0.5 N sodium pyrophosphateg Acid hydrolysish

OAdapted from Dolmat et al. (1980). Experimental plots received P (24.6 kglha) and K (46.5 kg/ha) fertilizers. bSignificant at p = .01. cSubbiah and Asija (1956). dwaring and Bremner (1964a). 'Keeney and Bremner (1%5). JStanford and DeMar (1%9). sta an ford (1968). "Purvis and Leo (l%l).

Table VI

Correlations between Values of Available M i Nitrogen by Different Methods with Dry Matter Yield and Nitrogen Uptake of Rice in a Greenhouse Pot Study0 ~~~

Correlation coefficient (r)

Dry matter Available N method

weight

N uptake

Organic C Total N Anaerobic incubation, 30°C (2 weeks) Anaerobic incubation, 40°C (1 week) Alkaline Khh04 Acid KMn04 Acid K2Cr207 H202

0.456 0.466 0.40C 0.466 0.40C 0.39= 0.39= 0.466

HzS04 (0.5 M )

0.104

0.826 0.84b 0.84" 0.826 0.816 0.75" 0.74" 0.826 0.42b

"n = 39; from Sahrawat (1982d). bSignificant at p = .01. cSignificant at p = -05. dNot significant at p = .05.

NITROGEN AVAILABILITY INDEXES FOR RICE

425

under flooded conditions on these soils in greenhouse pots (Sahrawat, 1982d; Table VI). Ponnamperuma and Sahrawat (1978) found that the nitrogen-supplying capacity of 506 diverse Philippine wetland rice soils measured by anaerobic incubation ranged from 13 to 637 pg/g of airdried soil and was highly correlated with the soil organic carbon content. It was further shown in field tests that soils with a nitrogen-supplying capacity exceeding 150 pg/g , measured by anaerobic incubation, gave yields of 4.5-7.0 tons/ha without nitrogen fertilizer. It is clear from the preceding discussion that ammonium released in soils under waterlogged conditions is a useful index for predicting soil N availability to wetland rice. Many more examples emphasizing the capability of anaerobic incubation tests for predicting N availability to rice can be found in an earlier paper by Sahrawat (1982d). B. FACTORSAFFECTING RESULTSOF LABORATORY INCUBATION TESTS

It should be emphasized here that the methods of preparing soil samples before incubation affect the mineralization of soil organic N. For example, Shiga and Ventura (1976) found that the pattern of NH,+ release in a clay soil was influenced by whether the soil was incubated in moist field conditions or after airdrying. It has been shown that soil drying prior to flooding enhances mineralization of organic N (Shioiri er al., 1941; Shioiri, 1948; Ventura and Watanabe, 1978; Sahrawat, 1980a). Sahrawat (1981a) found that airdrying produced a very quick release of NH,+ in four Philippine organic soils, although there was a virtual absence of ammonification in the continuously wet Histosols (Fig. 1). Soil mesh size also has a profound effect on soil N mineralization. It has been observed that grinding of soil samples results in increased respiration and mineralization of nitrogen under aerobic and anaerobic conditions (Waring and Bremner, 1964b; Chalk and Waring, 1970; Craswell and Waring, 1972a,b; Hiura er al., 1976; Powlson, 1980). Rovira and Greacen (1957) showed that the enhanced respiration in soil samples following compression and shearing resulted from exposure of the organic matter which had been inaccessible to soil microorganisms, rather than from enhanced aeration. Grinding and disruption of soils increases the mineralization of soil N more in clay soils where clay had protected organic matter from microbial attack (Rovira and Greacen, 1957; Craswell and Waring, 1972a). Powlson (1980) found that the grinding of two soil samples destroyed about one-fourth of the biomass in each sample, which resulted in increased mineralization of soil N. These examples stress the need for careful preparation of soil samples for studying them for potentially mineralizable N in laboratory incubation tests.

426

K. L. SAHRAWAT

In a study of 19 diverse Philippine rice soils, Sahrawat (1982~)showed that either the addition of a nitrification inhibitor (Nitrapyrin) or the exclusion of air increased the amounts of NH4+ released in soils under waterlogged conditions within 2 weeks with near-neutral or alkaline pH soils, probably by halting losses from nitrification-denitrification. These results suggest that the exclusion of air is essential for anaerobic incubation tests used for determining mineralizable N to avoid losses resulting from nitrification and denitrification. The addition of nitrification inhibitors to the soil-water system could also help halt these losses. Another important point which should be considered while measuring NH,+ released following waterlogged incubation of soils is that the soil samples should not be directly distilled with MgO; instead, filtered extracts of the samples should be used. This conclusion is based on the findings of Sahrawat and Ponnamperuma (1978), who found that direct distillation of soil samples gave inflated NH,+ values because of hydrolysis of organic matter at high pH (9.9-10.7) caused by the boiling MgO suspensions. Recent work has also suggested that carbon dioxide evolved during direct distillation with MgO of anaerobic soils causes a negative error in ammonium determination compared to distillation of the soil extracts using the steam distillation-titration methods (Sahrawat, 19828). Earlier work indicated that direct distillation of soils resulted in inflated values for NH,+nitrogen, presumably as a result of hydrolysis of organic matter at the high pH attained by soil-MgO suspension (Sahrawat and Ponnamperuma, 1978). In fact, these studies emphasize the complex interaction that occurs between positive error caused by organic matter hydrolysis and negative error caused by carbon dioxide evolved during direct distillation of anaerobic soils with MgO for estimation of ammonium. The final result obtained will thus be the resultant of the positive and negative errors, especially in anaerobic soils following’their incubation underwaterlogged conditions (Sahrawat, 19828). A recent report by Clausen er al. (1981) has also indicated that the carbonates interfere with the determination of mineral nitrogen in soils using the direct steam distillation method. Smith er al. (1980) found that the aerobic leaching incubation procedure used for estimating soil nitrogen mineralization potentials in three mineral soils employing the first-order model resulted in the leaching of significant amounts (13-163% of total mineralized N in 11 weeks) of organic N when 0.01 M CaCl, was used. These authors concluded that organic N extracted with inorganic N should be considered to avoid serious errors in the determination of N mineralization potentials and mineralization rate constants for soils using the aerobic incubation technique with successive subsequent leachings. While studying the mineralization of organic N in four wetland Histosols from the Philippines, Sahrawat (1983a) found that 2 M KCl extracts of these soils in aerobic and anaerobic states contained significant amounts of organic N, but that the amounts were higher in anaerobic samples following 4 weeks of incubation

NITROGEN AVAILABILITY INDEXES FOR RICE

427

under waterlogged conditions. These results suggest that ignoring the amounts of organic N extracted by salt solutions used for measuring mineral N may cause serious errors in the estimation of potentially mineralizable N in organic soils. However, no data are available for mineral wetland rice soils to show whether any organic N will be extracted by the salt solutions commonly used for the determination of NH, following anaerobic incubation. Results with organic soils also indicated that the amounts of organic N extracted by 2 M KCl increased in these soils following submergence. Because low-molecular-weight N compounds are synthesized in submerged soils (Ponnamperuma, 1972), and because whether these are extracted by salt solutions in soils will affect the estimation of potentially mineralizable N in wetland rice soils, this aspect needs further investigation. Thus, it is suggested that the methods of soil preparation, and the procedures used for anaerobic incubation and extraction of NH; following anaerobic incubation, should be clearly defined because all these steps affect the results obtained for potentially mineralizable N in submerged soils. It is recommended that the soil samples should be incubated anerobically (by excluding air) under moist field conditions with a 2-3 cm layer of standing water. Filtered extracts of the incubated soil samples, rather than the soil-KC1 suspensions, should be used for the estimation of ammonium. +

c. AMMONIUM CONTENT IN SOIL SOLUTION When a soil is kept submerged, a considerable portion of ammonia (NH, + NH,OH NH4+) may be found in the soil solution phase, especially in coarsetextured soils with a fair content of organic matter. This highly mobile form of ammonia is the direct source of nitrogen for the rice plant (Ponnamperuma, 1965, 1972). Studies made at the IRRI in the Philippines have shown that as with ammonium production in soils, the ammonia in soil solution also followed widely different asymptotic courses that were related to the rate of ammonium production and cation-exchange reactions. For example, a sandy loam rich in organic matter had a concentration of 70 ppm water-soluble (soil solution) ammonia during 75 days of submergence; on the other hand, a neutral clay loam, low in organic matter, accumulated only 5 ppm ammonia in the same period (IRRI, 1964). The soil solution of submerged soils is a dynamic phase and its composition relative to plant nutrients can be a useful tool for fertility evaluation (Ponnamperuma, 1965; Cho and Ponnamperuma, 1971). Mangaraja et al. (1976) tested several modifications of the Waring and Bremner (1964a) waterlogged incubation test for predicting the nitrogen-supplying capacity of 16 soils from Orissa (India) to rice grown under submerged conditions in greenhouse. It was found that the anaerobic incubation method

+

428

K. L. SAHRAWAT

involving incubation of the soil samples with 75 or 100 ppm N (as ammonium sulfate) for 7 days and measuring the ammonium nitrogen present in the supernatant solution gave the best correlations with the relative yield or N uptake of rice. D. SOILBIOMASSNITROGEN

Jenkinson and Powlson (1976) developed a method for measuring microbial biomass N and mineralization of soil N that involves fumigation of soil samples with chloroform. This results in the release of mineral N (which is then measured) from the killed microorganisms. Ayanaba et al. (1976) suggested that the mineral nitrogen released following chloroform fumigation of soils was related to their nitrogen-supplying capacity. The fumigation technique suggested by Jenkinson and Powlson is very sensitive and has been extensively used to assess the effects of agricultural practices on the size of microbial biomass (see Powlson and Jenkinson, 1981; Lynch and Panting, 1982). However, the method has not been evaluated for assessing the nitrogen-supplying capacity of soils as assessed by plant performance in greenhouse or field experiments. Also, this technique has not been used to evaluate the nitrogen-supplying capacity of wetland rice soils. It will be useful to evaluate this technique for determining the nitrogensupplying capacity of submerged soils, because these differ from upland soils in that their microbial populations are predominantly bacterial and thus less diversified than those of their upland counterparts.

IV. CHEMICAL INDEXES Numerous chemical methods have been proposed for assessing soil nitrogen availability to crops, including rice. Earlier work on this subject is reviewed by Bremner ( 1963, Robinson ( 1975), and Chang ( 1978). More recent references on chemical methods used for predicting soil N availability to wetland rice in greenhouse and field studies are listed in Table VII. A. ORGANIC CARBONAND TOTALNITROGEN CONTENTOF SOILS

The chemical indexes proposed for availability of soil nitrogen to rice include simple methods such as measurement of organic C and total N contents of soils (Bajaj etul., 1967; Sims et al., 1967; Ghosh and Hasan, 1980; Sahrawat, 1980c, 1982d). The logic behind using organic C or total N as an index of soil N availability is that the available N pool in soils ultimately comes from organic matter through the mineralization process. Soil-testing laboratories in India use

NITROGEN AVAILABILITY INDEXES FOR RICE

429

Table W Chemical Methods Used for Predicting Soil Nitrogen Availability to Rice in Greenhouse and Field Studies Method

Reference

Organic C (organic matter)

Bonner (1946); Sims et al. (1967); Bajaj et al. (1967); Singh and Tripathi (1970); Singh and Pasricha (1977); Bajaj and Hasan (1978); Ponnamperuma (1978); Ponnamperuma and Sahrawat (1978); Bajaj and Singh (1980); Sahrawat (198Oc, 1982d) Sims et al. (1967); Ponnamperuma and Sahrawat (1978); Dolmat et al. (1980); Sahrawat (1980c, 1982d) Sims and Blackmon (1967); Sims et al. (1967); Singh and Pasricha (1977); Bajaj and Hasan (1978); Bajaj and Singh (1980) Subbiah and Asija (1956); Tamhane and Subbiah (1960); Bajaj et al. (1967); Singh and Tripathi (1970); Rajamannar et al. (1970); Ranganathan et al. (1972); Ramanathan and Krishnamoorthy (1973); Gopalswamy et al. (1973); Singh and Pasricha (1977); Bajaj and Hasan (1978); Ponnamperuma and Sahrawat (1978); Velayutham (1979); Bajaj and Singh (1980); Dolmat et 01. (1980); Sahrawat (198Oc, 1982d) Sahrawat (1978, 1982d)

Total N Mineral N

Alkaline permanganate digestion

Acid permanganate extraction Acid dichromate extraction Hydrogen peroxide oxidation Extraction with dilute H2S04 Autoclaving in CaClz (0.01

Sahrawat (1982d) Sahrawat (198Oc, 1982d) Dolmat et al. (1980); Sahrawat (1982d) Dolmat et al. (1980)

M)

Boiling CaCIz (0.01 M) extraction Sodium pyrophosphate (0.5 N) extraction Alkali hydrolysis with Ca(OH), or NaOH

Dolmat et al. (1980) Dolmat et al. (1980) Chu (1962); Bajaj and Hasan (1978); Bajaj and Singh (1980)

organic C content as the index of nitrogen availability to rice (see Patnaik, 1970; Ghosh and Hasan, 1980; Sahrawat, 1980c, 1982d). In a study of 39 diverse Philippine rice soils, Sahrawat (1982d) found that organic C and total N contents of soils were good indexes of soil N availability to rice grown in pots under flooded conditions. Further studies indicated that potentially mineralizable N released under waterlogged conditions was highly significantly related to organic C and total N (Sahrawat, 1983b). These results, together with earlier evidence

430

K. L. SAHRAWAT

(for review see Robinson, 1975; Chang, 1978), suggest that simple tests like organic C and total N contents of soils could be used for predicting soil N availability to rice. However, there are reports in literature in which organic C or total N contents of soils have been found to be poor indexes of nitrogen availability to crops, including rice (see Bremner, 1965; Sims etal., 1967; Chang, 1978). In a field evaluation of several indexes of nitrogen availability at 34 locations in Louisiana, Dolmat et d. (1980) found that the rough yields of rice were correlated with total N content of soils, but the prediction was poor (Table V). Based on this discussion, it is concluded that organic C and total N contents of soils merit evaluation as possible indexes for N availability to rice. B. ALKALINE PERMANGANATE METHOD

Alkaline permanganate oxidation was first developed for determiningthe easily mineralizable N in organic manures (Assoc. Official Agricultural Chemists, 1930). Later Truog (1954) used the release of NH,+ from soil by the action of permanganate and N%CO, for predicting the potentially mineralizable N pool in soils. In his method, the soil samples were boiled with KMnO, and N%CO, for 5 min to release easily mineralizable N, mainly NH,+nitrogen. This method has been greatly modified (see Stanford, 1978; Sahrawat and Burford, 1982). The alkaline permanganate digestion method has been widely used for assessing the potentially mineralizable N pool in soils, especially in India for paddy rice (Subbiah and Asija, 1956; Tamhane and Subbiah, 1960; Bajaj et al., 1967; Patnaik, 1970; Rajamannar et al., 1970; Singh and Tripathi, 1970; Ramamoorthy et al., 1971; Dolmat etal., 1980; Ghosh and Hasan, 1980; Sahrawat, 1982d; also see Table Vn). Stanford (1978) has summarized previous work pertaining to the use of the alkaline permanganate method as an index of soil N availability to upland crops. Although the chemistry of the method is not clearly understood (Bremner, 1965), it is evident that alkaline permanganate releases nitrogen from soil organic matter by both oxidative and hydrolytic action (Stanford, 1978). Sahrawat and Burford (1982) found that the alkaline permanganate method could assess the availability of soil N, included nitrogen from amino acid and NH,+ but not nitrogen from urea, NO,--, or NO,-in the available N pool (Table VIII). Based on the results obtained with upland soils, which accumulate nitrate nitrogen, these authors suggested that the noninclusion of nitrate nitrogen in the available N pool might be the reason for the poorer predictability obtained by this method for upland crops than for paddy rice; nitrate formed in wetland soils may be lost because of denitrification. Boswell et al. (1962) found a good correlation between the values of available N obtained by the Truog method and the nitrifying capacity of 30 soils. In a study

NITROGEN AVAILABILITY INDEXES FOR RICE

43 1

Table VIU Recovery of Different Forms of N on Treating Pure Solutions of SpecifiedCompounds with the Alkaline Permanganate Digestion and Distillation Method0

Form of N

Compound added

Amino acid

L-Aspartic acid m-Leucine L-Glutamic acid L-Lysine monohydrochloride Mean Urea (NHdzS04 NaN02

Urea Ammonium Nitrite Nitrate

wo3

93 94 95 80 91 2 95 1

I

“From Sahrawat and Burford (1982). bBased on the addition of 2500 (*g N in solution as specified compounds

of 17 Indian soils, Srivastava (1975) found that the available N in soils determined by the alkaline permanganate method was correlated with organic matter, total N, amino sugar-N, hydrolyzable ammonium + amino sugar-N, hydrolyzable NH,+-N, and amino acid-N contents of soils. The NH,+ released by this method correlated with the potentially mineralizable N in soils, but the correlation coefficients were poor (Stanford, 1978). Sahrawat (1982e) found that the amounts of available N in Philippine rice soils determined by the alkaline permanganate method (alkaline KMnO,-N) were significantly correlated with the organic C, total N, and potentially mineralizable N contents determined by the anaerobic incubation methods (Table IX). It was also found that alkaline KMn0,-N was highly significantly correlated with N uptake or the dry matter weight of rice grown under flooded conditions in pots with 39 diverse soils (Table VI). These results and evidence obtained by other researchers (Tamhane and Subbiah, 1960; Bajaj et al., 1967; Chang, 1978), suggest that this method can be useful in predicting the availability of soil N to wetland rice. However, in a field study of 31 soils at 34 locations in Louisiana, Dolmat et al. (1980) found that the alkaline permanganate method was a poor predictor of rice yields (Table V). Bajaj and Singh (1980) evaluated several nitrogen availability indexes for rice in two field experiments and found that the available N determined by the alkaline permanganate method and ammonium released during 1 week of waterlogged incubation were significantly correlated with the yields of rice, but these correlation coefficients (r)were also low (0.40 and 0.25, respectively).

432

K. L. SAHRAWAT

T a b IX Conrlptions between Chemical Indexes of Available N and Organic C, Total N, and Mimrplizpble N r e l e a d during Aaperobic Incubatioa of 39 Philippine Wetland Rice SOW ~~

~

Correlation coefficient

(r)b

Chemical index

compared

Organic C

Total N

Mineralizable Nc

Mineralizable Nd

Alkaline KMn04-N

0.855 0.839 0.440 0.830 0.814

0.882 0.845 0.461 0.855 0.840

0.859 0.800 0.541 0.858 0.855

0.812 0.788 0.457 0.828 0.795

Acid KMn04-N HzS04-N Acid K2Cr207-N H202-N

“From Sahrawat (198%). bSignificant at p = .01. cMiralizable N p m d d in soils under waterlogged incubation at 30°C for 2 weeks. dMineralizable N pmduced in soils under waterlogged incubation at 40°C for 1 week.

C. ACIDPERMANGANATE METHOD

Stanford and Smith (1978) found that the ammonium extracted by an acid permanganate solution (0.02 M KMnO, in 0.5 M H,SO,) was a good index of potentially mineralizable nitrogen in diverse soil samples collected from many scattered locations in the United States. Based on the comparative evaluation of the acid permanganate and alkaline pennanganate methods, they suggested that the release of ammonium from soil organic nitrogen by the oxidative action of acid permanganate was a better index of potentially mineralizable N than the alkaline permanganate method (Stanford, 1978; Stanford and Smith, 1978). Sahrawat (1982d) showed that the ammonium released by acid permanganate (acid KMn0,-N) was highly correlated with N uptake or dry matter yield of rice grown under submerged conditionsin pots (Table VI). As observed with alkaline KMn0,-N, acid KMn0,-N correlated highly significantly with organic C,total N, and potentially mineralizable N produced under waterlogged conditions (Sahrawat, 1982e; Table m). D. OTHERCHEMICAL METHODS

In addition to the chemical indexes previously described, a few other chemical methods have been proposed for predicting soil N availability to rice. Sahrawat (198Oc, 1982d) found that the ammonium released from the soil organic N pool by hydrogen peroxide oxidation (H,O,-N) correlated highly with the N uptake or dry matter yield of rice grown under submerged conditions in pot experiments

NITROGEN AVAILABILITY INDEXES FOR RICE

433

(Table VI). Additionally, it was indicated that H,O,-N was highly correlated with organic C, total N, and potentially mineralizable N released under flooded conditions in 39 diverse rice soils from the Philippines (Sahrawat, 1982e; Table IX). The results of this study also showed that NH,+ extracted with 0.5 M H,SO, (H,SO,-N) was correlated with organic C, total N, and NH2-N released under flooded conditions, although the correlations were poor (Table IX). Also, H,SO,-N was found to be a poor predictor of the soil N availability to rice grown in submerged soils in a greenhouse study (Table VI). Evaluations of alkalihydrolyzable N as an index for soil N availability to rice have met with limited success, as did methods employing dilute acid solutions for hydrolysis of soil organic N (Chu, 1962; Bajaj and Hasan, 1978; Dolmat et al., 1980; Bajaj and Singh, 1980). In a study with a large number of diverse surface soil samples collected from the different rice-growing areas of the Philippines, Sahrawat (1982e) found that the ammonium extracted with a 0.02 M K,Cr,O, (in 0.5 M H2S04) solution was highly significantly correlated with the organic C and total N contents, as well as with the amounts of ammonium released under waterlogged conditions at 30°C within 2 weeks or at 40°C within 1 week (Table IX). In a greenhouse pot experiment, when rice was grown under flooded conditions on these soils the ammonium extracted with acid dichromate (K,Cr,O,-N) was found to be a good index of soil nitrogen availability to rice; the acid K,Cr,O,-N was highly positively correlated with the dry matter and N uptake of the rice (Sahrawat, 1982d; Table VI). Sahrawat (19820 then developed a method which can be used for the simultaneous determination of organic C and acid dichromate-oxidizable N in soils. The method is based on a simple modification of the Walkley-Black method used for determination of organic C. The amount of ammonium released by the oxidative action of acid dichromate during organic C determination is measured by distilling an aliquot after titration of excess chromate with ferrous sulfate [rather than ferrous ammonium sulfate (to avoid ammonium contamination)] with alkali. This method will provide opportunities for testing organic C and acid dichromate-N alone or in combination as indexes for soil N availability. Initial results obtained with 15 soils from the Philippines showed that acid dichromateN was highly positively correlated with the amounts of NH4+ released during anaerobic incubation tests of these soils (Sahrawat, 19820. Several other methods have been proposed for measuring the pool of soil available N but have not been evaluated for predicting soil N availability to wetland rice. Based on their chemistry and evaluation for upland crops, these techniques have potential for predicting soil N availability to rice crops. Jenkinson (1968) suggested that the glucose extracted by shaking soil samples with 0.05 M Ba (OH), was a good index of potentially mineralizable N in British soils. His studies and investigations by Whitehead et al. (1981) and Whitehead

K. L. SAHRAWAT

434

(1981) have shown that this method provides a good index for soil N availability to grasses in greenhouse and field studies. However, this method has been found to be less suitable for routine use (Whitehead, 1981). Stanford et al. have developed several laboratory techniques for measuring potentially mineralizable N (Stanford, 1977). For example, the method based on the extraction of NH, by autoclaving soil samples with CaC1, has been widely tested and found to be useful in predicting potentially mineralizable N in soils and its availability to upland crops (Stanford and Smith, 1972, 1976; Stanford, 1977; Whitehead et ul., 1981; Jones et al., 1982). Jones et al. (1982) have evaluated this technique to estimate potentially mineralizable N and rate constants for mineralization under field conditions with 90 samples representing eight soil orders; the prediction was greatly improved when chemical and taxonomic criteria were considered simultaneously. These authors developed regression equations for the estimation of potentially mineralizable N based on the use of a multiple regression of soil pH, organic C, total N, and soil taxonomy, which accounted for 83% of the variation of potentially mineralizable N in 90 samples from 67 soils representing eight soils orders. Mac Lean (1964) suggested a chemical method based on the extraction of soil samples with a dilute (0.01 M ) NaHCO, solution and the measurement, by digestion, of the organic N extracted by the reagent. In subsequent studies of the methd, Fox and Piekielek (1978a) developed a method which measures the organic N content in the bicarbonate extract by measuring the optical absorbance of the extract at 260 nm, which is more adaptable than the Kjeldahl procedure used by Mac Lean (1964). Michrina et al. (1981) evaluated this method in greenhouse and field studied for predicting soil N availability to corn; the availability ofN was significantly correlated with N uptake in the greenhouse, but the correlation was not significant in the field study with 10 soils common to greenhouse and field experiments. It was also observed that the organic matter of soils accounted for 72% of the variation in greenhouse relative N uptake but for only 23% of the field relative uptake variation. Based on extensive evaluation of several chemical indexes in laboratory, greenhouse, and field studies, Fox and Piekielek (1978b) and Michrina et ul. (1981) concluded that field experiments are necessary for evaluating laboratory indexes for available N, and that greenhouse testings are no substitute for field evaluation of the nitrogen availability indexes because of entirely divergent results obtained under the two situations with a given set of soils. Michrina et al. (1982) found that the chemical indexes based on extraction of N by 0.01 M NaHCO, (Mac Lean, 1964) and boiling 0.01 M CaC1, (Stanford, 1968) extracted specific fractions of the soil organic N which differed significantly in their chemical composition. They found that 43-92% of the N extracted by NaHCO, was in protein form and 8-3096 was detectable as ninhydrin. In the case of the boiling CaCl, extract, about 25-3096 of the N was detectable as +

NITROGEN AVAILABILITY INDEXES FOR RICE

435

ninhydrin. For the 10 soils studied, ninhydrin-detectable N values in CaCl, extracts were closely correlated to greenhouse and field relative N uptake values, whereas the ninhydrin-detectableN values in NaHCO, extracts were unrelated to both greenhouse and field relative N uptake by corn. However, the protein N and protein N plus ninhydrin-detectable N values of the NcHCO, extract were closely correlated to the relative N uptake values in the greenhouse study. This study indicates the usefulness of chemical characterization of the extracts obtained by chemical indexes for understanding which fractions of organic N contribute to the soil available N pool for different crops. More such studies may be needed to determine the applicability of different chemical indexes to different crops. Some of the extraction methods employing heating and boiling (such as the CaCl, method) evidently result in a greater organic nitrogen fraction than those employing extraction at room temperature (e.g., see Greenland, 1965). Jenkinson (1968) also pointed out that the amounts of polysaccharides extracted from soils increased markedly with boiling extraction methods. In addition, boiling causes the molecular hydrolysis of organic N compounds and the destruction of soil aggregates which may facilitate the extraction of increased amounts of organic N (Michrina ef al., 1982).

V. SIMPLE MODELS OF NITROGEN-SUPPLYING CAPACITY BASED ON BIOLOGICAL AND CHEMICAL INDEXES In this section, simple equations describing the relationships between the amounts of NH,+ -N released under waterlogged conditions and environmental factors such as temperature and other soil characteristics are discussed. These relationships have been formulated from studies with diverse soils from a particular region and are limited in that they have not been widely tested. However, these simple models in this or modified form might help in making approximate predictions of mineralizable N in wetland rice soils. Studies at the International Rice Research Institute in the Philippines by Ponnamperuma and his associates have shown that soils rich in organic matter release ammonium rapidly, following a logarithmic increase with time which can be represented by the following equation (IRRI, 1965): Y =a

+ b log n

(2)

where y is the total amount of NH:-N released, a is the amount of NH:-N content 1 day after flooding, b is the parameter highly correlated with the organic matter and total N contents of the soil, and x is weeks submerged. Further studies using anaerobic incubation of wetland soils indicated that the

K. L. SAHRAWAT

436

soils low in organic matter released much smaller amounts of ammonium at comparatively slower rates; the increases in NH, -Nwere not appreciable after 4 weeks, in contrast with soils rich in organic matter where concentrations of NH,+ continued to increase after 30 days. In soils low in organic matter, soil N mineralization followed an asymptotic course described by Eq. (1) or by the 1964; Ponnamperuma, 1972): following equation (nW, +

(3)

(A-VIA = e-='

where A is the mean maximum amount of NH, -N, Y is the actual amount of NH,+ -N f days after submergence, and c is the parameter controlled by the soil. Because temperature greatly influences the release of ammonium in submerged soils, Yoshino and Dei (1974) suggested the concept of efecn've temperufure for predicting mineralizable nitrogen in incubation tests by correcting for the soil temperatures prevalent during the growing season (for a review see Dei and Yamasaki 1979). According to this concept, the potentially mineralizable nitrogen is calculated from the following equation: +

Y

= k[(T- 15)DI"

(4)

where Y is the amount of nitrogen mineralized, T is the mean temperature of the treatment or daily soil temperature in "C, D is the number of days, (T- 15) is the effective temperature (in "C), and k is the coefficient relating to the potentials of mineralized N. The exponent n is a constant relating to the pattern of N mineralization. In this model, 15°C is considered to be the threshold temperature for the Occurrence of the mineralization process in submerged soils. It can be derived from Q. (4); if the effective temperature is the same in two soils, the amount of N mineralized will be the same irrespective of the temperature regimes. In other words, the amount of N mineralized at 30°C during 20 days will be the same as that mineralized at 35°C during 15 days because the summation of effective temperature is same in the two cases. This also indicates that if the temperature during an anaerobic incubation test is higher, the incubation period can be reduced accordingly to obtain similar amounts of mineralizable N as released during longer incubation periods at lower temperatures. Onikura el al. (1975), to calculate mineralizable N in submerged Japanese soils incubated for 70 days, developed a regression equation by considering the soil properties that affect the release of NH,+ . According to these authors the best prediction of mineralizable N (Nmin)was obtained by the following equation: N,,

= 1.70

+ 17.57(total N) + 0.444(CEC) - 1.58(Fe20,) - 0.233(Ca) (5)

which accounted for 87% of the variability in mineralizableN for a large number of soils. It should be mentioned here that total N and CEC combined accounted for 62% of the variation observed in the mineralizable N values, and that a combination of total N, CEC, and free Fe20, accounted for 82% of the variation.

NITROGEN AVAILABILITY INDEXES FOR FUCE

437

Sahrawat (1983b) developed regression equations to predict potentially mineralizable N as affected by soil properties using 39 diverse rice soils from the Philippines: N, where r2 =

+ 85.15 (organic C)

(6)

82.8%, and N,

where r2 =

= -59.96

= -44.4

+ 792.3 (total N)

(7)

88.8%.

A combination of organic C and total N accounted for 89.4% variability in the mineralizable N by the following equation:

Ndn = -29.9

+ r1336.1 (total N)] - 61.0 (organic C)

(8)

The most variability (9= 91.8%) was found by the regression equation Nmin= -68.8

+ [1%9.6 (total N)] - 128.8 (organic C) + [9.9 (C/N)]-9.2 (pH) + 0.88 (CEC) - 0.75 (clay) (9)

It is clear from regression Eqs. (6)-(9) that in these soils organic matter, as measured by the organic C and total N contents of the soil, accounted for the most variations in mineralizable N. It also seems, from these equations, that organic matter may be used (or at least shows potential for use) for predicting mineralizable N in these soils under submerged conditions. It has been suggested that the amount of soil N mineralized in time t is proportional to the square root of the period of incubation and can be expressed by the following equation (Stanford and Smith, 1972): N, = a$

(10)

where a is the N mineralization rate and t is the time. However, this equation has the limitation that at infinite time it will give an infinite amount of mineralized N, which is incorrect (Houng, 1980). Stanford and Smith (1972) suggested the following equation for calculating potentially mineralizable N: 1/N, = l/No = b/t

(1 1) where No is the potentially mineralizable N, N, is the amount of N mineralized in time t, and b is the constant related to soil properties. This equation was fitted into the fmt-order kinetic equation log(N,-N,)

= log No - b/2.303t

(12)

Lin et al. (1973) evaluated Eqs. (10)-(12) to calculate the potentially mineralizable N in 20 submerged Taiwanese soils and compared these values with the actual amounts of soil N mineralized during up to 12 weeks of anaerobic incuba-

Table X Correlations Between N Uptake and Yield Index of Rice in a Greenhouse Studya Soil N mineralization rateb

N mineralization Parameter compared

0-2

0-4

0-6

0-8

0- 10

0- 12

N uptake from no N pots Yield index [ ( N d d ) X l00Ie

0.834d 0.739d

0.868d 0.729d

0.877d 0.713d

0.841d 0.6Wd

0.846d 0.694d

0.850d 0.704d

~

“Data of Lin et al. (1973) adapted from Houng (1980). bBased on Eq. (10). Numbers represent weeks of incubation. ‘Based on 4 s . (11) and (121, respectively, with 12 weeks of incubation. dStatistically significant at p = .01. ‘N, is the yield of no N pot; N, is the yield of the N fertilizer pot. double application. fstatistically significant at p = .05.

~~~~

potential= 0.7418d

0.7893d

0.523f

0.5W

NITROGEN AVAILABILITY INDEXES FOR RICE

439

tion. They found, in a greenhouse pot study, that the actual amounts of NH, -N released within a certain period of incubation, for example, during 2 or 4 weeks of submergence, were better correlated with the rice yield index or N uptake than with the values of potentially mineralizable N obtained (Houng, 1980; Table X). Stanford and Smith (1976) estimated potentially mineralizable soil nitrogen (No) from the amounts of NH4+-N produced by hydrolysis of soil organic nitrogen during 16 hr of autoclaving (N,) using 475 diverse soil samples from the United States representing 54 soil types; they developed the following regression equation: +

No = r(4.1 21.0) N,]

+ 6.6

(13)

where No is the potentially mineralizable N and N, is the amount of NHZ-N released during autoclaving. Acid permanganate extraction has been proposed for the estimation of potentially mineralizable nitrogen in soils (Stanford and Smith, 1978) and of availability of soil N to wetland rice (Sahrawat, 1982d). The following regression equations were developed for the estimation of potentially mineralizable nitrogen (No) from values of NH, -N released by the oxidative action of acid permanganate (0.02 M KMnO, in 0.5 M H,SO,) using 62 diverse soils from the United States (Stanford and Smith, 1978). For 43 noncalcareous soils +

No = 3.2 (acid KMn0,-N) - 19.8

(14)

Using 19 calcareous soils,

No = 3.0 (acid KMnO,-N) - 7.5

(15)

No = 3.1 (acid KMn0,-N) - 8.5

(16)

For all the 62 soils,

The correlation coefficients (r) for these equations were 0.91, 0.86, and 0.89, respectively. It is suggested, because results obtained in laboratory and greenhouse studies with some of these indexes have given encouraging results, that some of the simple models based on statistical relationships between potentially mineralizable soil nitrogen and other indexes of soil N availability should be evaluated for their ability to predict soil nitrogen availability to rice under field conditions (Stanford and Smith, 1976, 1978; Sahrawat, 1982,d,e,f).

VI. A VALUES

In 1952, Fried and Dean suggested that the nitrogen-supplying capacity of soils could be determined by using 15N-labeledfertilizers, which would make it

440

K. L. SAHRAWAT

possible to differentiate the contributions of soil and fertilizer N to plant uptake. The underlying principle of the A-value concept is that the amount of a given nutrient in the soil and the amount that is applied as fertilizer will have equal availability to plants. The main assumptions in using the A-value technique are that reactions between soils and fertilizer are minimal, and that if reactions do occur between fertilizer N and soils they must be similar for the soils under comparison (Rennie and Fried, 1971). Because rice is grown on diverse types of soils, these assumptions probably will not always be met in wetland rice cultwe. However, it is a useful concept which takes into consideration the effect on soil nitrogen contribution by cropping history. For example, based on the data obtained by the International Atomic Energy Agency ( M A , 1970, reported by Broadbent, 1978), from field experiments in 13 countries, it was found that there was a linear relationship (r = .851) between the A value and the uptake of soil N by a rice crop (Fig. 2). The average A value from these experiments was 21% of the total N content of soils. Based on the data obtained for nitrogen uptake for soil and fertilizer nitrogen soun'~s,the A-value concept can be used to evaluate nitrogen available in the soil. The following equation is used to calculate A values: A = [B (1-Y)]/Y (17) where A is the amount of nitrogen available in the soil, B is the amount of nitrogen in the standard, and Y is the proportion of nitrogen in the plant derived from the standard. In cases where I5N-labeled fertilizer is used, Y is the percentage of the total nitrogen taken up by the plant from the fertilizer. The prospects and problems in the use of A values for characterizing available soil nitrogen pools have been critically discussed by several authors (Broadbent, 1970, Hauck and Bremner, 1976; Hauck, 1978, 1979). It is suggested that the use of A values as an index of soil nitrogen availability will require careful characterizationof the conditions under which this estimate is made. Studies indicate that A values in rice cultures are markedly affected by fertilizer N placement (Broadbent and Mikkelsen, 1968; Broadbent, 1970) and the time of fertilizer application (Koyama er al., 1973). Studies conducted in Japan on 10different soils showed A values ranging from 98 to 443 kg Nlha, and soil N contributed two-thirds of the total N taken by rice crops fertilized with varying rates of fertilizer N (Koyama, 1981). In pot study under flooded conditions with four Philippines rice soils it was found that the A values for these soils ranged from 13.5 to 30.5, with a mean value of 20.3. These A values are considerably higher than the reported average A values ranging from 4.3 to 8.7 for some upland soils (Hunter and Carter, 1965; Legg and Stanford, 1967; Herlihy and Sheehan, 1979). Broadbent and Keyes (1971) suggested that the difference in A values for wetland rice and upland soils could be partly because the available N pool in wetland rice culture is maintained mostly in the ammonium form, as compared to upland soils where it is maintained in both the ammonium and

NITROGEN AVAILABILITY INDEXES FOR RICE

oL

I

100

, 200 A VALUE

300

441

I

I

400

500

(KGIHA)

FIG. 2. Correlation of A values with uptake of soil nitrogen by rice plant in field. Y = 20.3 0.212X r = ,851. (Broadbent, 1978.)

+

nitrate forms. Also, loss of fertilizer N is higher under flooded conditions of rice culture. Broadbent (1978) has suggested that A values obtained from future field experiments should be measured more often under different soil cropping and environmental conditions so that a large body of data obtained under a set of defined conditions could improve the predictive value of this technique. With the availability of 15N-depleted materials, field experimentation has become less expensive, and this should help in generating more field evaluations of the nitrogen-supplying capacity of wetland paddy soils under diverse soil and environmental conditions than are presently available. A values determined in a large number of field experiments in Japan between 1973 and 1983 have shown that A values vary greatly with the rate of fertilizer N, the type of fertilizer, and the method of application (Koyama, 1981).

VII. ELECTRO-ULTRAFILTRATION The electro-ultrafiltration (EUF)technique is based on extracting different fractions of N from soils with varying voltages and temperatures. Electro-ultra-

442

K. L. SAHRAWAT

filtration thus offers a method to fractionate soil N into fractions which are instantaneously (intensity) available, released at 20°C and 200 V, as well as into those which represent potentially available N fractions or reserves (capacity), released at 80°C and 400 V (Nemeth, 1979). Using the EUF technique, NH4+ is discharged from soils at the cathode within 10-15 minutes. Dilute HCl is used to avoid the loss of ammonia through volatilization (Nemeth ef al., 1979). In addition to mineral N, the EUF technique also measures organically bound N in soils. For example, Nemeth ef al. (1979) found that low-molecular-weight N compounds in the form of amino acids-nitrogen were present in the filtrates. Analysis by EUF using high voltage (400V) and high temperature (80°C) have shown that soils rich in plant residues such as roots have a higher capacity to release reserve N. Studies made in Germany and Austria have indicated that the EUF technique gives a good prediction of potentially mineralizable N and soil N availability to upland crops (Nemeth, 1979; Wiklicky, 1982). The EUF technique, as used for determining the nutrients (including N) available in soils, is discussed in detail by Nemeth (1979). It has been found, using EUF, that NH4+ can be measured precisely in soils rich in clay and low in K, whereas it is relatively difficult to measure NH4+ in soils high in K using this technique. Wanasuria et al. (1980) found that EUF-extractable NH4+ was related to the N uptake and grain yield of rice in Philippine wetland rice soils. Mengel (1982) has suggested that the intensity factor is important for assessing NH4+ availability in wetland rice soils which in this respect resembles K availability.

VIII. PLANT ANALYSES Plant analysis is an accepted means of predicting fertilizer requirements and of diagnosing nutrient deficiencies based on critical nutrient values. Excellent reviews on the subject are available (Ulrich, 1952; Chapman, 1971; Jones and Steyn, 1973). Studies made at the University of California have clearly demonstrated that the total N content of the most recent fully developed leaf, also referred to as the “Y leaf,” of the rice plant at tillering is closely related to the grain yield of the rice (Reisenauer, 1978). It has also been indicated that the total N content of the leaf tissue is a better index than any nitrogen fraction. Tanaka and Yoshida (1970) made an extensive survey of rice fields in Asian countries and found that N deficiency was most notable in the rice plant. Their studies in the Philippines suggested that 2.5% was the critical concentration in the leaf blade of the rice plant at the tillering stage. Wallihan and Moomaw (1967) monitored the concentration of total N in the last four leaves formed in three varieties of rice grown under submerged condi-

NITROGEN AVAILABILITY INDEXES FOR RICE

443

tions in the field from the formation of flower primordia until flower emergence. They found that the concentration of total N changed with the leaf age and with the variety and growth stage of rice. Based on the stability of the nitrogen concentration and the sensitivity of the nitrogen supply, the blade of the next-tolast leaf (second leaf below the panicle), sampled at the time of flowering, was found to be suitable as the index leaf. No nitrate was detected in the leaf tissue. These observations corroborate earlier work done in Japan (Tanaka, 1961). Thenabadu (1972) made greenhouse and field studies, arriving at critical values of 2.2-2.3% N in the first and second most recently matured leaf of the rice plant sampled 67 days after transplantation. Krishnamoorthy et al. (1971) made an extensive survey of rice fields, and based on the analyses of different leaf tissues of the rice plants they concluded that the third leaf from top was suitable as the index leaf. They found that the critical concentration of total N in the index (third) leaf sampled between 3 and 6 weeks after transplantation was 1.4%. Hafez and Mikkelsen (1981) found that the Udy dye binding method (Udy, 1971), which is used for rapid estimation of protein in feeds, is also suitable for the analysis of rice plant leaf tissue for nitrogen content. Nitrogen content in leaf tissue determined by the dye binding method was highly correlated with the Kjeldahl nitrogen. This method, being simple and rapid, should aid in promoting plant tissue analyses for predicting nitrogen requirements of rice. It is thus clear that plant analysis gives a good index of the nitrogen status of the rice plant, and can be a useful guide for top-dressing requirements for fertilizer nitrogen if the tissue analysis is made at the tillering stage (e.g., see Chang, 1978). It is also clear that much standardization work on plant analyses is required for different cultivars grown in different environments in order to obtain critical and sufficient range concentrations for N by selecting an index plant part. Perhaps a combination of soil and plant tissue testing might provide a better index for nitrogen requirements of rice culture where indexes based on soil analysis alone may not be sufficient because nitrogen losses are high. Plant analysis has not been widely employed because often it is too late to alleviate any nitrogen deficiency discovered. The concentration of nitrogen in plants is greatly affected by genotype, the growth stage of the rice, and environmental conditions, and it is difficult to make a general recommendation even for different regions in a large country such as India.

IX. NITROGEN-SUPPLYING CAPACITY AND FERTILIZER RECOMMENDATIONS Soil tests have not been as successful for predicting the N-supplying capacities

of soils and for making N fertilizer recommendations as they have been for other

444

K. L. SAHRAWAT

nutrients such as P and K. The main problems seem to stem from the fact that nitrogen availability to plants is governed by several environmental and soil factors which are not taken into account whenempirical procedures for determining available N are used. However, tests for predicting the nitrogen-supplying capacities of soils are important for the efficient use of fertilizer N. It is always good to have a test which can provide a rough estimate of the pool of available N in soils so that fertilizer N can be applied to achieve a given yield of rice. This can be illustrated by an example taken from soil test crop-response project work in India (see Velayutham, 1979; Randhawa and Velayutham, 1982). The alkaline permanganate digestion method (Subbiah and Asija, 1956) was used for estimating the available N in soils for several crops, including rice (Ramamoorthy and Velayutham, 1976; Venkateswarlu, 1976). Based on the data obtained for rice grain, it was established that approximately 1.5-1.8 kg N/ha is needed for every 100 kg of grain in the alluvial soils of Delhi. From past experience it is known that about 26% of the available N (determined by the alkaline permanganate method) is taken up by the rice crop. For example, if the available N in a soil as measured by this method is 250 kg/ha, only about 65 kg of this pool will appear in the rice plants. Based on the yield target, the amount of fertilizer nitrogen (corrected for a use efficiency of 3040%) required for wetland rice can be calculated. This test gives a rough guide for making fertilizer N recommendations which should result in the more efficient use of fertilizer than in cases where the nitrogen-supplying capacity of the soil is not taken into considerdtion. Velayutham (1979) has summarized the Indian work on soil test crop response for rice. Briefly, the yield target and the required fertilizer nitrogen for achieving the yield target can be calculated from the following equations:

T = ns/(m-r) and F = rns/(m-r)

where T is the yield target in 100 kg/ha, n is the ratio of percentage conmbution from soil and fertilizer N, r is the N requirement in kg/ha of grain production, m is the ratio of N requirement and contribution from fertilizer N, s is the soil test N value in kg/ha, and F is the fertilizer nutrient rate in kg/ha. This scheme seems to provide a fair degree of approximation for efficient use of fertilizer N considering the N-supplying capacities of soils. Another example for fertilizer N recommendation based on the N-supplying capacity of rice soils is from the work done at the International Rice Research Institute in the Philippines by Ponnamperuma and his colleagues, who used the anaerobic incubation method to measure levels of available N. They sampled rice fields in 13 provinces in the Philippines and, based on the analysis of 483 soil samples, the available N in these soils ranged from 10 to 637 ppm. It was

NITROGEN AVAILABILlTY INDEXES FOR RICE

445

possible to separate low- and high-N-supplying capacities using the anaerobic incubation test (Ponnamperuma, 1978; Castro, 1979). Based on the results of potentially mineralizableN, Ponnamperuma (1978) formulated a rough guide for the fertilizer nitrogen requirements of a crop of 5 tons/ha for Philippine wetland rice soils. 1. Soils that needed no fertilizer nitrogen (available N > 150 ppm) 2. Soils that needed 50 kg N/ha at panicle primordia initiation (available N = 100-150 ppm) 3. Soils that needed about 50 kg N/ha at planting and about 50 kg N/ha again at panicle primordia initiation (available N=50-100 ppm)

It was further observed that at eight locations in a province, rice yields of 4.5-7 tons/ha were obtained on soils containing more than 155 ppm available N; zinc was applied but N was not (IRRI, 1974; Castro, 1979). These two examples illustrate the principle of basing the recommendation of fertilizer nitrogen needs of rice on the available N results, and it seems to be a step in the right direction. It is, however, recognized that these recommendationsmust be modified from time to time to reflect experience gained.

X. PERSPECTIVES The high cost of fertilizer nitrogen combined with the need for increased yields of rice has stimulated research on methods of using soil and fertilizer nitrogen efficiently. For the judicious and efficient use of fertilizer, a measure of the nitrogen-supplying capacity of soils is prerequisite because rice soils vary widely in their capacity to release ammonium nitrogen when submerged. Our fertilizer recommendations will be only as precise as our methods for measuring the amounts of available soil nitrogen. Because of the fact that one-half to two-thirds of the nitrogen used by the rice plant, even in well-fertilized paddies, comes from soil nitrogen through mineralization, research on methods for measuring the nitrogen-supplying capacities of wetland rice soils assumes still greater importance. For devising effective methods for measuring available nitrogen in soils, it is essential that the factors that affect mineralization and the availability of soil nitrogen to the rice plant be well understood. Soil and environmental factors that affect the mineralization of soil organic nitrogen are fairly well documented. However, with the present state of knowledge it is not possible to quantify (1) how the texture and the mineralogical makeup of a soil affect the release of nitrogen under submerged conditions, or (2) how the mineralization of soil organic nitrogen is affected by the presence of the rice plant. Attempts should be made in the future by using stepwise regression analyses to separate the effects of

446

K. L. SAHRAWAT

different soil characteristics on nitrogen mineralizationin flooded soils for different regions with due consideration of taxonomic criteria. It is envisaged that a knowledge of the environmental (such as temperature and moisture) and soil factors that affect mineralization will be useful in developing improved anaerobic incubation tests and, eventually, in developing models for measuring soil nitrogen mineralization rates and hence nitrogen-supplying capacity under field conditions, No data are presently available on the measurement of soil nitrogen mineralization rates in the field or on the comparative evaluation of mineralization rates as measured in the laboratory and in the field. Anaerobic incubation tests have shown potential as indexes for soil nitrogen availability to rice in a large number of greenhouse pot experiments and a few field experiments. Improved incubation tests should be devised by consideringthe climatic conditions of a region, especially soil temperature. The anaerobic incubation method is quite versatile in that it is very responsive and amenable to ranges in temperature. Research is also needed for the development of standardized methods for (1) determining the optimal time of soil sampling, (2) sampling soil (especially for submerged paddy fields after land-preparatory operations), (3) preparing soil samples for laboratory and greenhouse work, and fmally (4) use in laboratory and greenhouse expriments. Recent work has revived interest in using organic matter (as measured by organic C and total N) as the index of soil nitrogen availability to wetland rice (Ponnamperuma, 1978; Ponnamperuma and Sahrawat, 1978; Sahrawat, 198Oc, 1982d, 1983b). However, researchers usually have not obtained consistent results. It is known that both the quantity and the quality of organic matter affect the mineralization and availability of soil nitrogen. Little is known about the quality of organic matter (except for the C/N ratio) in wetland rice soils, which is indicated by our inability to answer simple questions even with our present knowledge about organic matter. For instance, (1) What are the criteria for characterizing the quality of organic matter in relation to its contribution to mineralizable nitrogen in submerged soils? and (2) How does the quality of organic matter affect mineralization and soil nitrogen availabilityto rice? More knowledge about the quality of organic matter, and especially about the fraction that contributes to soil mineralizable N pools, should improve our capability to use this simple index for predicting soil nitrogen availabilityto rice. Considerableresearch efforts have been devoted to the development of chemical indexes for assessing available soil nitrogen in soils, but these indexes have not been tested extensively for rice soils. Ideally, chemical indexes that extract the soil organic nitrogen fraction, which is the source of mineralizable nitrogen through the biological process, should be satisfactory in assessing the nitorgensupplying capacity of a soil. However, these conditions are usually not met and, in the case of many chemical indexes, their chemistry is not fully known. Alkaline pennanganate digestion is the most extensively used chemical method, especially in India, for assessing the availability of soil nitrogen to rice. Recent

NITROGEN AVAILABILITY INDEXES FOR RICE

447

knowledge about the chemistry of the method has shown that it exhibits a better potential for predicting soil nitrogen availability to wetland rice than to upland crops (Sahrawat and Burford, 1982). Simple models based on regression equations relating potentially mineralizable nitrogen (as measured by biological indexes) with chemical indexes and/or soil characteristics should be tested for their suitability to predict nitrogen availability to rice, since they have a sound basis (Stanford, 1977; Stanford and Smith, 1978; Sahrawat, 1983b). In view of the diverse soil and climatic conditions (where rice is grown) that affect soil nitrogen availability, it is quite probable that a single index of nitrogen availability will not find universal acceptance. Research on nitrogen availability indexes for wetland rice soils compared to arable soils is still in infancy, and it is hoped that this article will stimulate research in this area which holds considerable promise for the efficient use of fertilizer nitrogen as well as for devising conservative soil management and cultural practices to regulate soil nitrogen release in connection with nitrogen uptake by the rice plant. International cooperation is desirable for extensive testing of the promising indexes of nitrogen availability.

ACKNOWLEDGMENTS Part of this work was done at, and supported by, the International Rice Research Institute, Los Banos, Philippines. I am grateful to Dr.F. N. Ponnamperuma, Principal Soil Chemist, IRRI, for his valuable suggestions. I thank Dr.C. W. Hong for his helpful review of this article.

REFERENCES Acharya, C. N. 1935. Biochem. J . 29, 1116-1120. Association of Official Agricultural Chemists 1930. “Official and Tentative Methods of Analysis of the AOAC.” Washington, D.C. Ayanaba, A,, Tuckwell, S. B., and Jenkinson, D. S. 1976. Soil Biol. Biochem. 8, 519-523. Bajaj, J. C., and Hasan, R. 1978. Plum Soil 50, 707-710. Bajaj, J. C., and Singh, D. 1980. Commun. Soil Sci. Plant Anal. 11, 93-104. Bajaj, J. C., Gulati, M. L., and Tamahane, R. V. 1967. J. Indiun SOC. Soil Sci. 15, 29-33. Bonner, J. 1946. Bot. Gaz. (Chicago) 108, 267-279. Borthakur, H. P., and Mazumdar, N. N. 1968. J. Indiun SOC. Soil Sci. 16, 143-147. Boswell, F. C., Richer, A. C., and Casida, L. E. 1962. Soil Sci. SOC. Am. Proc. 26, 254-257. Bremner, J. M. 1965. In “Methods of Soil Analysis” (C. A. Balack, ed.),pp. 1324-1345. Am. SOC. Agron., Madison, Wisconsin. Broadbent, F. E. 1970. Soil Sci. 110, 19-23. Broadbent, F. E. 1978. In “Soils and Rice,” pp. 543-559. Int. Rice Res. Inst., Los Barios, Philippines. Broadbent, F. E. 1979. In “Nitrogen and Rice,” pp. 105-118. Int. Rice Res. Inst., Los Baftos, Philippines. Broadbent, F. E., and Mikkelsen, D. S. 1968. Agron. J . 60,674-677.

448

K. L. SAHRAWAT

Broadbent, F. E., and Reyes, 0. C. 1971. Soil Sci. 112, 200-205. Castro, R. U. 1979. Saturday Seminar, March 24, 1979 (mimeograph). Int. Rice Res. Inst., Los Ba~ios,Philippines. Chalk, P. M., and Waring, S. A. 1970. A m . J . Exp. Agric. Anim. Husb. 10, 306-312. Chang, S. C. 1978. In ‘‘Soilsand Rice,” pp. 521-541. Int. Rice Res. Inst., Los Baiios,Philippines. Chapman, H. D. 1971. Proc. Int. Symp. Soil Fert. Evaluation, New Delhi 1, 165-197. Cho, D. Y., and P o ~ m p e mF., N. 1971. Soil Sci. 112, 184-194. Chu, C. L. 1962. Tu Jang Hsueh Pa0 (Acra Pedol. Sin.) 10, 55-72. Clausen, C. R., Russelle, M. P., Flowerday, A. D. and Olson, R. A. 1981. Soil Sci. Soc.Am. J . 45, 1238- 1240. Craswell, E. T., and Waring, S. A. 1972a. Soil Biol. Biochem. 4,427-433. Craswell, E. T., and Waring, S. A. 1972b. Soil Biol. Biochem. 4,435-442. DeDatta, S . K., and Barker, R. 1978. In “Soils and Rice,” pp. 623-648. Int. Rice Res. Inst., Los Batios, Philippines. DeDatta, S. K., and Kerim, M.S.A.A.A. 1974. Soil Sci. Soc. Am. Proc. 38,515-518. Dei, Y., and Yamasaki, S. 1979. In “Nitrogen and Rice,” pp. 451-463. Int. Rice Res. Inst., Los Baiios, Philippines. Dolmat, M. T., Patrick, W. H., lr., and Peterson,F. I. 1980. Soil Sci. 129, 229-237. Fox, R. H., and F’iekielek, W. P. 1978a. Soil Sci. Soc. Am. J . 42, 747-750. Fox, R. H., and Piekielek, W. P. 1978b. Soil Sci. Soc.Am. J. 42, 751-753. Fried, M., and Dean, L. A. 1952. Soil Sci. 73,263-271. Gaballo, R. C. 1973. M.S. Thesis, Univ. of the Philippines, Los Bruios, Laguna. Gasser, J. K. R. 1969. In “Nitrogen and Soil Organic Matter,” pp. 71-77. Ministry of Agric. Food and Fisheries Tech. Bull. No. 15. HM Stationery Office, London. Ghosh, A. B., and Hasan, R. 1980. Fert. News 25, 19-24. Gopalswamy, A., Kumaraswamy, K., Durairaj Muthiah,N., and Nagalakshmi, K. 1973. Madrus Agric. J . 60, 707-709. Gotoh, S., and Onikura, Y. 1971. Soil Sci. Planr Nutr. (Tokyo)17, 1-8. Greenland, D. J. 1%5. Soils Fert. 28,415-424. Hafez, A. A. R., and Mikkelsen, D. S. 1981. Commun. Soil Sci. Planr Anal. 12, 61-69. Harada, T., Hayashi, R., and Chikamato, A. 1964. Nippon Dojo Hiryogaku Zasshi (J. Sci. Soil Manure Jpn.) 31, 21-24. Hauck, R. D. 1978. In “Nitrogen in the Environment” (D. R. Nielsen and I. G. Mac Donald, eds.), Vol. 1 , pp. 63-77. Academic Press, New York. Hauck, R.D. 1979. In “Nitrogen and Rice,” pp. 73-94. Int. Rice Res. Inst., Los Barios, Philippines. Hauck, R. D., and Bremner, J. M. 1976. Adv. Agron. 28, 219-266. Herlihy, M., and Sheehan, P. 1979. Plant Soil 53, 269-275. Hiura, K., Hattori, T., and Furusaka, C. 1976. Soil Sci. Planr Nu@. (Tokyo)22, 459-465. Houng, K. H. 1980. In “Increasing Nitrogen Efficiency for Rice Fertilization,” pp. 39-52. Food and Fertilizer Technology Center, Taipei, Taiwan. Hunter, A. S., and Carter, D. L. 1965. Soil Sci. 100, 112-117. IAEA 1970. Joint FAOlIAEA Division of Atomic Energy in Food and Agric. Tech. Rep. Ser. No. 108. Int. Atomic Energy Agency, Vienna. IAEA 1978. Joint FA0 IAEA Division of Atomic Energy in Food and Agric. Tech. Rep. No. 181. Int. Atomic Energy Agency, Vienna. IRRI 1964. “Annu. Rep. for 1963.” Int. Rice Res. Inst., Los Baiios, Philippines. IRRI 1965. “Annu. Rep. for 1964.” Int. Rice Res. Inst., Los Baiios, Philippines. IRRI 1967. “Annu. Rep. for 1966.” Int. Rice Res. Inst., Los Baiios, Philippines. IRRI 1973. “Annu. Rep. for 1972.” Int. Rice Res. Inst., Los Babs. Philippines.

NITROGEN AVAILABILITY INDEXES FOR RICE

449

IRRI 1974. “Annu. Rep. for 1973.” Int. Rice Res. Inst., Los Ba~ios,Philippines. IRRI 1978. “Annu. Rep. for 1977.” Int. Rice Res. Inst., Los Batios, Philippines. Jenkinson, D. S. 1968. J . Sci. FoodAgric., 19, 160-168. Jenkinson, D. S., and Powlson, D. S. 1976. Soil Biol. Biochem. 8, 209-213. Jones, C. A., Ratliff, L. F.,and Dyke, P. T. 1982. Commun. Soil Sci. Plant Anal. 13, 75-86. Jones, J. B., Jr., and Steyn, W. J. A. 1973. In “Soil Testing and Plant Analysis” (L. M. Walsh and J. D. Beaton, eds.), pp. 249-270. Soil Sci. Soc. Am. Inc., Madison, Wisconsin. Kai, H., Ahmad, Z., and Harada, T. 1969. Soil Sci. Plant Nutr. (Tokyo)15, 207-213. Kawaguchi, K., and Kyuma, K. 1969. “Lowland Rice Soils in Thailand,” p. 187. Center for Southeast Asian Studies, Kyoto Univ., Japan. Keeney, D. R., and Bremner, J. M. 1965. Soil Sci. SOC. Am. Proc. 31, 34-39. Koyama, T. 1971. Soil Sci. Pfant Nutr. (Tokyo) 17, 210-220. Koyama, T. 1981. Ferr. Res. 2, 261-278. Koyama, T., Chammek, C., and Niamsrichand, N. 1973. Trop. Agric. Res. Cent. Tech. Bull. No. 3, pp. 1-79. Min. Forestry and Agriculture, Tokyo. Krishnamoorthy, Ch., Venkateswarlu, J., Reddy, M. G. R., and Rao, G. V. M. M. 1971. Proc. Inr. Symp. Soil Fert. Evaluation, New Delhi 1, 954-967. Legg, J. O . , and Stanford, G. 1967. Soil Sci. SOC.Am. Proc. 31, 215-219. Lin, C. F., Chang, A. H., and Tseng, C. C. 1973. Nung Yeh Yen Chiu (J. Taiwan Agric. Res.) 22, 186-203.

Lopez, A. B., and Galvez, N. L. 1958. Philipp. Agric. 42, 281-290. Lynch, J. M., and Panting, L. M. 1982. J . Sci. Food Agric. 33, 249-252. MacLean, A. A. 1964. Nature (London) 203, 1307-1308. Mangaraja, S. C., Panda, D., Panda, N., and Misra, C. 1976. J . Res. Orissa Univ. Agric. Technol. 6, 43-52.

Mengel, K. 1982. Plant Soil 64, 129-138. Michrina, B. P., Fox, R. H., and Piekielek, W. P. 1981. Commun. Soil Sci. Plant AM^. 12, 5 19-535.

Michrina, B. P., Fox, R. H., and Piekielek, W. P. 1982. Plant Soil 64, 331-341. Moormann, F. R. 1978. In ”Soils and Rice,” pp. 255-272. Int. Rice Res. Inst., Los Barios, Philippines. Myers, R. J. K. 1975. Soil B i d Biochem. 7, 83-86. Nemeth, K. 1979. Adv. Agron. 31, 155-188. Nemeth, K., Makhdum, I. Q., Koch, K., and Beringer, H. 1979. Plant Soil 53, 445-453. Onikura, Y.,Yoshino, T.,and Maeda, K. 1975. Nippon Dojo Hit-yogakuZasshi ( J . Sci. Soil Manure Jpn.) 46,255-259. Patnaik, N. 1970. Indian Farming 20, 5-8. Patrick, W. H.,Jr., and Mahapatra, 1. C. 1968. Adv. Agron. 20, 323-359. Ponnamperuma, F. N. 1958. Int. Rice Comm. Newsl. 7 , 10-13. Ponnamperuma, F. N. 1965. In “Mineral Nutrition of Rice Plant,’’ pp. 295-328. Johns Hopkins Press, Baltimore, Maryland. Ponnampemma, F. N. 1972. Adv. Agron. 24, 29-96. Ponnamperuma, F. N. 1978. Paper presented at the Int. Rice Conf. on Rainfed Lowland Rice, April 1978. Int. Rice Res. Inst., Los Baiios, Philippines. Ponnamperuma, F. N., and Sahrawat, K. L. 1978. Agron. Abstr., p. 160. Powlson, D. S. 1980. J . Soil Sci. 31, 77-85. Powlson, D. S., and Jenkinson, D. S. 1981. J . Agric. Sci. 97, 713-721. Purvis, E. R., and Leo, W. M. M. 1961. J. Agric. Food Chem. 9, 15-17. Rajamamar, A., Ramamoorthy, S., Varadarajan, V., and Khadir, A. A. 1970. Madras Agric. J . 57, 139- 140.

450

K. L. SAHRAWAT

Raammoorthy, B., and Velayutham, M. 1976. In “Soil Fertility-Theory and Practice F’ractice” (J. S. Kanwar, ed.), pp. 156-201. Indian Council of Agric. Res., New Delhi. Ramamoorthy, B.,Pathak, V. N., and Bajaj, J. C. 1971. I n “Soil Survey and Soil Fertility Research in Asia and the Far East,” pp. 85-90. FAO-UNDP World Soil Resources Rep. No. 41. Ramanathan, K. M., and Krishnamoorthy, K. K. 1973. Madras Agric. J. 60, 720-723. Randhawa, N. S., and Velayutham, M. 1982. Fert. News 27, 35-64. Ranganathan, V.,Balasundaram, S. C., Govindaraj, K., and Samboomaraman, S. 1972. Int. Rice C O ~N m . d . 21, 18-26. Reddy, K. R.,and Patrick, W.H.,Jr. 1980. Planf Soil 57, 375-381. Beisenauer, H. M. ( 4 . ) 1978. Div. Agric. Sci. Univ. Calf. Bull. No. 1879. Rennie, D. A., and Fried, M. 1971. Proc. In?. Symp. Soil Fert. Evaluation, New Delhi 1,639-656. Robinson, J. B. D. 1975. Spec. Bull. No. 1. Common W. Bureau Soils, Harpenden, England. R o v h , A. D., and Greacen, E. L. 1957. Ausr. J. Agric. Res. 8,659-673. Sahrawat, K. L. 1978. Rep. Int. Rice Res. Inst. L a Barios, Philippines. Sahrawat, K. L. 1979. Fert. News 24, 38-48. Sahrawat, K. L. 198Oa. P h r Soil 55, 225-233. Sahrawat, K. L. 1980b. Planr Soil 57, 143-146. Sahrawat, K. L. 1980~.Planr Soil 55, 181-187. Sahrawat, K. L. 198Od. Agrochimica 24, 149-153. Sahrawat, K. L. 1981a. Soil Biol. Biochem. 13, 323-324. Sahrawat, K. L. 1981b. Commun. Soil Sci. Plant Anal. 12, 919-932. Sahrawat, K. L. 1982a. Oryza 19,43-46. Sahrawat, K. L. 1982b. Planr Soil 65, 281-286. Sahrawat, K. L. 1982c. Comun. Soil Sci. P h i Anal. 13,67-73. Sahrawat, K. L. 1982d. P h r Soil 65, 111-121. Sahrawat, K. L. 198%. Commun. Soil Sci. Plant Anal. 13, 363-377. Sahrawat, K. L. 1982f. Phnr Soil 69, 73-77. Sahrawat, K. L. 19828. Planf Soil 69, 283-285. Sahrawat, K. L. 1983a. Geoderma 29, 77-80. Sahrawat, K. L. 1983b. A m . J. Soil Res. 21, 133-138. Sahrawat, K. L., and Burford, I. R. 1982. Soil Sci. 133, 53-57. Sahrawat, K. L., andPonnampemma, F. N. 1978. Soil Sci. Soc. Am. J . 42, 282-283. Sakanoue, Y., and Matsubara, K. 1967. Nippon Dojo Hiryogaku Znsshi (J. Sci. Soil Manure Jpn.) 38370-73. Sanchez, P. A. 1973a. Soil Sci. 115, 149-158. Sanchez, P. A. 1973b. Soil Sci. 115, 303-308. Shiga, H., and Ventura, W. 1976. Soil Sci. Plant Nutr. (Tokyo) 22, 387-399. Shioixi, M. 1948. Bull. Agric. Expt. Stn., Ministry of Agric. (Japan) 64, 1-24. Shioiri, M., Aomine, S., Uno,Y., and Harada, T. 1941.Nippon Dojo Hiryogaku Zasshi (J. Sci. Soil Manure Jpn.) 15, 331-333. Sims, J. L., and Blackmon, B. G. 1967. Soil Sci. SOC. Am. Proc. 31, 676-680. Sims, J. L., Wells, J. P., and Tackett, D. L. 1%7. Soil Sci. SOC.Am. Proc. 31, 672-676. Singh, R. M., and Tripathi, B. R. 1970. J. Indian SOC.Soil Sci. 18, 313-318. Singh, T., and Pasricha, N. S. 1977. Indian J. Agric. Sci. 47, 460-464. Smith, J. L., Schnabel, R. R., McNeal, B. L., and Campbell, G. S. 1980. Soil Sci. SOC.Am. J. 44, 996-1000. Srivastava, 0.P. 1975. J. Indian Soc. Soil Sci. 23, 349-352. Stanford,G. 1968. Soil Sci. 106, 345-351. Stanford, G. 1977. Proc. Int. Semin. Soil Environ. Soil Fert. Management Intensive Agric., Tokyo, p ~ 412-418. .

NITROGEN AVAILABILITY INDEXES FOR RICE

45 1

Stanford, G . 1978. Soil Sci. 126, 244-253. Stanford, G., and DeMar, W. H. 1969. Soil Sci. 107, 203-205. Stanford, G., and Smith, S. J. 1972. Soil Sci. Soc. Am. Proc. 36, 465-472. Stanford, G., and Smith, S. J. 1976. Soil Sci. 122, 71-76. Stanford, G., and Smith, S. J. 1978. Soil Sci. 126, 210-218. Subbiah, B. V., and Asija, G. L. 1956. Curr. Sci. 25, 259-260. Subbiah, B. V., and Bajaj, J. C. 1962. Curr. Sci. 31, 196. Tamhane, R. V., and Subbiah, B. V. 1960. Inr. Rice Comm. Newsl. 9, 1-10. Tanaka, A. 1961. J. Fac. Agric. Hokkaido Univ. 51, 449-550. Tanaka, A., and Yoshida, S. 1970. Int. Rice Res. Inst., Tech. Bull. No. 10. Int. Rice Res. Inst., Los Ba~ios,Philippines. Thenabadu, M. W. 1972. Plant Soil 37, 41-48. Truog, E. 1954. Commer. Fert. 88, 72-73. Udy, D. C. 1971. Am. OilChem. SOC. J . 48, 29A-33A. Ulrich, A. 1952. Annu. Rev. Plant Physiol. 3, 207-228. Velayutham, M . 1979. Fer?. News. 24, 12-20. Venkateswarlu, J. 1976. In “Soil Fertility-Theory and Practice” (J. S. Kanwar, ed.), pp. 410-456. Indian Council of Agric. Res., New Delhi. Ventura, W., and Watanabe, I. 1978. Soil Sci. Planr Nu@. (Tokyo) 24,535-545. Wallihan, E. F., and Moomaw, J. C. 1967. Agron. J. 59, 473-474. Wanasuria, S., Mengel, K.,and DeDatta, S. K . 1980. Proc. Int. Symp. Applic. EUF Agric. Production, Budapest. Waring, S . A., and Bremner, J. M. 1964a. Nature (London) 201, 951-952. Waring, S. A,, and Bremner, J. M. 1964b. Nature (London) 202, 1141. Whitehead, D. C. 1981. J. Sci. FoodAgric. 32, 359-365. Whitehead, D. C., Barnes, R. J., and Morrison, J. 1981. J. Sci. Food Agric. 32, 211-218. Wiklicky, L. 1982. Planr Soil 64, 115-127. Yamane, I. 1967. Rep. Inst. Agric. Res. Tohoku Univ. 18, 87-108. Yanagisawa, M., and Takahashi, J. 1964. Nogyo Gijutsu Kenkyusho Hokoku B (Bull. Natl. Inst. Agric. Sci.) 14, 41-171. Yoshino, T., and Dei, Y. 1974. Jpn. Agric. Res. Q. 8, 137-141. Yoshino, T., and Dei, Y. 1977. Noji Shikenjo Kenkyu Hokoku ( J . Cent. Agric. Exp. Stn.) 25, 1-62.

This Page Intentionally Left Blank

Index A Absorbed-ion activity, 224-226 Acetochlor, 268, 276, 282, 284, 287 AD-2, 268, 280 AD-67, 268, 280, 305 Aegilops, 107 Aegilops bicornis, 167 Aegilops bicuncialis, 167 Aegilops caudata, 158, 167, 171, 172 Aegilops columnaris, 167 Aegilops comosa, 167 Aegilops crassa, 167 Aegilops cylindrica, 167 Aegilops heldreichii, 167 Aegilops juvanalis, 167 Aegilops kotschyi, 163-164, 167 Aegilops longissima, 167 Aegilops mutica, 167 Aegilops ovata, 159, 167, 172 Aegilops sharonensis, 167 Aegilops spelroides, 165, 167 Aegilops squarrosa, 163, 167 Aegilops triaristata, 167 Aegilops tricuncialis, 167 Aegilops umbellulata, 167 Aegilops uniaristata, 167 Aegilops variabilis, 163, 167 Aegilops ventricosa, 167 Aegilotricum, 159 Alachlor, 268, 275, 276, 282-287, 297, 302 Alexandergrass, 270, 283 Alfalfa, 270 Allium cepa, 270 Alnus, 44 Amaranthus retrojlexus, 270 Amiirol, 268 Ammonium production, 417-421 Anabaena, 3 Anhydrite, 72 Ascorbic acid, 288 Asulam, 268, 290 Atrazine, 268, 369 Avena fatua, 270 Avena sativa, 270 Azotobacterin, 291

6 Barban, 268, 271, 277, 282, 299 Barley, 105, 108, 110, 114-115, 122-123, 270, 278, 281, 282 Bamyardgrass, 270, 286 Bean common, 116, 127-129 field, 129-130, 270, 275, 276, 281 Beet, 270, 291 Bensulide, 268, 285 Bentazon, 369 Bentonite, 237, 248 Beta vulgaris. 270 Bluegrass, Kentucky, 270, 276-278 Boron, 37 Brachiaria plantaginea, 270 Bromacil, 369 Butachlor, 268, 276, 285, 287 Butam, 268, 278 Buthidazole, 268, 278 Butylate, 268, 277, 281, 297, 306

C Calcite, 72, 86 Captan, 369 Carbon distribution, 26 Carboxin, 280 Carrot, 270, 291 Casuarina, 44 Cation exchange, thermodynamics, 2 15 -264 CCC, 268 CDAA, 268, 279 CDEC, 268, 282 Ceanothus, 44 Celestite, 72 Centrosema pubscens, 25 CGA-43089, 268, 272, 274, 280, 283-285, 290, 293-301, 308-309 CGA-92194, 268, 272, 283, 286, 293-301 Chenopodium album, 270 Cblormequat, 268 Chlornitrofen, 268, 289 453

454

Index

Chloroacetanilide, 280 2-Chlomthyl phosphoric acid, 181 Chlorsulfuron, 268, 278, 282-283, 285, 286 Cisanilide, 268, 275, 278 Claviceps purpurea, 194 Clay mineral potksium exchange, 215-264 Clover, 33 white, 42, 43 Colletia. 44 Competition selection, 105-1 11 Comptonia, 44 Concep I, 297 Concep 11, 297 Coriaria. 44 Corn, 270 herbicide antidote, 274-289, 297, 299-310 Cotton, 103, 117, 120, 131-133, 275, 286 Crop evolution, annual, 97-143 Cucurbita foetidissima, 317, 319-331 Cycloate, 268, 277, 280, 282 Cysteine, 288 Cytoplasmic sterility, 157-169

Electro-ultrafiltration, 441-442 Eleusine, 283 Epronaz, 268, 278 Eptam, 369 EPTC, 269, 271-281, 284-287, 291, 297-308 Eradicane, 297-299 Ergot, 194 Erosion, reclamation, 38 Ethephon, 181, 182 Ethofumesate, 269, 279, 280, 285

F

Feldspar, 67-68, 82 Flax, 270 Flowering behavior, 183-185 Fluazifop-butyl, 269, 279, 285 Foxtail, green, 270, 275, 286 Frankia, 3

G D 2,4-D, 269, 270-272, 288, 369 Dalapon, 370 Datisca, 44 Daucus carota, 270 DCPA, 268, 278 DDCA, 292 Diallate, 268, 276, 282 Dicamba, 369 Diclopop-methyl, 268, 278, 283, 285 Diethatyl, 268, 276, 282, 284 Dimefuron, 268, 278 2,3-Dimercaptopropanol, 288 Dinoseb, 369 Diphenamid, 268, 275, 278, 280, 283 Diquat, 369 Discaria. 44 Diuron, 268, 286 Dowco, 221, 268, 278 Dryas, 44

E Echinochloa crus-gali, 270 Eleagnus, 44

Genetics cross-fertilization factors, 183- 190 cytoplasmic differentiation, 164- 166 cytoplasmic sterility, 157-169 fertility restoration, 169-180 hybrid seed production, 191-196 rice mutations, 383-413 Glomus, 35 Glycine mar, 130, 270 Glycine soja, I30 Glyphosate, 269 Goethite, 73 Gossypium barbadense, 131 Gossypium hirsutum, 131, 270 Gourd, buffalo, 317, 319-331 Gypsum, 72, 73

H H-26910, 269, 276, 282 Hackberry, 74 Halite, 72 Haynaldia villosa. 167 Hedysarum coronarium, 33 Helianthus annuus, 133

455

Index Herbicide antidote, 265-316 Heterosis, 147-155 HMI, 289 Hordeum disrichum, 127 Hordeum vulgare, 270 Hyppophae, 44

I Itchgrass, 270, 283

J Jarosite, 73 Johnsongrass, 270 Jojoba, 317, 332-346

Methylbromide, 19 Metoiachor, 269, 272-274, 276, 282-286, 290, 297, 302, 308-309 Metribuzin, 269 Mica, 240-243 weathered, 70-71, 82 Microorganism, proteolytic, 357-359 Millet, 270, 283 Molinate, 269, 277, 282 MON-4606, 268, 272, 285-287, 293-297 Montmorillonite, 237, 240-243, 373 Monuron, 369 Mycorrhiza, nitrogen-fixing, 1-54 Myrica, 44

N

K Kaolinite, 240-243, 246, 248

L Lambsquarters, 270, 286 Laterite, 73 Legume, mycorrhiza, 23-44 Lepidociocite, 73 Lignin, 375 Linum usirarissimum, 270 Linuron, 269, 283 Lolium perenne, 270 Lotus pedunculatus, 25 Lycopersicon esculetum, 270

M Maize, 107, 108, 115-116, 124-126 Malathion, 369 Maleic hydrazide, 181 MBR-18337, 269, 285, 369 MCPA, 269, 289 Medicago sariva, 28, 30, 33, 42, 43, 270 Mefluidide, 269, 279, 280 3-Methoxy-5-methylisoxazole, 289

NA See 1.8-naphthalic anhydride I ,8-Naphthalic anhydride, 268, 272, 292, 295-304, 309 Nicotiana tabacum, 270 Nitrofen, 269, 289 Nitrogen biological indexes, 421-428 chemical indexes, 428-435 fertilizer, 37 plant analysis, 442-443 protein transformation, 35 1-382 rice soil, 415-451 Nitrogen fixation, nodulating mycorrhiza, 1-54 Nosroc, 3 NP 5 5 , 269, 279

0 Oat, 115, 275, 277, 288, 299 wild, 270, 271, 291 Olivine, 7 1, 83 Onion, 270 Oryza fatua, 160, 163 Ovza glaberrima, 160, 163, 183 Oryza japonica, 123 Oryzaperennis, 158, 160, 161, 163 Oryza rufipogon, 160, 163 Oryza sativa, 183, 270 Oryza sativa f. spontanea, 157, 158, 159-161

456

Index

P Panicwn milliaceum, 270 Paraquat, 269, 288, 291, 369, 370 Parasponia, 44 Pasture improvement, 38 P a , 130-131 Pebulate, 269, 282 Pendimethalin, 269, 278 Perfhidone, 269, 279, 283, 285 Pesticide, mycorrhiza affect, 19 Phaseolus vulgaris, 109, 116, 127, 270 Phlewn pratense, 270 Phosphobacterin, 291 Phosphorus, mycorrhiza affect, 11-13, 16, 18,

28-29, 31-36 Photosynthesis, mycorrhiza affect, 17 Pigweed. 270, 286 Pisum sativum, 37, 130 Poa pratensis, 270 Pollen suppressant, chemical, 180-182 Pollinator distance, 188-190 PoRu!aca oleracea, 270 Potassium, soil and clay exchange, 215-264 Potato, 270 Propachlor, 269, 276 Propanil, 269, 289 Protect, 297 Protein transformation, soil, 351-382 Pueraria, 28 Purshia. 44 Purslane, 270, 286

Q Quartz,

weathered, 68-70, 82

R R-25788,268, 272, 274, 279-283, 289, 292, 310 R-28725,268,280, 305 R-29148,268, 280 Redroot, 270, 286 RH531, 181-182 RH2956, 181-182 Rhizobiwn, 3, 23-44 Rhizosphere, 358, 368, 376

Riboflavin, 288 Rice, 105, 107, 114, 123-124, 275 breeding, 145-214 herbicide antidote, 275-289 mutations and genetics, 383-413 nitrogen availability, 415-45 1 Robus, 44 Roettboelia exaltata, 270 Root exudation, 30 Rubefaction, 79-80 Rye, 270 Ryegrass, 270, 291

S

S-449,268, 271 Sand dune stabilization, 38-39 Screen, 297 SD-58525,269, 285 SD-91779,269, 285 Secale cereale, 167,270 Setaria viridis, 270 Sethoxydim, 269, 279, 283, 285 Simetryne, 289 Simmondsia chinensis, 317, 332-346 Soil, protein transformation, 35 1-382 Sodium l-@-chlorophenyl)-l,2-dihydro-4, 6-dimethyl-2-oxonicotinate, 181 Sodium methyl arsenate, 182 Soil nitrogen-fixing mycorrhiza, 1-54 potassium exchange, 215-264 protein transformation, 35 1-382 rice nitrogen, 415-451 submicroscopic examination, 55 -96 Solanum tuberosum, 270 Sorghum, 108, 116, 126-127 herbicide antidote, 272-290, 297,299-302,

308-309 Sorghum, 286 Sorghum bicolor, 270 Sorghum halepense, 270 Sorghum sudanense, 270 Soybean, 33, 34, 106, 116, 130,286 Stylosanthes guyanensis, 25 Sugarbeet, 286, 291 Sunflower, 116, 133-134

SUTAN + , 297, 298 Swep, 269, 289

457

Index

T 2,4,6-T, 269, 271-272, 288, 369 Temperature, soil, 364-365 Terbutol, 269, 278 Thenardite, 73 Thiobencarb, 269, 277 Thiocarbamate, 279, 286, 287, 301, 306 Thiram, 369 Tillering, 108 Timothy, 270, 276, 278 Tobacco, 270, 288 a-Tocopherol, 288 Tomato, 270, 271, 275, 278, 281, 283 Triallate, 269, 276, 282, 297 Trifluralin, 269, 275, 278, 280, 283, 288 Trifolium repens, 25, 41 Trifolium subsrerraneum, 37 Triricum, 107 Triricum aesrivum, 163, 165, 167 Triricum araraticum, 163, 167 Triticum boeoricum, 162, 166, 167 Triricum dicoccoides, 167 Triticum dicoccoides, var. nudiglumis, 163,167 Triricum diococcum, 167 Triticum durum, 159, 167 Triticum macha, 167, 171 Triticum monococcum, 167 Triticum spelta var. duhamelianum, 164, 176 Triricum fimopheevi, 161, 162, 164, 165, 167, 171, 172, 197 Triricum zhukovskyi, 167

V Verdasan, 369 Vermiculite, 240 -243 VERNAM + , 297, 298 Vernolate, 269, 276, 282, 297 Viciafaba, 26, 129 Vitamin D. 288

W

Wheat, 105, 107, 108, 110, 114, 121-122 breeding, 145-2 14 herbicide antidote, 271, 275-278, 282-283, 291, 297

X Xylachlor, 269, 284

Y

Yield, comparison, 119-120

U UBI-S734, 269, 285 Ustilago tritice, 195

Zea mays, 270 See also Corn Zinc methyl arsenate, 182

This Page Intentionally Left Blank