Advances in Agronomy, Volume 22

ADVANCES IN

AGRONOMY VOLUME 22

CONTRIBUTORS TO THIS VOLUME

K. P. BARLEY J. R. BROWNELL J. w.CARY FRANCISE. CLARK J. P. EVENSON D. L. GRUNES PAUL R. HENSON E. M. HUTTON D. S. MCINTYRE H. F. MAYLAND M. M. MORTLAND ELDORA. PAUL D. L. PLUCKNETT W. G. SANFORD ROBERTR. SEANEY P. R. STOUT

ADVANCES IN

AGRONOMY Prepared under the Auspices of the

AMERICAN SOCIETY

OF

AGRONOMY

VOLUME 22 Edited by N. C. BRADY Roberts Hall, Cornell University, lthaca, New York

ADVISORY BOARD W. L. COLVILLE W. A. RANEY I. J . JOHNSON J . R. RUNKLES R. B. MUSGRAVE G. W. THOMAS 1970

ACADEMIC PRESS

New York and London

COPYRIGHT 0 1970, BY ACADEMIC PRESS,INC. ALL RIGHTS RESERVED.

NO PART OF THIS BOOK MAY BE REPRODUCED IN ANY FORM, BY PHOTOSTAT, MICROFILM, RETRIEVAL SYSTEM, OR ANY OTHER MEANS, WITHOUT WRITTEN PERMISSION FROM THE PUBLISHERS.

ACADEMIC PRESS, INC. 1 1 1 Fifth Avenue, New York, New York 10003

United Kingdom Edition published by ACADEMIC PRESS, INC. (LONDON) LTD. Berkeley Square House, London W I X 6BA

LIBRARYOF CONGRESS CATALOG CARDNUMBER:50-5598

PRINTED IN THE UNITED STATES O F AMERICA

CONTENTS CONTRIBUTORS TO

VOLUME 22 ....................................................................

.................................................................................................... PREFACE

ix xi

TROPICAL PASTURES

E . M. HUTTON 1. Introduction

II. 111. IV . V. VI . VII . VII1 . IX . X. XI . XI1 . XI11. XIV .

xv .

XVI .

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

Climate and Potential for Improved Pastures in the Tropics ................. Role of Plant Introduction ................................................................ Tropical Legumes ... ............................... ................................. Temperate Legumes d in the Subtropics .......................................... Nodulation and Nitrogen Fixation of Tropical Legumes .......................... Legume Nutrition Relative to Tropical Pasture Development ................... Undesirable Compounds in Tropical Legumes ....................................... Physiology of Tropical Legumes .......................................................... Breeding and Genetics of the Main Legumes ......................................... The Main Grasse .................................................... umes ............................................ Phosphorus and Nitrogen Fertilization of Grass Breeding and Genetics of Tropical Grasses Beef Production from Legume-Based Tropi ........................... Summary and Conclusions ........................................... References ...........................................................................

CLAY-ORGANIC

2 3 5 8 21

23 27 33 36 39 44 50 53 57 60 65 66

COMPLEXES A N D INTERACTIONS M . M . MORTLAND

I.

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

I 1 . Bonding Mechanisms in Clay-Organic Complexes .................................. I11 . Nature of Some Clay-Organic Complexes and Reactions ........................ IV . Nature and Importance of Some Clay-Organic Complexes in Soils and Sediments ........................................................................................ V . Conclusions .....................................................................................

................................................................................. V

75 77 95 108

113 I14

vi

CONTENTS

BIRDSFOOT TREFOIL ROBERTR. SEANEYA N D PAULR. HENSON

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

I. Introduction ..................

11. Morphology ................ ................................ 111. Physiology ..... ......................................................

IV. Culture .......... .................................................... V. Utilization ............................................................. V1. Genetics and Cytology............................................

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

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

120 122

125 127 139 143

147 153

THE CONFIGURATION OF THE ROOT SYSTEM IN RELATION TO NUTRIENT UPTAKE

K. P. BARLEY I. Introduction .....................................................................................

I59

11. Geometrical Description of the Root System ......................................... 111. Nutrient Transference in the Soil ......................................................... IV. Physiological Conditions Governing Uptake ..........................................

161

V. The Influence of Configuration on Uptake.. ........................................... VI. Conclusions .......................................... .................... References ................. ................ .....................

167 17 1 177 197 198

FROST AND CHILLING INJURY TO GROWING PLANTS

H. F. MAYLAND AND J. W. CARY I. Introduction Ill. IV. V. VI.

................................................................ f Protein Structure .....................................

Cold Lability of Enzymes ................................................................... Membrane Composition and Permeability ................... Protection from Freezing.. ........................................ Conclusions ...........................................................

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

203 206 2 15

vii

CONTENTS

THE PLATINUM MICROELECTRODE METHOD FOR SOIL AERATION MEASUREMENT

D . S. MC ~ NT YRE

I . The Method ..................................................................................... I1. Electrochemistry ............................................................................... 111 . Physical Effects of Electrode Insertion ................................................. IV . Models and Microelectrode Response .................................................. V . Conclusions ..................................................................................... V1 . O2 Flux and Plant Response ............................................................... VI1 . Summary ......................................................................................... References .......................................................................................

235 241 266 268 276 278 281 281

RATOON CROPPING D . L . PLUCKNETT. J . P . EVENSON. AND W . G . SANFORD 1. Introduction ..................................................................................... I1 . Genetic Aspects ............................................................................... 111. Botanical and Physiological Considerations .............................. IV Ratooning and Environmental Factors ..................................... V . Soil Relationships ............................................................................. VI . Pests and Disease ............................................................................. VII . Management .................................................................................... VIII . Future Outlook ................................................................................. References ................ ....................................................

.

286 293 294 300 305 312 317 323 326

GRASS TETANY OF RUMINANTS

.

D . L. GRUNES.P . R . STOUT.AND J . R BROWNELL 1. Introduction ..................................................................................... 11. Incidence. Climate, and Season ...........................................................

I11 . Symptoms of Animals ........................................................................ Iv . soils ............................................................................................... V . Forage ............................................................................................

332 332 336 340 342

...

CONTENTS

Vlll

.

v1 Animal Factors in Hypomagnesemia ........................................ VII . Treatment of Affected Animals ................................. VIII . Prevention of Grass Tetany ............................................. IX Magnesium Deficiency in Humans ....................................................... ..... .................................................... X . Summary References .......................................................................................

.

354 363 363 368 369 369

THE MICROFLORA OF GRASSLAND

FRANCIS E. CLARKA N D ELDORA . PAUL 1. Introduction ....................................... ..... I1 . The Microflora of the Living Plant ....................................................... I11 . The Microflora of Grassland Litter ...................................................... IV . The Microflora of Grassland Soils ....................................................... V . Biomass and Bioactivity Measurements ................................................ VI . The Humic Component of Grassland Soil ............................................. VI1 . Nitrogen Transformations in Grassland Soils ......................................... References .......................................................................................

375 376 385 395 403 409 416 426

Author Index .............................................................................................. Subject Index .............................................................................................

431 459

CONTRIBUTORS TO VOLUME 22 Numbers in parentheses indicate the pages on which the authors’ contributions begin.

K. P. BARLEY(159), Waite Agricultural Research Institute, Glen Osmond, South Australia J . R. BROWNELL* (331), Kearney Foundation of Soil Science, Davis, Calif0rn ia J. W. CARY(203), Snake River Conservation Research Center, Soil and Water Conservation Research Division, Agricultural Research Service, U.S. Department ofAgriculture, Kimberly, Idaho FRANCISE. CLARK (373, U. S. Department OfAgriculture, Fort Collins, Colorado J . P. EVENSON (28% University of Queensland, Brisbane, Australia D. L. GRUNES(33 11, U. S. Plant, Soil and Nutrition Laboratory, U.S. Department of Agriculture, Ithaca, New, York PAUL R. HENSON( 1 19), U. S. Department of Agriculture, Beltsville, Maryland E. M. HUTTON( l ) , Commonwealth ScientiJic and Industrial Research Organization, Division of Tropical Pastures, Cunningham Laboratory, Brisbane, Australia D. S . MCINTYRE(235), Commonwealth ScientiJic and Industrial Resea reh organization, Canberra,Australia H. F. MAYLAND (203), Snake River Conservation Research Center, Soil and Water Conservation Research Division, Agricultural Research Service, U.S. Department of Agriculture, Kimberly, Idaho M. M. MORTLAND( 7 3 , Michigan State University, East Lansing, Michigan ELDORA. PAUL (373, The University of Saskatchewan, Saskatoon, Canada D. L. PLUCKNETT (285), College of Tropical Agriculture, University of Hawaii, Honolulu, Hawaii W . G. SANFORD (285), College of Tropical Agriculture, University of Hawaii, Honolulu, Hawaii ROBERTR. SEANEY ( 1 19), Cornell University, Ithaca, New York P. R. STOUT(331), Kearney Foundation of Soil Science, Davis, California *Present address: Fresno State College, Fresno, California

ix

This Page Intentionally Left Blank

PREFACE

In keeping with one of the basic objectives ofAdvances in Agronomy, this volume covers a variety of subjects of concern to crop and soil scientists. Likewise, the sixteen North American and Australian authors who contributed these papers have a breadth of backgrounds and interests. They have covered topics vital to both public and scientific concerns. The continuing world-wide attention to tropical agriculture is recognized in this volume. An analysis is made in one chapter of the potential, and the problems, of ratoon cropping -a practice of considerable importance with tropical crops such as bananas, sugar cane, and pineapples. An extensive review of research to improve and utilize tropical pastures relates to the potential for forage production in the tropics. The problems involved in the production and utilization of a number of important tropical forage species are also emphasized. While one is impressed with the research contributions, the opportunities and .problems ahead present challenges which dwarf these accomplishments of the past. Two other chapters relate to the animal industry as well as crop production. One is a review of the research on birdsfoot trefoil (Lorus cornicularus L.), an important pasture and forage legume of especial value in the North Central and Northeastern states. The second deals with grass tetany, a ruminant animal malady associated with forages low in magnesium and often relatively high in nitrogen and potassium. Factors affecting the animal disorders, probable reasons for them, and therapeutic techniques are reviewed. A review of recent research findings on frost and chilling damage to plants includes evidence as to the mechanism of damage, and information on methods of preventing or reducing this damage. In another chapter the relationship of geometric configuration of roots to nutrient uptake is examined. Research is reviewed which identifies the conditions under which nutrient transfer in the soil and root system configuration 1im it nutrient uptake. Growing public concern for environmental quality has forced a realistic consideration of the part tne soil might play as a sink for various kinds of wastes. Included are pesticides and other exotic chemicals, sewage, and similar wastes. One chapter is addressed to a review of the reaction of soils (clays) with organic compounds. The increasing specificity of our knowledge is impressive but the need for greater understanding of these reactions in soil is even more obvious. The soil environment for plant roots and other living organisms is considered in two chapters. A critical review of the platinum electrode xi

PREFACE

xii

method for measuring soil aeration casts some doubt on the interpretation of earlier findings, especially those wherein so-called critical values of oxygen flux for root and plant growth were established. The microflora of grasslands and grassland soils is discussed in one chapter. These contributions will be especially helpful as background for those concerned with ecosystems and how man is modifying them.

ADVANCES IN

AGRONOMY VOLUME 22

This Page Intentionally Left Blank

TROPICAL PASTURES

.

E M. Hutton Commonwealth Scientific and Industrial Research Organization. Division of Tropical Pastures. Cunningham laboratory. Erisbane. Australia

Page

I . Introduction .................................................................................... I1 . Climate and Potential for Improved Pastures in the Tropics ..................... 111. Role of Plant Introduction ..................................................................

IV . Tropical Legumes .............................................................................

.............................................................................. ecies ....................................................................... The Desmodiums ....................................................................... Glycine ( G . wightti) .................................................................... Leucaena ( L. leucocephala) ...... ...... Miles Lotononis ( L. bainesii) ........................................................ Dolichos and Vigna Species .........................................................

C. D. E. F. G. H . Centro (Cenfrosemapubescens) ................................................... 1. Calopo (Calopogonium mucunoides) and Puero (Pueraria

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

V . Temperate Legumes Used in the Subtropics ....................................... A . Hunter River Lucerne .............. ...................................... ......................... B. Barrel Medic ................................. C . White Clover .......................... VI . Nodulation and Nitrogen Fixation of Tropical Legumes .......................... VII . Legume Nutrition Relative to Tropical Pasture Development .................. VIII . Undesirable Compounds in Tropical Legumes ...................................... A . Estrogens and Substances Causing Bloat and Milk Taints ..... B . Mimosine ................................................................................. C . Indospicine ................................................................................

IX .

.............................................................................. pica1 Legumes .................................

A . Temperature and Growth in Several Tropical Legum

B. Townsville Stylo ........................................................................ C . Glycine wightii ... .............................................................. D Siratro ...................................................................................... E. African Trifoliums ............................ ................................. Breeding and Genetics of the Main Legumes ................................. A Breeding Systems .............................. ................................. B. Townsville Stylo ........................................................................ C . Siratro ...................................................................................... D . Desmodiums ..............................................................................

.

X.

.

1

2 3

5 8 9 12 14 16 17 19 20 21 21

21 21 22 22 23 27 33 33 33 34 36 36 36 37 37 38 39 39 39 40 41 41

2

E. M. HUTTON

............ ,f............. ......... Leucaena ................. Indigofera ................. Lucerne ......................................... ................. Main Grasses ................. .................. A. Brachiaria ................. ................. B. Cenchrus ...................... ........................................ C. Chloris .................... ........................................ D. Cynodon ................... ........................................ E. Digitaria .................................................................................... F. Melinis ......................... ........................................ G. Panicum ..................................... ........................................ H. Paspalums ......................... ........................................ I. Pennisetum .............................................................................. J. Setaria ...................................................................................... K. Sorghum ................................................................................... L. Urochloa ................................................................ ......... XII. Feeding Value of Grasses versus Legumes .......................................... ~

F. G. H. XI. The

Phosphorus .................. B. Nitrogen ........................ ................ XIV. Breeding and Genetics of Tropical Grasses A. Setaria ............................................. A.

C. Coastal Bermudagrass .

................ A. Wet Tropics

C. Humid Subtropics ....................................................................... D. Subhumid Subtropics ........................ XVI. Summary and Conclusions .........................................,.......................

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

1.

42 42 43 43 44 44 45 45 46 46 47 47 48 48 49 49 50 50 53 53 54 51 58 58 59 59 60 61 61 63 64 65 66

Introduction

Improvement of tropical grasslands was neglected for many years because most of the areas involved are in developing countries with pressing sociological and economic problems. Also, it had been concluded (Whyte et al., 1953) that it would be very difficult to introduce a legume into tropical grasslands and establish legume-based pastures as productive as those in temperate areas. It was left to grassland scientists in several countries, including Hawaii (Takahashi, 1956), Jamaica (Motta, 19561, the Congo (Germain and Scaut, 1960),and Australia (J. G. Davies, 1960), to pioneer research on legume-based pastures for the tropics. This work was intensified and expanded mainly at experiment stations in north-

TROPICAL PASTURES

3

eastern Australia and Hawaii. At some of the centers significant progress has been made in solving the problems of tropical legume-grass pastures. The staff of the C.S.I.R.O. Cunningham Laboratory in Brisbane summarized their findings (C.A.B. Bull. 47, 1964) for the benefit of pasture workers in other tropical areas. More recently, in 1966, W. Davies and Skidmore edited a book, “Tropical Pastures,” which presents the modern approach to tropical pastures. Now that reliable tropical legumes and grasses are available as well as knowledge of their fertilization, management, and productivity, considerable interest has been stimulated in the tropics in the use of improved tropical pastures for animal production. A rapid expansion in the areas planted with tropical pasture species can be envisaged with a concomitant increase in production of beef and milk so that the “protein drought” in many countries will gradually disappear. The development of pasture ecosystems for the wide range of environments throughout the tropics will present many new problems. This will require increasing research activity in many countries on pasture and animal production and a swing away from the preoccupation with veterinary and animal health problems. II. Climate and Potential for Improved Pastures in the Tropics

The tropics and its agricultural development are discussed by Phillips (196 1) and Webster and Wilson ( 1 966). Wet equatorial climates with constant heat, rainfall, and humidity occur mainly within 5” to 1O”N and S of the equator over a large area of South America and in West Africa, Malaysia, Indonesia, the Philippines, New Guinea, and various Pacific Islands. Annual rainfall is usually 80- 120 inches, but a higher maximum is obtained in a number of places. The crops grown include rubber, oil palm, banana, coffee, coconut, cocoa, and rice, and experiments and some commercial plantings have shown that productive pastures can be established in the wet tropics. Between the low latitude zones of wet equatorial climate and the Tropics of Cancer and Capricorn (23 M”N and S) there are extensive areas with annual rainfalls of 20-80 inches and with a tropical monsoon climate in which there is an alternation of wet and dry seasons. These areas occur in South America, West and Central Africa, India, and in countries of Southeast Asia including Burma, Thailand, Laos, Cambodia, and Vietnam, and also in northern Australia. Four climatic zones can be distinguished in these monsoonal areas as follows: annual rainfall of 40-80 inches in two rainy seasons with short dry seasons; annual rainfall of 25-50 inches in two short rainy seasons with pronounced dry seasons;

4

E. M. HUTTON

annual rainfall of 30-50 inches in one fairly long rainy season with a long dry season; and annual rainfall of 20-40 inches with one short rainy season and a long dry season. Rainfall reliability decreases at the lower rainfalls. In the wetter parts, perennial tropical crops are grown as in the wet tropics. Where dry seasons are well defined, annual crops like rice, cotton, and maize are important, and in the driest areas sorghum, bulrush millet, and peanuts are grown. The tropical monsoonal areas have considerable potential for improved pastures and cattle production, as shown by some of the results obtained in different regions (vide Norman and Arndt, 1959; Shaw, 196 1 ; Stobbs, 1969a). The moist tropical climate extends from the Tropics of Cancer and Capricorn to about latitudes 35"N and S, respectively, to give zones with a humid subtropical climate (McIntyre, 1966; C. L. White et al., 1968). These encompass southeastern United States, the middle Orient, a zone including southern Brazil, southern Paraguay, northern Argentina, and Uruguay, and areas along the eastern coasts of South Africa and Australia. These humid subtropical transition zones have a rather variable rainfall (Griffiths, 1959), which is predominantly in summer but with a winter increment. Usually referred to simply as the subtropics, they are important agriculturally and grow the hardier tropical crops and, as shown in Australia (J. G. Davies and Eyles, 1965), have considerable potential for the development of cattle pastures based on tropical legumes and grasses. The subtropics of eastern Australia is described by Coaldrake (1 964) as the region where plants may be subjected to water stress or surplus any month of the year and where water, rather than energy, is more likely to limit plant production. Coaldrake also commented on the drastic reduction of herbage quality caused by a relatively small number of rather mild winter frosts in these areas. Their sudden onset allows no hardening of plants, and in any case most tropical pasture plants originated in frostfree areas and their above-ground portions, though usually not the roots, are killed by frost. Most of the recent cultivars that are having an impact on tropical pasture development in Australia and elsewhere were selected and developed at research centers in Australia's subtropics (J. G. Davies and Eyles, 1968). Thus, in Australia, the subtropics and Tropics form a continuum in which the factors governing pasture production vary in degree rather than in principle. As outlined, the tropical zones where there is distinct potential for increased cattle production on improved pastures include the wet equatorial, the tropical monsoonal, and the humid subtropical covering approximately 27% of the world's area. Of the moist tropics, 33% is wet tropics, 49% monsoonal, and 18% subtropics. The arid tropical zone,

TROPICAL PASTURES

5

with its deserts, the dry subtropical with xerophytic scrub, herbs, and grasses, and the semiarid tropical of grassland steppes are zones where the vegetation can be improved with range management techniques and not usually with the use of sown species. As pointed out by Hutton (1968a), 60% of the world’s cattle are in the moist tropics, where 10% is cropped, 20% is pasture, 35% is forest, and a third is wasteland. Forest areas and wasteland are often in hilly areas and attempts to crop them result in soil erosion. With the use of the new tropical pasture species and fertilizer, idle uplands and unimproved native pastures can be improved markedly with significant effects on beef and milk production. Ill. Role o f Plant Introduction

The vital role of plant introduction in the development of pastures has been recognized for many years in the United States and Australia, both of which are deficient in promising native pasture species in their temperate and also tropical areas. Although the tropics of Latin America and Africa are rich in indigenous legumes and grasses with potential value for improved pastures, Kenya is the only country in these continents where a study has been made of the important species of native grasslands (Edwards and Bogdan, 195 I). The U.S.D.A. plant introduction services, which commenced in 1898 (vide Yearbook of Agriculture, 1962) and which are now vested in the New Crops Research Branch, have not been particularly concerned with introducing tropical pasture plants. However, pangolagrass (Digitaria decumbens), which was introduced as vegetative material from South Africa in 1935 by the then U.S.D.A. Division of Plant Exploration and Introduction, has made a major impact on tropical pasture improvement (Oakes, 1960). Interest in Digitaria was stimulated and led to the collection by Oakes (1965) of an extensive range of species and ecotypes within the genus. The breeding of coastal bermudagrass (Burton, 1954), which has significantly increased pasture production in the southeastern United States, was achieved through the use of two tall-growing South African introductions of Cynodon dactylon. U.S.D.A. plant introduction work has also assembled species in the genus Paspalum, such as P. notaturn, and P . dilatatum and obtained species for the Hawaiian Experiment Station in the legume genus Desmodium. As a result of the work on indigenous grasses in Kenya (Edwards and Bogdan, 1951) the tropical world has obtained valuable ecotypes of a number of grasses including Rhodes (Chloris gayana), buffel (Cenchrus ciliaris), star (Cynodon dactylon), molasses (Melinis minutiflora),guinea

6

E. M. HUTTON

(Panicum maximum), kikuyu (Pennisetum clandestinum), and Setaria (Setaria sphacelata).The indigenous legumes were also studied in Kenya, and valuable ecotypes of Glycine wightii (formerly G.juvanicu)(Bogdan, 1966a) and species of Dolichos and Vigna have been made available to pasture scientists in other countries. Kenya workers have studied both native and introduced pasture plants at the Grassland Research Station, Kitale, since I95 1. The cultivated varieties of herbage plants resulting from this work were described by Bogdan ( 1965). The Kenyan example could well be emulated in the countries of Central and South America, where there is a wealth of indigenous legumes waiting to be collected, classified, and assessed. Latin America is the source of a few important grasses, notably in the genus Paspalum, but does not possess as valuable a grass flora as Africa. Although Africa has an extensive range of native legumes, it has not as yet contributed as many promising pasture legumes as Latin America. There is an awakening interest in many tropical countries in introduction of tropical legumes and grasses that have shown promise elsewhere. Most of them still show a reluctance to investigate their own rich native flora for promising pasture plants. As pointed out by Hutton (1970), Australia is singularly deficient in indigenous legumes and grasses that can be used as the basis for improved pastures and increased animal production. As a result, there has been a continuing interest in Australia in pasture plant introduction, which commenced on a random basis about the 1880’s and became organized in 1930 with the establishment of the Plant Introduction Section of the C.S.I.R. Division of Plant Industry (McTaggart, 1942). Up to the present, 50,000 introductlons have been brought into Australia and 6 I % of these are pasture and forage species. Much of the current pasture development in the Australian Tropics is based on the chance annual introduction Townsville stylo (Stylosanthes humilis) which was recognized at Townsville around 1900 and known formerly as Townsville lucerne. Introduction of tropical pasture species has been a major aim of C.S.I.R.O. plant introduction work since its inception, and a large number of legume and grass accessions from tropical countries have been evaluated over the years. The selection of grasses adapted to northern Australia has been relatively easy, whereas obtaining adapted legumes has proved difficult, particularly for the subtropics, where rainfall is variable and frosts can occur. Since the turn of the century, the Australian wet tropics of about 4 million acres of northeastern coastal country between Mossman and Mackay has had adapted introductions of tropical grasses such as guinea,

TROPICAL PASTURES

7

molasses, and para (Brachiaria mutica). Schofield ( 1941) eventually obtained successful legumes for this area including stylo (Stylosanthes guyanensis), centro (Centrosema pubescens), puero (Pueraria phaseoloides), and calopo (Calopogonium mucunoides). Much of the plant introduction work for northern Australia over the last thirty years has aimed at obtaining legumes and grasses for pasture development in the extensive tropical monsoonal and humid subtropical areas between latitudes 30"sand 11"sand comprising about 260 million acres (J. G. Davies and Eyles, 1965). Miles ( 1 949) made distinct progress with this problem by evaluating an extensive range of introduced legumes and grasses in central coastal Queensland from 1936 to 1946. He showed that the low mineral and protein status of the native pastures could be raised by perennial legumes in a number of genera including Arachis, Centrosema, Desmodium, Glycine, Indigofera, and Stylosanthes. The most promising grass introductions included ecotypes of Chloris gayana, Cenchrus ciliaris, Digitaria sp., Panicum maximum, Paspalum notatum, Setaria sphacelata, and Urochloa sp. Miles' results (1 949) stimulated the first work in overseas plant exploration by Australia. Hartley (1949) joined a U.S.D.A. expedition to subtropical South America and collected mainly ecotypes of species in the genera Arachis, Desmodium, Stylosanthes, and Paspalum. From these introductions the cultivars Oxley Fine-stem stylo and Hartley plicatulum (Paspalum plicatulum) (Bryan and Shaw, 1964) have been selected. Another ten important overseas collections of pasture plants have been made by Australians in tropical monsoonal and humid subtropical areas during the period 1952-1968 (Hutton, 1970). A range of material was collected, particularly in the legume genera Centrosema, Desmodium, Glycine, Phaseolus, and Stylosanthes, and the grass genera Cenchrus, Panicum, Paspalum, Setaria, and Urochloa. Only the introductions from J. F. Miles' visits to South Africa and east and west Africa in 1952 have been fully evaluated. These have yielded Miles Lotononis (L. bainesii) (Bryan, 1961), Rongai Dolichos lablab (W,ilsonand Murtagh, 1962), and Samford Rhodes grass. R. J. Jones' collections ( 1 964) of Setaria sphacelata from East Africa have already produced the frost-tolerant cultivar Narok setaria, and it is anticipated that further promising lines will come from these. The systematic exploration in 1965 of legumes and a few of the grasses by Williams ( 1966) in the main states of Brazil, and in Bolivia, Paraguay, and northern Argentina, has substantiated that these areas are rich in indigenous species potentially valuable as tropical and subtropical pasture plants. Williams found annual types of stylo similar to Townsville stylo in a number of regions.

E. M. HUTTON

8

Due to the progress made on plant exploration and introduction for the tropics, it is now more difficult to find native legumes and grasses which are superior to existing pasture cultivars. This is no reason to curtail this activity, as only a fraction of the almost unlimited variation in the indigenous flora of countries like South and Central America and Africa has been investigated. With the advances in knowledge of the feeding value of pastures relative to species and management, variants of present cultivars or even new species could be required and may well be found among the native plants of these and other countries. In any case, the pasture plant breeder needs a continual flow of new genetic material which can be obtained only through plant exploration and introduction. IV.

Tropical Legumes

Australian research centered in Queensland is now in the forefront on the introduction, selection, and development of legumes for tropical pastures. This has resulted from the realization that legume-based pasture is the most economical method for the development of the cattle industry in the vast unused coastal and subcoastal areas of northern Australia (J. G. Davies and Eyles, 1965). Maintenance of around 40% of a phosphate-responsive legume in a tropical pasture is the cheapest way to provide nitrogen for the pasture and grazing animal (Hutton, 1968b). In this section the origin and agronomic features of the principal tropical legumes commercialized in Australia will be discussed. For their detailed descriptions, see Barnard (1967). Some, like the drought-resistant Townsville stylo and siratro (Phaseolus utropurpureus), are adapted to a wide range of conditions in northern Australia, whereas those including centro, glycine, the desmodiums, and Miles lotononis, are less drought tolerant and more restricted in their adaptation. With the exception of Miles lotononis, aboveground growth of all the tropicals is killed by frost, which is a constant feature of the subtropics in winter. However, the perennial crown and root systems survive and regenerate unless subjected to intense and repeated frosting. As larger areas of the different tropical legumes are established, they will become hosts to various diseases and pests which could affect persistence of some cultivars. Fortunately the main legumes do not appear to be affected by rootknot nematodes, and siratro is highly field resistant (Hutton and Beall, 1957). The root-damaging Amnemus weevil (Amnemus quadrituberculutus) is a serious pest of glycine and the desmodiums on the north coast of New South Wales, whereas Miles lotononis and lucerne are re-

TROPICAL PASTURES

9

sistant and siratro is seldom attacked severely (Braithwaite, 1967; Mears, 1967). Other native weevils have damaged a number of the legumes in north Queensland. The bean fly (Melanagromyzaphaseoli) can seriously damage Murray lathyroides (Phaseolus lathyroides) throughout the season but siratro, affected only in the seedling stage, can be protected by seed treatment with dieldrin (R. J. Jones, 1965). The viruslike disease “legume little leaf” due to a mycoplasma (Bowyer et al., 1969) affects a number of the legumes and under relatively dry conditions markedly reduces stands of the desmodiums and Miles lotononis, and causes some loss in siratro. In high rainfall areas, varying amounts of defoliation is caused by Rhizoctonia solani in several legumes, particularly siratro. Commercial seed production of the different legume cultivars is increasing in Australia and several other countries. In Kenya Desmodium intortum, silverleaf desmodium ( D . uncinatum), Glycine wightii, Stylosanthes guyanensis, Dolichos lablab, and Trifolium semipilosum are sold whereas in South Africa and Brazil G . wightii is the one usually harvested for sale. A. STYLOS 1. Townsville Stylo (S. humilis)

The history and potential of the annual Townsville stylo is summarized by Humphreys ( 1967), who noted that its natural spread is confined to the north of Western Australia, the Northern Territory, and Queensland. It has thin, fibrous stems and narrow elongated and pointed leaves, and forms a dense stand under favorable conditions. Flowers are yellow and inconspicuous and arranged in a short compressed spike. The brown pods are hooked, have two segments, but usually contain only one true seed. The hooked pods cluster into small balls and comprise the commercial seed, the yields of which range from 400 to 700 Ib per acre. As a result of D. F. Cameron’s work (1965) with naturalized ecotypes, three vigorous upright cultivars, Gordon (late), Lawson (midseason), and Paterson (early black seeded) are being commercialized. Townsville stylo (Fig. 1) is grown widely in Australia, particularly from latitudes 1 1“ S to 24”sand where the annual rainfall is between 25 and 7 0 inches. It flourishes on poor sandy soils but does not establish readily on deep cracking clays and in waterlogged areas. Seed is sown at 3-4 lb per acre in conjunction with 1 cwt superphosphate per acre by aerial or ground methods into grazed open woodlands or cleared and cultivated areas. It is susceptible to shading from vigorous associate grasses, so it

10

E. M. HUTTON

FIG. I . Townsville stylo, Narayen Research Station, near Mundubbera, Queensland.

TROPICAL PASTURES

11

should be kept well grazed. Where there is strong grass competition, as in the Northern Territory, Stocker and Sturtz ( 1 966) and H . P. Miller ( 1967) obtained satisfactory establishment of Townsville stylo by sowing it immediately after a burn early in the wet season which rapidly destroyed growing native grasses. A fall of rain of an inch or more in early summer will cause rapid germination of seed. The young seedlings are quite drought resistant, but a prolonged dry period will cause their death. The high actual seed yields of 800-1000 Ib of seed per acre and the high percentage of hard seed ensure the persistence of Townsville stylo. Norman and Arndt ( 1959) in the Northern Territory and Shaw ( 196 1) in central Queensland proved the value of Townsville stylo in beef production and so paved the way for its widespread use in northern Australia. It is proving successful in the tropical monsoonal areas of Southeast Asia, the Philippines, Brazil, Central America, and East Africa as well as in southern Florida (Kretschmer, 1965). 2 . Schojeld Stylo (Stylosanthes guyanensis)

This perennial, which originates from Brazil (Schofield, 1941), is naturalized in the wet Tropics of northeastern Australia. It is tall and branched with hairy stems, narrow pointed leaflets, and compact spikes of small yellow flowers. Seeding is profuse, and the small brown singleseeded pods shed on ripening, which makes mechanical harvesting difficult. Schofield stylo grows in frost-free conditions in northern Australia where annual rainfalls are 35-160 inches, summer temperatures are high, and soils remain moist. It is sown at 2 Ib per acre, usually after cultivation, grows well on both poor and fertile soils, and is compatible with tropical grasses like guinea and molasses. Heavy grazing and fire will soon reduce stands of this legume. A few recent introductions of S . guyanensis from central and South America are superior to Schofield stylo, as they are prostrate and branch vigorously from the base under heavy grazing. Various ecotypes of perennial stylo are being increasingly grown in east Africa and will no doubt be grown more widely in the Tropics.

3. Oxley Fine-Stem Stylo This stylo came from Paraguay and was one of Hartley's collections (1949). It was selected as a result of Shaw's work (1967a,b) on granitic spear grass soils of southern Queensland. It is semiprostrate, well

12

E. M . HUTTON

branched, and has an underground crown. The leaflets are narrow and pointed, and the small yellow flowers are in compact spikes. The brown single-seeded pods fall very rapidly as they ripen, which makes harvesting particularly difficult. Oxley fine-stem stylo is frost and drought tolerant and is adapted to the ligher soils of Australia’s subtropics in annual rainfalls of 28-50 inches. It is sown at 1-2 Ib per acre and is compatible with buffel and Rhodes grasses.

B. PHASEOLUS SPECIES I . Siratro ( P . atropurpureus) Siratro (Fig. 2) is a perennial bred from two Mexican ecotypes of P. atropurpureus (Hutton, 1962). I t has deeply penetrating swollen roots and a high level of drought resistance. If the soil is moist for an extended

FIG.2. Siratro growing at Sarnford Pasture Research Station, near Brisbane, Queensland.

TROPICAL PASTURES

13

period, the trailing pubescent stems root at random along their length. Its twining habit allows it to reach the light in dense pastures. Leaves have three ovate leaflets with dense silvery hairs on the lower surface. The evidence indicates that siratro is a short-day plant with abundant flowering and seeding occurring when vegetative growth is checked by dry weather or cooler conditions in autumn. The relatively large flowers are in a raceme and are deep red, and age to dark purple before withering. The pods are narrow and cylindrical and they shatter as they ripen from the raceme base. Actual seed yields are around 800 Ib per acre under favorable conditions but commercial yields are 100- 150 Ib per acre because of shattering. Seed is usually scarified to reduce the percentage of hard seed. Siratro grows from latitude 1 1 "S to 30"s in Australia in annual rainfalls from 25 to 70 inches and thrives on a variety of soils, particularly those well drained. Seed is sown at 2-3 Ib per acre on well grazed native grassland or cultivated areas, and establishment is rapid under favorable conditions. It is compatible with a range of grasses including Rhodes, buffel, green panic (Panicum maximum var. trichoglume), guinea, and Nandi setaria and its growth is rapid at the height of the wet season. Siratro, due to its perenniality, high actual seed yield, and quick regeneration from seed, is usually quite persistent provided year round stocking rates are kept within reasonable limits of a cattle beast to 1.3-3.0 acres. Siratro is promising in tropical monsoonal areas in the countries of Central and South America and eastern Africa. In eastern Africa its growth is restricted at elevations above sea level of 4500 feet or more. Good growth of siratro is also reported from New Guinea, Fiji, Philippines, Rajasthan (Patil et al., 1967), southern Florida (Kretschmer, 1966), and Southeast Asia.

2 . Murray lathyroides ( P . lathyroides) Murray lathyroides was first reported in the Brisbane district toward the end of the last century (Bailey and Tennison-Woods, 1879) and is now naturalized along the coast of northeastern Australia. Ecotypes of P . futhyroides are widely distributed and grow wild in a number of countries of Southeast Asia, Central and South America, and Africa and also in New Guinea, the Pacific and Hawaiian Islands, southern Florida, West Indies, Philippines, and India. They vary from upright to prostrate and have either sparse or strong basal branching. Murray lathyroides is an erect vigorous annual or biennial with some branching and smooth

14

E.

M.

HUTTON

lanceolate leaflets. The conspicuous deep pink to red flowers are in a raceme, and the narrow cylindrical pods shatter as they ripen from the raceme base. Murray lathyroides was developed by Paltridge ( 1942) because of its vigor, palatability, and high protein content. It grows well in pasture mixtures, on a range of soils in annual rainfalls of greater than 30 inches and is seeded into cultivated land at 2-3 Ib per acre. Periodic cultivation is necessary to ensure its regeneration from fallen seed. An important attribute is its ability to persist on heavy-textured waterlogged soils.

C. THEDESMODIUMS The role of the desmodiums as pasture plants has been summarized by Bryan (1966, 1969). Greenleaf (D.intorturn) and silverleaf ( D . uncinaturn) are the two commercialized in Australia. A number of D. canurn and D . sandwicense ecotypes have been introduced, but none have shown real promise as yet, although kaimi Spanish clover (D. canurn) has proved of value in Hawaii (Hosaka, 1945; Younge et al., 1964). D . gyruides, a shrub used to prepare land for cocoa in Fiji and elsewhere, has potential for forage as it persists under grazing and produces green leaf in the dry season in frost-free areas. The strongly stoloniferous D. heterophyllurn, naturalized in a number of tropical countries, has performed well in association with aggressive grasses like pangola in the wet Tropics of north Queensland.

1 . Greenleaf (D.intorturn) Indigenous ecotypes of D. intorturn are common in Central America and Brazil (Williams, 1966), and greenleaf is a mixture of three similar introductions from El Salvador and Guatemala. It is rather a coarse trailing perennial which roots along the pubescent stems under moist conditions and has a fibrous root system (Fig. 3). The deep green, rounded leaflets often have a reddish brown to purple flecking on the upper surface. Short days induce flowering and the small deep lilac to pink flowers are in compact terminal and axillary racemes. The small narrow and recurved seed pods are segmented and have hooked hairs that cause them to adhere to clothing and animals. Seed yields of 100-120 Ib per acre are being obtained under irrigation in the dry season. Greenleaf thrives on a variety of soils in the coastal areas of northern Australia where annual rainfall is 40 inches or more and the dry season not too severe, as in the Northern Territory. It is not particularly drought resistant and grows well in moist elevated areas as the Atherton Table-

TROPICAL PASTURES

FIG. 3. Greenleaf desmodium, Samford Pasture Research Station, near Brisbane, Queensland.

land of north Queensland. It is sown at 2 Ib per acre in pasture mixtures into cultivated land, and establishment is often slow because of retarded

16

E. M. HUTTON

nodulation. Once established, it grows rapidly under warm moist conditions and is compatible with most of the tropical grasses. Leaves and shoots are quite palatable to cattle (Bryan, 1966), and it thins out at stocking rates in excess of a beast to the acre. Types of D . intorturn similar to greenleaf have performed well in trials in Tanzania and Uganda (Naveh and Anderson, 1967; Stobbs, 1969a) and a number of other tropical countries. Near the equator it grows from sea level to elevations of 6500 feet, so is very adaptable.

2. Silverleaf ( D . uncinatum) Silverleaf desmodium was introduced from Brazil in 1944, and more recently Williams (1966) collected similar ecotypes from there. It is a trailing perennial with thin, ovate, hairy leaflets which have a broad irregular silver band along the midrib. Moistness induces rooting along the pubescent stems, and swollen as well as fibrous roots are produced. Flowering occurs in short days, and the paired lilac to mauve flowers are borne in open terminal and axillary racemes. The segmented sickleshaped pods have hooked hairs, and the seeds are flat and larger than those of greenleaf. Commercial seed yields of 200-300 Ib per acre are obtained. Silverleaf is not as hardy as greenleaf and thrives only in moist coastal areas of northeastern Australia where annual rainfall is 40 inches or more. Growth is restricted by high summer temperatures so elevated areas with cooler nights often provide it with a better environment. It grows on a variety of soils and is sown in mixtures at 2 lb per acre on cultivated land. Establishment is usually rapid and it combines well with grasses like the setarias, panicums, and paspalums. Palatability of leaves and shoots is high (Bryan, 1966), and it is susceptible to overgrazing. Silverleaf desmodium has proved promising in several countries, notably in East Africa (Bogdan, 1965; Naveh and Anderson, 1967). More recent results have indicated that it will be replaced there by greenleaf or a similar type of D.intorturn. It is of interest that at Palmerston North, New Zealand, silverleaf was the only tropical legume surviving in the third season from a range which was sown. D. GLYCINE ( G . wightii) The perennial G . wightii is mainly indigenous to Africa, although there are some Southeast Asian forms. Descriptions are given by Verboom ( 1965) of the five main types in Zambia and by Bogdan ( 1966a) of the five distinct African types he assembled in his plots at Kitale in Kenya.

TROPICAL PASTURES

17

Bogden ( 1 966a) found that G. wightii occurred to an altitude of 7000 feet in Kenya and that the three outstanding varieties agronomically in his collection were Kenyan in origin. Glycine is a promising legume in a number of other countries including Brazil (Neme, 1958) and Tanzania (Naveh and Anderson, 1967). The three Australian cultivars of glycine have been selected from a series of African introductions. Clarence came from the South African Transvaal (Murtagh and Wilson, 1962), Cooper from Tanzania (Edye and Kiers, 1966), and Tinaroo from Kenya (Kyneur, 1960). They are trailing and twining, and root along stems in contact with the soil; Tinaroo and Cooper are more stoloniferous than Clarence. All have a strong taproot, and Tinaroo and Cooper have finer stems and are more branching than Clarence. They have ovate leaflets; those of Tinaroo are thin, smooth, and bright green, those of Cooper are softly hairy and ash green, and those of Clarence are coarse and hairy with brown pigmentation. Cooper and Clarence are early flowering, but Tinaroo is very late. The racemes are many flowered, and the flowers are small and white with violet streaks on the standards, which are pink tinged except in Tinaroo. The pods are straight and flattened, and commercial seed yields of 200-300 lb per acre are obtained. Glycine thrives only in northeastern Australia on kraznozems and black self-mulching clay soils where annual rainfall is 30-70 inches. It is not adapted to the long, hot dry seasons of the far north of Australia nor to the large areas of solodic and podzolic soils along the coast. Glycine is sown in mixtures at 2-4 lb per acre in cultivated soil. Although slow to establish in the first year, it combines well with a range of grasses if carefully managed during this period. E. LEUCAENA( L . leucocephala) Both Oakes (1968) and Gray ( 1 968) have reviewed the relevant facts concerning the origin, agronomy, and use of the promising high-protein forage tree, leucaena. It is indigenous to Central America and has spread to the Caribbean islands, the Philippines, and Southeast Asia, to the Hawaiian and other Pacific islands including Fiji and New Guinea, and to the northern Australian coast from Darwin around to Brisbane. No doubt its use as a shade tree for cocoa and coffee accelerated its spread. Interest in its use as a forage for cattle was first engendered by Takahashi and Ripperton ( 1 949), who obtained yields from it of 8-9 tons of dry matter per acre containing 1-1.5 tons of protein in a 50-60 inch rainfall in Hawaii. Since then it has been used for cattle fattening in the Hawaiian

18

E. M. HUTTON

Islands, and experimental plantings have been made in a number of countries including Australia. Dijkman ( 1950) has described its use in the control of soil erosion. Its commercial exploitation as a forage has been inhibited by the concern that its mimosine content (Yoshida, 1944) adversely affects reproduction despite Hawaiian Agricultural Experiment Station experiments (Anonymous, 1948) which showed no reduction in calving rate of dairy cows fed leucaena continuously. Hutton and Gray (1959) grouped leucaena introductions into three types. These are the short, bushy, early Hawaiian type, the tall, late, sparsely branched El Salvador type, and the tall, late, strongly branched Peru type. The low-yielding Hawaiian type is the one naturalized in northeastern Australia and around Darwin. El Salvador and Peru are the cultivars commercialized in Australia, but Peru is preferred because it has the highest dry matter and protein yields (Hutton and Bonner, 1960) and the best growth habit for grazing (Fig. 4). Leucaena is a deep-rooted tree with a high level of drought resistance. It has smooth bipinnate leaves with narrow leaflets, small white flowers in a globose head, brown strap-

FIG.4. A productive stand of Peru leucaena, Samford Research Station, near Brisbdne, Queensland.

TROPICAL PASTURES

19

shaped pods, and very hard glossy brown seeds. At present commercial seed is harvested by hand, and 300 Ib per acre is easily obtained. Leucaena grows successfully in northern Australia on a range of well drained soils where the annual rainfall is 30 inches or more and frosts are light or absent. It is one of the few tropical legumes adapted to highly calcareous soils (Oakes and Skov, 1967), and indications are that it is responsive to lime as well as to phosphate. Satisfactory establishment requires that seed be planted at 3-4 Ib per acre in rows 8-10 feet apart into well prepared soil and that weeds be controlled by interrow cultivation or herbicides. Seed needs to be treated by immersion for 4 minutes in water at 80°C to ensure a good germination. When the young trees are growing vigorously they are no longer susceptible to competition, and an associate grass like guinea, green panic, setaria, or pangola can then be planted between the rows. The resultant two-level pasture can be heavily stocked, as the pliable stems of leucaena are not damaged by grazing. If leucaena grows out of reach of the animal it can be controlled by a heavy mechanical slasher. Leucaena gives high liveweight gains and no deleterious effects in cattle if grazed in rotation with non-leucaena pastures (see Section VIII, A, p. 33). F. MILESLOTONONIS (L. bainesii) This perennial cultivar came from seed collected in the Worcester Veldt Reserve of South Africa by J. F. Miles in 1952. The agronomic features of Miles lotononis are described by Bryan (1961), who commented on the need to introduce further L. bainesii ecotypes and increase variability available for improvement in its adaptability. L. bainesii is indigenous over several million acres of the northern Transvaal and Rhodesia, and a number of variants have been collected (Smith, 1969). As yet, Miles lotononis has shown adaptation to only limited areas in various tropical countries. Miles lotononis is slender, prostrate, and stoloniferous and roots at the nodes only in sandy and self-mulching soils. It has strong taproots and a degree of drought resistance. Leaves are digitate trifoliate with smooth, narrow, and pointed leaflets. The small yellow flowers are in a raceme or open head and the small brown pods release the minute seeds from a basal opening. Commercial seed yields are 50-60 Ib per acre. Miles lotononis is adapted to friable soils, particularly sandy types, in humid subtropical areas where annual rainfall is 35-40 inches or more. It is the only frost-resistant summer legume, but grows slowly at low temperatures. The seed requires inoculation (Norris, 1958) and is sown

20

E. M. HUTTON

at 0.5-1.0 lb per acre on cultivated soil. Initially the seedlings are slow growing and sparse, but later they grow into a dense cover under warm moist conditions. It forms balanced mixtures with a number of grasses including pangola provided constant grazing pressure is maintained. Miles lotononis does not persist on tight soils. At times it almost disappears from sandy soils but usually recolonizes these.

G. DOLICHOS A N D VIGNASPECIES The genera Dolichos and Vigna are closely related and contain a high proportion of quick growing annual and biennial types. In dolichos there are three cultivars, Rongai lablab (D. lablab), Archer axillaris (D. axilh i s ) , and Leichhardt dolichos (D. uniforus). Dalrymple vigna (V. luteola) is the only vigna cultivar at present. Rongai lablab (Wilson and Murtagh, 1962) was collected in Kenya by J. F. Miles. It is a tall vigorous and well branched biennial with broad ovate leaflets and relatively large white flowers, and becomes rampant and trailing under favorable conditions. Seed yields of 500-600 lb per acre are obtained. It thrives on a variety of soils where the rainfall is 25 inches or more and is replacing cowpeas (Vigna sinensis) in many areas. The relatively large seed is sown at 10-15 lb per acre. Rongai lablab is drought and cold tolerant and is used as a forage crop and as a preparation for sowing perennial pastures. At maturity the stems are woody and unpalatable to animals. Archer axillaris from Kenya is a trailing and twining short-term perennial with glossy leaves and small greenish yellow flowers. Commercial seed yields are 200-500 Ib per acre. It is sown at 2-4 Ib per acre in well drained frost-free areas where annual rainfall is 40 inches or more and is valued for its rapid growth in both spring and summer (Luck, 1965). Leichhardt dolichos is a short-season twining annual with softly tomentose leaves and small greenish yellow flowers. Sown at about 10 lb per acre, it grows on a variety of soils where annual rainfall is 25-45 inches and produces heavy seed yields of 1000 Ib/acre or more held in the pod for dry season grazing (Staples, 1967). Dalrymple vigna, a biennial, was collected in Costa Rica by W. W. Bryan. It is trailing with shiny, bright green leaflets and roots along stems in contact with the soil. The flowers are large and yellow, and commercial seed yields of around 150 Ib per acre are obtained. Sown at 5-10 Ib per acre, it grows on most soils where annual rainfall is 35-40 inches or more and will grow under waterlogged or saline conditions. It is very palatable, and at Samford Pasture Research Station persisted only two years under grazing (R. J. Jones et al., 1967).

TROPICAL PASTURES

21

H. CENTRO (Centrosema pubescens) Centro is indigenous to South America and is an important perennial in wet tropical pastures around the world. It has a strong taproot and is trailing and twining with some nodal rooting. Leaflets are bright green and shiny, and it has large mauve flowers. Seed yields of 300-500 Ib per acre are obtained. It is seeded at 4 Ib per acre and thrives in combination with grasses like para and guinea on most soils in the wet tropics where annual rainfall is 50 inches or more. On the north Queensland coast, centrobased pastures have persisted and maintained high animal productivity with proper management and fertilization. A. W. Moore (1962) in Nigeria and Bruce (1965) in Australia have shown that centro significantly increases soil fertility.

I. CALOPO(Calopogonium mucunoides) A N D PUERO(Pueraria phaseoloides) Calopo is native in Central and South America and is naturalized throughout the wet tropics. It is a trailing short-term perennial with oval leaflets invested with brown hairs. The flowers are small and pale blue, and seed yields are 200-300 lb per acre. Sown at 2-3 lb per acre, it is a good pioneer on most soils and produces a thick stand not very palatable to cattle. It is not favored for general use. Puero, or tropical kudzu, is native to Malaysia and is a perennial. It is trailing and twining, with broad, dark green leaves and long, vigorous runners. The flowers are of medium size and white with a violet blotch. Seed yields are 300-400 Ib per acre. It is a pioneer legume in the wet Tropics and is sown at 2-3 Ib per acre. Puero is very palatable and disappears under heavy grazing. V. Temperate Legumes Used in the Subtropics

In Australia’s subtropics with its increment of winter rain, the winterand spring-growing lucerne (Medicago sativa), barrel medic ( M . truncatula), and white clover (Trifolium repens) are of increasing importance in pastures as complementary species to tropicals. Their frost resistance ensures that their quality .is not reduced by the light frosts that often occur in winter. Inclusion of temperates in subtropical pastures could be tried elsewhere, as it maximizes the period during the year when pasture of good feeding value is available. A. HUNTER RIVERLUCERNE

Hunter River, an old lucerne cultivar (Whittet, 1923), is the main type grown, and Christian and Shaw ( 1 952) first showed its importance in the

22

E. M. HUTTON

subtropics. 't Mannetje ( 1967) found that inoculation and lime-pelleting seed of Hunter River enabled it to grow in combination with tropical species in the extensive areas of acid granitic soils of the speargrass zone south of the Tropic. Hunter River lucerne is already important on the fertile brigalow soils and can be grown as far west as the 18-inch isohyet in southern Queensland. It is an erect perennial with wedge-shaped leaflets and violet flowers and has a deep taproot and the crown at the surface. Hunter River does not persist well in the subtropics because of intolerance to continuous grazing and rotting of old crowns in the moist summer. It may eventually be replaced by more productive cultivars like Siro Peruvian or African (D. G. Cameron, 1968) or with a lucerne bred specifically for subtropical pastures (Bray, 1967).

B. BARRELMEDIC The cultivars Cyprus and Jemalong have proved valuable on both the solodics and clay loams of the brigalow zone of southern Queensland (J. S. Russell, 1970). They are semiprostrate annuals with hairy leaflets, and each Jemalong leaflet has a prominent purple-brown blotch whereas the leaflets of only a few plants of Cyprus have this blotch. A number of runners are produced, and the flowers are small and yellow; Cyprus is about a month earlier than Jemalong. Profuse production of the barrelshaped pods occurs, and seed yields of 300-500 lb per acre are obtained. These cultivars usually persist after introduction into an area because of a high percentage of hard seed. Regeneration from seed is good under favorable conditions. C. WHITECLOVER A number of cultivars of white clover, including Grasslands Huia, Irrigation, and Ladino, have been successful in subtropical pastures along the moist coast from northern New South Wales to Maryborough in southern Queensland (Andrew and Bryan, 1955). They are also used in lower latitudes on moist tableland areas such as the Atherton Tableland of north Queensland. White clover is a valuable pasture component because of its high digestibility and feeding value. lt is mainly annual in habit in northeastern Australia, but abundant seed production ensures its persistence. Kenya white clover (Trifolium semipilosum), which has a strong perennial root system and greater drought tolerance than white clover ('t Mannetje, 1964), could replace white clover in some areas when its establishment problems are overcome.

TROPICAL PASTURES

23

VI. Nodulation and Nitrogen Fixation of Tropical Legumes

Rapid and effective nodulation of legumes is essential for their establishment and vigorous growth and for significant amounts of nitrogen and protein to be added to the pasture ecosystem. Henzell (1962) found that nitrogen fixation of several tropical legumes was substantial although not quite at the level of white clover and lucerne. By inclusion of centro in giant star grass (Cynodon dactylon), A. W. Moore ( 1 962) obtained an increase of 250 lb of nitrogen per acre foot per annum under the pasture and raised nitrogen content of the grass from 1.8 to 2.4%. R. J. Jones et al. (1967) showed that the increase in total nitrogen yield of pasture from inclusion of tropical legumes was directly related to legume yield. It is estimated (Henzell, 1968) that nitrogen fixation by tropical legumes in northern Australia is 20-260 Ib an acre a year depending on level of legume yield. At Beerwah, near Brisbane, silverleaf desmodium increased soil nitrogen by 30-77 Ib an acre a year relative to amount of applied superphosphate (Henzell et al., 1966). Addition of relatively high levels of superphosphate to Townsville stylo in the field raised the nitrogen yield to 3.3 times and the dry matter yield to 2.4 times those at nil superphosphate (Shaw et al., 1966). Supporting glasshouse and controlled environment experiments showed that sulfur produced the major increase in nitrogen fixation, and phosphorus the major increase in dry weight. Field nodulation of some tropical legumes was studied by Whiteman and Lulham ( 1970) and Whiteman ( 1 970a,b). In undefoliated plots, both silverleaf desmodium and siratro had marked summer (February-April) peaks in nodule and plant dry weight. Cutting and grazing reduced nodule number per plant in silverleaf desmodium and mean weight per nodule in siratro. In greenleaf, desmodium and siratro effects of defoliation were not evident for 18 days, but then reduction in nodule weight per plant was related to severity of the initial defoliation. Seasonal buildup and decline of the nodule population in three desmodiums were not related to onset of flowering. Peak nodulation occurred 3 months before flowering in greenleaf and 1 month before in silverleaf. Most of the tropical legumes will grow and nodulate freely on acid soils: this is partly because they have the ability to extract calcium from these low-calcium soils (Andrew and Norris, 1961). This is of considerable significance in the economic development of pastures in the extensive areas of solodic and podzolic soils low in calcium in the moist tropics and subtropics of northern Australia. Costly lime applications are unnecessary for vigorous legume growth, and only superphosphate with or without molybdenum is usually needed for high production of legume-

24

E. M . HUTTON

based pastures. Norris ( 1956) concludes that the valuable attributes of tropical legumes are a reflex of their origin and that of their associated slow-growing cowpea type of Rhizobium under tropical rainforest conditions of low fertility. He regards tropical legumes and the cowpea Rhizobium as the progenitors of temperate legumes and their more specialized rhizobia. A study by Norris ( 1965) of the acid production of a large number of Rhizobium strains resulted in their separation into slow growing alkali-producers mainly from tropical legumes and fast growing acid-producers mainly from temperate legumes. These findings help to explain the adaptation of tropical legumes to acidic soils of low nutrient status and of temperate legumes to fertile alkaline soils. Another interesting finding by Norris ( 1959) was that Rhizobium is not calcium-sensitive and requires trace amounts of calcium only, but has a much greater need for magnesium. This has resulted in the use of dolomite in the mixtures used for pelleting clover seed (Hastings and Drake, 1962). Present knowledge concerning the inoculum and lime requirements of tropical legumes as summarized by Norris ( 1966, 1967) is given in Table I. Among the legumes having cowpea Rhizobium, there is a range from unspecialization to strain specificity and this represents different stages of development. The promiscuous legumes as shown in Table I include siratro, Townsville stylo, and cowpea, and these nodulate freely and grow without inoculation and lime additions. However, if they are inoculated with a selected strain of Rhizobium, a marked increase in growth and nitrogen fixation often occurs (Norris, 1966). The more specialized types like centro, the desmodiums, glycine, and Miles lotononis require inoculation with selected rhizobia but no lime. Silverleaf desmodium has inferior ability to extract calcium from poor soils compared with Stylosanthes species (Andrew and Norris, 196 l ) , and the same has been found for glycine, but lime is only occasionally beneficial, and the selected rhizobia they require are still cowpea types. Miles lotononis is an interesting example of a legume which grows successfully in sandy soils as acid as pH 4.0 and which requires a highly specific red Rhizobium originally obtained from South Africa (Norris, 1958). Finally Table I lists the highly specialized types, including leucaena, lucerne, and ordinary and Kenya white clovers which need acid-producing and fastgrowing rhizobia and respond to lime. Norris (1959) and Norris and 't Mannetje (1964) showed that central African Trifolium species like Kenya white clover are highly specific and will not nodulate with common clover rhizobia. Leucaena is exceptional among the tropicals in this respect, which explains its adaptation to calcareous soils and soils of high

25

TROPICAL PASTURES

TABLE I A Guide to lnoculum and Lime Requirement of Legumes Used in Tropical Pastures

Species

Common name

Calapogonium mucunoides Centrosema pubescens Desmodium intortum Desmodium uncinaturn

Calopo Centro Greenleaf desmodium Silverleaf desmodium

Dolichos axillaris Dolichos biforus Dolichos lablab Glycine wightii

Archer dolichos Leichardt dolichos Rongai dolichos Cooper, Clarence, or Tinaroo glycine Peru leucaena Miles lotononis Lucerne

Leucaena leucocephala Lotononis bainesii Medicago sutiva

Phaseolus atropurpureus Phaseolus aureus Phaseolus lathyroides Phaseolus mungo Pueraria phaseoloides Stylosanthes guyanensis Stylosanthes guyunensis Stylosanthes humilis Trifolium repens Trifolium semipilosum Vigna luteola Vigna sinensis

Siratro Golden gram Murray lathyroides Mung bean Tropical kudzu Schofield stylo Oxley fine-stem stylo Townsville stylo White clover Kenya white clover Dalrymple vigna Cowpea

Expected lime response

lnoculum requirement

No No No Rarely, in extreme conditions No No No Occasionally at pH below 5.5 Yes No Yes, lime is obligatory if pH is 5.5 or lower No No No No No No No

Cowpea" Specific Desmodium Desmodium

Cowpea" Cowpea(' Cowpea" Cowpea" Cowpea'' Cowpea" Specific

No Yes Yes No No

Cowpea" Clover Specific Cowpea" Cowpea"

Cowpea" Cowpea" cowpea Cowpea Specific Specific Lucerne

~

''A promiscuous species that will normally nodulate from native cowpea Rhizobium even if not inoculated.

base status as in the Caribbean and Hawaiian islands. However, in northern Australia areas of leucaena have grown quite satisfactorily in moderately acid soils, an observation which indicates that lime may be necessary only during the establishment phase. The question of lime-pelleting tropical legume seed for preservation of the applied rhizobia and their successful establishment on the roots during germination and early growth is discussed by Norris ( 1 966, 1967).

26

E. M. HUTTON

The successes with preinoculation and lime-pelleting have been with the fast-growing acid-producing rhizobia and legume species like lucerne and white clover. The calcium carbonate coating is stuck on with a harmless adhesive of gum arabic or methyl cellulose. The calcium carbonate overcomes soil acidity and preserves the rhizobia on the seed. Most tropical legumes with their tolerance to soil acidity require pelleting only for preservation of their alkali-producing rhizobia. With the exception of leucaena and perhaps silverleaf desmodium and glycine, extra alkalinity added by lime-pelleting is unnecessary and may depress nodulation in some tropical legumes. Rock phosphate is advocated as a pelleting material for the tropicals, and preliminary results are promising. The collection of Rhizobium at the Cunningham Laboratory comprises large numbers of strains isolated from nodules collected from indigenous legumes growing throughout the tropical world as well as strains from other laboratories and from the nodules of the various field-grown cultivars. The strains are stored by absorbing them on unglazed porcelain beads, which are then placed on slagwool pads in small McCartney bottles half filled with silica gel desiccant (Norris, 1963). Rhizobium strains are usually supplied to the farmer as peat cultures which are now being used widely for inoculation of tropical legumes. Testing for effectivness is done by the Leonard bottle-jar technique in the glasshouse (Norris, 1964). Just how effective and persistent the inoculants are in the field is not known in the case of the promiscuous legume species. Recently research on the various aspects concerning the ecology and field effectiveness of inoculant strains has been commenced at the Cunningham Laboratory using serotyping techniques. This may result in the selection of more aggressive and persistent inoculant strains and so lead to better field performance of a number of the tropical legumes. Heritable variation in nodulating ability was found in centro by Bowen and Kennedy (1961), and it appears to be a feature in at least two other tropical pasture legumes, greenleaf desmodium and glycine. In spite of thorough inoculation with a fully effective strain of Rhizobium, yellow, retarded, and poorly nodulated plants appear in populations of these species. A study of the inheritance of nodulating ability in greenleaf desmodium and glycine is now in progress at the Cunningham Laboratory. Since the plant dominates the symbiosis, genetical work aimed at improving current cultivars must include a check of the nodulating ability of the selections made from each generation. Unless this is done, improved agronomic types with poor field nodulation could easily be produced (vide Gibson, 1962).

TROPICAL PASTURES

VII.

27

Legume Nutrition Relative to Tropical Pasture Development

Unless plant nutrient deficiencies in the soil are corrected and the legumes given adequate quantities of the various elements essential for optimum growth, poor production will result from legume-based pastures. Also, testing of all potential pasture species needs to be done in the presence of adequate fertilizer. These important aspects are often overlooked in the rapidly developing countries of the tropics, with the result that attempts to establish improved pastures often have been a failure. On a highly manganiferous Low Humic Latosol in Hawaii, Younge and Takahashi (1953) obtained a marked response to Mo in dry matter and protein yields of thirteen lucerne varieties given adequate basal fertilizers including phosphate, K, sulfate, and borax. Further work by Takahashi ( 1956) showed the importance of lime, phosphate, K, sulfate, borax, and Mo in maintaining about 40% kaimi clover (D. canurn) with kikuyu- (P. clandestinum), pangola-, or dallisgrass (P. dilatatum) in the humid lowlands of Hawaii. On the island of Kauai, revegetation of bauxite substrate (stripsoil) with and without return of 8 inches of topsoil was studied by Younge and Moomaw (1960) after heavy dressings of all elements thought to be deficient and planting a mixture of D.intortum, kaimi clover, and pangolagrass. The controls without fertilizer were a failure, but topsoil and stripsoil (with extra P) in the first two years gave high dry matter yields, particularly of D.intortum. Younge et al. (1964) summarized results from several old and recent experiments on the Hawaiian islands of Oahu, Kauai, and Maui. D.intortum-pangolagrass adequately fertilized with lime and N, P, K, maintains 30-40% legume and gives up to 20,000 Ib dry matter per acre per annum and a corresponding beef liveweight gain of 800 Ib. Kaimi clover is tolerant of lower soil fertility and not as productive as D . intortum, and both legumes respond markedly to P, K, Mo, and Zn. The experiments also showed that lime, Mo, and Zn significantly increased seed yield of these legumes. These Hawaiian experiments demonstrate the importance of adequate nutrients, notably P and Mo, for satisfactory growth and seed production of legumes in the tropics. In northern Australia, the great arc of solodic and podzolic soils in coastal and subcoastal areas from northern New South Wales to Cape York Peninsula across the Gulf of Carpentaria to the Northern Territory and Kimberleys of Western Australia are very deficient in P and S, both vital to legume growth. A deficiency of Mo, an essential for functioning of Rhizobium, occurs in a number of areas, but its extent is not known at present.

28

E. M. HUTTON

The initial detailed research on soil nutrient deficiencies and nutrient responses of legumes in coastal areas of northeastern Australia was done by Andrew and Bryan (1955, 1958; Bryan and Andrew, 1955) in the southern Wallum on very poor sandy acid soil near Beerwah, 40 miles north of Brisbane. These soils are very deficient in P (3-6 ppm), Ca, and K, as well as N. Depending on site, the limiting nutrients for plant growth in descending order are P, N , Ca, K, S, Cu, Zn, Mo, and B. Effective strains of rhizobia were obtained for white clover, P . luthyroides, and the other test legumes which resulted in efficient nodulation and elimination of the need for N. Relatively low applications of calcium carbonate promoted good legume growth, as the response was due to Ca nutrition, not to the soil pH factor. Overliming could adversely affect pH and availability of elements like Cu and Zn. White clover needs more Ca than silverleaf desmodium and responded significantly to soil additions of Cu which did not increase growth of silverleaf. On the lateritic podzolic soil type, maximum response of white clover to S occurred only when 200 lb of Ca per acre was applied. Under grazing and with adequate fertilizer, white clover persists on the wet gley soils and not on the dry podzolics whereas silverleaf persists on both soil types. For pasture establishment on these poor southern Wallum areas, the fertilizer mixture needed per acre comprises 5 cwt superphosphate, 5 cwt calcium carbonate, 1 cwt potassium chloride, 7 Ib copper sulfate, 7 Ib zinc sulfate, 7 Ib borax, and 2 oz elemental Mo. This is a heavy initial dressing, but the subsequent annual requirement is only 2 cwt superphosphate and 1 cwt potassium chloride per acre. A similar fertilizer regime is needed to develop improved legume-based pastures in the northern Wallum extending in Queensland from Maryborough to just north of Bundaberg (T. R. Evans, 1967). Altogether the sandy infertile coastal lowlands of the Wallum cover some 2 million acres. Cost of developing improved pastures here is high, but beef production is profitable because of a good rainfall (4065 inches), long growing season, and maintenance of a high stocking rate of a beast per acre. The Wallum is only a small fraction of the vast area in northern Australia awaiting development through improved pastures (J. G. Davies and Eyles, 1965). Much of the area can be developed more cheaply than the Wallum and requires only a suitable legume-based pasture and annual application of 1-2 cwt of superphosphate per acre, use of Mo where deficient, and the occasional inclusion of a potassic fertilizer. Following the preliminary Wallum work, Andrew ( 1960) found that additions of sodium phosphate, potassium chloride, and calcium carbonate each increased the yield of white clover in a humic gley soil and increased its P, K, and Ca contents, respectively. Yield of clover was

TROPICAL PASTURES

29

closely correlated with its P content and also its K and Ca contents. The critical percentage (sufficiency in the plant for maximum growth) was 0.23 for P, 1.1 for K, and I .O for Ca. There was no correlation between P deficiency symptoms and P content. A close correlation between K deficiency symptoms and yield and K content was obtained, and Ca deficiency symptoms appeared only when Ca was omitted. Responses of a number of legumes to Cu were studied by Andrew and Thorne ( I 962) and Andrew ( 1963a). Srylosanrhes guyanensis (incorrectly S. bojeri) was one of the most sensitive to Cu deficiency, and silverleaf desmodium and white clover were the least sensitive. For maximal growth, sensitive species required 4 lb of copper sulfate per acre and insensitive species, 0.5 lb per acre. The sensitive species were less efficient in extracting Cu. A Cu concentration in legumes above 5 ppm is satisfactory and below 4 ppm indicates deficiency. With the exception of S. guyanensis, Cu deficiency symptoms occurred first in young growth as partial wilting and necrosis of younger leaves and shoots. Concave curling of leaflets and tip necrosis was general in all Cu-deficient species. Since P is so vital in legume nutrition and development of tropical pastures, Andrew and Robins (1 969a,b) examined its effect on growth and chemical composition of the main tropical legumes. Hunter River lucerne, a temperate legume, was included because of its increasing use in the subtropics. All species responded to P, and Cooper glycine and greenleaf desmodium were most responsive; although Townsville stylo and Miles lotononis were least responsive, they accumulated most P in plant tops. Townsville stylo, Hunter River lucerne, and Miles lotononis gave maximum yields at approximately 4 cwt superphosphate per acre, whereas the more responsive species Cooper glycine, greenleaf and silverleaf desmodium, siratro, and Murray lathyroides required up to 10 cwt per acre to achieve maximum production. Critical P percentages in the tops of Murray lathyroides, siratro, Townsville stylo, centro, Cooper glycine, Miles lotononis, Hunter River lucerne, silverleaf desmodium, greenleaf desmodium, and Dalrymple vigna at immediate preflowering were 0.20, 0.24, 0.17, 0.16, 0.23, 0.17, 0.24, 0.23,0.22, and 0.25, respectively. The N concentrations in plant tops were increased by P supply, and there was a good correlation between N and P levels in plant tops. Phosphorus applications beyond that necessary to produce maximum dry matter production continued to increase N concentration in plant tops; this poses the question, should a pasture be fertilized for maximum dry matter or nitrogen production? When sodium dihydrogen phosphate was applied, Na concentration in Dalrymple vigna, Hunter River lucerne, and Miles lotononis was increased. Use of monocalcium phosphate did

30

E. M. HUTTON

not increase plant Ca levels but increased Mg in Murray lathyroides and siratro. Increases in P supply reduced K concentration in most species, which was partly compensated by increased Mg and Ca concentrations. Throughout siratro and Murray lathyroides were relatively high in Mg, Miles lotononis and greenleaf desmodium in K, Dalrymple vigna, Miles lotononis, Hunter River lucerne, and Murray lathyroides in Na, and centro and Townsville stylo in Ca. In Andrew and Robins’ experiments (1969a,b) relative growth rate of Townsville stylo was superior to that of the other species when grown in soils low in available P. This was due to its ability to absorb greater quantities of P from such soils and supports the work of Andrew (1966a). He made a kinetic analysis of P absorption from solutions by excised roots of Townsville stylo, Murray lathyroides, silverleaf desmodium, Hunter River lucerne, and barley. Townsville stylo absorbed greater quantities of P per unit weight of root per unit time than the other species at both low and high P concentrations. Figure 5 gives the relationship between P uptake and time at a low phosphate substrate concentration (1 X 10+ M KH2POJ. The relatively low P critical percentage and greater efficiency of Townsville stylo in extracting P is clearly shown and explains its ability to grow and spread on soils with only 3- 10 ppm of available P. Field evidence indicates that other Stylosanthes species possess this characteristic. Andrew and Robins (1969c,d) studied the effects of potassium chloride on the growth and chemical composition of eight tropical and four temperate pasture legumes. They all gave a marked dry matter response to K but increase in K concentration of the tops occurred only at medium to high rates of application. Miles lotononis, greenleaf desmodium, and the four temperates were high in K and Townsville stylo was low. Critical percentages of K in tops of Murray lathyroides, siratro, greenleaf, and silverleaf desmodiums, Townsville stylo, Miles lotononis, centro, Cooper glycine, Hunter River lucerne, Jemalong barrel medic, irrigation white clover, and Palestine strawberry clover were, respectively, 0.75, 0.75, 0.80,0.72,0.60,0.90,0.75,0.80,1.2, 1.0, 1.0,and 1.0. Application of potassium had little effect on total cation content. The effect on plant Ca in Murray lathyroides, siratro, and Hunter River lucerne was small compared to that in the others. Townsville stylo had the highest Ca concentration, and the four temperates, greenleaf desmodium, and centro had low concentrations. Except in centro, Jemalong barrel medic, and Palestine strawberry clover, potassium chloride depressed Mg uptake. Murray lathyroides and siratro were high in Mg; centro and Palestine strawberry clover were low. Substantial reductions in Na concentrations occurred

31

TROPICAL PASTURES

70

60

-

--

"'05 0 x

c 0

e em

40-

L

a,

a

U

300

m

n

ln

a,

s

20-

10

-

5

0

2

4

6

8

1

20

0

Time (min)

FIG. 5. Relationship of phosphorus uptake to time in excised roots of Stylosanthes humilis ( A ) , Phaseolus lathyroides (0). Desmodium uncinatum lucerne (O), and barley (0).Substrate concentration, 1 X M KH2P04. (From Andrew and Robins, I969a,b).

(a),

in all species except the desmodiums and centro. The temperates, particularly Palestine strawberry clover, accumulated more Na than the tropicals; Miles lotononis, Murray lathyroides, and Townsville stylo were the highest in the tropicals. Species with marked cation interactions were those with high uptakes of Mg and Na. With potassium chloride application plant concentration of N was unaffected, P was decreased, and chloride increased. The desmodiums accumulated chloride to high levels which depressed growth. Manganese and Al excess is often present in the poor acid soils of northern Australia. In preliminary experiments Andrew (1963b, 1966b) found

32

E. M. HUTTON

that Townsville stylo and Miles lotononis were quite tolerant to excess Mn and A1 compared with the sensitive glycine and Hunter River lucerne. Andrew and Hegarty ( 1969) compared the response of eight tropical and four temperate legumes to excess Mn in water culture and found that the tropicals were as much affected as the temperates. The tolerant species produced no more dry matter than the less tolerant. Manganese concentrations reached in Townsville stylo were about twice those in Miles lotononis, yet both species were among the most tolerant. Among the least tolerant, Mn uptake by P . atropurpureus greatly exceeded that of Tinaroo glycine. Of the other species centro was tolerant while Murray lathyroides. leucaena, and silverleaf desmodium were intermediate in response. The results indicated that the relative tolerance of species depended partly on retention of Mn within the root system. Manganese treatments had little effect on the Ca and N levels in the tops of most species. Since superphosphate is the principal fertilizer used to correct the gross deficiency of P and S in coastal and subcoastal soils of northern Australia, studies on its effects on pasture components at different sites in this extensive area are necessary. These aspects were investigated by Truong et al. (1967) and R. K. Jones (1968) on solodic soils in southeastern and northeastern Queensland, respectively. Truong er al. ( 1967) found that on poor solodic soil (5-9 ppm P) at Beaudesert with nil P, siratro was stunted and the tops contained 0.18% P and that with P applications up to the equivalent of 4 cwt superphosphate per acre, dry matter yields increased and the tops contained 0.22% P. White clover also responded to P, and the P levels in the tops at nil P and the equivalent of 4 cwt superphosphate per acre were 0.15% and 0.2 I %, respectively. Siratro and white clover also responded to Mo. In addition, white clover responded to calcium carbonate and S but the application of small amounts of Mo removed the need for calcium carbonate. With white clover interactions occurred between Mo and S and also S and calcium carbonate. In R. K. Jones’ experiments ( 1 968), Townsville stylo growing on a solodic soil responded markedly to superphosphate. Applying 3 cwt per acre of superphosphate in the first year only rather than annual amounts of 1 cwt per acre gave higher yields of dry matter over a threeyear period. Splitting the application into annual dressings gave slightly higher yields of N and P per acre for the three years. It is apparent that superphosphate has a marked residual effect in this soil. Critical value for P in Townsville stylo tops varied from 0.16 to 0.17% similar to that found by Andrew and Robins (1969a). It was of interest that seed P levels appeared to be related to fertilizer history and P status of the plants.

TROPICAL PASTURES

33

Andrew ( 1965, 1968) reviewed diagnostic techniques for determining mineral status of tropical pasture plants and the problems in their use. He emphasized that foliar analysis is a relative approach, the observed value being related to a standard critical percentage which must be determined for the different species under optimum growth conditions in which secondary deficiencies and toxicities are corrected. With legumes, unless they are well supplied with nitrogen, either mineral or symbiotic, there is no point in establishing their critical values for elements like P, K, and Mg. The critical percentages for the various nutrients can be applied to field-grown material only under strict sampling conditions which relate to stage of growth, part of plant, environmental factors, and freedom from undue insect and pathogen damage. For correct interpretation of the foliar anslysis of field material relative to critical values, an appreciation of the interactions of N , P,S, and K in plant nutrition is required. The results coming forward from research on nutrition of pasture legumes if properly applied will ensure maximum production of dry matter and protein from legume-based pastures in the different tropical environments. Also, they will give a lead to animal nutritionists concerned with the relationship between the mineral balance of pastures and animal production. Plant breeders and geneticists could well accept the challenge presented by the efficient utilization of P by Townsville stylo and endeavor to breed this character into a number of the other tropical legumes. VIII. Undesirable Compounds in Tropical Legumes

A. ESTROGENS A N D SUBSTANCES CAUSING BLOATA N D MILKTAINTS Analyses and biological tests have indicated that estrogens and allied compounds are at a low level in the tropical legumes and are unlikely to reduce fertility. The only tropical legume known to cause bloat is Rongai lablab (Hamilton and Ruth, 1968), and then only occasionally when it is young and growing rapidly. Perhaps tropical legumes are free from the bloat-inducing 18 S protein of McArthur and Miltimore ( 1966). Milk from dairy cows fed only Rongai lablab or Nandi setaria had a strong odor and taint and was considered unacceptable on receipt, but pasteurization made it acceptable (Hamilton et al., 1969). The milk from a sole diet of leucaena was acceptable without treatment.

B. MIMOSINE Yoshida ( 1944) showed that leucaena contained mimosine, but no effects on reproduction were obtained in dairy cows fed only leucaena

34

E. M . HUTTON

and concentrates for several years at the Hawaiian Agricultural Experiment Station (Anonymous, 1948). Mimosine is an undesirable depressant of cell division (Hegarty et al., 1964a) and accounts for about 0.5% of the N of leucaena herbage. It has been known for some time that feeding leucaena affects reproduction in monogastric animals like rabbits (Willet et d., 1947)and sows (Wayman and Iwanaga, 1957). Hegarty et al. ( 1 964b) developed methods for the extraction and determination of mimosine present in leucaena leaves and urine. This enabled them to study the reaction of sheep to consumption of leucaena and mimosine. Sheep shed their fleeces on a sole diet of leucaena because the mimosine in it suppressed mitotic activity in the follicle bulb of the growing wool fiber and caused follicle degeneration. Follicle regeneration occurred when sheep were taken off leucaena but occurred also in animals on a continuing leucaena diet. During leucaena feeding, only small quantities of mimosine were excreted, most being degraded by rumen flora to 3,4dihydroxypyridine, the main urine component. Sheep cannot detoxicate mimosine after absorption beyond the rumen. It was found that sheep could be conditioned to a sole diet of leucaena without ill effect due to increased detoxication from adaptation of rumen microorganisms. In Queensland continuous grazing of leucaena pastures for extended periods has adversely affected steers. Symptoms include shedding of hair on the rump and tail and loss of weight indicating incomplete breakdown of mimosine in the rumen. Hamilton er al. (1 970) made a close study of reproduction in dairy heifers fed a complete diet of this legume. Leucaena did not affect estrus cycle length, conception rate, gestation length, calving rate, milk production or composition. However, mild incoordination occurred briefly during gestation in some cows and birth weight of calves from cows fed leucaena was lower than of control calves. There was no residual effect of leucaena on calf growth rate, as the resulting calves grew at the same rate as the controls. C. INDOSPICINE Indigofera spicata (syn. I . endecaphylla) is found in a number of areas including India, Ceylon, Indonesia, Philippines, Hawaii, Central America, Brazil, and west Africa, and is regarded as a promising pasture legume because of its vigorous, prostrate, stoloniferous habit and high level of N fixation (Henzell, 1962). Trials in the early 1950’s with several introductions at coastal sites in southeastern Queensland confirmed the potential of this legume. However, its widespread use was prevented by the work of Emmel and Ritchey (1941) and Nordfeldt et al. (1952), who fed

TROPICAL PASTURES

35

it to rabbits, cows, and sheep, and found that it caused liver degeneration and that pregnant animals aborted. In studies of Nordfeldt et al. (1 952), the guinea pig was less susceptible, and Freyre and Warmke (1952) showed that guinea pigs survived indefinitely on I. spicata but pregnant females aborted. All these findings indicated the presence of an unidentified hepatotoxin in 1. spicata. The chick test of Rosenberg and Zoebisch (1952) for investigating toxicity of forage legumes was used by Morris et al. (1 954) to study I. spicatu, in which they identified hiptagenic acid (3-nitropropionic acid) which was considered to be the toxin. Cooke (1 955) also supported thi; view. Britten et al. (1959a) using the chick test concluded that 3-nitropropionic acid was probably the sole toxic agent in I . spicatu. A high correlation between the chick test and a chemical test for 3-nitropropionic acid was found by Britten el al. (1959b), who recorded differences in toxicity between I. spicata plants. Later Britten et al. (1963) showed a positive correlation between the amount of 3-nitropropionic acid in the ration and toxicity to chicks. Hutton et al. (1958a) obtained a similar type of liver damage in rabbits whether green leaf, dried leaf, or seed of I . spicata was fed. 3-Nitropropionic acid did not appear to be the hepatotoxin involved, as it was not present in the seed and force-feeding the pure compound did not produce liver damage in rabbits. Further studies with mice (Hutton et d., 1958b) indicated that 3-nitropropionic acid was not the hepatotoxin in 1. spicatu, so preliminary work was commenced to isolate the compound implicated (Coleman et ul., 1960). Hegarty and Pound ( 1 968) finally reported the isolation from I. spicata of the first naturally occuring hepatotoxic amino acid, which they named indospicine. It is ~-2-amino-6amidinohexanoic acid, and when injected subcutaneously into mice it produces fat accumulation and cytological changes in the liver (Hegarty and Pound, 1970). Fat accumulation was inhibited by simultaneous injection of arginine but not by canavanine, so indospicine may produce its hepatotoxic effects by interference with arginine metabolism. This could explain the results obtained by previous workers with chicks which are uricotelic and may have a different metabolic pathway from mammals. Hegarty and Pound (1970) found that a substantial part of the hepatotoxicity of extracts of I. spicatu seed was accounted for in terms of indospicine. Now that the hepatotoxin in I. spicata has been identified, breeding a nontoxic line of this valuable legume is possible. Perhaps mutagenic techniques may be the most appropriate to achieve this result.

36

E. M. HUTTON

D. TANNIN The desmodiums have a high tannin content (Rotar, 1965) and are the only important tropical legumes containing this chemical complex (Hutton and Coote, 1966). Whether tannin is an undesirable component in them has yet to be determined. No doubt their tannin content would preclude them causing bloat. However, the tannin may reduce their digestibility, as the mean in v i m digestibility of 30 bred lines of D . intorturn was 53.1% whereas that of the same number of P . atropurpureus lines was 64.9%. R. J. Jones (1969) found a similar difference in digestibility between these species. IX. Physiology of Tropical legumes,

All the legumes need to be characterized physiologically to quantify their growth potential for any particular environment in the subtropics and tropics. This involves studies on the effects of energy input, photoperiod, temperature, and moisture and the interrelationships between these on growth, maturation, and seed production of the different legumes. Interactions between these environmental factors and nodulation and application of essential mineral elements should also be studied because of their fundamental importance in the pasture environment. Proper physiological characterization of the legumes will not only explain their adaptation to the various conditions, but indicate how they can be improved and how they can be used more profitably in the pasture system. Also it should be possible to correlate data from controlled environment facilities with climatic data and so predict the field behavior of a legume in any particular region. Research in controlled environments needs to be linked with studies on the physiological reactions of both tropical legumes and grasses in pastures under grazing. In this way the factors involved in legume-grass competition, dry matter and protein production, and persistence could be elucidated so that ways of improving output from the pasture ecosystem could be devised. Research on physiological parameters of tropical pasture plants in laboratory and field has barely commenced and needs to be intensified. A. TEMPERATURE A N D GROWTH I N SEVERAL TROPICAL LEGUMES Whiteman (1968) studied the effects of temperature on growth in a long day (16 hours) of the six tropical legumes Murray lathyroides, siratro, silverleaf and greenleaf desmodiums, D . sandwicense, and Tinaroo glycine. At the first harvest (14 days) seedling dry weight (stems, leaves)

TROPICAL PASTURES

37

was highly correlated with mean seed weight of each species, siratro having the greatest seed and seedling weights and greenleaf desmodium the smallest. At the second harvest, the real differences between the legumes in growth rate were expressed. Growth was abnormal and reduced markedly at the lowest temperatures 15110°C and 18/ 13°C (daylnight). Optimum temperature for growth of all the legumes was 30/25 2 3°C which is lower than for tropical grasses and higher than for temperate legumes and grasses. Above 33/28"C growth rate declined, particularly in the Desmodium species, but not so markedly in siratro.

B. TOWNSVILLE STYLO In experiments with Townsville stylo, 't Mannetje (1965) extended the photoperiod of 8 hours sun with incandescent light. He found that it was a short-day plant in temperatures of 30°C (day) and 25°C (night) and that dry matter yields in 12- and 14-hour photoperiods were greater than those in 8- and 10-hour ones. Sweeney's results, quoted by Humphreys (1967), showed optimum dry matter production at 33/28"C. In a study of the flowering behavior of seven selections (early to late) of Townsville stylo, D. F. Cameron ( 1967a) showed that day length was the main factor controlling flowering and that they all had a strong short-day response. At normal temperatures, maximum day lengths (critical day lengths) at which all plants flowered were 13 hours for the early selections, 12 hours for the midseason and late midseason, and 1 1.5 hours for the late. Both high night temperature and low day temperature delayed or inhibited flower initiation in the early and midseason selections, and these effects were greater at the critical day length. D. F. Cameron's field and shadehouse experiments (1967b) with different sowing dates and locations gave similar results to his controlled environment studies. In the early December sowing, the range in flowering time between maturity groups was 56 days because the longer day lengths promoted flowering in early types and prevented floral initiation in late types. With the late March sowing, day length was short enough to promote flowering in all maturity types so the range in flowering time was only 8 days. At the southerly locations, most selections flowered later because the longer day lengths delayed flowering time. C. Glycine wightii Edye and Kiers observed ( 1966) variation in maturity, stolon development and frost resistance in 50 accessions ofglycine at Lawes, southeastern Queensland. The discontinuous variation in flowering enabled definition

38

E. M. HUTTON

of very early, early, midseason, and late maturity types. Generally, collections from below 15" latitude are early, midseason, or late types and are more strongly stoloniferous and more frost susceptible whereas collections from above this latitude are predominantly early types, which are less stoloniferous and more frost resistant. Tow ( 1967) compared the growth of Tinaroo glycine and green panic in controlled conditions. He found that green panic produced much more dry matter than glycine per unit of intercepted light and per unit of water transpired and that green panic had higher shoot-root ratios than Tinaroo glycine. He investigated five other varieties of glycine and all had higher shoot-root ratios than Tinaroo, some with similar ratios to green panic. Wutoh et al. (1968a) showed that flowering in five glycine introductions (very early to late) is affected by both photoperiod and temperature. Temperatures of 27/2216°C (daylnight) appeared to be best for growth and seed production. All accessions studied were short-day types and in the sensitive ones temperature had little effect on flowering. In others, lowering day or night temperature or both induced flowering in long (16-hour) days and could prevent it in short days. Seed formation did not occur in day temperatures above 27°C. Response to salinity in glycine was investigated by Gates el al. ( 1 966a, b) as soils with a high salt content occur in the brigalow lands of northeastern Australia, where this legume has potential. With the highest salt level (240 meq of NaCl per liter) yield of Tinaroo glycine dropped to 25% of the control but nitrogen content was unaffected. Phosphorus content rose by up to 100% in the roots but did not change in the tops. Percentage of soluble nitrogen increased by more than 50% as salinity rose, indicating impairment of protein synthesis. Glycine can adapt to high levels of sodium chloride unaided by divalent ions provided the increase in salinity is gradual. In their experiments with 22 glycine accessions, Gates et al. (1966b) used four salinity levels up to 140 meq of NaCl per liter. The highest salinity level had a relatively greater impact on growth than the others. Differences in dry weight of the glycines at all salinity levels were of similar proportions to those at the control level. The normal capacity for growth of an accession seemed to be an important feature in determining its response to salinity. D. SIRATRO In a controlled environment experiment (Hutton, 1964), growth of siratro was very poor at 18/ 13°C and 2 1 / 16°C in short (8-hours) and long (8-hours sun 8-hours incandescent light) days. Over the temperature range 24/ 19"C, 27/22"C, 30/25"C, and 33/28"C, dry matter production

+

TROPICAL PASTURES

39

of siratro in a short day averaged 30% of that in a long day. Maximum dry matter yield in the long day was at 27/22"C and 30/25"C and was reduced by 30% at 33/28"C and by 7% at 24/19"C. Flowering occurred in short and long days at all temperatures except 18/ 13°C and was best at 24/19"C, 27/22"C, and 30/25"C. At a constant temperature of 28"C, Whiteman (1969) found that siratro flowered in daylengths of 8, 10, and 12 hours but not at 16 and 24 hours, indicating that siratro is a short-day plant. In a comparison between siratro and hamilgrass ( P . maximum) in a controlled environment, Ludlow and Wilson ( 1968) found that relative growth rate of the grass was almost twice that of siratro. This was due to a much higher net photosynthetic rate in the grass which also had a higher respiration rate than siratro.

E. AFRICAN TRIFOLIUMS Although not strictly tropical, the African trifoliums have some potential in the subtropics. 't Mannetje and Pritchard (1968) studied the reactions of 13 species and varieties of these trifoliums in controlled environment and glasshouse experiments. T. baccarinii, T. pseudostrictum, and T . usambarense behaved as sensitive short-day plants, and T. burchellianum var. burchellianum and T. africanum behaved as sensitive long-day plants. The other species were either day-neutral or could not be clearly classified because of strong effects of night temperature on flowering. T. semipilosum flowered in all treatments except the 10-hour photoperiod at 25/20"C and better in a low than high night temperature. Conditions favorable for flowering were usually the best for growth. X. Breeding and Genetics of the Main legumes

This was reviewed by Hutton ( 1965), and only the more recent findings will be discussed here. The complement of characters present in the current range of legume cultivars in Australia, with the exception of siratro (Hutton, 1962), have resulted from natural selection in various native habitats overseas. Thus a number of the legumes are not fully adapted to conditions in northern Australia and need further breeding or selection to adapt them more closely to the soil-pasture-animal system. A. BREEDINGSYSTEMS Breeding systems vary (Hutton, 1960) from close pollination as in Townsville stylo, siratro, Indigofera, glycine, and Miles lotononis (Byth 1964) through a combination of self- and cross-pollination in the desmodi-

40

E. M. HUTTON

ums (Rotar et al., 1967) to cross-pollination in lucerne, white clover, and some of the African trifoliums including T. semipilosum (Pritchard and 't Mannetje, 1967). In the self-pollinating tropicals a high relative humidity is needed for pollen-tube growth on the stigmas, and seeds are set only if the humidity is adjusted to the optimum level around the emasculated and hand-pollinated flowers. In leucaena the round heads of small flowers are self-pollinated but can be emasculated after anthesis by washing in a very weak solution of a nontoxic wetting agent and then hand-pollinated.

B. TOWNSVILLE STYLO D. F. Cameron (1965) found that significant variation occurred in flowering time, growth habit, plant growth, and seed yields among a large number of collections made from the naturalized populations in Queensland and the Northern Territory. There was a continuous range in flowering time from early to late, and plant growth was related to time of flowering. Late types gave higher dry matter yields than midseason types, which gave higher yields than early types. The midseason types gave the highest seed yields. Time of flowering and seed setting are very important in Townsville stylo to ensure production of large amounts of seed for reestablishment in the following wet season. D. F. Cameron (1967~)related flowering time of 58 ecotypes with their collection site characteristics. In general the late-flowering types were collected from areas with an annual rainfall over 45 inches, and early types came from drier areas with a rainfall of 23-35 inches. Distribution was also correlated with latitude. The five collections obtained south from Rockhampton were early or midseason and all collections from the northern parts of Cape York Peninsula, the Northern Territory, and Western Australia were late flowering. Late types usually give higher yields than early types in areas with a long growing season, because they continue vegetative growth after early types have set seed and stopped growing. With a short growing season, late types are unable to flower and seed and fail to produce a dense sward in the following season whereas early types regenerate well because they flower and set seed before soil moisture is exhausted. Erect types usually give higher yields than prostrate, particularly when associate grasses are present in the pasture. Inheritance of flowering time has been studied (D. F. Cameron, 1968) in a diallel cross between nine lines ranging from early to late. In all crosses involving two late-flowering types, late flowering was strongly dominant and the 3 : 1 segregation indicated a major single gene difference

TROPICAL PASTURES

41

between these and other maturity types. A few hybrid populations contained some very early segregates indicating the presence of a recessive gene at another flowering locus. A three-locus model involving genes governing late, midseason, and early maturities was proposed. Stylosanthes species comprise a polyploid series ( x = 10). S. humilis, S . guyanensis, and S . hamata are diploids, S . mucronata is a tetraploid and S . erecta a hexaploid, and all are perennials except S . humilis ( D . F. Cameron, 1967d). Sterile hybrids resulted from the crosses S . humilis x S . guyanensis and S . humilis X S . hamata, but fertile tetraploids were induced by colchicine (D. F. Cameron, 1968). Of these the S . humilis X S . guyanensis (cv. Schofield) progeny was the most perennial. C. SIRATRO This cultivar was bred from two Mexican ecotypes of P. atropurpureus and combines the stoloniferous habit of one with the higher yield and better seeding ability of the other (Hutton, 1962). The aims of breeding work with siratro (Hutton, 1965) are greater yield, persistence, and stoloniferous development, a longer period of active growth, higher seed yield, and slower shattering pods. Two main cycles of crossing and selection have been completed. The first produced higher-yielding and more stoloniferous lines than siratro. The second has recently given lines with higher yield of dry matter and seed, and a longer period of active growth due to ability to grow at temperatures of 50-70°F. Stoloniferous development is not as pronounced in the latest lines, but this may not be disadvantageous, as under drier conditions this character is rarely expressed. A high level of root-knot nematode resistance is being retained, and improvement in tolerance to Rhizoctonia solani and “legume little leaf’ is sought.

D. DESMODIUMS Crosses were made between greenleaf and silverleaf desmodiums and between greenleaf and D . sandwicense with the objective of introducing valuable genes into greenleaf and generating new variation (Hutton and Gray, 1967). Progeny from the first cross had a high degree of sterility (95.7%) and progeny from the second cross were fully fertile. Both progenies gave promising segregates. No useful lines resulted from the cross D . sandwicense x silverleaf. Park and Rotar ( I968a,b) studied inheritance of flower and stem color, leaflet markings, and petal anthocyanin concentration in Spanish clover ( D . sandwicense). Flower and internode color were each controlled by a single gene, and these genes were linked. Silver midrib markings on the

42

E. M. HUTTON

leaflet and differences in anthocyanin concentration of petals also appeared to be under monogenic control. McWhirter (1 968) established the occurrence in Desmodium of a system of determination of the male-sterile, fertility-restored phenotypes similar to cytoplasmic male sterility systems found in other angiosperms. A scheme was presented utilizing genetic stocks for commercial production of F1 seed of the interspecific hybrid D . sandwicense X D . intorturn. E. GLYCINE As a basis for glycine improvement variation of important characters in a range of accessions was studied by Edye and Kiers ( 1966), and inheritance of some of these characters was determined by Wutoh et al. (1968b, c) in selected crosses. The introductions varied in maturity from very early to late, and in seed set and yield, stolon development, and frost resistance. In a diallel cross of five glycine introductions Wutoh et al. ( 1968b) estimated general and specific combining ability components of variance for nine characters. Genetic advance is possible in flowering time, maturity date, and seed weight traits, but would be limited for yield, stolon length, stolon number, and percentage of stolons rooted. Fortunately, yield and flowering time were not correlated. Tinaroo had high general and low specific combining ability for most traits associated with forage yield, so should be a valuable parent in breeding for increased yield. Using six introductions, Wutoh et al. (1968~)studied the inheritance of seven traits in seven hybrids and their progeny. Hybrid vigor in the F1was not found in flowering time, stolon number, or percentage of stolons rooted, but occurred in the other traits followed by inbreeding depression in the FI. All traits were polygenically controlled, and genotypic correlations indicated that early flowering, high yield, and good stolon development could be combined j n one variety. Slow establishment of glycine appears to be related to poor nodulating ability. Mode of inheritance of nodulation in glycine is currently being studied at the Cunningham Laboratory with the aim of developing lines with rapid and effective nodulation in the field. F. LEUCAENA Breeding and genetical work at the Cunningham Laboratory is aimed at developing leucaena lines with high forage yield and a dense, compact, branching habit suited to grazing. From a diallel cross with four ecotypes, Gray (1 967a) found that Peru and Guatemala had a greater general combining ability than Hawaii and El Salvador for main stem length. Peru

TROPICAL PASTURES

43

and Hawaii were superior to El Salvador and Guatemala in general combining ability for stem number. Gray ( 1 967b) followed inheritance of a number of characters in all possible crosses between five leucaena ecotypes. In the F1, leaf size, stem length, and flowering date exhibited significant genotypic effects, and most F1 means were similar to those of the higher parent. In the Fe, stem length and number appeared to be controlled by multiple genes affecting vigor, and Fz and FS segregations indicated that branching habit is controlled by two pairs of disomically inherited genes. Length of main stem and stem number were highly correlated in Fz and F3 progenies of the cross Bald Hills X Guatemala (Gray, 1967c), which suggested that these are not inherited independently but are related to plant vigor. As predicted from the diallel data, the cross Peru x Hawaii combining genes for dense branching and high vigor has given a number of promising new lines. There is a need to breed mimosine-free lines of leucaena since mimosine is not fully destroyed in the rumen under some conditions. It may be possible, as Gonzalez et al. ( 1 967) have observed wide variation in mimosine contents in segregating populations from interspecific leucaena crosses. Plants were selected with less than 30% of the mimosine values of the leucaena cultivars usually grown.

G. INDIGOFERA Breeding and genetical work is warranted with the potentially valuable I. spicata, especially now that its toxin has been identified by Hegarty and Pound (1 968). Hutton and Guerassimoff ( I 966) used a red-stemmed I. spicata type from Ceylon as the male parent in crosses with several green-stemmed types from the Republics of Congo and Ghana. A number of F, seedlings died and progeny was obtained from only one F1 plant. This showed genetic diversity between the parental ecotypes. In the FP and Fs there was a high degree of sterility, but fertile plants giving fertile progeny were obtained. From these, early flowering lines have been selected with higher dry matter and seed yields than the parents. H. LUCERNE The breeding work with lucerne at the Cunningham Laboratory is aimed at developing a subtropical grazing lucerne which will grow with tropical legumes and grasses in a pasture. Selections from the creepingrooted variety Rambler (Heinrichs, 1954) were crossed with the cultivars Hunter River, Hairy Peruvian, Indian, Pampa, and Saladina. In successive generations Edye and Haydock (1967) found that on a fertile clay

44

E. M. HUTTON

loan creeping-rooted,ness increased from 2 to 59% and that three times more creeping-rooted than noncreeping plants persisted. They were also able to increase both winter and summer yields by mass selection. Subsequent work in two other subtropical environments by Bray ( 1967,1969) with creeping-rooted clones from the preceding program has confirmed that creeping-rootedness increased survival. In two polycross populations high positive genotypic correlations were established between creeping-rootedness and yield at all times of the year, indicating successful selection in early generations. Combinations of the best parental clones should give synthetic creeping-rooted cultivars with high yield and persistence. XI. The Main Grasses

Much of the important cattle country in northeastern Australia is dominated by the perennial native speargrass (Heteropogon contortus) (Shaw and Bissett, 1955; Tothill, 1966), which is improved markedly in feeding value by introduction of Townsville stylo (Shaw, 1961). The counterpart of speargrass in a number of monsoonal areas of Africa and Central and South America is jaraguagrass (Hyparrhenia rufa), which can also be upgraded by a Stylosanthes species such as S . guyanensis (syn. S . grucilis) (Stobbs, 1966, 1969b). Most of the grasses planted widely in improved tropical pastures are African in origin, eastern Africa being a particularly important source (mgdan, 1966b). Exceptions are the paspalums, native to South America, notably Brazil. Australia and other countries now have large collections of promising tropical grasses, and the problem is to sort out the species and ecotypes that will contribute the most digestible dry matter to the pasture ecosystem. In this section the relevant facts concerning the agronomy of the main grasses will be systematized under the different grass genera involved. With few exceptions they are perennials, and detailed descriptions of most of them are given by Barnard ( 1967). A. BRACHIARIA The important species in this genus are adapted to the wetter tropical areas with annual rainfalls of 40-60 inches or more. Paragrass ( B . mutica) from tropical Africa is the best known and most valuable because of its tolerance to swampy conditions present in many parts of the tropics. I t is a perennial with coarse runners rooting at the nodes, which produce erect shoots with broad hairy leaves. Seed production is poor, and it is usually planted vegetatively. Under waterlogged conditions it will give a dense,

TROPICAL PASTURES

45

almost pure stand, its nitrogen supply being derived from drainage or blue-green algae. Murray lathyroides is one legume that will grow with it in this type of environment. Where drainage is better, centro and Schofield stylo are suitable associate species. Feeding value of paragrass is high, and it can be stocked heavily. Signalgrass (B. decumbens) from Uganda is vigorous and strongly stoloniferous with bright green leaves: it has distinct promise in northern Queensland (Schofield, 1944) and other tropical areas. Seed production is good, but the hard seed needs sulfuric acid treatment to give satisfactory germination. With drainage and fertile soils or nitrogen applications, it will give high liveweight gains per acre. There is a range of types of B. ruziziensis and also of B. brizantha (Bogdan, 1955). An ecotype of the first species has been commercialized in Australia as Kennedy ruzigrass: although it is very hairy, it is very palatable to stock. B. brizantha has not been used widely but was found suitable for growing with coconuts in Ceylon. 9. CENCHRUS

Buffel (C. ciliaris) is the most important grass in northeastern Australia because of its drought resistance, adaptability to a range of soils, persistence under heavy grazing, good feeding value, and ease of establishment from seed. I t will become more widely grown in the drier monsoonal areas of the world. In Australia a number of cultivars have been commercialized: Biloela (Grof, 1957) and Molopo (Flemons and Whalley, 1958) are representative of the tall vigorous rhizomatous types, and Gayndah (Marriott and Anderssen, 1953) of the shorter nonrhizomatous types. Seed production of buffelgrass is good, and it is compatible with legumes like siratro and responds to superphosphate (Edye et al., 1964). The less vigorous birdwood grass (Cenchrus setigerus) is a better associate for Townsville stylo than buffel (Norman, 1962).

C. CHLORIS Rhodes (C. gayana) is the only important grass in the genus; its botany, distribution in Africa, and salient agronomic features are discussed by Bogdan ( 1969). It seeds prolifically and is a valuable pasture component south of the Tropic in Australia because of its quick establishment, stoloniferous habit, adaptability, and cornpatability with a range of legumes including lucerne (Christian and Shaw, 1952: 't Mannetje, 1967). Rhodesgrass is not as drought resistant as buffet, is salt tolerant (Teakle, 19371, and responds significantly to high fertility (Henzell,

46

E. M. HUTTON

1963). It is a variable species, and Bogdan (1969) describes a number of ecotypes differing in vigor, leaf size, and thickness of stems and stolons. In Australia the cultivars grown include pioneer, an early type; Katambora, with thin stems and stolons and narrow leaves; Callide (Grof, I96 I), a coarse vigorous late type; and Samford, which is late, leafy, and palatable. Milford and Minson (1968b) found no differences in feeding value between the various Rhodesgrass types. D. CYNODON Common stargrass ( C . dactylon) is widely distributed in east Africa and is very variable, the ecotypes differing in'size, color (yellowish green to bluish green), and texture of stems and leaves (Edwards and Bogdan, 1951). Most types are poor seeders but easily spread vegetatively because they are strongly stoloniferous. Some are also rhizomatous. The forage of stargrass has a relatively high dry matter content and is of high feeding value. Giant stargrass (incorrectly C. plectostachyus), from the warmer moist region near Lake Victoria, is giving better animal production than pangolagrass with nitrogen fertilizer in some tropical areas. A number of stargrass ecotypes have been introduced into Australia, but none have shown definite promise. Perhaps more work is needed, as Burton (1947, 1954) successfully developed coastal bermuda which has become the most important grass in the southeastern United States.

E. DIGITARIA Pangolagrass ( D . decumbens), an aneuploid (2n = 27), is a most important and widely grown species in the tropics; it originated in the Nelspruit district of the eastern Transvaal (Chippindall, 1955). Oakes (1960) described its introduction into the United States and its use there and in the Caribbean, and Hodges et al. ( I 967) discussed its adaptation to Florida pastures. It is very adaptable and thrives where annual rainfall is over 40 inches; it is established vegetatively from pieces of its thin many-noded and vigorously rooting stems. Because of its strongly stoloniferous habit, it rapidly forms a dense stand which resists heavy grazing. Its high feeding value makes it an ideal grass for a system of nitrogen fertilization or for growing with a compatible legume like the tree leucaena. Pangolagrass has been seriously damaged in several countries by a virus disease first noted in Surinam by Dirvin and Van Hoof ( 1960). It is also attacked by a number of insects including spittlebug (Prosapia bicincta), which is a problem in southern Florida (Mead, 1962). Bryan and Sharpe (1 965) found that growth of pangolagrass in the

TROPICAL PASTURES

47

Wallum of southeastern Queensland was retarded in winter when average night temperature was below 58°F. It is possible that, as a result of Oakes’ (1 965) Digitaria collections from South Africa and studies on cold tolerance in Digitaria (Oakes and Langford, I967), pangolagrass may be replaced in some areas with better-adapted types. There is little doubt that ecotypes of Digitaria smutsii have definite potential and warrant further work. F. MELINIS Molassesgrass ( M . minut@ora), indigenous to Africa, is the only species grown, and ecotypes show wide variation in vigor, leafiness, and growth habit (Bogdan, 1955). It is used extensively in Brazil, and the type used in Australia came from there. Molassesgrass is a useful pioneer on well drained areas where annual rainfall is 40 inches or more; it does not persist under heavy stocking. Leaves are covered with soft short hairs that exude a sticky substance with a molasseslike odor.

G. PANICUM After the buffel cultivars, those of guineagrass or panic (P. maximum) are the most important grasses in northeastern Australia at present. The guineagrasses are grown and naturalized throughout the tropical world and make a significant contribution to animal production in many areas. In Brazil the tall robust coloniao is valued because of its palatability and feeding value and its ease of establishment from cuttings and seed. Both coloniao and the similar hamilgrass are grown in wetter parts of northern Queensland. Other types grown in northern Australia include the shorter-growing Gatton and Sabi panics, and Petrie green panic. Of these, Petrie green panic (Marriott and Winchester, 1951) is the most important, as it is drought resistant and can be grown in an annual rainfall as low as 22 inches. It is compatible with a range of legumes and has a good feeding value because of accumulation of surplus carbohydrate in the aboveground parts (Humphreys and Robinson, 1966). Bogdan ( I 955, 1965) in his studies of African ecotypes of P. maximum found wide variation in vigor and leafiness and classified them into larger fodder types and smaller grazing types. One interesting type was Embu creeping guinea which forms a continuous sward because of prostrate stems which root freely at the nodes. The closely allied P. coloratum complex contains a range of types with potential in the subtropics, and several cultivars including the makarikari grasses Bambatsi, Burnett, and Pollock have been developed (Barnard, 1967).

48

E.

M.

HUTTON

H. PASPALUMS Paspalums are mainly grown in the humid subtropics where annual rainfall is 30-35 inches or more. Those widely used are common paspalum or dallisgrass ( P . dilatatum) and bahiagrass ( P . notatum) both natives of South America. In northern Australia the paspalums used in pastures include common paspalum, Hartley and Rodd’s Bay plicatulums ( P . plicatulurn) (Bryan and Shaw, 1964) from Brazil and Guatemala, respectively, and Paltridge scrobic ( P . commersonii) (Paltridge, 1955) from Rhodesia. Shaw et al. (1965) compared 17 introductions of Paspalum species and common paspalum in a cutting trial in which optimum growth was maintained by supplementary irrigation and liberal application of fertilizers containing nitrogen, phosphorus, and potassium. They found that P . yaguaronense, P . notatum var. saureae, P . notatum var. latijlorum, and Rodd’s Bay plicatulum were markedly superior in yielding ability and length of growing season to common paspalum. Common paspalum, grown with white clover and fertilized with superphosphate, is an important pasture for dairying on the alluvials of northern New South Wales and southern Queensland. Paltridge scrobic and the two plicatulums are grown with a range of legumes in coastal areas south of the Tropic and have proved valuable under waterlogged conditions.

I. PENNISETUM Kikuyugrass ( P . clandestinum), native to a restricted area of the Kenyan highlands (Edwards and Bogdan, 195 1), has proved to be a very valuable grass because of its high feeding value, tolerance to heavy grazing, and marked response to nitrogenous and other fertilizers. It requires an annual rainfall of 35-40 inches or more and thrives on basaltic tablelands throughout the tropics, but at low elevations it is restricted to the subtropics. Kikuyugrass is strongly stoloniferous and rhizomatous and is usually established vegetatively. In southeastern Queensland it grows well with white clover provided it is adequately fertilized with molybdenized superphosphate and potassium chloride (R. E. White, 1967). There are several ecotypes of kikuyugrass, and commercial seed production of a superior type has been successfully developed in northern New South Wales (Wilson, 1970). Elephant- or Napiergrass ( P . purpureum) is indigenous to Africa, particularly Uganda (Edwards and Bogdan, 1951) and is grown in most tropical countries because of its high yields of palatable forage. It is vegetatively propagated and needs a high rainfall for best results. Most types are tall and robust, but Grof ( 1 96 1) developed Capricorn, a shorter grazing type for north Queensland conditions.

TROPICAL PASTURES

49

Bulrush millet (Pennisetum typhoides) is a valuable annual fodder crop grown in the monsoonal tropics. Norman and Begg ( 1 968) have found cultivars of bulrush millet which give high yields of standover forage for dry season feeding of cattle in the Northern Territory. The deep rooting habit of this species enables it to withstand drought and recover nitrate nitrogen at depth in the soil. J. SETARIA

The S . sphacelata complex was reviewed by Hacker and Jones (1 969), who discussed its botany, natural distribution, agronomy, cytology, and breeding and its future in Australia. It is widespread in Africa and is very variable because of its cross-pollinating habit. Various ecotypes have been studied in Kenya (Bogdan, 1955, 1965); in Australia, Nandi, selected by Bogdan (1959), and Kazungula, developed in South Africa (Chippindall, 19551, have been commercialized. These cultivars establish readily from seed, are palatable, and persist under grazing on a wide range of soils, but they need 35-40 inches or more of annual rainfall for good production. They combine well with a range of legumes and with proper management form a stable pasture. A difficulty with setaria is its oxalic acid content (Dougall and Birch, 1967), which under some conditions can reach a level high enough to cause oxalate accumulation in the kidneys and death of cattle (Jones et al., 1970). These conditions need study, as they are liable to occur occasionally in the field, particularly with cultivars, like Kazungula, which have a high oxalic acid content. K. SORGHUM There has been considerable interest in perennial tetraploid (2n = 40) sorghums for forage and pasture since the advent of S. afmum (L. R. Parodi, 1943). Johnsongrass (S. hafepense) in spite of its weed potential (Roseveare, 1948) has been used as a pasture plant in some countries, and at the Texas Agricultural Experiment Station perennial sweet sudan was bred from a cross between it and sudangrass (Hoveland, 1960). R. A. Parodi and Scantamburlo ( 1 954) continued their work on S. afmum and developed strains with better yields and less aggressive rhizomes than the common types. Krish is a perennial diploid (2n = 20) sorghum selected by Pritchard ( 1 964) at the Cunningham Laboratory from progeny of S. halepense X S . roxburghii (Krishnaswamy et al., 1956). S . afmum is adapted mainly to the subtropics and thrives on fertile soils in annual rainfalls of 20-35 inches in a number of countries including Argentine, South Africa, the United States, and Australia. It is a

50

E. M . HUTTON

useful pioneer and establishes and grows quickly, producing a bulk of forage of high feeding value. J. G. Davies and Edye ( 1 959) discussed the agronomy and potential of S . almum in Australia, its genetics and breeding, and the various ecotypes introduced. It is no more cyanogenetic than other sorghums, so care with grazing is needed only when growth is stunted. Gates et al. (1966~)showed that three sorghum species, including S . almum, had a high resistance to salinity stress. Crooble (Boyle, 196I), the main cultivar grown in northern Australiz,, is particularly valuable in the brigalow region. After clearing and burning brigalow (Acacia harpophylla) and associated trees, a sowing of Crooble in the ash gives a highly productive pasture for cattle fattening over two years. Combined with lucerne and barrel medics it provides a pasture for about four years in rotation with crops like grain sorghum and wheat. Krish (Pritchard, 1964) combines high yield and palatability with a degree of frost tolerance and disease resistance. It can be grown in the wetter coastal areas of southern Queensland, where Crooble is badly affected with leaf diseases.

L. UROCHLOA The Urochloa species with potential as pasture grasses are U.bolbodes, U . mosambicensis, and U.pullulans (Miles, 1949; Chippindall, 1955; Bor, 1960), the last two being naturalized in northeastern Australia. All are African in origin, and U.mosambicensis is native also to Burma. U . mosambicensis is considered to have value as an associate for Townsville stylo north of the Tropic in Australia, and commercial seed of one ecotype is produced. It is very palatable even at maturity when some greenness is retained. The study of a new range of U . mosambicensis ecotypes from Africa may result in better-adapted cultivars. XII.

Feeding Value of Grasses versus Legumes

French (1957) in a review on the nutritional value of tropical grasses notes that the data showed them to be high in fiber and low in crude protein and that work was required on the effects of crude fiber and lignification on organic matter digestibility. The factors involved in the feeding value of tropical grasses and legumes as determined by indoor feeding experiments with sheep are discussed by Milford and Minson (1966a, 1966b). Level of animal production from pasture is directly related to its feeding value, the most important factor being the voluntary intake of dry matter, which is correlated with dry matter digestibility. Prime determinants of intake are species or variety of pasture plant and its

TROPICAL PASTURES

51

maturity. Intake of 30-day-old growth of tropical grasses was relatively high, whereas that of 150-day-old mature growth fell off markedly except in a cultivar like Callide Rhodesgrass. By contrast, intake of Cooper glycine and siratro remained high to maturity. Milford (1960, 1967) found that intakes of digestible dry matter of 17 tropical grasses were lower than those of seven tropical legumes and lucerne, the difference being greatest in autumn and winter. Dry matter digestibility of tropical grasses is generally lower than that of temperate grasses. Minson and McLeod (1 970) found a high negative correlation ( r = - 0.76) between dry matter digestibility and temperature in both tropical and temperate grasses. They concluded that the difference in digestibility between tropical and temperate grasses is mainly due to the conditions under which each is normally grown, the former in warm to hot and the latter in cool. Legumes did not behave in the same way since respective digestibilities of the temperate white clover and tropical siratro were similar in both summer and winter. Milford and Minson ( I 966b) showed that decline in digestibility with age was rapid in tropical grasses compared with tropical legumes which retain relatively high digestibility at maturity and even after frosting (Milford, 1967). Cowpea and Rongai lablab are examples of annual legumes which maintained a high feeding value (digestibility and intake) with age but cowpea was superior at the first harvest (Milford and Minson, 1968a) and both had very high voluntary intakes compared with Rhodesgrass of the same digestibility (Milford and Minson, 1968b). The fact that feeding value of tropical legumes is superior to that of tropical grasses at all but the earliest growth stages has important implications in the development and management of legume-based tropical pastures. Among a range of tropical grasses in Trinidad, Butterworth ( I 964) found that pangola, jaragua, and Kazungula setaria had the highest digestible energy contents. In a study of pangolagrass, S.almum, and siratro, Minson and Milford ( I 966) found that age was the most important factor determining the digestible energy content. Rate of fall in energy digestibility percentage with maturity was least in siratro. Also siratro gave a higher daily intake of digestible energy than the two grasses. These results indicate that the relatively rapid decline in feeding value of tropical grasses with age is due to increasing shortage of digestible energy from a rapid buildup in crude fiber. Milford and Minson (1966a) consider that crude protein content does not limit intake of tropical grasses until it falls to 7%. Young growth of tropical grasses usually provides enough protein for the animal, but Milford and Haydock ( I 965) showed that content and digestibility of crude

52

E. M. HUTTON

protein declined more rapidly with age in tropical grasses than legumes like siratro. Among the grasses the decline was least in kikuyu and greatest in buffel and pangola. Thus, although feeding value of pangolagrass is high, animal intake near maturity can be limited by a deficiency of crude protein. Minson (1967) showed that intake of pangola can be increased 54% when crude protein content is raised from 3.7 to 7.2% by urea application a month before harvesting. Voluntary intake of mature pangola was also increased (Minson and Milford, 1967a) by including 10-20% lucerne or white clover in the diet. Playne (1969a,b) studied the effect of supplements on intake and digestibility of speargrass and Townsville stylo. Control sheep fed Townsville stylo gained weight, and those fed speargrass and urea lost weight. Dicalcium phosphate supplement increased intake and liveweight of sheep fed Townsville stylo, but there was no response in those fed speargrass and urea. Supplementing sheep with sodium sulfate, gluten, or sodium sulfate plus gluten markedly increased intake and dry matter digestibility of hay comprising four parts speargrass, one part Townsville stylo, and resulted in substantial liveweight gains compared with losses in the control sheep. Thus the phosphorus, sulfur, and sodium contents of tropical forages appear to significantly affect their feeding value. Several important points emerge from the research on feeding value of tropical species. The higher intake of digestible energy and protein of tropical legumes at all but the early growth stages makes it evident that legume-based pastures should be managed to contain a high proportion of legume. Also to further increase the feeding value of tropical pastures, every effort should be made to select grasses with a relatively high intake and crude protein content at maturity. This will not be easily achieved as illustrated by the quite small differences in feeding value obtained by Milford and Minson (1968b) between a number of Rhodesgrasses. It is now apparent that the physiological and chemical factors associated with the relatively rapid increase in structural carbohydrates in tropical grasses with time need intense study. Most of the work on intake and feeding value has been done in indoor feeding experiments with dried and chaffed material, and there could well be important differences in intake between housed and grazing animals, as shown by T. B. Miller’s results (1969). Selective grazing occurs in the field and species in the green state could have a different palatability and intake fed dry although Minson (1 966) found no difference in intake between fresh and dried material of three tropical grasses. Indoor trials have allowed rapid progress to be made in knowledge of the feeding value of tropical pasture species, but the stage has been reached when intake of the grazing animal animal needs investigation.

TROPICAL PASTURES

53

XIII. Phosphorus and Nitrogen Fertilization of Grass

Relatively cheap nitrogenous fertilizers are now becoming more generally available. In the subtropics and tropics it does not appear that they will be used to any extent except in favorable situations with an extended growing season and proximity to profitable markets. Regular and heavy application of nitrogen to grass will not achieve high animal production unless adequate amounts of other essential nutrients, such as phosphorus, sulfur, potassium, and calcium, are also supplied. Dry matter responses of tropical grasses to nitrogen have received much attention. However, more studies of the interaction of nitrogen and other elements on composition and yield are needed to define mineral requirements of the various grasses. Burton et al. ( 1 969) found that omitting phosphorus and potassium from fertilizer which supplied 600 Ib of nitrogen per acre to coastal bermudagrass growing on a sandy loam reduced its yield 45% without affecting its protein, carotene, and xanthophyll contents. It appeared that inadequate potassium was responsible for this response and that applying large quantities of phosphorus and potassium to coastal bermuda scarcely increases protein, carotene, and xanthophyll contents. Mineral composition of molasses, pangola, napier, kikuyu, bermuda, and guinea grasses growing on a sandy loam in central Brazil was studied by Gomide et al. ( I 969a). Adequate amounts of ammonium sulfate, superphosphate, and potassium chloride were used at planting, and different levels of nitrogen were applied at regular intervals subsequently. Nitrogen fertilizer had no effect on any of the minerals studied except manganese, which increased. With increasing age there were significant decreases in potassium, phosphorus, magnesium, copper, and iron. Average potassium content of all grasses at 4 weeks of age was 1.42%: and at 36 weeks, 0.30%. Kikuyu had the highest potassium content at 36 weeks. Average phosphorus content of all grasses was 0.26% at four weeks, and 0.12% at 36 weeks. Pangolagrass was a poor source of phosphorus at all ages. In the rest of this section specific responses of tropical grasses to phosphorus and nitrogen will be discussed with emphasis on the animal production resulting from nitrogen fertilization. A. PHOSPHORUS

E. W. Russell ( 1 966) investigated the effects of different amounts of superphosphate on yield and composition of stargrass in Kenya; he found that yield was not always increased and that phosphorus content increased most in the wet season. Other studies have been made of phosphorus levels in tropical grasses, but Birch (1 953) attempted to define

54

E. M. HUTTON

the requirement for this element. He found that if the phosphorus content of kikuyu and Rhodesgrass exceeded 0.33% there was no response to added phosphate, but if it was less that 0.23% a significant response was obtained. In Trinidad Ahmad et al. (1969a) found that phosphorus content of pangolagrass tended to increase with age and that 0.26% in the grass did not limit growth. Andrew and Robins (1970) investigated the effect of phosphorus on growth and chemical composition of a number of tropical grasses; among these, molassesgrass was the most responsive and kikuyu the least responsive. Critical percentages of phosphorus in the tops of molassesgrass, Gayndah buffel, common paspalum, green panic, pioneer Rhodes, S . almum, Nandi setaria, pangola, and kikuyu were 0.18, 0.26, 0.25, 0.19, 0.23, 0.20, 0.22, 0.16, and 0.22, respectively. Apart from increasing phosphorus concentration in the tops, phosphate application decreased nitrogen and potassium, did not affect calcium, increased magnesium in most species, and increased sodium in four grasses. Pangola had the lowest phosphorus and nitrogen concentrations, and Nandi setaria the highest. Rhodesgrass, green panic, and pangola had high sodium and Gayndah buffel was intermediate; these species had relatively low potassium and magnesium. Nandi setaria had high potassium and low sodium, and molassesgrass, S . almum, and kikuyu had high magnesium, low sodium, and intermediate potassium. S . almum and common paspalum were relatively high in calcium, and pangola was low.

B. NITROGEN Most tropical grasses have a capacity for high photosynthetic rates (Ludlow and Wilson, 1968), and high dry matter production of 10,000 Ib or more per acre in response to nitrogen fertilization is usual in the humid subtropics and tropics (vide Prine and Burton, 1956; Romney, 1961 ; Vicente-Chandler et al., 196 1 ; Oakes and Skov, 1962; Ahmad et al., 1969b). In southeastern Queensland Henzell(l963) found that tropical grasses (Rhodes, paspalums, S . almum) given adequate superphosphate and potash yielded only 1000-5000 dry matter per acre per year, but up to 20,000 Ib per acre in wet years when as much as 400 lb of nitrogen was also applied. Nitrogen content of the grasses was relatively low unless nitrogen was applied in excess of the requirements for maximum growth. The best nitrogen recoveries in plant tops were 40-50% obtained at the higher nitrogen applications. It was considered that efficiency of dry matter production relative to applied nitrogen and recovery of nitrogen in the tops could both be improved. Henzell and Stirk ( 1 963)

TROPICAL PASTURES

55

concluded that nitrogen rather than soil moisture is the primary limiting factor to grass growth under natural rainfall in southeastern Queensland. The relatively low nitrogen content of tropical grasses compared with temperate is thought by Henzell and Oxenham (1964) to be due in part to a greater degree of nitrogen deficiency in warm climates. Pangolagrass, because of its adaptability and productivity, is being used increasingly throughout the humid tropics and subtropics (Nestel and Creek, 1962). In the Wallum of southeastern Queensland, Bryan and Sharpe ( 1965) applied increasing rates of nitrogen as urea to pangola and obtained maximum annual yields of dry matter of 2 1,000 Ib per acre and of nitrogen of 200 Ib. The mean yield of nitrogen in plant tops was 45% of that applied, and the nitrogen content was generally low. In Guadeloupe, with dry season irrigation and 1735 lb of nitrogen per acre, Salette (1 966) produced in a year 35,159 Ib of dry matter per acre from pangola, the highest yield recorded for this grass. The uptake of nitrogen by Rhodesgrass was followed with labeled nitrogen fertilizers by Martin et al. ( I 963), Henzell et al. ( I 964, 1968), and Vallis et al. (1967). A mean of 93.6% of added total nitrogen and 94.0% of added isotopic nitrogen was recovered from a soil-Rhodesgrass system. Nitrogen uptake by the grass was a linear function of the quantity of labeled ammonium nitrate applied for rates up to the equivalent of 400 Ib of nitrogen per acre, but the proportion of fertilizer nitrogen recovered in the plants fell significantly when 800 lb of nitrogen per acre was used. Fertilizer nitrogen was distributed between tops and roots in the ratio of 5.2 : 1 for total nitrogen and 4.5 : 1 for isotopic nitrogen. Rhodes grass when grown in association with tropical legumes took up considerably more labeled nitrogen than the legumes. I t is apparent that tropical grasses give a high dry matter response to fertilizer nitrogen, but there are problems in uptake and recovery of applied nitrogen as well as in the maintenance of a sufficiently high nitrogen content. Also, high yields of dry matter are of limited value if they cannot be efficiently converted into high yields of animal products. Factors involved in feeding value of tropical grasses given adequate N , P, and K were studied in central Brazil by Gomide et al. (1969b). They found that mean dry matter percentage of molasses, pangola, napier, kikuyu, bermuda, and guinea grasses increased from 2 I .2 to 5 1.4%with time. Napier had the lowest dry matter percentage, and bermuda the highest. Nitrogen decreased dry matter percentage in the early stages, but not later. Cellulose percentage of molasses, napier, and guinea increased with age while that of the other three grasses remained almost constant. Crude fiber

56

E. M. HUTTON

percentage increased with age, and in vitro cellulose digestibility decreased. The overall effect of nitrogen fertilization was to decrease in vitro digestibility except in the young growth. Grasses which give high yields of digestible nutrients with nitrogen fertilization are required. Two of the most promising are pangola and kikuyu, shown by Milford and Minson ( 1 966b) to retain a relatively high digestibility with age. Kikuyu has the added advantage of maintaining high crude protein with age (Milford and Haydock, 1965). Nitrogenfertilized signalgrass ( B . decumbens) is giving high animal production in the Queensland wet tropics (Grof, 1969), so it is another promising species for a nitrogen system. A number of experiments on animal production from nitrogen-fertilized grass have been reported, and only a few of these will be quoted. Suman et al. (1962) in the southeastern United States measured beef liveweight gains on coastal bermuda, common bermuda, and Pensacola bahia grasses fertilized annually with 100, 200, and 400 Ib of nitrogen per acre; they obtained the highest gains from coastal bermuda. During three years, with 400 Ib of nitrogen coastal bermuda produced a mean of 91 3 to 938 Ib liveweight gain per acre per annum. Over four years in the humid tropics of Puerto Rico, Caro-Costas et al. (1965) grazed young beef steers on separate pastures of five grasses which annually received 280 Ib of nitrogen per acre and also phosphate and potash. Pangola, guinea, and napier grasses produced similar yields averaging 1058 Ib liveweight gain and 7559 Ib total digestible nutrients per acre per annum at 2.5 steers per acre. Para and molasses grasses were inferior and produced an average of only 636 Ib liveweight gain and 5030 Ib total digestible nutrients per acre per annum at 1.6 steers per acre. On the Wallum of southeastern Queensland, T. R. Evans ( 1 969a) obtained, over two years, mean liveweight gains of 1 139 Ib and 12 I5 Ib per acre per annum from pangolagrass fertilized annually with 400 Ib and 800 Ib of nitrogen per acre, respectively, and grazed at a mean of three beasts per acre. Using 400 Ib of nitrogen per acre per annum, T. R. Evans ( 1969b) has improved these results markedly by attention to timing of the split applications of nitrogen. In north Queensland at the Parada Research Station of the Queensland Department of Primary Industries, annual beef liveweight gain averaged 1800 Ib per acre in the last three years on irrigated and nitrogen fertilized (300 Ib N per acre per annum) pangolagrass stocked at a mean of three beasts per acre (J. Evans, 1967). Both the Wallum and Parada experiments also received adequate annual amounts of superphosphate and potash. At Wollongbar Research Station in the subtropics of northern New

TROPICAL PASTURES

57

South Wales, Holder ( I 967) reported that fertilizing kikuyugrass with 300 lb of nitrogen per acre per annum increased both stocking rate and butterfat production per acre per annum two to three times. In the humid tropics at Turrialba, Costa Rica, Blydenstein et al. ( 1969) obtained over a year a mean milk yield of 5367 Ib per acre from pangolagrass fertilized with 205 lb of nitrogen together with phosphate and potash. Efficiency of dry matter conversion in this experiment was about 12%. In Jamaica, Nestel and Creek (1966) have shown that intensive beef production or dairying on nitrogen-fertilized pangolagrass gives a good return on money invested provided attention is given to stocking rate and effective managerial practices. Henzell ( 1 968) has outlined the problems associated with use of nitrogen fertilizers on grass. These include requirements for other nutrients such as superphosphate and potash, loss of nitrogen to the air, acidifying effects, management needed, and the relatively low feeding value of the bulk of grass produced. This last problem is the most important in animal production. Although nitrogen fertilization markedly increases dry matter yield of tropical grasses, all the evidence indicates that except in the young growth stage it does not increase concentration of digestible nutrients and efficiency of conversion into animal products. XIV. Breeding and Genetics of Tropical Grasses

A major aim in breeding tropical grasses is to produce varieties that do not decline rapidly in feeding value as the plants mature. In vitro digestibility (Minson and Milford, 1967b) and crude protein content could be used to select from large numbers of samples the few with potentially higher feeding value. These selections could then be grown in larger plots where their response to nitrogen fertilization could be measured and their feeding value determined from a bulk of dried material fed to sheep (Minson and Milford, 1968). The final criterion for selection is the animal production obtained from the one or two superior selections grown in a pasture either associated with a standard legume or fertilized with nitrogen. Compatibility with a legume is essential in most grasses, as for many years legume-based pastures will be the main source of cattle feed in a majority of the areas now being developed in the Tropics. It may be necessary to breed slower growing grasses which do not compete strongly with legumes at the peak of the season to achieve a balanced pasture. Standard breeding programs can be adopted with cross-pollinating grasses, which include setaria, S. almum, Rhodes, P. coloratum, and spe-

58

E. M. HUTTON

cies of Digitaria. Setaria is almost entirely cross-pollinated (Gildenhuys, 1960), which facilitates hybridization, and although S. almum is selfcompatible, hot water can be used to emasculate whole inflorescences (Pritchard, 1965a). A number of the important tropical grasses are apomictic. These include buffel (Snyder et al., 1955), guinea (Warmke, 1954), green panic, molasses, paspalums (Bashaw and Holt, 1958), and species in the genera Brachiaria (Pritchard, 1967), and Urochloa. In obligate apomicts, the only variation available for selection is that which exists between the accessions collected from different ecological niches. Most produce some functional pollen, so crosses are possible if sexual forms of the apomicts can be found. Burton and Forbes (1960) overcame the apomictic barrier in P . notatum by crossing the common apomictic tetraploid and fertile induced tetraploids from the sexual diploid Pensacola bahiagrass. The search for a sexual type in other important apomicts has been successful only in buffelgrass (Bashaw, 1962), and the resultant crosses have released considerable variation and have shown apomixis to be recessive to sexuality. A. SETARIA The main objectives of the breeding work with setaria at the Cunningham Laboratory are to produce cultivars with frost resistance, high feeding value, low oxalic acid content, and an extended growing season. The diploid Nandi (2n = 18) and tetraploid Kazungula ( 2 n = 36) belong to the S . sphacelata complex, which also contains pentaploid, hexaploid, octoploid, and decaploid races (Hacker, 1966). Crosses have been obtained by Hacker ( 1967) between all proximate ploidy levels except diploid and tetraploid, and also between high and low levels. Thus for seed production, lines of setaria should be isolated from each other. Hacker (1968a) has cast doubt on the validity of the separation of species in the S.sphacelata complex as he has been able to hybridize diploid forms of S . anceps and S . trinervia, and S.anceps and S . splendida, and hexaploid lines of all three species. From Hacker's work (1 968b) it appears that the S . sphacelata complex forms an autopolyploid series. B. Sorghum almum The aim is to breed lines of S . almum with higher yield and persistence than the Australian cultivar Crooble and possessing juicy stems, distinctive brown glumes, late flowering, and tolerance to leaf diseases. Pritchard ( 1965a) crossed S . almum and perennial sweet sudangrass (Hoveland, 1960) and found that juicy stem and brown glume and plant color of the latter were linked and mainly tetrasomically inherited. Selection was

TROPICAL PASTURES

59

facilitated by an association between translucent midrib, juicy stem, and high soluble carbohydrate, and some of the advanced lines have a 20% higher soluble carbohydrate in the stem than Crooble. Using a tetrasomically inherited albino seedling character in S . almum, Pritchard ( 1965b) studied natural crossing in S . almum and between S . almum and the weed S . halepense and concluded that there was a degree of genetic isolation between these species. From cytological examination of aneuploid plants of S . almum and S . halepense I’ritchard (196%) suggested that these species were autotetraploids. Pritchard ( 1965d) crossed diploid Sorghum with S . almum and obtained tetraploids and triploids. Segregation in tetraploid progeny resulted in tramfer of certain characters from the diploid to tetraploid level. Slight fertility of the triploids also enabled transfer of characters from the diploid to tetraploid (or near tetraploid) by backcrossing and selfing. The triploids could also be used to transfer such characters as perenniality from tetraploid to annual diploid sorghums. C. COASTAL BERMUDAGRASS Burton’s notable development ( 1 947, 1954) of (coastal bermudagrass has been followed by further work aimed at improving its agronomic characters and feeding value. Dry matter digestibility of the many genotypes was determined by Burton er al. ( 1 967), who found that quality of a number of clones decreased as age of forage increased and that genotype X age interactions were not significant. A coastal >: Kenya 56 F1 hybrid averaged 12.3% more digestible dry matter than coastal over a four-year period. Several Midland X Kenya 6 1 hybrids had higher yields and better digestibility than either parent. This work has indicated that the quality of C . ductylon may be improved by breeding.

D. BUFFELGRASS At the Cunningham Laboratory, Pritchard (196’7) crossed the sexual buffelgrass from Texas (Bashaw, 1962) with the main apomictic cultivars and obtained more variation than has been assembled in over thirty years of introduction. A number of promising leafy lines which flower later than the parents and have greater cold tolerance have been selected. The in vitro digestibility at maturity of some of these is siiperior to that of the parents, so it may be possible to improve the feeding value of buffelgrass. Pritchard (1 967) found that the original sexual plant has a chromosome number of 2n =36 and that selfed progeny numbers range from 2n = 35 to 2n = 38. Chromosome numbers of the apomictic cultivars are 2n = 36 for Molopo and Lawes and 2n = 43 for Tarewinnabar, Nunbank, and Biloela.

60

E. M. HUTTON

Progeny of crosses between these apomicts and the sexual type have chromosome numbers from 2n = 34 to 2n = 45. The examples given indicate that agronomic characters and feeding value of tropical grasses can be improved by breeding. Of the grasses not mentioned in this context, pangola and kikuyu merit attention although they pose special difficulties for the breeder. There is a need to increase their adaptability and it is possible that even their relatively high feeding value and response to nitrogen could be improved. In future it is hoped, as more information becomes available on the physiological and biochemical characters which control growth and adaptation of grasses, that more precise selection in breeding populations will be possible. Xv. Beef Production from Legume-Based Tropical Pastures

The degree to which tropical pasture research and development is successful can be measured only in terms of animal production and its profitability. Productivity of nitrogen-fertilized grass systems has already been dealt with, and in this section animal production from legume-based pastures is discussed. Because of their flexibility and cheap production of N for pasture and grazing animal, legume-based pastures will continue to predominate in development of tropical areas. Tropical legumes are not able to produce enough N for the associate grasses to attain their potential dry matter production, but does this matter? Feeding value of the pasture is the main determinant of animal production, so a higher proportion of legume and smaller bulk of less digestible grass is an advantage. In Norman and Stewart’s experiments (1 964) (see Table 2) at Katherine in the Northern Territory, liveweight gain of cattle was directly related to the proportion of Townsville stylo in the pasture. TABLE I1 Dry Season Performance of Cattle on Sown Pastures with Varying Proportions of Grass and Legume Composition of pasture (%) Birdwood grass

Annual grasses

Townsville stylo

Nitrogen content of pasture at start of grazing (%)

51.5 9.9 -

25.1

45.4 31.4

22.8 44.7 62.6

0.75 1.12 1.34

Liveweight gain (Ib/head) 20 99 196

Period of gain (weeks)

8 20

22

TROPICAL PASTURES

61

In general, progress in tropical beef production depends on persistent and adapted legumes, regular application of superph,osphate (giving both P and S), and use of adapted tropical cattle with tick. resistance and heat tolerance (Schleger and Turner, 1965). Significant advances have been made despite Whyte’s pessimistic conclusions ( 1962) concerning improvement of tropical grasslands. The following examples from areas in the main tropical climates will make this clear. A. WET TROPICS Younge et al. (1964) estimated that about one-quarter of the Hawaiian rangelands are unproductive low wetlands but capable of trebling the current annual beef production. On the island of h4aui, pangolagrassD . intortum pastures fertilized once per acre with lime at about 3000 Ib and a starting fertilizer comprising 44 Ib N, 84 P, 104 K, 3.5 B, and 2.5 Mo gave a mean over two years of 764 Ib liveweight gain per acre per annum at a stocking rate of about two beasts an acre, which was highly profitable. Also on Maui, pangola, dallis, kikuyu, and native grasses each mixed with kaimi clover and given one dressing per acre of 6 tons of calcium carbonate and starting fertilizer as in the previous experiment produced over four years annual liveweight gains per acre of 720, 630, 575, and 524 Ib, respectively. At the Kauai Branch !station on an aluminous-ferruginous latosol, Younge and Plucknett (1966) with a pangolagrass-D. intortum pasture given an application of basic fertilizers and four rates of P produced as high as I164 Ib liveweight gain per acre per annum with yearling steers. Mean stocking rate varied from 1.18 beasts an acre at the lowest P level to 2.38 an acre at the highest. There was a curvilinear response to P, but the highest liveweighi: gain was obtained from the heaviest P application. Unimproved pastures produce about 30 Ib liveweight gain per acre per annum.

B. MONSOONAL TROPICS Over two dry seasons at Katherine in the Northern Territory, liveweight of steers at a beast to 2 acres of Cenchrus-Townsville stylo pasture increased and that of steers at a beast to 17 acres on native pasture declined substantially (Norman and Arndt, 1959). Shiiw (196 I ) at Rodd’s Bay, Queensland, showed that year-round productivity of native speargrass pasture could be increased markedly by oversowing with Townsville stylo and topdressing annually with 1 cwt molybdenized superphosphate an acre. Carrying capacity of native pasture was trebled from a steer to 9 acres to a steer to 3 acres, annual liveweight gain per acre was

62

E. M . HUTTON

increased five times, and steers were marketed one to two years earlier than those on unimproved pasture. Shaw’s work ( 1 96 I ) was followed up by Edye and Ritson ( 1 969) with a four-year study of cow fertility and beef production in Droughtmasters (Shorthorn X Brahman) continuously grazed on Townsville stylo-speargrass pasture (at Lansdown Pasture Research Station) near Townsville, Queensland (Fig. 6). Pasture treatments were annual superphosphate applications per acre of nil, I cwt, and 3 cwt, respectively. The breeders were carried at a cow to 3 acres or 6 acres, and application of superphosphate raised mean calving percentage from 6 1 to 8 1%. Weaner steers were carried at a beast to 2 acres or 4 acres and at 28-30 months of age were around 800 Ib liveweight without superphosphate and 1000-1 100 Ib with superphosphate.

FIG. 6. Droughtmaster cattle grazing on Townsville stylo-speargrass pasture at Lansdown Pasture Research Station, near Townsville, Queensland.

In the Northern Territory Norman and Stewart (1964) found that pastures dominated by Townsville stylo gave the best liveweight gains. Following this finding Norman (1968) studied the performance of beef cattle on different sequences of native pasture and fertilized Townsville

TROPICAL PASTURES

63

stylo and showed a linear inverse relationship between the time spent on Townsville stylo and the total time from weaning to slaughter. The group maintained on Townsville stylo continuously from weaning reached slaughter weight at 30 months of age. All these results have stimulated the use of Townsville stylo fertilized with superphosphate in many areas of the monsoonal tropics of northern Australia. Woods (1969) documented the rapid expansion in area of Townsville stylo pastures in the Northern Territory since 1966; up to 10,000 acres a year are being sown on individual properties. The practice has been shown to be profitable (Haug and Hirst, 1967) in the speargrass zone and is apparently economic in the other areas where it is used. In Nigeria, Okorie et al. (1965) demonstrated that ]productivepastures of giant stargrass and centro fertilized with superphosphate could be maintained under rotational grazing and were suitable for fattening N’Dama cattle. Crude protein level of the pasture varied from 7.2 to 16.4% and organic matter digestibility from 68 to 77%. At a stocking rate of 1.7 to 2.7 steers an acre, mean annual liveweight gains per acre were 320 Ib. In Uganda, Stobbs (1 966) compared production from small east African Zebu cattle on grass and grass-legume pastures at stocking rates of 1 to 1.5 beasts an acre. Annual liveweight gain per acre was increased from 188 lb to 470 Ib by the addition of centro and Stylosanthes guyanensis and fertilizers. Introduction of legumes without ferl ilizer gave smaller increases and the pasture became dominated by H. rhfa and S . guyanensis. In further elaboration of his results, Stobbs (1 969c) noted that grasslegume (centro, S . guyanensis) pastures receiving P and S produced liveweight gains similar to those from grass pastures given 140 lb N per acre per annum in addition to P and S. Stobbs (1 969b,d) outlined the effects of stocking rate and grazing frequency on legume-based pastures in Uganda; increased liveweight gain per acre resulted from heavy stocking and from rotational grazing only in the dry season. Milk production from a Setaria sphacelata-D. intorturn pasture was studied by Deschuytener (1 967) in the Republic of Rwanda. At a stocking rate of a cow per 1.2 acres without a supplement, he obtained 445 gallons of milk per acre.

c. HUMIDSUBTROPiCS The development of beef production on the Wallum of southeastern Queensland with its very poor sandy and acid soils and annual rainfall varying from 70 inches in the south to 40 inches in the north resulted from successful integration of research on plant nutrition, pasture species, and

64

E. M . HUTTON

grazing management. Bryan ( 1968a) described the initial grazing trials at Beerwah in the southern Wallum of four pastures with a mean legume content of 13% and based, respectively, on common paspalum, Paltridge scrobic, Rodd’s Bay plicatulum, and pangolagrass. Heavy dressings of fertilizer as described previously were applied and the main legumes were silverleaf desmodium, Miles lotononis, and white clover. Stocking rate varied from a beast to 1 to 1.5 acres and mean annual liveweight gain was 244 Ib an acre while the best pastures produced 262 Ib an acre. Crude protein content of the pastures was probably a limiting factor in autumn and winter. Bryan (1 968b) found that common grazing of eight legumes and eight grasses in complex mixtures resulted in stabilized pastures with a legume content of 25 to 30% which produced a mean annual liveweight gain of 300 Ib an acre. The most persistent species were the legumes white clover, silverleaf desmodium, and Miles lotononis and the grasses common paspalum, Rodd’s Bay. plicatulum, bahia, pangola, and Rhodes. Bryan and Evans (1 968) showed that three pastures at Beerwah with legume contents from 13 to 30% gave increased animal production per acre with increasing legume content and that the least depression in liveweight gain in winter occurred with the highest legume content. Using the legumes white clover, Miles lotononis, and greenleaf desmodiurn with pangolagrass, Bryan and Evans (1968, 1969) obtained 450 Ib annual liveweight gain an acre over a three-year period at a stocking rate of 1 to 1.5 beasts per acre and with a legume content of 27-30% in the pasture. Economic studies in the southern Wallum by W. L. Moore ( 1 967) have shown that investment there in beef production would be profitable. In the northern Wallum where annual rainfall is 40-45 inches, T. R. Evans (1 968) produced vigorous pastures carrying a steer an acre and giving annually 350-400 Ib liveweight gain an acre. The pastures have been developed from siratro, greenleaf desmodium, or Miles lotononis and one of the grasses Nandi setaria, Rhodes, P . plicatulum or P . colorarum. After .a heavy initial application of fertilizer annual dressings of 2 cwt superphosphate and 1 cwt potassium chloride an acre were needed. At Mississippi in the southeastern United States with a 50-inch annual rainfall, Hogg ( 1966) developed highly productive pastures of coastal bermudagrass and white and red clovers. These produced quality forage more economically than nitrogen-fertilized grass. Annual liveweight gain per acre on coastal bermuda was 5 17 lb with white clover and 282 Ib with 120IbN. D. SUBHUMID SUBTROPICS

In a relatively dry but fertile brigalow area of southern Queensland, a mean cattle liveweight gain of 174 Ib an acre was obtained during a year

TROPICAL PASTURES

65

of below average rainfall from grass-legume pastures stocked at a beast to 2.3 acres (Coaldrake and Smith, 1967). Animal production on sown and native pastures in the brigalow was measured for three years by Coaldrake et al. (1969) during conditions which changed from moderate to severe drought. Sown pastures of a grass and Hunter River lucerne were grazed continuously at a beast to 2.3 acres while native pasture was stocked at a beast to 4.6 acres, but results were meaningful only in the first two years. Annual liveweight gain per acre was lowest on native pasture, highest on S. almum at 198 Ib in the firs1 year, and highest on green panic at 166 lb in the second year when S. almum died. Lucerne died out under continuous grazing, and buffel was the most droughtresistant grass. Drier subtropical areas like the brigalow (annual rainfall 24-27 inches) present difficult problems in grazing management and legume persistence. XVI.

Summary a n d Conclusions

In the last decade tropical pasture research has culminated in the provision of persistent and productive legumes and grasses for use in pastures throughout the tropics. It has shown that, with proper fertilization and management of these cultivars, the tropics can produce considerably more beef and milk than at present. Although nitrogen fertilization of high quality grasses like pangola and kikuyu leads to high stocking rates and high production, the costs involved in this system are high, so it will be used only in special situations. Legume-based pasture will be the main system used in development of the extensive unused areas of the tropics for beef cattle as it is capable of markedly increasing production at relatively low cost. The studies on the interrelationships between legume, rhizobium, plant nutrients, and soil type have made it possible to successfully establish legumes in pastures in most tropical areas. The finding that the majority of tropical legumes and their assoicated rhizobia are: adapted to acid soils is of special significance in tropical development. Determination of critical percentages of the major elements sufficient for maximum growth of the different legumes has given a sound basis for fertilization of tropical pastures. Characterization of the associate grasses in this respect has commenced but requires much more attention if a proper understanding of the mineral balance in tropical pastures is to be obtained. One question which needs to be resolved is whether the mineral content of pasture giving maximum dry matter yield is at a suffiiciently high level for maximum animal production. More work is required on the physiological and biochemical character-

66

E. M. HUTTON

ization of the different species so that their deficiencies relative to pasture and animal production in various tropical environments can be determined. This will provide a better basis for plant introduction and breeding activities and for management studies aimed at maximizing yields of digestible nutrients and animal products from the pasture. A proportion of the physiological work needs to be done in the field so that it is possible to ascertain the factors involved in the persistence and compatibility of cultivars in grazed pastures. Determination of the biochemical and physiological bases for the relatively low feeding value of tropical grasses is an important problem. Until this is solved, it will be difficult to improve the efficiency of conversion of tropical pasture into animal products. Future progress in tropical pasture research is dependent on maintaining the ecosystem approach, in which the interactions between soil, pasture, and animal are intensively studied in pastures in different environments. Use of modern computer techniques are essential to fully assess these interactions and devise methods for improving output from tropical pastures.

REFERENCES “After a Hundred Years” 1962. The Yearbook of Agriculture, U.S.D.A. Ahmad, N., Tulloch-Reid, L. I., and Davis, C. E. 1969a. Trop. Agr. Trin. 46, 179-186. Ahmad, N., Tulloch-Reid, L. I., and Davis, C. E. 1969b. Trop. Agr. Trin. 46, 173-178. Andrew,C. S. 1960.Aust. J.Agr. Res. 11,149-161. Andrew, C. S. l963a. Aust. J . Agr. Res. 15,654-659. Andrew,C. S. 1963b.Aust.C.S.l.R.O.. Rep.Div. Trop. Past., 1962-1963pp. 35-36. Andrew, C. S. 1965. J . Aust. lnst. Agr. Sci. 31,3- 10. Andrew, C. S. l966a. Aust. J . Agr. Res. 17,611-624. Andrew,C. S. 1966b.Aust. C.S.I.R.O., Rep. Div. Trop. Past., 1965-1966 p. 63. Andrew, C. S. 1968. J . Aust. lnst. Agr. Sci. 34,154- 162. Andrew, C. S., and Bryan, W. W. 1955.Aust. J . Agr. Res. 6,265-290. Andrew, C. S., and Bryan, W. W. 1958. Aust. J . Agr. Res. 9,267-285. Andrew, C. S., and Hegarty, M. P. 1969.Aust. J . Agr. Res. 20,687-696. Andrew, C. S., and Norris, D. 0. 196 I . Aust. J . Agr. Res. 12,40-55. Andrew, C. S., and Robins, M. F. 1969a. Ausr. J . Agr. Res. 20,665-674. Andrew, C. S., and Robins, M . F. 1969b. Aust. J . Agr. Res. 20,675-685. Andrew, C. S., and Robins, M. F. 1969~.Aust. J . Agr. Res. 20,999-1007. Andrew, C. S., and Robins, M. F. 1969d. Aust. J . Agr. Res. 20, 1009-1021. Andrew, C. S., and Robins, M. F. 1970. Aust. J . Agr. Res. (in press). Andrew, C. S., and Thorne, Peggy M. 1962. Aust. J . Agr. Res. 13,82 1-835. Anonymous. 1948. Hawaii, Agr. Exp. Sta., Rep. Biennium End. June 30 pp. 38-39. Bailey, F. M., and Tennison-Woods, J. E. 1879. Proc. Linn. Soc. N.S.W. 4,152. Barnard, C. 1967. “Australian Herbage Plant Register,” Div. Plant Ind., C.S.I.R.O., Australia. Bashaw, E. C. 1962. Crop Sci. 2,4 12-4 15.

TROPICAL PASTURES

67

Bashaw, E. C., and Holt, E. C. 1958.Agron.J . 50,753-756. Birch, H . F. 1953. J . Agr. Sci. 43,329-335. Blydenstein, J . , Louis, S., Toledo, J., and Camargo, A. 1969. J . Brit. Grassl. SOC.24,7 1-75. Bogdan, A. V. 1955. East Afr. Agr. J . 20, 151-165. Bogdan, A. V. 1959. East Afr. Agr. J . 24,206-2 17. Bogdan, A. V. 1965. East Afr. Agr. Forest. J . 30,330-338. Bogdan, A. V. I966a. Trop. Agr. Trin. 43.99- 105. Bogdan, A. V. 1966b. I n “Tropical Pastures” (W. Davies and C. L. Skidmore, eds.), Chapter 5, pp. 75-88. Faber & Faber, London. Bogdan, A. V. 1969. Herb. Abstr. 39, 1-13. Bor, N. L. 1960. In “The Grasses of Burma, Ceylon, India, and Pakistan.” pp. 372-373. Permagon, Oxford. Bowen, G. D., and Kennedy, Margaret M. 1961. Queensl. J. Agr. Sci. 18, 161-170. Bowyer, J. W., Atherton, J. G., Teakle, D. S., and Ahern, Gabriellt: A. 1969. Aust. J. Biol. Sci. 22,27 1-274. Boyle, A. J . 1961. Agr. Gaz. N.S.W. 72,338-344 and 379. Braithwaite, B. M. 1967. N.S.W. Dep. Agr., Rep. Wollongbar Agt. Res. Sta., 1964-1967 pp. 31-33. Bray, R. A. 1967. J . Aust. Inst. Agr. Sci. 33,46-47. Bray, R. A. 1969. Aust. J. Agr. Res. 20,47-55. Britten, E. J . , Matsumoto, H., and Palafox, A. L. 1959a. Agron. J . 51,462-464. Britten, E. J., Palafox, A. L., and Matsumoto, H. 1959b. Agron. J . 51,651-654. Britten, E. J . , Palafox, A. L., Frodyma, M. M., and Lynd, F. T. 1963. Crop. Sci. 3,415416. Bruce, R. C. 1965. Queensl. J . Agr. Anim. Sci. 22,221-226. Bryan, W. W. 1961. Aust. J . Exp. Agr. Anim. Husb. 1,4-10. Bryan, W. W. 1966. Proc. 10th Int. Grassl. Congr., 1966 Sect. 1 , Pap. No. 54. Bryan, W. W. 1968a. Aust. J . Exp. Agr. Anim. Husb. 8,5 12-520. Bryan, W. W. 1968b.Aust. J . Exp. Agr. Anim. Husb. 8,683-690. Bryan, W. W. 1969. Herb. Abstr. 39, 183-191. Bryan, W. W., and Andrew, C. S. 1955. Aust. J . Agr. Res. 6,291-298. Bryan, W. W., and Evans, T. R. 1968. Aust., C.S.I.R.O., Rep. Lliv. Trop. Past., 1967I968 pp. 20-26. Bryan, W. W., and Evans, T. R. 1969. Aust., C.S.I.R.O., Rep. D i v . Trop. Past., 19681969 p. 23. Bryan, W. W., and Sharpe, J. P. 1965.Aust. J . Exp. Agr. Anim. Husb. 5,433-441. Bryan, W. W., and Shaw, N. H. 1964. Aust. J . Exp. Agr. Anim. Husb. 4,17-21. Burton, G . W. 1947. J. Amer. SOC.Agron. 39,551-569. Burton, G . W., 1954. Georgia, Agr. Exp. Sta., Bull. [N.S.] 2. Burton, G. W., and Forbes, 1. 1960. Proc. 8th Int. Grassl. Congr., IS60 pp. 66-7 1. Burton, G. W., Hart, R. H., and Lowrey, R. S. 1967. Crop. Sci. 7.,329-332. Burton, G . W., Wilkinson, W. S., and Carter, R. L. 1969. Agron. J. 61,60-63. Butterworth, M. H. 1964. J. Agr. Sci. 64, 319-322. Byth, D. E. 1964. Nature (London) 202, 830. Cameron, D. F. 1965. Aust. J . Exp. Agr. Anim. Husb. 5, 49-51 Cameron, D. F. 1967a. Aust. J . Exp. Agr. Anim. Husb. 7, 489-494. Cameron, D. F. 1967b. Aust. J . Exp. Agr. Anim. Husb. 7, 495-500. Cameron, D. F. 1967c. Aust. J . Exp. Agr. Anim. Husb. 7, 501-.508. Cameron, D. F. 1967d. Aust. J . Agr. Res. 18, 375-379. Cameron, D. F. 1968. Ph.D. Thesis, University of Queensland.

E. M. HUTTON

68

Cameron, D. G. 1968. Queensl. Agr. J . 94, 534-543. Caro-Costas, R., Vicente-Chandler, J., and Figarella, J. 1965. J. Agr. Univ. P . R. 49,99111.

Christian, C. S., and Shaw, N. H. 1952. Aust. J. Agr. Res. 3,277-299. Chippindall, Lucy, K. A. 1955. “The Grasses and Pastures of South Africa,” pp. 412414. Central Newsagency, Parov, Cape Province, South Africa. Coaldrake, J. E. 1964. Bull. Commonw. Bur. Past. Field Crops 47,17-36. Coaldrake, J. E., and Smith, C. A. 1967. J . Ausr. Inst. Agr. Sci. 33, 52-54. Coaldrake, J. E., Smith, C. A., Yates, J. J., and Edye, L. A. 1969. Aust. J . Exp. Agr. Anim. Husb. 9,47-56. Coleman, R. G., Windrum, G. M., and Hutton, E. M. 1960. J. Nutr. 70, 267. Cooke, A. R. 1955. Arch. Eiochem. Biophys. 55, 114. Davies, J. G. 1960. Proc. 8rh Inr. Grassl. Congr., 1960 pp. 381-385. Davies, J. G., and Edye, L. A. 1959. J. Ausr. Inst. Agr. Sci. 25, 117-127. Davies, J. G., and Eyles, A. G. 1965. J . Aust. Inst. Agr. Sci. 31,77-93. Davies, J. G., and Eyles, A. G. 1968. J. Aust. Inst. Agr. Sci. 34, 130-145. Davies, W., and Skidmore, C. L., eds. 1966. “Tropical Pastures.” Faber & Faber, London. Deschuytener, G. 1967. Ann. Gembloux 73,61-76. Dijkman, M. J. 1950. Econ. Bot. 4,337-349. Dirven, J. G. P., and Van Hoof, H. A. 1960. Tidschr. Plantenziekten 66,344-349. Dougall, H. W., and Birch, H. F. 1967. Plant Soil 26,85-98. Edwards, D. C., and Bogdan, A. V. 195 1. “Important Grassland Plants of Kenya.” Pitman & Sons Ltd., Nairobi. Edye, L. A., and Haydock, K. P. 1967. Aust. J . Agr. Res. 18,891-901. Edye, L. A., and Kiers, H. J. 1966. Aust. J . Exp. Agr. Anim. Husb. 6,380-387. Edye, L. A., and Ritson, J. B. 1969. Aust., C.S.I.R.O., Rep. Div. Trop. Past., 1968-1969 pp. 94-102. Edye, L. A., Humphreys, L. R., Henzell, E. F., and Teakle, L. J. H. 1964. Pap., Queensl. Univ. Dep. Agr. 1, 133-172. Emmel, M. W., and Ritchey, G. E. 1941.5.Amer. Soc. Agron. 33,675-677. Evans, J. 1967. Queens. Dep. Prim. Ind., Agrostol. Tech. Ann. Rep. p. 27. Evans,T. R. 1967. Trop. Grassl. 1,143-152. Evans, T. R. 1968.Aust., C.S.I.R.O., Rep. Div. Trop. Past., 1967-1968 pp. 26-27. Evans, T. R. 1969a. Aust. J. Exp. Agr. Anim. Husb. 9,282-286. Evans, T. R. 1969b. Personal communication. Flemons, K. F., and Whalley, R. D. 1958. Agr. Gat. N.S.W.69,449-460. French, M. H. 1957. Herb. Abstr. 27,l-9. Freyre, R. H., and Warmke, H. E. 1952. P. R. Fed. Exp. Sra., Rep. p. 21-22. Gates, C. T., Haydock, K. P., and Little, 1. P. l966a. Aust. J . Exp. Agr. Anim. Husb. 6,261265. Gates, C. T., Haydock, K. P., and Claringbold, P. J. 1966b. Ausr. J . Exp. Agr. Anim. Husb. 6,374-379. . J. Exp. Agr. Gates, C. T., Haydock, K. P., Claringbold, P. J.,and Robins, M. F. 1 9 6 6 ~Aust. Anim. Husb. 6,161-169. Germain, R., and Scaut, A. 1960. Proc. 8th Int. G r a d Congr., 1960 pp. 37 1-374. Gibson, A. H. 1962. Aust.J. Agr. Res. 13,388-399. Gildenhuys, P. J. 1960. Sci. Bull., Govt. Pr. Prrroria 314. Gomide, J. A., Noller, C. H., Mott, G. O., Conrad, J. H., and Hill, D. L. 1969a. Agron. J. 61.121-123.

TROPICAL PASTURES

69

Gomide, J. A., Noller, C. H., Mott, G. O., Conrad, J. H., and Hill, D. L. 1969b. Agron. J . 619 116-120. Gonzalez, V., Brewbaker,J. L.,and Hamill, D. E. 1967. CropScl'. 7,140-143. Gray, S. G . l967a. Aust. J . Agr. Res. 18,7 1-76. Gray, S. G . l967b. Aust. J . Agr. Res. 18,63-70. Gray, S. G . I967c. Aust. J . Agr. Res. 18,77-83. Gray, S. G . 1968. Trop. Grassl. 2,19-30. Griffiths,J. 1959. Nature (London) 183,132. Grof, B. 1957. Queensl. Agr. J . 83, I 1 1 - 1 18. Grof, B. I96 I . Queensl. Agr. J . 87,74 1-742. Grof. R. 1969. Personal communication. Hacker, J. B. 1966.Aust. J . Agr. Res. 17,297-301. Hacker, J. B. 1967. J . Aust. Inst. Agr. Sci. 33,265-266. Hacker, J. B. 196th. Aust. J . Bor. 16,55 1-554. Hacker, J. B. 1968b. Ausr. J . Eot. 16,539-544. Hacker, J. B., and Jones, R. J. 1969. Trop. Grassl. 3,I3-34. Hamilton, R. I.,and Ruth,G. 1968. Trop. Grassl. 2,135-136. Hamilton, R. I., Fraser, J.,and Armitt, J. D. 1969.Aust. J . Dairy 'Technol. 24,62-65. Hamilton, R. I., Donaldson, L. E., and Lambourne, L. J. 1970. Aust. J . Agr. Res. (in press). Hartley, W. 1949. Aust., C.S.I.R.O., Div. Plant Ind., Div. Rep. '7. Hastings, A., and Drake, A. D. 1962. N.Z. J. Agr. 104,33 1-333. Haug, N. F., and Hirst, G. B. 1967. Quart. Rev. Agr. Econ. 20,113-126. Hegarty, M. P., and Pound, A. W. 1968. Nature (London) 217,354-355. Hegarty, M. P., and Pound, A. W. 1970.Aust. J . Biol. Sci. (in press). Hegarty, M. P., Schinckel, P. G., and Court, R. D. l964a. Aust. J . Agr. Res. 15, 153-167. Hegarty, M. P., Court, R. D., and Thorne, Peggy M. l964b. Ausi. J . Agr. Res. 15,168-179. Heinrichs, D. H. 1954. Can. J. Agr. Sci. 34,269-280. Henzell, E. F. 1962. Aust. J . Exp. Agr. Anim. Husb. 2, 132-140. Henzell, E. F. 1963.Ausr. J. Exp. Agr. Anim. Husb. 3,290-299. Henzell, E. F. 1968. Trop. Grassl. 2,l-17. Henzell. E. F., and Oxenham, D. J. 1964.Ausr. J . Exp. Agr. Anim. Husb. 4,336-344. Henzell, E. F., and Stirk, G. B. 1963. Aust. J . Exp. A g r . Anim. Husb. 3, 300-306. Henzell, E. F., Martin, A. E., Ross, P. J., and Haydock, K. P. 1964. Aust. J . Agr. Res. 15, 876-884. Henzell, E. F., Fergus, 1. F., and Martin, A. E. 1966.Aust. J . Exp. Agr. Anim. Husb. 6,157160.

Henzell, E. F., Martin. A. E., Ross, P. J., and Haydock, K. P. 1968. Aust. J . Agr. Res. 19, 65-77. Hodges, E. M., Killinger, G. B., McCaleb, J. E., Ruelke, 0. C., Allen, R. J., Jr., Schank, S. C., and Kretschmer, A. E., Jr. 1967. Florida Agr. Exp. Sra., Bull. 718. Hogg, P. G. 1965. Proc. 9thInt. Grassl. Congr., 1965. pp. 1099-1 101. Holder, J. M. 1967. Trop. Grassl. 1,135- 14 1. Hosaka, E. Y. 1945. Huwaii, Agr. Exp. Sta.. Circ. 22. Hoveland, C. S. 1960.Alabama, Agr. Exp. Sta.. Progr. Rep. Ser. '18. Humphreys, L. R. 1967. J . Aust. Inst. Agr. Sci. 33,3-13. Humphreys, 1.. R., and Robinson, A. R. 1966. Queensl. Dep. Prim. Ind., Div. Plant Ind. Bull. 354. Hutton, E. M. 1960. Emp. J . Exp. Agr. 28,117-125. Hutton. E. M. 1962.Aust.J . Exp.Agr.Anim. Husb. 2, 117-125. Hutton, E. M . 1964. Unpublished experiment.

70

E. M. HUTTON

Hutton, E. M. 1965. J.Aust. Inst. Agr. Sci. 31,102-109. Hutton, E. M. 1968a. Agr. Sci. Rev. 6,7- 17. Hutton, E. M. 1968b.J . Aust. Insr. Agr. Sc i 34,203-2 18. Hutton. E. M. 1970. Proc. / / r h Int. Grass/. Cong., 1970 (in press). Hutton, E. M., and Beall, L. B. 1957.J . Aust. Inst. Agr. Sci. 23,158-159. Hutton, E. M., and Bonner, 1. A. 1960.5.Aust. Inst. Agr. Sci. 26,276-277. Hutton, E. M., and Coote, J. N. 1966.J . Aust. I n s t . Agr. Sci. 32,139- 140. Hutton, E. M., and Gray, S. G. 1959. Emp. J . Exp. Agr. 27,187-196. Hutton, E. M., and Gray, S. G. 1967.J . Aust. Inst. Agr. Sci. 33,122- 123. Hutton, E. M., and Guerassimoff, J. 1966. Euphytica 15,353-36 I . Hutton, E. M., Windrum, G. M.,and Kratzing, C. C. 1958a. J . Nutr. 64,321-338. Hutton, E. M., Windrum, G. M., and Kratzing, C. C. 1958b. J. Nutr. 65,429-440. Jones, R. J. 1964. J . Aust. Inst. Agr. Sci. 30,278. Jones, R. J. 1965.Ausf.J. Exp. Agr. Anim. Husb. 5,458-465. Jones, R. J. 1969. J . Aust. Inst. Agr. Sci. 3562-63. Jones, R. J., Davies, J . G., and Waite, R. B. 1967. Aust. J . Exp. Agr. Anim. Husb. 7, 5765. Jones, R. J., Seawright, A. A., and Little, D. A. 1970. J . Ausr. Inst. Agr. Sci. 36.41-43. Jones, R. K. 1968.Aust. J . Exp. Agr. Anim. Husb. 8,52 1-527. Kretschmer, A. E. 1965. Proc. Soil Crop Sci. SOC.Florida 25,248-262. Kretschmer, A. E. 1966. Proc. Soil Crop Sci.SOC.Florida 26,238-245. Krishnaswamy, N., Raman, V. S., and Chandrasekharan, P. 1956. Curr. Sci. 25,195-197. Kyneur, G . W. 1960. Queensl. Agr. J . 86,507-5 13. Luck, P. E. 1965. Plant. Inrrod. Rev. 1,No. 3,50a-5la. Ludlow, M. M., and Wilson, G. L. 1968.Ausr. J . Agr. Res. 19,35-45. McArthur, J. M., and Miltimore, J. E. 1966. Proc. 10th Inr. Grassl. Congr., 1966 pp. 518521. Mclntyre, M. P. 1966. “Physical Geography.” Ronald Press, New York. McTaggart, A . 1942. C.S.I.R.O., Ausr., Pamph. 114. McWhirter, K. S. 1968.Aust. J . Agr. Res. 20,227-24 I . Marriott, S. J., and Anderssen, K. B. 1953. Queensl. Agr. J . 76,3-9. Marriott, S. J., and Winchester, W. J. 195 1 . Queensl. Agr. J . 73,19-24. Martin, A. E., Henzell, E. F., Ross, P. J., and Haydock, K. P. 1963.Ausr. J . Soil Res. 1, 169-184. Mead, F. W. 1962. Florida, Dep. Agr., Div. Plant Ind., Entomology Circ. 7. Mears, P. T. 1967. Trop. Grassl. 1,98- 105. Miles, J. F. 1949. Aust., C.S.I.R.O., Div. Plant Ind., Div. Rep. 6. Milford, R. 1960.Aust.J . Agr. Res. 11,138-148. Milford, R. 1967.Ausi. J . Exp. Agr. Anim. Husb. 7,540-545. Milford, R., and Haydock, K. P. 1965.Ausr. J . Exp. Agr. Anim. Husb. 5,I3-17. Milford, R., and Minson, D. J. 1966a. Proc. 9th Inr. Grass/. Congr., 1965 pp. 815-822. Milford, R., and Minson, D. J. 1966b. I n “Tropical Pastures” (W. Davies and C. L. Skidmore, eds.), Chapter 7, pp. 106-1 14. Faber & Faber, London. Milford, R.,and Minson, D. J. 1968a. Aust. J . Exp. Agr. Anim. Husb. 8,409-412. Milford, R., and Minson, D. J. 1968b. Ausr. J . Exp. Agr. Anim. Husb. 8,413-418. Miller, H. P. 1967. Aust. J . Exp. Agr. Anim. Husb. 7,5 15-5 17. Miller,T. B. 1969. J . Brii. Grassl. SOC.24,158-162. Minson, D. J. 1966. J . Brit. Grassl. SOC.21,123- 126. Minson, D. J. 1967. Brit. J . Nutr. 21,587-597.

TROPICAL PASTURES

71

Minson, D. J., and Milford, R. 1966.Aust.J. Agr. Res. 17,411-423. Minson, D. J., and Milford, R. l967a. Ausr. J. Exp. Agr. Anim. Hrrsb. 7,546-55 1. Minson, D. J., and Milford, R. 1967b. J. Brit. Crassl. SOC.22,170- 175. Minson. D. J., and Milford, R. 1968.5.Agr. Sci. 71,38 1-382. Minson, D. J., and McLeod, M. N. 197O.Proc. 11th Int. Grassl. Congr. 1970pp. 719-722. Moore, A. W. 1962. E m p . J. Exp. Agr. 30,239-248. Moore, W. L. 1967. Trop. Grassl. 1,2 1-36. Morris, M. P., Pagan, C., and Warmke, H. E. 1954. Science 119,322-323. Motta, M. S. 1956. Proc. 7th Inr. Grassl. Congr., 1956 pp. 539-546. Murtagh, G. J . , and Wilson, G. P. M. 1962. Agr. Gaz. N.S.W.73,634-637. Naveh. Z., and Anderson, G. D. 1967. East Afr. Agr. Forest. J. 32!, 282-304. Neme, N. A. 1958. Agronomica 10.20-30. Nestel, B. L., and Creek, M. J. 1962. Herb. Absrr. 32,265-27 1. Nestel, B. L.. and Creek, M. J. 1965. Proc. 9rh Inr. Grassl. Congr 1965 pp. 157 1-1574. Nordfeldt, S., Henke, S. L., Morita, K.,Matsumoto, H., Takahashi, M., Younge, 0. R., Willers, E. H., and Cross, R. F. 1952. Hawaii, Agr. Exp. Sra., Te,:h. Bull. 15. Norman, M. J. T. 1962. Aust. J. Exp. Agr. Anim. Husb. 2,22 1-22’7. Norman, M. J . T. 1968. Ausr. J. Exp. Agr. Anim. Husb. 8,21-25. Norman, M. J. T., and Arndt, W. 1959.Aust., C.S.I.R.O., Div. LirndRes.Reg. Surv., Tech. Pap. 4. Norman, M. J. T., and Begg, J. E. 1968.J.Aust. Inst. Agr. Sci. 34,59-68. Norman, M. J. T., and Stewart, G. A. 1964. J.Ausr. Inst. Agr. Sci. 30,39-46. Norris, D. 0. 1956. Emp. J . Exp. Agr. 24,247-270. Norris, D. 0. 1958.Ausr. J. Agr. Res. 9,629-632. Norris, D. 0. 1959.Ausr.J. Agr. Res. 10,65 1-698. Norris, D. 0. 1963. E m p . J . Exp. Agr. 31,255-258. Norris, D. 0. 1964. Bull. Commonw. Bur. Past. Field Crops 47,1116- 198. Norris, D. 0. 1965. PIanrSoil22,143-166. Norris. I).0. 1966. I n “Tropical Pastures” (W. Davies and C. L . Skidmore, eds.), Chapter 6. pp. 89-105. Faber & Faber, London. Norris, D.O. 1967. TropGrassl. 1,107-121. Norris, D. 0.. and ’1 Mannetje, L. 1964. East Afr. Agr. Forest. J. 29,2 14-235. Oakes, A. J . 1960. Proc. 8th Int. Grassl. Congr., 1960 pp. 386-38’2. Oakes, A. J. 1965. Trop. Agr. Trin. 42, 323-331. Oakes, A. J. 1968.Advan. Front. Plant Sci. 20,I-114. Oakes, A. J., and Langford, W. R. 1967.Agr0n.J. 59,387-388. Oakes, A. J.,and Skov. 0. 1962.Agron.J.54,176-178. Oakes, A. J.,and Skov, 0. 1967.J.Agr. Univ. P . R . 51,176-181. Okorie, I . I., Hill, D. H., and Mcllroy, R. J. 1965.5.Agr. Sci. 64,235-245. Paltridge. ‘1‘. B. 1942. Aust., C.S.I.R.O., Pumph. 114. Paltridge, T. B. 1955.Aust.. C.S.I.R.O.,Bull. 274,p. 108. Park, S. J., and Rotdr, P. P. 1968a. Crop Sci. 8,467-470. Park. S. J . , and Rotar, P. P. 1968b. Crop Sci. 8,470-474. Parodi, L. R. 1943. Rev. Argent. Agron. 10,361-372. Parodi, R. A., and Scantamburlo, J. L. 1954. Quarta Reun. Plantas Forrajeras pp. 52-60. Patil, B. D., Singh, S. V., and Singh, 0. N. 1967. Indian Farming 17,36-37. Phillips, J. 1961. “The Development of Agriculture and Forestry in the Tropics.” Faber & Faber, London. Playne. M. J. 1969a. Aust. J. Exp. Agr. Anim. Husb. 9,192-195.

72

E. M. HUTTON

Playne, M. J. l969b. Aust. J . Exp. Agr. Anim. Husb. 9,393-399. Prine, G . M., and Burton, G. W. 1956.Agron. J . 48,296-301. Pritchard, A. J. 1964. Aust. J . Exp. Agr. Anim. Husb. 4,6- 14. Pritchard, A. J. 1965a. Aust. J . Agr. Res. 16,525-532. Pritchard, A. J. l965b. Aust.J. Agr. Res. 16,533-540. Pritchard, A. J. 1 9 6 5 ~Aust. . J . Agr. Res. 16,5 17-523. Prichard, A. J. 1965d. Euphytica 14,307-314. Pritchard, A. J. 1967.Aust., C.S.I.R.O., Rep. Div. Trop. Past., 1966-1967pp. 67-68. Pritchard, A. J., and ’t Mannetje, L. 1967. Euphytica 16,324-329. Romney, D. H. 1961. Trop. Agr. Trin. 38, 39-47. Rosenberg, M. N., and Zoebisch, 0. C. 1952.Agron. J . 44,3 15-3 18. Roseveare, G. M. 1948. Imp. Bur. Past. Field Crops, Aberystwyth, Gt. Brit., Bull. 36. Rotar, P. P. 1965. Trop. Agr. Trin. 42, 333-337. Rotar, P. P., Park, S. J., Bromdep, A.. and Urata, U. 1967. Hawaii, Agr. Exp. Sta., Tech. Progr. Rep. 164,l-13. Russell, E. W. 1966 In “Tropical Pastures” (W. Davies and C. L. Skidmore, eds), Chapter 2 , pp. 30-45. Faber & Faber, London. Russell,J. S. 1970. Trop. Grassl. 3,123-135. Salette, J . E. 1965. Proc. 9th I n t . G r a d Congr., 1965 pp. 1 199-1203. Schleger, A. V., and Turner, H. G. 1965. Aust. J . Agr. Kes. 16,92-106. Schofield,J. L. 1941. Queensl. Agr. J . 56,378-388. Schofield,J. L. 1944. Queensl. J . Agr. Sci. 1,2-58. Shaw, N. H. 1961.Aust. J . Exp. Agr. Anim. Husb. 1 , 7 3 4 0 . Shaw,N. H. 1967a.Aust.,C.S.I.R.O., Rep.Div. Trop.Past., 1966-1967pp. 15-16. Shaw, N. H. 1967b. Trop. Grassl. 1,76-77. Shaw, N. H.,and Bisset, W. J. 1955. Aust. J . Agr. Res. 6,539-552. Shaw, N . H., Elich, T. W.. Haydock. K. P., and Waite, R. 9 . 1965.Ausr. J . Exp. Agr. Anim. Husb. 5,423-432. Shaw, N. H., Gates, C. T., and Wilson, J. R. 1966. Aust. J . Exp. Agr. Anim. Husb. 6, 150156. Smith, A. 1969. Personal communication. Snyder, L. A., Hernandez, Alice R., and Warmke, H. E. 1955. J . Agr. Univ. P . R. 39, 150164. “Some Concepts and Methods in Sub-tropical Pasture Research.” 1964. Bull. Commonw. Bur. Past. Field Crops 41. Staples, I . 9. 1967. Queensl. Agr. J . 92,388-392. Stobbs, T. H. 1966. Proc. 9rh I n t . Grassl. Congr. 1965 pp. 939-942. Stobbs, T . H. I969a. J . Brit. Grassl. SOC.24, 8 1-86. Stobbs, T. H. 1969b. Trop. Agr. Trin. 46, 187-194. . Brit. Grassl. SOC. 24, 177-183. Stobbs, T. H. 1 9 6 9 ~J. Stobbs, T. H. 1969d. Trop. Agr. Trin. 46, 195-200. Stocker, G. C., and Sturtz, J. D. 1966. Aust. J . Exp. Agr. Anim. Husb. 6, 277-280. Suman, R. F., Woods, S. G., Peele, T. C., and Godbey, E. G. 1962. Agron. J . 54,26-28. Takahashi, M. 1956. Proc. 7th Int. Grassl. Congr., 1956 pp. 547-555. Takahashi, M., and Ripperton, J. C. 1949. Hawaii, Agr. Exp. Sta., Bull. 100. Teakle, L. J. H. 1937. J . Agr., West. Aust. 14, [2], 313-324. ‘t Mannetje, L. 1964. Aust. J . Exp. Agr. Anim. Husb. 4, 22-25. ’t Mannetje, L. 1965. Aust. J . Agr. Res. 16,767-77 I .

TROPICAL PASTURES

73

‘t Mannetje, L. 1967. Trop. Grassl. 1, 9-19. ’t Mannetje, L.. and Pritchard, A. J. 1968. New Phytol. 67,257-2133. Tothill, J. C. 1966. Aust. J. Bot. 14, 35-47. Tow, P. G . 1967. Nerh. J . Agr. Sci. 15, 141-154. Truong, N. V., Andrew, C. S., and Skerman, P. J. 1967. Ausr. J . Exp. Agr. Anim. Husb. 7, 232-236. Vallis, I., Haydock, K. P., Ross, P. J., and Henzell, E. F. 1967. Aust. J. Agr. Res. 18, 865-877. Verboom, W. C. 1965. Trop. Agr. Trin. 42,229-242. Vicente-Chandler, J . , Figarella, J., and Silva, S. 1961. J. Agr. Univ. P. R. 45,37-45. Warmke, H. E. 1954. Amer. J . Bot. 41,5-11. Wayman, 0.. and Iwanaga, 1. 1. 1957. Amer. SOC.Anim. Prod., Proc. West. Sect., Misc. Pap. 81. Webster, C. C., and Wilson, P. N. 1966. “Agriculture in the Tropics.” Longmans, Green, New York.. White, C. L.. Renner. G. T., and Warman, H. J. 1968. In “Geography: Factors and Concepts” (K. F. Mather, ed.), p. 208 Appleton, New York. White, R. E. 1967. Ausr. J . Exp. Agr. Anim. Husb. 7,509-5 14. Whiteman, P. C. 1969. Personal communication. Whiteman, P. C. 1968. Aust. J. Exp. Agr. Anirn. Husb. 8,528-532. Whiteman, P. C. 1970a. Aust. J . Agr. Res. 21,207-214. Whiteman, P. C. I970b. Aust. J. Agr. Rex. 21, 2 15-222. Whiteman, P. C . , and Lulham, Ann. 1970. Aust. J. Agr. Res. 21, 195-206. Whittet, J. N. 1923. In “Grasses and Fodder Plants of N.S.W.” (E. Breakwell, ed.), pp. 25 1-27 I . N.S.W. Govt. Printer. Whyte, R. 0. 1962. Trop. Agr. Trin. 39, 1 - 1 I . Whyte, R. 0.. Nilsson-Leissner, G., and Trumble, H. C. 1953. F A 0 Agr. Studies 21. Willet, E. L., Quisenberry, J. H., Henke, L. A., and Maruyama, C. 1947. Hawaii, Agr. Exp. Sta., Biennial Rep., 1944-1946 p. 46. Williams, R. J. 1966. PIanr Inrrod. Rev. 3,No. 2, 18. Wilson, G . P. M. 1970. Proc. 11th Inr. Grassl. Congr. 1970 p. 312. Wilson, G. P. M., and Murtagh, G. J. 1962. Agr. Gaz. N.S.W. 73,460-462. Woods, L. E. 1969. Trop. Grassl. 3,91-98. Wutoh. J . G . , Hutton, E. M., and Pritchard, A. J. 1968a. Aust. J . Exp. Agr. Anim. Husb. 8, 544-547. Wutoh, J . G., Hutton, E. M., and Pritchard, A. J. 1968b. Aust. J . Agr. Res. 19,411-418. Wutoh, J. G., Hutton, E. M., and Pritchard, A. J. 1968 ~.Ausr. J . Exp. Agr. Anim. Husb. 8 , 3 17-322. Yearbook of Agriculture. 1962. “After a Hundred Years.” U.S. Dep. Agr., Washington, D.C. Yoshida. Ruth K. 1944. PhD. Thesis, University of Minnesota. Younge, 0. R.,and Moomaw, J. C. 1960. Econ. Bot. 14,3 16-330. Younge, 0. R., and Plucknett, D. L. 1965. Proc. 9rh Inr. Grassl. Congr., 1965 pp. 959963. Younge, 0. R., and Takahashi, M. 1953. Agron. J. 45,420-428. Younge, 0. R., Plucknett, D. L. and Rotar, P. P. 1964. Hawaii Agr. Exp. Sta., Tech. Bull. 59.

This Page Intentionally Left Blank

CLAY-ORGANIC

COMPLEXES A N D INTERACTIONS M. M. Mortland

Michigan State University, East Lansing, Michigan

Introduction ........................................ ................................. Bonding Mechanisms A. Cationic .................................................................. B. Anionic .............................................................................. C. Ion-Dipole and Coordination ................................................. D. Hydrogen Bonding .................... E. Van der Waals F F. Pi Bonding ........ G. Entropy Effects .................................... H. Covalent Bonding .......................................................................... 111. Nature of Some Clay-Organic Complexes and Reactions ........................... A. Adsorption of Organic Acids, Amino Acids, Proteins, Purines, Pyrimidines, and Nucleosides .......................................................... B. Interaction of Organic Pesticides with Clays ...................................... C. Interaction of Polymers and Surfactants with Clays ............................ D. Catalysis Reactions ....................................................................... E. Diffusion of Organic Compounds in Clays ......................................... IV. Nature and Importance of Clay-Organic Complexes in Soils 1. II.

Page 75 77 77 85 85 89 93 93 94 94 95 95 99 101

105 107 108

V. Conclusions ............................. References .........................................................................................

1 I3 1 I4

1. Introduction

Clays interact with many organic compounds to form complexes of varying stabilities and properties. These interactions are of great importance in nature and in industry. The clays in soils and sediments often have organic material with which they are intimately associated. This association of clay and organic material has a multitude of consequences that are reflected in the physical, chemical, and biological properties of the matrix in which they occur. In soils, for example, some organic compounds derived from plant and animal remains and their decomposition products are strongly adsorbed by clay minerals. In this interaction with the clay, some kinds of compounds may bridge between neighboring clay 75

76

M. M. MORTLAND

particles creating relatively stable aggregates, thus greatly influencing the moisture and aeration properties of the soil. Other clay-organic interactions result in the protection of the organic compound from biological degradation. This protective effect has its consequence in that organic matter content of soils is usually positively correlated with clay content, all environmental factors being equal. Biologically active compounds, such as herbicides and insecticides, may be adsorbed by clay minerals and rendered inactive or, when conditions have changed, be released to become reactive; or yet again undergo catalytic degradation at the clay surface and completely lose their toxicities. In the sediments of the geologic column as in coal under-clays and shales there are usually very significant amounts of organic materials strongly bound to the clays. Industry has long utilized clay-organic complexes on a large scale, as in lubricants, paper, cosmetics, medicinals, paints, etc. Thus the clayorganic complexes are obviously of importance not only in soils, but also in other areas. The nature of clay-organic complexes has long been a subject for research. As far as soils and sediments are concerned, this work has been handicapped by incomplete knowledge of the composition of the organic matter itself, knowledge that is essential for a real understanding of those complexes. Progress is being made in this area, however, and will continue. The work summarized by Kononova ( I 96 1) should be consulted for information on this subject. Two different approaches have been made in studying clay-organic complexes. On the one hand, various extracts and derivatives of organic matter were allowed to react with clays, and the properties of the resulting complex were studied. The other approach has been to utilize organic compounds of known constitution, and to deduce the nature of their interaction with clays from their known properties. The latter direction has resulted in considerable fundamental knowledge about the binding mechanisms involved between various functional groups of organic molecules and the clay mineral surfaces. Logic would suggest a somewhat similar reactivity for these same functional groups where they exist in soil organic matter interacting with clay. The scope of this review is primarily with the more recent developments. A review up to 1965 was made by Greenland (1965a,b); another extensive review made by Bailey and White (1 970) is concerned primarily with pesticide interactions with clay but also covers much of organicclay relationships in general. Both of these reviews should be consulted by the interested reader. The author has not attempted to recapitulate what various scientists have done, but has tried to relate the observations

CLAY-ORGANIC

COMPLEXES A N D INTEWCTIONS

77

of recent work to fundamental concepts. In addition, the writer admits to a bias toward infrared techniques and results devleloped over a number of years of research, these the reader will quickly recognize. No apology if made for this; we merely point out its existence. II.

Bonding Mechanisms in Clay-Organic

,Complexes

Over the years many experimental techniques have been used in the study of clay-organic complexes, and all of them have made contributions of one sort or another. Recently, however, the availability of highquality double-beam infrared spectrophotometers coupled with the development of auxiliary techniques of sample preparation have led to major breakthroughs in the investigation of clay-organic complexes. V. C. Farmer (1968) has summarized some of the applications of infrared absorption to clay mineral research. Extremely thin self-supporting films of many clay minerals can be made that can be placed directly in the beam of the spectrophotometer. The films are so thin that most of the infrared radiation is transmitted through the film except in regions of the spectra where the clay itself absorbs infrared radiation. When organic molecules are adsorbed on the films, spectra of the adsorbed species may easily be obtained. Shifts in frequency of various diagnostic bands may then give information on the mode of binding of the organic molecule by the clay. The conclusions from such observations are often unambiguous because direct observations are being made of the molecule in the adsorbed state. Infrared absorption in combination with X-ray diffraction and other physical-chemical techniques made possible a rigorous approach to the study of clay-organic complexes. The infrared technique is obviously at its best when compounds of known composition and whose infrared absorption spectra are well unders.tood, are utilized in clay-organic studies. There is much less rigor when it is applied to investigations involving compounds whose nature and spectra are less well defined. Workers utilizing the technique on clay-organic complexes would be well advised to be extremely careful in their interpretations since sometimes more than one explanation will account for observed changes in spectra. A.

CATIONIC

1. Ion Exchange Organic cations will be adsorbed at clay mineral surfaces by ion exchange with cations neutralizing the negative electrical charges responsible

78

M. M. MORTLAND

for the cation exchange capacity of the mineral: RNHT + M+-clay ~2 RNHT-clay

+ M+

(1)

where RNHf is some organic cation and M+ another species of cation. Many organic cations are positively charged because of protonation of an amine groups as in the case of alkyl amines and amino acids. However, some compounds such as urea and amides are protonated on the oxygen of the carbonyl group. Certain properties of the organic cations set them apart from metal ions and will influence their adsorption on or displacement from clay mineral surfaces. These properties are as follows: (1) The ionic property is usually pH dependent. (2) Other forces influence adsorption to the clay surface. These other forces include hydrogen bonding, ion-dipole, and physical forces, and their importance depends upon such factors as molecular weight, nature of the functional groups present, and configuration of the molecule. (3) The interaction of the organic ion with water will be quite variable, depending on the nature of the ion. This interaction of the organic cation with water may be of crucial importance in its interrelations with the clay surface and exchangeable metal ions residing there. Numerous scientists have contributed to the literature on the interaction of organic cations with clays; the earlier work will be found cited by Greenland (1 965a). More recently, studies on the replacement of exchangeable sodium and calcium from montmorillonite by various alkylammonium cations were made by Theng et al. (1967). They found that the affinity of the clay for the organic cation was linearly related to molecular weight with the exception of the smaller methylammonium and the larger quaternary ammonium ions. Thus, the more the length of the alkylammonium chain increases, the greater is the contribution of physical, noncoulombic forces to adsorption. Within a group of primary, secondary, and tertiary amines, the affinity of the alkylammonium ions for the clay decreased in the series R3NH+> R2NH,+ > RINH3+.These differences were explained in terms of size and shape of the ions. In general, Theng et al. found that the Na+ was much more easily exchanged by the alkylammonium ions than was the Ca2+,as would be expected from previous work on metal ion exchange in clays. In studies in which the alkylammonium ion is replaced by metal cations, Mortland and Barake (1964) showed that the order of effectiveness in replacing ethylammonium ion was A13+ > Ca2+ > Li+. In addition, it was noted in X-ray diffraction studies on partially exchanged systems that the organic and inorganic cations were not distributed uniformly throughout all the sur-

CLAY-ORGANIC COMPLEXES A N D INTEIMCTIONS

79

faces of montmorillonite, but that a segregation of the two kinds of ions took place in various layers. This suggests that when the displacement of ethylammonium ion by the metal ion from one interlamellar position begins, it is completed before ethylammonium ions from other layers are exchanged. Barrer and Brummer ( I 963) have made similar observations and suggest that this segregation of organic and metal cations may be explained on the basis of accompanying water layers in which homoionic cation layers tend to give regular continuous and stable monolayer or double-layer arrangements. Mixtures of cations in any one interlamellar position may render it impossible to form this type’ of geometrically regular water layer. Thus it appears that, in montomorillonite partially saturated with organic and metal cations, interstratified layers occur in which each layer contains mainly one type of cation, a situation which must be the most thermodynamically stable. The effect of charge density of the clay mineral on the competitive adsorption by ion exchange of two divalent organic cations was studied by Weed and Weber (1 968). The two cations diquat (6,7-dihydrodipyrido( 1,2-a :2’, 1 ’-c)-pyrazidiinium dibromide) and paraquat (1,l’dimethyl-4,4’-dipyridium dichloride) differ in one respect in that the charge centers of the cations are 3-4 A apart in the former and 7-8 A apart in the latter. The preference for one or the other of the cations by layer silicates was related to the geometric “fit” between the charges on the cation and those on the clay in that the cation whose charge centers could most nearly approach the adsorption sites on the mineral surface would be preferred. These results indicate that the negative charge on the layer silicate lattices are discrete and relatively fixed and are not smeared out as has been suggested by some workers. If the latter were the case, the differential selectivity by various layer silicates for these two organic cations would not have been obtained. Charge density of the clay mineral may also affect the orientation of adsorbed organic cations through steric effects. Thus, Serratosa ( I 966) showed by infrared absorption technique that in pyridinium-montmorillonite the organic cation assumed an orientation where the plane of the pyridine ring was parallel with the platelets of the clay mineral and a resulting 00 1 spacing of 12.5 A.On the other hand, pyridinium-vermiculite has the pyridinium cations vertically positioned with respect to the clay platelets and a 001 spacing of 13.8 A.Apparently the close proximity of the cation exchange sites one to another prevents the pyridinium from assuming the parallel position because of the restricted area permitted for each pyridinium. Where neutral but polar organic molecules are bound to the clay surface by other mechanisms, such as ion-dipole interaction,

80

M. M. MORTLAND

charge density would also be expected to affect their orientation within the interlamellar regions of swelling clay minerals. Certain organic cations such as butylammonium, when placed on the cation exchange sites of vermiculite have been shown by Walker and Garrett (1 967) to cause gross one-dimensional swelling of vermiculite crystals in water. Hundreds of angstroms may separate the individual platelets. These workers have shown dispersions of these crystals to be stable over long periods in distilled water, but very sensitive to electrolytes which cause their flocculation. The macro swelling of vermiculite is apparently activated by those organic ions which form clathrate structures with water.

2. Protonation of Organic Molecules at Clay Surfaces In addition to adsorption of organic compounds by ion exchange of organic with inorganic cations on the exchange complex of clays, many compounds may become cationic after adsorption at the clay surface through protonation. The sources of the protons for such a reaction are: (1) exchangeable H+ occupying cation exchange sites, (2) water associated with metal cations at the exchange sites, or (3) proton transfer from another cationic species already at the clay surface. It is thus quite obvious that the existence of an organic compound in cationic or molecular form is dependent upon the acidity or proton-supplying power of the clay surface. The reaction of a compound with exchangeable H+ to form the cationic species is quite straightforward:

where R is some alkyl group. This reaction goes to completion because it is essentially one between a strong acid and a relatively strong base to give a salt. It is therefore characterized by high heats of reaction and a high degree of irreversibility. Considerable energy must be put into the system to drive the reaction back to the left. It has been shown calorimetrically by Mortland et al. (1963) that when R N H 2 is N H 3 , the heat of reaction (enthalpy) is about 35 kcal per mole. Similar heats of reaction would also result when R is an organic group of some kind. Where the hydrogen is part of a hydroxyl group at the edges of clay minerals or on amorphous material, the ability to protonate organic bases would be very much pH dependent. The second process by which organic molecules may become protonated is by proton donation from water at the clay mineral surface.

CLAY-ORGANIC

COMPLEXES A N D INTERACTIONS

81

Ordinary water is not likely to be acidic enough to protonate many organic molecules. However, when water is associated with metal cations, hydrolysis of this complex produces more or less H+, depending upon the properties of the metal ion involved. The more electronegative the metal cation, the more acidic will be the complex with water. Thus aqueous solutions of AI3+are quite acidic, and those of cations like Na+ are much less so. When such hydrated metal ions are present on the cation exchange sites of clay minerals, they impart differential proton-donating powers to that mineral surface. The hydrolysis of such hydrated metal cations can be described in the following equation: [M (H*O)s]+" S [MOH ( H ~ O ) ~ - I ] + + " -H ' i Acid Conjugate base

(3)

where M is the metal cation in question. The degree to ,whichthis reaction goes to the right is described by the equilibrium constant and the acid properties of alarge number of hydrated metal ions have been summarized by Hunt ( 1 963). It would be expected then that the ability of a clay surface to protonate compounds would be dependent upon the nature of the metal cations saturating the exchange sites on the clay, and this has been shown for NH3 by Russell (1965) and Mortland and Raman (1968). The overall reaction involved when an organic molecule is protonated by such a process is:

where M is the exchangeable metal cation and B is the base in question. It has been observed by a number of workers ming spectroscopic methods (Mortland et al., 1963; Fripiat et al., 1965; Russell, 1965; Mortland, 1966; Mortland and Raman, 1968; Harter and Ahlrichs, 1967; Swoboda and Kunze, 1968) that the acidity or proton-donating properties of the clay surface are greater than would be expected from pH measurements of the clay in water. This conclusion has been reached on the basis of infrared absorption studies, where the protonated condition of a compound can be differentiated from the neutral form. For example, when NH3 is adsorbed on Ca2+and even Na+ saturated montmorillonite films, considerable quantities of N H t were observed to be formed (Mortland et al., 1963; Russell, 1965; Mortland and Raman, 1968). Urea and various amides, which are extremely weak bases, have been observed by Mortland ( 1 966) and Tahaun and Mortlaiid ( 1 966a) to become protonated at montmorillonite surfaces where the exchangeable

82

M. M. MORTLAND

cation was H+, AI3+,or Fe3+,but not in Na+ or Ca2+saturated clays. On the other hand, some substituted ureas were not observed to protonate on montmorillonite surfaces (Kim and Weed, 1968; W. J. Farmer and Ahlrichs, 1969). Ordinarily a pH of less than 0.5 is required for a molecule like acetamide to become protonated. These excessive acidic effects have been ascribed to increased dissociation of the water associated with the exchangeable metal cations above and beyond that predictable from their hydrolysis constants. An important point in proof that the water associated with exchangeable cation is involved is that the proton donating process of the clay surface is greatly dependent upon the kind of exchangeable cation present. Another factor is the water content itself. The surface acidity of the clay increases as the water content decreases. This has been explained by Mortland and Raman (1 968) on the basis that when a great deal of water is present, polarization forces of the exchangeable cation may be said to be distributed among a large number of water molecules. However, as the water content decreases, these polarization effects become more concentrated on the fewer remaining water molecules, causing an increase in hydrolysis and so in their proton-donating capabilities. An example of this process was shown by V. C. Farmer and Mortland ( 1 966); they found that pyridine when adsorbed by hydrated Mg-montmorillonite was coordinated to the Mg2+ by bridging through directly coordinated water molecules. When this system was dehydrated, pyridinium ions were observed, suggesting an increase in the acidity of the clay surface. In other work, Mortland and Raman (1968) found that when there was 40.7% water (oven-dry basis) in Ca2+montmorillonite, only 16 meq of N H i per 100 g of clay was formed in the presence of NH,. On the other hand, when the water content was 5.9% the NHi formed was 80 meq per 100 g, thus showing in a dramatic way the effect of water content on the proton-donating properties of the clay surface. The third mechanism by which organic compounds may become protonated after adsorption at a clay mineral surface is by proton transfer from a protonated species already present. The general reaction is: AH+

+ B % BH++ A

(5)

where AH+ represents the protonated species (proton donor) on the surface of the clay, and B represents the base (proton acceptor) with which it reacts. The degree to which Eq. ( 5 ) will go to the right will depend largely on two factors: the relative magnitude of the dissociation constants of the two interacting species, and the relative concentration or activities of the reactants and products. The protonated species can

CLAY-ORGANIC

COMPLEXES A N D INTERACTIONS

83

be considered an acid capable of donating a proton while the uncharged species is a base capable of accepting a proton. Thus, Russell, Cruz, and White ( I 968a) showed by infrared spectroscopy the formation of 3aminotriazolium cation when NH2 montmorillonite was treated with 3aminotriazole. As the infrared spectrum for the 3-aminotriazolium cation appeared, that of NH2 tended to decrease in intensity thus establishing the proton transfer process. Raman and Mortland (1 969) have confirmed the above reaction and in addition have observed proton transfer between a number of organic compounds on montmorillonite surfaces. The systems in which proton transfer was observed are contained in Table I, the degree of transfer being quite variable but generally in accordance with the relative basicities of the interacting species and the concentrations of reactants and products. TABLE I Proton Transfer Reactions on Montmorillonite Observed by Raman and Mortland ( I 969) Exchangeable cation on clay

Molecule observed to accept a proton

NH i Pyridinium Ethylammonium Methylammonium (NH&COH+ (urea)

Pyridine, methylamine, 3-aminotriazole NH3, methylamine, 3-aminotriazole N Ha, pyridine, 3-aminotriazole NH3, pyridine, 3-aminotriazole NH3, pyridine

While relatively simple clay-organic systems have been described above, it is reasonable to expect that similar proton transfer would occur in other clay-organic systems, such as soil organic matter, and that this is an important reaction to be recognized in natural systems. An example might be reactions that occur when ammonia is applied to soils. In the zone of ammonia injection there is a tremendous sink for protons. Ammonia would accept protons from available sources in organic matter rendering those sites no longer electopositive. This would change the bonding characteristics of the organic matter within the organic colloids themselves as well as with the inorganic colloids. 3. Hem isal t Forma tion

When the amount of an adsorbed base (B) on a clay exceeds the number of protons available for cation formation, one of the following situations occurs: (1) the protonated molecule retains its proton against attraction

84

M. M . MORTLAND

by the nonprotonated molecule; (2) two molecules compete for the proton on an equal basis, and it does not identify with either one but belongs equally to both, forming a strong symmetrical hydrogen bond and a cation of the [Bz- H]+ type. Many organic bases form symmetrical hydrogen bonds or hemisalt type complexes which are well documented in the chemical literature. Examples of these complexes observed in clays are ethylammonium-ethylamine montmorillonite by V. C. Farmer and Mortland (1969, pyrindinium-pyridine by V. C. Farmer and Mortland ( 1 966), urea-montmorillonite by Mortland ( 1966) and various amide-montmorillonitecomplexes by Tahoun and Mortland ( 1 966a). Hemisalt formation has quite a striking effect upon the infrared absorption spectra of the organic cation, so that if excess base is present beyond the number of protons available there might be a mistaken conclusion that no protonated species is present. For example, in ethylammonium-montmorillonite a strong infrared absorption band appears at 1510 cm-' in spectrum 1 (Fig. I), which is the symmetric deformation vibration of NH;. When ethylamine is adsorbed on this system, the 15 10 cm-l band disappears and strong, broad, featureless absorption appears from 3300 to at least 1200 cm-l as shown in spectrum 2, which has the Wavenumbers (Cm-')

0 c u)

.-

: e

k

FIG. I . Infrared spectra o f ( I ) ethylammonium-saturated montrnorillonite, degassed 20 minutes; (2) ethylammonium-saturated montmorillonite after adsorbing ethylamine (50 cm pressure) for 15 minutes and then degassed for 45 minutes; (3) copper-saturated montmorillonite, after adsorbing ethylamine (50 cm pressure) for 2 hours and then degassed for 30 minutes. Curves 1 and 2 have the same baseline; curve 3 is displaced. (Reprinted from J . Phys. Chem. 69,684. Copyright (1965) by American Chemical Society. Reprinted by permission of the copyright owner.)

CLAY-ORGANIC

COMPLEXES A N D INTERACTIONS

85

same baseline as spectrum 1. The ethylammonium-ethylamine complex is stable in a vacuum, but it breaks up and ethylamine is lost when exposed to water in the atmosphere with the concomitant reappearance of the infrared spectrum of ethylammonium-montmorillonite.The formation of the B2H+ type of cations has been observed in the experience of the author to be very common rather than the exception in clay-organic systems containing excess organic base and must be considered in interpretations of infrared spectra.

B. ANIONIC While anions are normally expected to be repelled from the surface of the negatively charged clay minerals, their presence at the clay surface has been observed by Yariv et al. (1966) utilizing infrared absorption. In studying interactions of benzoic acid with montmorillonite, they observed benzoate anion formation in relatively dry clay films as a result of the following reaction: Mii+-clay+ nHOB,+ nH+-clay+ M(OB,),,

(6)

The amount of benzoate anion present depended greatly on the kind of exchangeable metal ion (M) present being greatest for the polyvalent cations. These observations were made on systems where benzoic acid was adsorbed from the vapor phase on relatively dry clay or where aqueous solutions of benzoic acid had been utilized and the water evaporated away. Probably little benzoate anion would have been adsorbed from an aqueous solution, and points up the differences in surface chemistry of clay minerals between aqueous suspensions and air-dry environments, both of which are possible in nature. c .

ION-DIPOLE AND

COORDINATION

The classical view of adsorption of polar but nonionic organic molecules by clay minerals has been to attribute a major function to the oxygen atoms or hydroxyl groups of the silicate surface. This interaction has been said to be one of hydrogen bonding between them and functional groups on organic molecules. This idea has developed mainly as an extension from earlier concepts of the mode of water adsorption at clay mineral surfaces as being mainly one of hydrogen bonding between oxygen atoms of the silicate surface and the water molecules. Such concepts arose from indirect evidence obtained many ways and from knowledge regarding the chemical characteristics of the compounds in question.

86

M. M. MORTLAND

With the advent of rigorous infrared absorption techniques, however, it has often been possible to view the condition of the adsorbed molecule directly and to sometimes draw relatively unambiguous conclusions regarding the mechanisms by which they are held at the clay mineral surface. Thus, Russell and Farmer (1964) were able to distinguish water which is directly coordinated to the exchangeable cations from more labile water in outer spheres of coordination. From the studies of a large number of polar molecules adsorbed on clay minerals, it is quite evident that the nature of the saturating cation on the exchange complex plays a decisive role in the adsorption process. This was evident in the preceding section on protonation, where it was shown that the kind qf exchangeable cation with its associated water molecules determined the acidity of the clay surface and therefore protonation processes. So also they serve as adsorption sites for polar nonionic molecules by ion-dipole or coordination types of interaction. An example is given in Fig. 1, where the infrared absorption spectrum of ethylamine on Cu-montmorillonite is shown in curve 3. This particular complex had an intense blue color, was stable in the air, and the proportion of amine to copper was 4 : 1 indicating a square-planar complex. The greater affinity the exchangeable cations have for electrons, the greater will be the energy of interaction with polar groups of organic molecules capable of donating electrons. Thus, transition metal cations on the exchange complex having unfilled d orbitals will interact strongly with electron supplying groups. In the case of molecules such as water and ammonia, the solvation of the exchangeable cations on the clay surface is the most energetic and therefore the primary mechanism of adsorption. Where there is not an exchangeable cation in the interlamellar regions to solvate, there is no expansion of 2: 1 type minerals, i.e., talc and pyrophyllite. In accordance with this the heats of wetting of clay minerals are generally in relation to the solvation energy of the exchangeable cation (Keay and Wild, 196 1;Kijne, 1969). The preceding discussion does not exclude other types of interaction with the silicate surface as additional mechanisms of adsorption, but it is apparent that they are generally weaker and come about after ligand positions with exchangeable cations are occupied. These weaker interactions are of greatest importance in clay systems where ions of relatively low solvation energy are on the exchange complex. The classical view of the adsorption of alcohols by clay minerals has been one of hydrogen bonding to the oxygens of the silicate surface. In fact, the use of ethylene glycol and glycerol for the measurement of specific surface of clays was predicated on this mode of interaction. Some

CLAY-ORGANIC

COMPLEXES A N D INTERACTIONS

87

earlier workers, however, have recognized the possible influence of the exchangeable metal ions, as for example, Glaeser ( 1 954), who observed retention of methanol and ethanol by clays to be a function of the exchangeable cation. McNeal ( 1 964) in working with ethylene glycol showed significant effects of exchangeable cation on the retention by clay minerals. So also Bissada etal. (1967) showed the effect of exchangeable cation on clay interactions with alcohol. Direct observation of alcohol-clay complexes with infrared absorption were made by Dowdy and Mortland ( 1967,1968), Ovcharenko etal. ( I967), and Tarasevich etal. ( 1967). The complex formed between ethylene glycol and exchangeable Cu(I1) ions in montmorillonite showed OH stretching vibrations of the glycol displaced from around 3360 cm-I in the liquid to 2650 and 2750 cm-l (Dowdy and Mortland, 1968). These bonds result from direct coordination of the glycol to the Cu(I1) through the oxygen atoms. The result of this interaction is to lower the force constant between the 0 and H with the resulting lowering of OH stretching frequencies. The degree to which OH stretching vibrations were lowered was found to be related to the solvation energy of the cation. These results prove the importance of ion-dipole reactions for alcohols on clay surfaces since if hydrogen bonding were the adsorption mechanism, the infrared spectrum of the adsorbed alcohol should remain the same regardless of the exchangeable cation. In looking at the interaction between water and alcohols on clay surfaces it was shown by Dowdy and Mortland ( 1 967, 1968) that they both compete for ligand positions around the cation. Ethanol or ethylene glycol can completely dehydrate the clay mineral or on the other hand, water can displace the alcohol according to the mass action requirements:

where M is the exchangeable cation and ROH is some alcohol. Thus while water has been shown to be retained on clays to high temperatures especially by the more highly polarizing cations (Fripiat et af., 1960; Russell and Farmer, 1964), the clay surface can easily be dehydrated at low temperatures by introducing a polar molecule which competes with water for ligand sites around the exchangeable cation. Ion-dipole or coordination types of interaction on clays have been noted for a wide group of other polar molecules, i.e.: NH3 (James and Harward, 1962; Cloos and Mortland, 1964; Russell, 1965); ketones (Rios and Rodrigues, 196 I ; Bissada et af., 1967; Parlitt and Mortland, 1968); urea and amides (Mortland, 1966; Tahoun and Mortland, 1966b; W. J. Farmer and Ahlrichs, 1969); pyridine (V. C. Farmer and Mort-

88

M. M. MORTLAND

land, 1966); nitrobenzene (Yariv et al., 1966); amino acids (Fripiat et al., 1966); amines (V. C. Farmer and Mortland, 1965); 3-aminotriazole (Russell et al., 1968a); ethyl N,N-di-n-propylthiolcarbamate(Mortland and Meggitt, 1966). For molecules of the amide or urea type, there are two most likely sites of interaction with an exchangeable cation, the oxygen of the carbonyl group and the amide nitrogen. It has been possible with infrared absorption to distinguish between these two bonding sites. The structure of these compounds involves resonance between the following forms: 0

0-

When the formation of oxygen-to-metal bond occurs, the contribution of structure (I) will decrease, and this will result in more double-bond character for the CN bond and more single-bond character for the CO bond. The result of this is to decrease the CO stretching frequency and to raise the CN stretching frequency. On the other hand, when coordination occurs through the nitrogen the contribution of structure (11) will decrease with resulting increase in the double-bond character of the C O bond and increasing single-bond character of the CN linkage. The chief interaction for this group of compounds seems to be through the oxygen of the carbonyl, although for urea there is some indication that when alkali metal or alkaline earth cations saturate the exchange complex of the clay, interaction may be through the nitrosen, but definitely through the carbonyl for transition metal cations. The amount of decrease in the CO stretching frequency is usually in proportion to the electrophilic nature of the cat'ion. Thus, greatest shifts occur when the molecules interact with transition metal cations and least for alkali metal and intermediate for alkaline earth types. Water has its influence on the coordination of urea and amide type molecules at clay surfaces. Usually little or none of these compounds will be adsorbed on clay minerals from a water suspension. However, as the amount of water is reduced in the system, the polar molecules can compete with water for ligand positions around the metal cation, first through a water bridge (to be discussed in a following section), then in many cases a direct metal-organic interaction. Thus it is obvious that the chemistry of clay surfaces can be quite different in water suspensions as

CLAY-ORGANIC

COMPLEXES A N D INTERACTIONS

89

contrasted with drier systems, yet the relatively dry situation is as natural a one as the very wet. Research workers must learn to appreciate these differences in surface chemistry, and not assume that observations made on a suspension system necessarily apply in all environments.

D. HYDROGENBONDING This is an extremely important bonding process in many clay-organic complexes. While it is less energetic than coulombic interactions, it becomes very significant particularly in large molecules and polymers where additive bonds of this type coupled with a large molecular weight may produce a relatively stable complex. I.

Water Bridge

This is a bonding mechanism recognized only recently with the advent of detailed infrared absorption studies of clay-organic complexes. It involves the linking of a polar organic molecule to an exchangeable metal cation through a water molecule in the primary hydration shell. An example would be a ketone interacting with a hydrated exchangeablecation M4-11:

I

R

This kind of bond has been demonstrated for montmorillonite complexes with pyridine (V. C. Farmer and Mortland, 1966), ketones (Parfitt and Mortland, 1968), benzoic acid and nitrobenzene (Yariv el al., 1966), amides (Tahoun and Mortland, 1966b), and organic polymers (Parfitt, 1969). It chiefly manifests itself in the infrared spectrum by a lowering of the OH stretching frequencies of water due to the hydrogen bond formation. It is of greatest importance where the cation in question has a high solvation energy and so retains its primary hydration shell in spite of the importunities of neighboring polar organic molecules. An example of this point is in the adsorption of acetone by Na+ and Mg2+montmorillonite where Parfitt and Mortland (1968) showed for a given environment the ketone to be directly coordinated to the Na+, but indirectly coordinated to the Mg2+ through a linking water bridge. Of

90

M. M. MORTLAND

course, where special treatments have been applied to dehydrate the cation, direct ion-dipole interaction will take place.

2 . Organic-Organic Hydrogen Bonding When the exchangeable cation on the clay is an organic cation, the possibility exists of interaction with another species of organic compound through hydrogen bonding. The antecedent of this was the situation discussed earlier under hemisalt formation where it was noted that when excess organic base is present, two base molecules may share the the same proton on an equal basis. If the molecule associating with the organic cation is a dgerent species of lower basicity, the proton remains with the original organic, but may still be perturbed by interaction with negative polar groups of the other compound as below:

An example of this type of interaction was demonstrated by Mortland (l968b) when he showed by infrared absorption that pyridinium montmorillonite complexed with ethyl N,N-di-n-propylthiol carbamate (EPTC) through hydrogen bonding with the carbonyl of the EPTC. In later work Doner and Mortland (1969) complexed a series of dialykl amides of varying basicities with trimethylammonium montmorillonite and showed a positive linear relationship between the change in the NH stretching frequency of the cation and the change in the C O stretching frequency of the amides which is shown in Fig. 2. This established without ambiguity the nature of the organic-organic interaction. Lailach and Brindley ( 1969) have hypothesized a hydrogen bonding mechanism in the coadsorption of purines and pyrimidines by montmorillonite, on the basis of X-ray diffraction data. The concept and importance of organicorganic interactions at clay surfaces is worthy of future study. 3. Clay Mineral Oxygens and Hydroxyls

Interaction of molecules capable of hydrogen bonding with oxygens or hydroxyls of the clay mineral surface has been considered to be the primary mode of interaction and the basis for many models of adsorption in the past. Johns and Sen Gupta (l967) suggested that hydrogen bonding between N H groups of alkylammonium ions and oxygens of the silicate

CLAY-ORGANIC 400

-

470

-

460

-

450

-

'Z

0

COMPLEXES A N D INTERACTIONS

91

r'

P

4

440-

430

-

420

-

3 20

25

30

4'0~5

Avco,

CM-'

FIG.2. Change in the NH and CO stretching frequencies of some complexes of dialkylamides with trimethylammonium-montmorillonite compared with the NH frequency of untreated trimethylamrnonium-montmorilloniteand the CO frequency of liquid dialkylamides. (Reprinted from Clays Clay Miner. 17,265-270. Copyright ( 1 969) by Clay Minerals Society. Reprinted by permission of the copyright owner.)

sheets is responsible for stabilizing the orientation of these organic cations within the structure of vermiculite. However, for the adsorption of neutral but polar organic molecules by metal cation saturated clays, ion-dipole interactions are overriding in their effect. The very weak attraction between surface oxygen atoms and groups capable of hydrogen bonding has been the conclusion of a number of workers using infrared absorption. In the case of water, it has been shown by V. C. Farmer and Russell (1967) that the hydrogen bond between the water and oxygens of the silicate lattice is weaker than water-water hydrogen bonding with the OH stretching frequency falling at 3630 cm-' for the former and a bond at 3425 cm-I attributed 'to the latter. This work was done with nonpolar tetramethylammonium ion on the exchange sites to minimize watercation interaction. In support of these observations, Mortland (1 966) observed that urea adsorbed in the interlamellar regions of montmorillonite exhibited NH stretching vibrations at higher frequencies than

92

M. M. MORTLAND

solid urea dispersed in KBr or in a paraffin mull. Thus, a greater freedom of the amide group is indicated when the urea is in the montmorillonite, and any hydrogen bonding between NH2 groups and silicate oxygens is at least weaker than intermolecular hydrogen bonding. W. J . Farmer and Ahlrichs (1969) had similar results for urea in montmorillonite. Ledoux and White (1966) studied by means of infrared absorption the hydrogen bonding of hydrazine, formamide, and urea intercalated in kaolinite. Again they observed that the hydrogen bonding with the hydroxyls and oxygens of the kaolinite surfaces was weaker than the intermolecular hydrogen bonds. Earlier concepts of clay-organic interaction have involved the possibility of hydrogen bonding between CH groups and oxygens of the silicate sheets. Infrared results by a number of workers show no evidence for such a bond. While the primacy of the exchangeable cation in determining reactions at the clay surface has been established, the oxygens and hydroxyls of the silicate lattice will play their role with hydrogen bonding albeit a weak one. Their role in adsorption will increase in importance when the cation exchange complex is occupied by cations of low solvation energy. In the opinion of the author, hydrogen bonding through the water bridge mechanism described above is often a much more important interaction in binding polar organic molecules to clay surfaces. There is some evidence from infrared absorption and other data that stronger hydrogen bonds to oxygens of the silicate sheets of 2 : 1 minerals exist when there is tetrahedral charge than when charge arises from the octahedral layer. Negative charge would be more localized in the three oxygens attached to A P , while it would be more diffusely dispersed in the case of charge arising from substitution in the octahedral layer. Thus V. C . Farmer and Russell ( 1967) reported that N H t in saponite, vermiculite, and beidellite has an absorption band in the 3025-3050 region which they attribute to this interaction, whereas octahedrally charged smectites had no such band. Swoboda and Kunze (1968) attributed a stronger surface acidity of tetrahedrally charged over octahedrally charged smectites to stronger hydrogen bonding of coordinated water associated with exchangeable metal cations. The result would be to increase the ionization of adsorbed water molecules. Additional results obtained by Doner and Mortland (unpublished) indicate that for Cu(I1) saturated smectites, coordinated water is much more stable in tetrahedrally charged species than in those possessing octahedral charge as measured by resistance to extraction by methanol. The stronger

CLAY-ORGANIC

COMPLEXES A N D INTERACTIONS

93

hydrogen bond in conjunction with coordination with the Cu(I1) confers additional stability on the adsorbed water.

E. VAN

DER

WAALSFORCES

Van der Waals or physical forces operate between all atoms, ions or molecules, but are relatively weak. They result from attraction between oscillating dipoles in adjacent atoms. They decrease very rapidly with distance between the interacting species. They do become quite significant in clay-organic complexes, particularly for organic compounds of large molecular weight. This is because these interactions are additive, and Greenland (1965a) suggested an increment of 400 cal per mole for each CH2 segment of n-alkylammonium ions. With the large cations he suggests that van der Waals interactions dominate the adsorption process. Many workers have shown that long-chain alkylammonium ions project in a plane at high angles to the mineral surface (Weiss, 1963; Walker, 1967b). The principal van der Waals interactions are therefore suggested to be between the adsorbed molecules rather than between the adsorbed molecules and the surface. This suggests another organic-organic interaction in addition to hydrogen bonding of functional groups described in the preceding section.

F. PI BONDING A characteristic feature of transition metals is their ability to form complexes with a variety of neutral molecules, particularly unsaturated hydrocarbons. The unique characteristics of d orbitals allow certain types of unsaturated hydrocarbons and their derivatives to be bound to these metals through donation of 7r electrons of the organic compound. In a real sense these may be described in surface chemistry as Lewis bases. Solomon ( 1968) has described a variety of complexes of this type at clay surfaces and, in particular, catalytic reactions that will be described later. Doner and Mortland (1970) have observed very specific 7r bonding between benzene, xylene, toluene, and chlorobenzene and Cu(I1) montmorillonite. No other metal cation on the exchange complex gave such a complex. In order to make the complex, it was necessary to remove water which was accomplished by evacuation over PzOS. The Cu-clay was then exposed to benzene vapor. Under this environment it was observed that only octahedrally charged swelling clays made the complex and none of those tetrahedrally charged would do so. The tetrahedrally charged clay minerals apparently have much more stable

94

M. M. MORTLAND

hydration than the octahedrally charged varieties. It is felt that the water coordinated to the Cu(I1) has a hydrogen bond to oxygens around the charge site which is much stronger than that in octahedrally charged clays. The result of this is to confer additional stability on the hydration complex. G. ENTROPYEFFECTS Adsorption of some organic polymers from solution on clay minerals is apparently favored if there is a positive entropy change in the system. At first thought, this might seem inexplicable, but the solvent molecules must be considered as well as the adsorbate. Apparently many solvent molecules may be desorbed when one polymer molecule is adsorbed which results in an overall increase in translational entropy contributing to the free energy of adsorption. Greenland et al. (1 965b) found a positive entropy effect for a glycine polymer on montmorillonite. Parfitt (1969) in studying adsorption of polymers on Ca-montmorillonite observed that positive entropy changes contributed to adsorption of polymers with a molecular weight > 200 and was the dominant factor when the molecular weight was > 400. He attributed this largely to increase in translational degrees of freedom when several water molecules were desorbed as a polymer molecule was adsorbed and was able to show that water was in fact transferred from the surface to the bulk phase during the adsorption process .

H. COVALENT BONDING It is possible to create bonding between silicates and organic groupings by special techniques in the laboratory (Uytterhoeven and Fripiat, 1960). For example, Si- 0- C type bonds may be formed by reacting with acid anhydrides with the following reaction

where R is some alkyl group. Thus, exposed active hydroxyl groups may be the site of organic attachment. Another reaction involving two steps utilizes thionyl chloride as follows: = Si-OH = Sic1

+ SOCle = Sic1 + SOn+ HCI + RLi + = Si-R+ LiCl

where R is an alkyl or aryl group. Methyl derivatives of chrysotile have been made by Fripiat and Mendelovici (1968). The degree to which the

CLAY-ORGANIC COMPLEXES A N D INTERACTIONS

95

above types of reactions and resulting covalent bonding occurs in soils and sediments is unknown, but it seems likely that in the geologic column under relatively high pressures and temperatures and over geologic time some covalent bonding between organic matter and silicates may come about. Reactions of active hydroxyl groups of silicates with certain organic compounds have been used by Uytterhoeven (1962) to distinguish the water associated with exchangeable cations and the hydroxyl groups of the kaolinite crystal. For example, methyl lithium, CH,Li reacts with active hydrogen from either source as follows:

so that the amount of methane evolved is proportional to the amount of active hydrogen. When the clay surface contains water of hydration, the methane evolved is the sum of the two above reactions. When the surface is dehydrated, the methane evolved represents only structural hydroxyl groups. Grignard reagents have been used in a similar fashion. 111.

Nature of Some Clay-Organic Complexes a n d Reactions

A. ADSORPTION OF ORGANIC ACIDS,AMINOACIDS,PROTEINS, A N D NUCLEOSIDES PURINES,PYRIMIDINES, Negatively charged organic species would be expected to be repelled by the negatively charged clay minerals with little or no adsorption and this has been reported by several workers (Frissel, 1961; Arlidge and Anderson, 1962; Law and Kunze, 1966). Since the anionic nature of organic acids depends upon the H+ concentration of the system, the adsorptive properties of organic acids on clays would be expected to be pH dependent. Thus, Bingham et al. (1965) found acetate adsorption on montmorillonite to be greatest at low pH and least at high pH. Schnitzer and Kodama (1966, 1967) have also demonstrated this with a fulvic acid of molecular weight 670. They found considerable adsorption on montmorillonite at pH 2.5, but much less at high pH. Although some aluminum was dissolved from the clay at the low pH, they did not think it contributed to the adsorption by bridging through the polyvalent cation. They observed a 001 spacing of 17.5 A and a fulvic acid content of 31 mg/40 rng of clay at pH 2.5 but only a 10.5 A 001 spacing and a 15.8 mg of of fulvic acid per 40 mg of clay at pH 6.0. These results, in addition to showing the pH effect, also prove that the fulvic acid is adsorbed in the

96

M . M . MORTLAND

interlamellar regions of the montmorillonite at least at the low pH. The infrared data of Schnitzer and Kodama ( 1967) indicated that the adsorbed fulvic acid was mainly in an undissociated form. While they gave no specific mechanism it seems to the author, in view of the water present, that probably a water bridge between the carbonyls of the fulvic acid and mainly A13+on the exchange complex would be a likely interaction. It has been shown by Yariv et al. (1966) that benzoic acid is coordinated through water molecules to the more highly polarizing exchangeable cation in the interlayer space of montmorillonite, but directly coordinated to N H i and K+. On dehydration of the complex, benzoic acid became directly coordinated to all the exchangeable cations investigated. The effect of exchangeable cation on organic acid adsorption by clays has been demonstrated by Yariv et al. ( 1966). They showed by infrared absorption that the amount of benzoate ion in the clay complex depended upon the kind of exchangeable cation. The conclusion of this must be that the benzoate ion is associated primarily with the interlayer cation, not with the silicate lattice. They also observed that the infrared spectra of benzoate salts of the cations involved were not the same as in the clay complex, thus the benzoate does not form a discrete crystalline phase but presumably lies in the interlayer space. In addition, Yariv et al. (1966) showed that the 00 1 spacing of montmorillonite-benzoic acid complex varied depending upon the nature of the exchangeable cation. They conclude that benzoic acid is adsorbed to montmorillonite through the water bridge described above, by direct ion-dipole interaction depending upon the kind of cation and the hydration status of the system, or as benzoate anion associated with polyvalent cations. Kodama and Schnitzer ( 1968) also show a marked effect of exchangeable cation on fulvic acid adsorption on montmorillonite which may be explained in the same way as described above for benzoic acid. In later work, Kodama and Schnitzer ( 1969) suggested that adsorption of fulvic acid on montmorillonite is related to the ease with which the fulvic acid can displace water from between the silicate layers. This is in accord with the above discussion in that the fulvic acid and water compete for ligand positions around the exchangeable cation either for direct coordination or by bridging through directly coordinated water molecules. The adsorption of amino acids by clays has received a great deal of attention in the past. Adsorption as the cation, of course, takes place below the isoelectric point of the amino acids. When present in the cationic form, the amino acid may then be adsorbed by ion exchange with other cations on the exchange complex of the clay (Greenland et al., 1965a; Cloos et al., 1966; Sieskind, 1963; Fripiat et al., 1966; and

CLAY-ORGANIC

COMPLEXES A N D INTERACTIONS

97

earlier work referred to by Greenland, 1965a, in his review). When adsorbed at the clay surface as a cation, the amino acids resist displacement when washed with water or other solvent but may be replaced with salt solutions (Greenland et al., 196Sa; Cloos et al., 1966). In addition to adsorption by ion exchange, amino acids may become cationic after adsorption and protonation at a clay surface capable of supplying protons. As discussed in an earlier section, these protons may arise from exchangeable H + or from water associated with highly electronegative exchangeable cations. Fripiat et al. (1966) have observed by infrared absorption the zwitterion as' well as the cationic form of some amino acids and peptides adsorbed on montmorillonite. Greenland et al. (1965a) have shown that other forces also contribute to the adsorption process, which they describe as an interaction between the dipoles of the amino acids and the exchangeable cation and charged surface sites onethe clay, and of dispersion forces between the surface and the amino acid molecules. Evidence given for this was the adsorption without protonation or cation exchange by Na+ and Ca2+-montmorillonite at or near the isoelectric point of the amino acids (Greenland et al., 196Sb). Cloos et al. ( I 966) propose an adsorption process which involves salt formation between polyvalent exchangeable metal ions with the functional groups of the amino acid. This would be similar to cation bridging effects proposed by several workers and extensively reviewed by Greenland ( I96Sa), in which polyvalent exchangeable cations form a bridge between the anion and the clay surface. As mentioned above, Yariv et al. ( 1966) using infrared methods noted benzoate anion in montmorillonite when benzoic acid was adsorbed by clays, particularly when polyvalent cations were occupying the exchange sites. The function of exchangeable cations in amino acid adsorption by clays would seem to be paramount and would include the following interactions: ( 1 ) determine the proton supplying power of the clay surface and therefore the possibilities of protonation of the amino acid and thereby its cationic nature; (2) direct coordination of polar groups (carbonyl or amino) to the exchangeable cation: (3) indirect coordination through water bridges composed of water molecules directly coordinated to the cation: (4) bridging by polyvalent cations where an amino acid anion neutralizes one of the positive charges on the exchangeable cation. The applicability of the above discussion to peptides and proteins is obvious, and the same kinds of interaction would be expected. Several workers referred to by Greenland in his review (1965a) have shown that both coulombic and physical forces are involved in the adsorption of proteins by clays. Certain other factors become important in polymers as

98

M. M. MORTLAND

shown by Greenland et al. (1965b) in studies on several amino acids and their peptides. They found adsorption on acid montmorillonite resulted in protonation of amino acids and their peptides as would be expected. Under neutral conditions, the free energy of adsorption was related to molecular weights, dielectric constant, and shapes of the adsorbed molecules. They found for adsorption of glycine and its peptides on Camontmorillonite that as molecular weight increased, the entropy factor became more favorable for adsorption. The desorption of several water molecules accompanies the adsorption of the polymer thus leading to a favorable entropy effect for adsorption. For proteins of large molecular weight, the entropy factor must be important in promoting adsorption. Uncoiling and shape alterations as the protein becomes adsorbed at the clay mineral surface will have entropy effects. A number of workers referred to in Greenland’s review (1 965a) have shown that globular proteins may or may not uncoil upon adsorption in montmorillonite. Very large amounts of proteins may be adsorbed by montmorillonite resulting in interlamellar spacings of tens of angstroms. The adsorption of purines, pyrimidines, and nucleosides by montmorillonite have been studied by Lailach et al. ( 1968a,b)and Lailach and Brindley (1969). They investigated the relationships of adsorption to the exchangable cations on the clay and the pH of the medium. Generally it was found that adsorption takes place primarily by a cation exchange reaction between the inorganic cations on the clay and protonated organic molecules when the ambient pH was near the pK, value for the organic compound, thus under acidic conditions. For the alkali metal and alkaline earth cation saturated clays the effects of the metal cation on adsorption were considered to be secondary except for the nucleosides, which exhibited considerable differences. The greater adsorption of the nucleosides on the alkali metal-saturated clays was thought to result from their greater dispersion and therefore a greater accessibility of the internal surfacesof the clay mineral. Where transition-type metal cations saturated the exchange capacity of the clay, adsorption of the organic compounds at low pH took place by cation exchange processes after protonation of the organic compound. As pH was increased, complex formation with the inorganic cations became increasingly important. Considerable differences between various compounds were observed, and thus the original work should be consulted by the interested reader. It was observed by Lailach and Brindley (1969) that thymine and uracil were not adsorbed from aqueous solutions by Na- and Ca-montmorillonitein a pH range of 1-6. However, when such a compound as adenine was present too, appreciable adsorption occurred. Lailach and Brindley attributed this

CLAY-ORGANIC COMPLEXES A N D INTERACTIONS

99

coadsorption to hydrogen bond formation between the two species of molecules, the protonated one being adsorbed by cation exchange at the clay surface. This agrees with the infrared absorption observations of Mortland ( 1 968b) on pyridinium-ethyl N,N-di-n-propylthiolcarbamatemontmorillonite complexes, and of Doner and Mortland ( 1 969) on trimethylammonium-dialykl amide-montmorillonite complexes.

B. INTERACTION

OF

ORGANIC PESTICIDES WITH CLAYS

The subject of clay interactions with organic pesticides has been reviewed by Bailey and White (1964) and again in 1970. The interested reader should consult these excellent reviews for detailed information. The discussion here will therefore be limited in scope. Most, or probably all, of the specific bonding mechanisms discussed in Section I 1 apply to organic pesticides, which are extremely diversified in their structures and properties. Some pesticides are cationic and so may be adsorbed by clays by ion exchange processes. Examples of this kind are the herbicides diquat and paraquat, which are quaternary ammonium compounds and therefore strong bases. Their salts ar.e very soluble in water and are completely ionized. Weber et al. ( 1 965) studied their adsorption by montmorillonite and found them to be preferentially adsorbed by the clays up to their cation exchange capacities. Knight and Tomlinson ( I 967) studied the interaction of paraquat with a number of mineral soils and found it to be strongly adsorbed. Weed and Weber ( 1969) found that the kind of exchangeable cation markedly affected adsorption of diquat and paraquat by vermiculite but had much less effect on adsorption by montmorillonite. Once adsorbed, the two organic herbicides were much more difficult to exchange with salt solutions from montmorillonite than from vermiculite. Bioassay studies (Weber and Scott, 1966; Weber et al., 1969) revealed that paraquat adsorbed by montmorillonite exhibited very little phytotoxicity while that adsorbed by kaolinite and vermiculite became available for bioactivity with time. For most other organic pesticides which are weaker bases, their existence as cations and therefore their ability to exchange with metal ions on the clay will depend upon their ability to accept a proton from the medium, which in turn is determined by the pH. Thus, the surface acidity of clay minerals may provide the source of H+for protonating pesticides. As described in an earlier section, these protons may exist at exchange sites on the clay mineral or be generated from water associated with exchangeable metal cations. Some s-triazines were shown to become pro-

100

M. M . MORTLAND

tonated at clay mineral surfaces by Russell et al. ( 1 968b) utilizing infrared absorption. Weber ( 1966) demonstrated for a series of s-triazine compounds that the maximum adsorption on montmorillonite occurred at a pH in the vicinity of the pKo value of each compound, that is, the pH at which the compound became protonated. A further lowering of the pH resulted in some desorption of the s-triazine compounds which was attributed to competition of the protonated species with H+. An alternate explanation would be that the low pH released A13+from the clay lattice which would be a much better competitor than would H+ to displace the protonated organic cation from the exchange complex. In studying the adsorption of 3-aminotriazole (a herbicide) by montmorillonite Russell et al. ( I 968a) found by infrared absorption that it would protonate to form the 3-aminotriazolium cation. In the case of montmorillonite saturated with polyvalent cations (Ca2+,Cu2+,Na+, A13+) protonation was thought to be due to highly polarized water molecules in direct coordination with these cations. The decreasing order of extent of protonation (Ca < Mg < Al) reflects the order of decreasing polarizing power of the cations. Infrared absorption results of Russell et al. (1 968a) indicated coordination of 3-aminotriazole to Ni2+ and Cu2+ cations on montmorillonite; thus binding of a pesticide to the clay mineral surface by coordination interaction is established. Mortland and Meggitt ( 1966) showed that ethyl N,N-di-n-propylthiolcarbamate(EPTC) complexes to montmorillonite by ion-dipole interaction between the carbonyl of EPTC and the exchangeable metal cation on the clay. The decrease in CO stretching and increase in the CN frequency was related to the electron affinity of the cation. EPTC was also shown by Mortland (1968b) to be capable of being bonded to montmorillonite through a hydrogen bond provided by an organic cation on the exchange complex. Such organic-organic complexes at clay surfaces as demonstrated by this model system undoubtedly exist in nature. Complexation of pesticides with clays must affect their bioactivity to a greater or lesser extent depending upon the energy of adsorption and the ease of displacement. An example of the effect of exchangeable cation on montmorillonite on the relative ease of release of the adsorbed pesticide, are some results of the author in Fig. 3, where the differential release of 3-aminotriazole from Ca-, Cu-, and Al-montmorillonite is shown. As indicated in the figure, the Cu-clay-pesticide complex was very stable, showing only a gradual release, while the Ca-clay-pesticide system was almost immediately completely released, the A1 system being intermediate in release properties. Table I1 shows the relative phytotoxicity of these

CLAY-ORGANIC

COMPLEXES A N D INTERACTIONS

101

Number of washes

FIG.3. Release of 3-aminotriazole cornplexed with homoionic montmorillonites (Ca, At, Cu) upon extraction with water (dashed curves) and with 0.01 N MgCL (solid curves).

TABLE I I Yield of Ryegrass in Pots Treated with 5 ppm of 3-Aminotriazole, Free and Complexed with 3 Different Homoionic Montmorillonite Clays"

Treatment Control 3 AMT Ca-clay 3 AMT AI-clay 3 AMT Cu-clay + 3 AMT

+

+

Yield (dry weight, rng/pot)* 522 191

234 256 545

"Bioassay by courtesy of G. R. Stephenson and S. K. Ries, Horticulture Department, Michigan State University. "Average of 3 replicates.

102

M.

M. MORTLAND

complexes to ryegrass in comparison with the uncomplexed compound. These results are in accord with the extractions in that the toxicity is directly related to the ease of extraction. Similar results have been obtained by the author for a number of other pesticides, and bioassay data on the complexes are generally in accord with extraction results. Since it is possible to make specific cation-clay-pesticide complexes which exhibit differential release properties, it would seem a fruitful area for pesticide formulation for controlling pesticide release where such is desirable. Many pesticides are not adsorbed from water onto the clay mineral surface. These are molecules that are negatively charged (anionic) or electrically neutral and not polar enough to compete with water for adsorption sites on the clay. Anionic pesticides are often negatively adsorbed; that is, the concentration of the organic compound in the solution is increased in the presence of clay due to the repulsion of the negatively charged clay surface. On the other hand, many of these materials may be complexed on the clay surface when water is removed merely by drying in the air. Evidence for these complexations has come from infrared adsorption studies, i.e., benzoic acid (Yariv et al., (1966), and EPTC (Mortland and Meggitt, 1966). It was further shown that when EPTC-clay complexes were reintroduced to water, the herbicide was quantitatively displaced from the clay surface to the solution. The above observations suggest that water content of the system is a very important factor in determining whether or not some clay-pesticide interaction takes place and that conclusions drawn from suspension systems may not be valid at low water contents. Dry as well as wet conditions exist in nature, and the condition of the pesticide in both environments must be taken into account. A N D SURFACTANTS WITH CLAYS C. INTERACTIONOF POLYMERS All the binding mechanisms described earlier apply to polymer adsorption on clays if the appropriate functional groups are present. However, because of their large size and shape properties, entropy effects may play an important role in their adsorption by clays, as indicated by Parfitt (1969). Changes in shape of the molecules on adsorption as well as solvent-polymers and solvent-clay relationships will have their contributions to entropy effects. The adsorption of polymers is usually characterized by slow adsorption rates and a relatively small effect of temperature on adsorption. The nature of the solvent is of importance in adsorption of the polymer, being more strongly adsorbed by a surface from poor solvents and less strongly from better solvents.

CLAY-ORGANIC COMPLEXES A N D INTERACTIONS

103

In the study of the adsorption of polyoxyethylated alcohols of molecular weight around 1000 on montmorillonite, Schott (1964) found stronger adsorption on Ca2+than on Na+ clay. Rapid coverage of the Ca2+clay up to a monolayer on each silicate surface was observed. Direct interaction between the ether oxygens and Ca2+or through a bridging water molecule may have been involved. Adsorption of polyethylene glycols from water was found by Howard and McConnell ( 1967) to take place on silica. A plateau in the adsorption curve occurred at a value corresponding to slightly more than a monolayer for polymers ranging in molecular weight from 1400 to 18,000. Clapp et af. ( I 968) investigated the adsorption of a bacterial polysaccharide on montmorillonite. They found that the material was adsorbed within the interlamellar regions of the clay giving a maximum 001 spacing of 16.9 A. Adsorption depended upon the degree of dispersion of the clay, which in turn was a function of the saturating cation and the salt concentration. The polymer could be removed by washing with NaZS04 solutions. Since the amount of release was a function of the salt concentration, it was suggested that displacement of the polymer from the interlamellar positions of Na-montmorillonite was caused by a salt-controlled reduction in double-layer swelling. Periodate oxidation studies indicated that the polymer held in interlamellar positions was partially shielded from periodate reaction. Apparently the release of the polymer from the clay with salt solutions results from contractive forces between the clay layers overcoming the binding forces between the polymer and clay surface and the polymer is squeezed out. In this connection it would be interesting to compare polymer-clay complexes prepared with various exchangeable cations for the purpose of comparing release properties with the salt solution. If direct ion-dipole interactions were predominant, considerable differences in release properties would result, while if only water bridges to metal ions and hydrogen bonding to the silicate surface were the chief binding mechanisms, much less effect of exchangeable cation could be observed. Also, the effects of dehydration on the stability of the clay-polymer systems would be very interesting since it often results in much more stable clay-organic complexes. Parfitt ( 1 969) has studied the adsorption of a number of polymers and polysaccharides by montmorillonite. In addition to speciffc bond interactions described earlier, he emphasized the point that conformational changes in the polymers are important. A polymer which is coiled in the solvent may become uncoiled or the coils compressed upon adsorption at the clay surface, which can result in large entropy effects. Parfitt (1969) was able to show adsorption of high molecular weight dextran,

104

M . M . MORTLAND

amylose, soil polysaccharide by montmorillonite. An aminopolysaccharide was only slightly adsorbed by Ca-montmorillonite. Work of Schott ( 1 964) and Parlitt ( I 969) suggests salting-out effects as being of possible importance in the adsorption of certain polymers: that is, the polymer may be less soluble in the vicinity of the clay-water interface than in the bulk of the solvent due to the presence of the exchangeable cations. Other recent workers showing adsorption of polysaccharides by montmorillonite are Finch et al. (1966) and Swincer ( 1968). The binding of polyethyelene to clay surfaces through the intermediation of ionizing radiation has been studied by Nahin (1966). Many polymerization reactions may be induced through effects of ionizing radiation, and Nahin was able to promote cross-linking between polyethylene and clay surface by such a process. He proposed that the radiation caused momentarily positively charged polyethylene radicals to form, which then exchanged with H+ ions at exchange sites. Such a treatment resulted in a much more mechanically stable clay compared with samples that were not irradiated. In addition, in the irradiated samples, much less organic material could be extracted with toluene than in samples that were not irradiated. Both gamma and beta radiation were employed on kaolinite and montmorillonite clays, and a number of other organics in combination with the clay besides polyethylene were utilized as well. The results generally indicated that polyethylene could be directly bonded to the clay surface and that it was more effective if the clay surface already contained an organic such as polyvinyl alcohol than if the surface were completely inorganic. These results suggest that, over geologic time, similar bonding in natural systems between saturated and unsaturated hydrocarbons and clay materials may come to pass. The adsorption of amine-terminated polystyrene by kaolinite and montmorillonite was investigated by Dekking (1964). He found that adsorption on the clay could be accomplished by ion exchange processes if the salt of the polystyrene-amine was used or by protonation by acid-base reaction if the polystyrene-amine was reacted directly with an acid clay. The clay-polymer complexes so formed had greatly different properties from the inorganic clays as would be expected, the complexes being hydrophobic and organophilic. X-ray diffraction data indicated a polystyrene one layer thick in the montmorillonite. Van Olphen (1967) prepared a number of clay gels in combination with polyelectrolytes (macromolecular chains of polyions) and found an increase in strength of the clay gels probably because of bridging between the clay particles. The increase in strength permitted various forms of the clay gel-polyelectro-

CLAY-ORGANIC COMPLEXES A N D INTERACTIONS

I05

lyte to be created. For example, uniform thin sheets could be prepared. It was further shown that such complexes could be used for chromatographic separations. As an example, ortho, meta, and para xylenes could easily be separated by use of one of the clay-organic gels in the column of a chromatograph. Surface-active chemicals (surfactants) have come into broad use for agricultural applications. They may be generally classified as ionic and nonionic; the ionic may be subdivided into cationic and anionic. Law and Kunze ( 1 966) have studied the adsorption of compounds representing all above categories on montmorillonite and kaolinite. They concluded that the cationic species were adsorbed on the clay as cations by ion exchange and that the anionic varieties were not appreciably adsorbed. Electrically neutral but polar surfactants containing hydroxyl groups were said to be adsorbed by hydrogen bonding to oxygen atoms of the silicate surface. However, as noted in earlier discussion, infrared data indicates that hydrogen bonding to surface oxygens by alcoholic groups is very weak and more important interactions are undoubtedly ion-dipole type directly with exchangeable metal cations or indirect coordination through bridging water molecules of the primary hydration shell of the cations. Valoras et al. (1969) reported adsorption of several nonionic surfactants by montmorillonite, vermiculite, and kaolinite as well as by soil and its constituents.

D. CATALYSIS REACTIONS Clays and clay minerals have been shown to catalyze reactions of various kinds in molecules adsorbed at their surfaces. Most of the work on catalytic reactions at surfaces of silica, silica-alumina, alumina, zeolites, and clay minerals has been on systems extensively -dehydrated and at high temperatures. More recent work, however, has demonstrated catalytic reactions at relatively low temperatures and sometimes at appreciable hydration levels, and it is this kind of environment that is of principal interest in soil clays and in sediments. McAuliffe and Coleman ( 1955) demonstrated the catalytic effect of acid clays on the hydrolysis of ethyl acetate and the inversion of sucrose. They found that Al:’+ saturated clay had less catalytic effect than did clay which was predominately H saturated. In other hydrolysis studies, Mortland and Raman ( I 967) found that Cu(I1) was very effective in catalyzing the breakdown of several organic phosphates. When Wyoming bentonite had Cu( 11) as the exchangeable cation, catalytic activity was very great,

106

M. M. MORTLAND

while Cu(I1) saturated nontronite, vermiculite, and beidellite were much less active, and Cu(I1)-organic soil had no catalytic effect at all. These results demonstrate the marked influence of the exchanger on the Cu(I1) activity and in consequence its ability to catalyze the hydrolysis of the organic phosphates. In studies of other reactions at clay mineral surfaces, Chaussidon and Calvet (1965) noted that alkylammonium cations on the exchange sites of montmorillonite would decompose at a relatively low temperature to give ammonium ion plus some hydrocarbons of various chain lengths. In similar work with protonated lysine on the cation exchange sites of montmorillonite, Mortland (unpublished) found decomposition to ammonium ion, but at a relatively high temperature of 250°C. Fripiat et al. ( 1966) reported that when glycine and p-alanine adsorbed on montmorillonite was dehydrated and heated, amide linkages were observed to form. Weiss ( 1963) reported that proteins adsorbed on highly charged mica surfaces with H 3 0 + were cleaved into peptides and amino acids. Mortland (1966) observed the catalytic decomposition of urea to ammonium on montmorillonite films in air-dry condition and at 20°C when the exchangeable cation was Cu(1I). Ni(I1) and Mn(I1) clays showed smaller amounts of urea decomposition but alkali and alkaline earth saturated clays showed none at all. Russell et al. (1968a) observed the formation of hydroxy atrazine when atrazine was adsorbed on Hmontmorillonite. This reaction involves the substitution of a hydroxyl group for a chlorine atom on the ring structure of the triazine and is an example of a nonbiological mechanism of degradation of a pesticide. Skipper (1 970) confirmed this observation and showed that another acid clay material, allophane, was not able to catalyze this reaction. This undoubtedly is because the surface acidity of the allophane is not as low as that of the clay mineral montmorillonite. Catalytic decomposition of a polyalcohol, glycerol, by layer silicates has been reported by Walker (1967a). Carbon was apparently the product and was observed when the clays were immersed in boiling glycerol. His conclusions were that: ( 1) the glycerol decomposition occurs only when two silicate surfaces are in simultaneous contact with the molecule; and (2) small and highly charged cations in the exchange complex greatly enhance the effect. Other work which shows the effect of metal cations on catalytic reactions at clay surfaces is that of Solomon (1 968), who observed that aluminum at crystal edges and transition metals in the higher valency state on the exchange complex act as electron acceptor sites, while transition metals in the lower valency states could act as electron doner sites. Thus, as required for a given polymerization reac-

CLAY-ORGANIC

COMPLEXES A N D INTERACTIONS

107

tion, the clay minerals, through the appropriate metal cation, donate or accept electrons in catalyzing the reaction. Also the clay may function as an inhibitor if it converts reactive organic intermediates to nonreactive species through the electron donation or acceptance mechanism. The effect of the high electric field at clay mineral surfaces in catalyzing various reactions has been discussed by Fripiat (1968). It is proposed that the greater acidity of water at clay surfaces as found by several workers promotes catalytic react ions requiring protons. Among these reactions are decomposition of ammines to N H i , the formation of triphenylcarbonium ion from triphenylcarbinol, polymerization of amino acids, and the decomposition of amines. I n connection with the latter, Fripiat ( 1968) reported at least eight different hydrocarbons determined chromatographically which result from heating of ethylammonium montmorillonite at 300°C for 6 hours. The initial products of the breakdown are N H $ C2H5+,the latter being very active then entering into a variety of reactions resulting in the observed products. It seems reasonable to suggest that catalytic reactions at clay surfaces in soils and in sediments play a much more important role in conbersions and alterations of organic compounds adsorbed at their surfaces than one would conclude from the other literature where almost all such changes are attributed to a biological agency. In the geologic column, high pressures and relatively high temperatures may promote catalytic reactions requiring such conditions. Thus, the kinds of reactions described here are undoubtedly of importance in diagenetic changes which take place in sediments in geologic time.

+

E. DIFFUSION OF

ORGANIC C O M P O U N D S IN C L A Y S

Diffusion of organic compounds in clays has relevance to natural systems such as soils and sediments but very few data on this subject appear in the literature. For example, diffusion rates of various pesticides in soils and sediments may be of importance in contributing to their dissemination in the environment. Ehlers et al. ( 1969a,b) have considered the water content, bulk density, and temperature effects on lindane diffusion in soils and the theoretical implications of such movement. The diffusion rates of organic compounds in clays will depend upon a number of variables; these include (1) mechanism of binding of the organic compound to the clay surface (If strongly bound, for example a cation, surface diffusion would likely predominate. On the other hand, if it is only weakly bound, most diffusion will take place away from the clay surface in adjacent water films or in the vapor phase in voids of unsatu-

108

M. M. MORTLAND

rated clay matrices.); (2) other properties of the organic compound, such as molecular weight, solubility in water; (3) water content of the system; (4)nature of the clay; (5) temperature; and (6) bulk density. The self-diffusion of various alkylammonium ions in several montmorillonites and kaolinite have been studied by Gast, Lewis, and Mortland (unpublished). In montmorillonites equilibrated at zero relative humidity, the self diffusion coefficients for methyl, ethyl, and propyl ammonium ions were of the order of cm2 sec-I while at 93% relative humidity it was of the order lo-* cm2 sec-'. It was of interest to note that the first increment of water adsorbed, increased the diffusion rates much more than later increments. The major effect of hydration in promoting the alkylammonium ion diffusion is attributed to an unkeying of the cation from the clay lattice resulting in a much freer condition of the cation. The water hydrogen bonds with the N H i groups in the hydration process. Diffusion rates of the organic cations in kaolinite were generally much faster at given relative humidities than in the swelling clays. The diffusion of three anionic dyes, eosin yellowish, bromophenol blue, and naphthol green bluish in montmorillonite were studied by Anderson et al. (1967). They assumed the compounds were moving in adsorbed water layers on the clay surface. They found apparent diffusion coefficients of the order of lo-' cm2 sec-l at a relative humidity of 97.5%, which they felt was 100-I000 times lower than would be expected in aqueous solution on the basis of molecular weight. At 80% relative humidity, there was no diffusion at all. They concluded that the adsorbed water was liquidlike in its behavior and that dissolution and diffusion of solutes in adsorbed water films is of importance at water contents far below the wilting point. They also recognized that organic-clay interactions may have played a role in affecting diffusion rates. There is no doubt that these compounds would interact with the clay mineral through ion-dipole interaction with exchangeable metal ions especially at lower relative humidities and through water bridges at higher water contents. IV.

Nature and Importance of Some Clay-Organic Complexes in Soils and Sediments

The existence of such a powerful adsorption medium as clay minerals in the natural environment of soils and sediments must exert some influence on biological processes occurring there. Mention was made at the beginning of the high correlation between organic matter and clay contents of many soils and the fact that most clays have organic matter strongly adsorbed to them. Thus, as the biological processes proceed, the

CLAY-ORGANIC COMPLEXES A N D INTERACTIONS

109

clay fraction of the soil will exert its influence by interaction with undecomposed organic matter, intermediate products of decomposition, and end products of this process. Not only that, the clay may exert direct influence on the biological agents themselves, such as enzymes, microorganisms, plant roots. The work summarized by McLaren and Peterson ( 1965) is particularly interesting in this regard and should be consulted by the reader. In studying the relationship of soil clay mineralogy to the occurrence of Fusarium wilt in bananas in Panama, Stotzky et al. (1 96 1 ) found a high correlation between resistance to the disease and soils which contained montmorillonite in the clay fraction. Soils which did not contain montmorillonite had short lifetimes with respect to the production of bananas because of the buildup of the Fusarium wilt. It is not clear exactly what the function of the montmorillonite was in reducing the disease. Several possibilities exist; for example, the clay might act as an absorbent for toxins from the Fusarium or perhaps interact with the Fusarium, or even interfere in the interactions between the plant and the pathogen in a beneficial way. It was noted earlier that a number of workers utilizing infrared absorption techniques have found clay mineral surfaces to be more acidic than would be expected from pH measurements of water suspensions of these clays. In anticipation of these findings, biologists utilizing enzymes in clay mineral systems have noted greater acidic properties at clay surfaces than the pH measurement would indicate. Thus Peterson ( 1957), McLaren and Seaman ( 1968) found that for some enzyme actions in clay systems may be shifted by as much as two pH units toward the alkaline region suggesting that the acidity at the clay surfaces is lower than the ambient buffer solution. McLaren and Peterson (1965) suggest that apparently the enzymes respond to the concentration of hydrogen ions rather than activity, which would be the same at all points in the system at equilibrium, but it is also likely that the true activity may not be reflected by pH measurements with glass electrodes in colloidal systems. Some enzyme systems maintain activity after adsorption at clay mineral surfaces while others are apparently deactivated or at least reduced in activity (McLaren and Peterson, 1965; Galstyan et al., 1968). Bacteria-clay interactions are of importance as suggested by the work of Marshall (1968). He found that r. trifolii survived high temperatures much better in the presence of montmorillonite or illite than without them. He suggested an edge-to-face association between the clay platelets and the bacteria. The extrapolation of these results were to suggest that the ability of these bacteria to survive exposure to high temperatures in dry

110

M. M. MORTLAND

soils was related to a protective effect of the soil colloids. It was suggested that a clay covering might protect the bacteria from exposure to high temperatures by modifying the rate of water loss from the cells. Jenny and Grossenbacher ( 1963) utilizing electron microscopy methods observed very intimate contact between mucigel surrounding plant roots and clay surfaces. They propose that the mucigel and clay particles are actually bound together by chemical interaction, perhaps through carboxyl-polyvalent cation linkages, for example, A13+on the minerals. The ramifications of the plant-mucigel-mineral matrix regarding diffusion, transpiration, and metabolic processes are obvious. The organic materials associated with clays in nature are often extremely complex materials. In this connection the work of Stevenson (1969), Degens (1967), and Murphy er al. (1969) should be consulted. The end products of the biological and diagenetic processes are such materials as kerogen, coal, and petroleum. As diagenesis proceeds, oxygen decreases, carboxyl and alcoholic groups are lost, there is a lowered nitrogen content, and a concomitant increase in carbon occurs. The relative distribution of various kinds of organic matter between the clay mineral interfaces and the bulk of the soil or sediment has apparently not received great attention. Some compounds, such 3s polysaccharides, amino acids, peptides, proteins, which normally would be decomposed by microorganisms, may be protected to some extent when they are adsorbed within the interlamellar regions of clay minerals. However, they may undergo chemical alteration at the clay surfaces as indicated in the section on catalysis. This may account for the fact that ammonium ion often occurs in abundance in illites which could have arisen from diagenetic changes of a swelling clay, vermiculite, or montmorillonite that had adsorbed considerable quantities of amino acids or peptides. The adsorption of freshly formed humic substances from aqueous extracts of humified clover by several types of clays was studied by Wada and lnoue ( I 967) and lnoue and Wada (1968). They found much greater adsorption by allophane than by such crystalline clay minerals as montmorillonite and halloysite; this seems to concur with the observation of the high organic matter contents of soil clays high in allophane. They also found that the humified clover extract was not affected by the nature of the exchangeable cation on the clay and that it did not penetrate the interlamellar spaces of montmorillonite to any great degree. This is in contrast to results with soil humic or fulvic acids, where Kodama and Schnitzer ( 1968) have shown a great effect of the exchangeable cation on adsorption of a soil humic compound by montmorillonite. In addition, it

CLAY-ORGANIC COMPLEXES A N D INTERACTIONS

111

was shown by Schnitzer and Kodama ( 1 966) in the laboratory that a fulvic acid from soil organic matter would penetrate the interlamellar regions of montmorillonite. Adsorption and penetration of the fulvic acid was very much pH dependent, being greatest at the low pH of 3 and decreasing as pH increased to 7. Thermal studies on fulvic acid-montmorillonite complexes by Kodama and Schnitzer (1970) suggest that it is possible to differentiate between externally adsorbed and internally adsorbed fulvic acid on montmorillonite. The externally adsorbed material decomposed before the combustion of that retained in the interlamellar spaces. Isothermal experiments also showed that the fulvic acid complexed with the montmorillonite delayed the thermal decomposition as compared with uncomplexed fulvic acid, which relates to the observed stability of clay-organic matter complexes in nature. In contrast to this, McLaren and Peterson ( I 965) pointed out that they were unable to find proof that organic matter has penetrated the internal surfaces of montmorillonites isolated from soil. Changes in the nature of organic matter when clay (Wyoming bentonite) was applied to sand soils was studied by Colom and Wolcott (1967). In these experiments clay application rates up to 50 tons per acre had been applied to Plainfield sand, mixed in the top 6 inches, then cropped for several years. Colom and Wolcott found positive correlations between clay rates and various categories of acid- and alkali-soluble organic fractions. Total nitrogen increased and the carbon-nitrogenratio decreased with increasing clay rates. While it was obvious that the clay had affected the development of organic matter in the soil, it was not clear what the relationship was between the organic matter and clay, that is, whether the resulting material was in fact an organic-clay complex, discrete organic material, or a combination of these two categories. The role of clay-organic complexes in soil structure has long been appreciated. The adsorption of organic materials by clays modifies the relationship of the clay to the surrounding environment in a fundamental way. These modifications include the interaction with water and salts as well as effects on swelling properties. Inter- as well as intraparticle bonds result, and clay-organic complexes form aggregates of varying size and stability which are of major importance in determining the physical nature of soil and, therefore, the environment it provides for plant growth. Greenland (1965b) has well documented the work in this area. The recent work of Edwards and Bremner (1967) is of particular interest because it attempts to relate bonding mechanisms described earlier in this work that have been established on a fundamental basis,

112

M. M. MORTLAND

with actual aggregate formation and stability. The proposal is made that organic materials and clay particles are linked via polyvalent metal cations and that microaggregates consist of groupings of these complexes resulting from bridging through polyvalent metals in .various combinations with clay and organic matter. These workers point out that a Nasaturated resin shaken in water with the soil aggregates has a dispersing effect which probably results from the adsorption of the polyvalent ions such as Ca, Mg, Fe, and A1 by the resin, and a substitution of sodium in the aggregate weakens the interparticle bonds which promotes bond rupture required for dispersion. Dispersive effects of alkali metal cations has usually been explained in the past on the basis of zeta-potential effects. But Edwards and Bremner ( 1 967) pointed out that the aggregates described above could easily be dispersed by sonic or ultrasonic vibration without previous saturation with an alkali metal ion and showed no tendency to flocculate on standing after vibration treatment. The bonds formed in the aggregate formation are suggested to be weak enough to be broken by the vibrations imposed upon them and that stable microaggregates are formed by a mechanism which is a reversal of the process by which they may be dispersed. The concepts of Edwards and Bremner (1967) on microaggregates of soils are extremely intriguing because they coincide to a degree with the basic mechanisms of adsorption of polar organic molecules by clays established through infrared and other studies. That is, that the nature of the exchangeable cation is of paramount importance in determining the nature and energy of the adsorption process. Where protonation is not involved, ion-dipole or coordination-type interactions, either direct or through a water bridge, are decisive, as pointed out in earlier sections. Polyvalent cations are generally much more electrophilic than monovalent cations like the alkali metals and so form much firmer bonds with functional groups of organic compounds which are able to furnish electrons (i.e., carboxyl and amino groups). Also, electrostatic adsorption through neutralization of positive charges on polyvalent cations by anionic groups on organic matter is a possibility. It would seem to the author that the most important bonding mechanism which could be easily broken by vibration might be organic matter-water bridge-polyvalent cation-clay, since direct coordination between functional groups and polyvalent cations would be quite energetic and less likely to be broken by sonic or ultrasonic vibrations. Also, under natural conditions many such polyvalent cations would be likely to retain their primary hydration shell. The water bridge bond would be a much more likely candidate for disruption than would direct coordinate bonds. It would seem that the

CLAY-ORGANIC COMPLEXES A N D INTERACTIONS

113

above concepts may provide a better explanation for microaggregate formation and stability than those based on electrostatic considerations. V.

Conclusions

The dominant factors determining the nature of clay-organic interactions are the properties of the organic molecule, the water content of the system, the nature of the exchangeable cation on the clay surface, and the unique properties of the clay mineral structures. The exchangeable cations determine the surface acidity and therefore the possibilities of protonation of the organic compound. However, even in a homoionic clay the surface acidity will vary with hydration level, becoming more acidic with decreasing water content. Where protonation of the organic molecule is not involved, the exchangeable metal cations act as electron acceptors by which they interact with electron-donatingfunctional groups of organic compounds. Such ion-dipole or coordination type of bonding will vary greatly in energy depending upon the nature of the exchangeable cation. Here again the hydration level of the system may be a factor because the exchangeable metal cation may retain its primary hydration shell, in which case functional groups of organic materials may be absorbed to the cation by hydrogen bonding via a water bridge. Where organic cations occupy the exchange sites on the clay mineral, other organic compounds may interact with them by hydrogen bonding to form an organic-organic complex on the mineral surface. It should be remembered that for most nonionic interactions with clay minerals, organic compounds are in competition with water for adsorption sites. It must be recognized that the surface chemistry of clays, with regard to interaction with organics or other molecules, is different in water suspensions compared with air-dry environments, that results obtained in one situation may or may not apply to the other, and that both extremes may exist in nature. The clay mineral surface makes its contribution to adsorption of organic molecules through hydrogen bonding between its oxygens or hydroxyls and appropriate functional groups of the organic material. The contribution of the clay surface itself to the total adsorption energy is maximal when ions of low solvation energy occupy the cation exchange sites and is minimal when they are occupied by cations of high solvation energy. In addition, the charge density of the clay minerals will sterically affect the position and orientation of organic molecules associated with exchange sites either as cations or indirectly through ion-dipole interactions. Much remains to be learned regarding the clay-organic systems and the reactions taking place at this interface. In particular, it is quite likely that

114

M. M. MORTLAND

many reactions, which have not yet been recognized, are catalyzed by the clay minerals. In the past, almost all alterations of organic materials in soils and sediments were attributed to a biological agency, but the author is confident that the future will record important effects of clay surfaces on some processes. REFERENCES Anderson, D. M., Brown, R. L., and Buol, S. W. 1967. Soil Sci. 103,28 1-287. Arlidge, E. Z., and Anderson, G . 1962. J. Soil Sci. 13,2 16. Bailey, G. W., and White, J. L. 1964. J. Agr. Food Chem. 12,324-332. Bailey, G . W., and White, J. L. 1970. ResidueRev. (in press). Bailey, G. W., White, J. L., and Rothberg, T. 1968. Soil Sci. Soc. Amer., Proc. 32,222-234. Barrer, R. M., and Brummer, K. 1963. Trans. Faraday SOC.59,959-968. Bingham, F. T., Sims, J. R., and Page, A. L. 1965. Soil Sci. SOC.Amer., Proc. 29,670-672. Bissada, K., Johns, W. D., and Cheng, F. S. 1967. Clay Miner. 7,155-166. Chaussidon, J., and Calvet, R. 1965.5. Phys. Chem. 69,2265. Clapp, C. E., Olness, A. E., and Hoffman, D. J. 1968. Trans. 9th Znt. Congr. Soil Sci., 1968, VOI.I , pp, 672-634. Cloos, P. and Mortland, M. M. (1965). Clays Clay Miner. 13,23 1-246. C~OOS, P., Calicis, B., Fripiat, J. J., and Makay, K. 1966. Proc. Znt. Clay Conf., 1966, pp. 223-232. Colom, J., and Wolcott, A. R. 1967. Plant Soil 26,261-268. Degens, E. T. 1967. In “Diagenesis in Sediments” ( G . Larsen and G. V. Chilingar, eds.), pp. 343-390. Elsevier, Amsterdam. Dekking, H. G. G. 1964. Clays Clay Miner. 12,603-616. Doner, H. E., and Mortland, M. M. 1969. Clays Clay Miner. 17,265-270. Doner, H. E., and Mortland, M. M. 1970b. Science 166,1406- 1407. Dowdy, R. H., and Mortland, M. M. 1967. Clays Clay Miner. 15,259-27 1. Dowdy, R. H., and Mortland, M. M. 1968. SoilSci. 105,36-43. Edwards, A. P., and Bremner, J. M. 1967. J. Soil Sci. 18,64-73. Ehlers, W., Farmer, W. J., Spencer, W. F., and Letey, J. 1969a. Soil Sci. SOC. Amer., Proc. 33,505-508. Ehlers, W., Letey, J., Spencer, W. F., and Farmer, W. J. 1969b. Soil Sci. SOC. Amer., Proc. 33,50 1-504. Farmer, V. C. 1968. Clay Miner. 7,373-387. Farmer, V. C., and Mortland, M. M. 1965. J . Phys. Chem. 69,683-686. Farmer, V. C., and Mortland. M. M. 1966. .I. Chem. SOC.,A pp. 344-35 1. Farmer,V.C.,and Russel1.J. D. 1967.ClaysCIuvMiner. 15,121-142. Farmer, W. J.,and Ahlrichs, J. L. 1969. SoilSci. SOC.Amer., Proc. 33,254-258. Finch, P., Hayes, M. H. B., and Stacy, M. 1966. Trans. Comm. 11 and IV, Znt. Soil Sci. SOC., 1966 pp. 19-32. Fripiat, J. J. 1968. Trans. 9th Znt. Congr. Soil Sci., 1968 Vol. 1, pp. 679-689. Fripiat, J. J., and Mendelovici, E. 1968. Bull. SOC.Chim. Fr. pp. 483-492. Fripiat, J. J., Chaussidon, J., and Touillaux, R. 1960. J. Phys. Chem. 64,1234- 1241..

CLAY-ORGANIC COMPLEXES A N D INTERACTIONS

115

Fripiat, J. J., Servais, A., and Leonard, A. 1962. Bull. SOC.Chim. Fr. pp. 635-644. Fripiat, J. J., Jelli, A., Poncelet, G., and Andre, J. 1965. J . Phys. Chem. 69,2 185-2 197. Fripiat, J. J., Cloos, P., Calicis, B., and Makay, K. 1966. Proc. I n t . Clay Conf., IY66 pp. 233-245. Frissel, M. J. I96 I . Versl.Landbouwk. Onderzoek 67.B, 3-54. Galstyan, A. Sh., Tatevosian, G. S., and Havoundjian, Z. S. 1968. Trans. 9th Inr. Congr. SoilSci., 1968Vol. 3,pp. 281-288. Glaeser, Rachel. 1954. Ph.D. Thesis, University of Paris. Greenland, D. J. 1965a. SoilsFerf. 28,415-425. Greenland, D. J. 1965b.SoilsFerf. 28,521-532. Greenland, D. J., Laby, R. H., and Quirk, J. P. 1965a. Trans. Faraday SOC.61, 20132033. Greenland, D. J., Laby, R. H., and Quirk, J. P. 1965b. Trans. Faraday SOC.61, 20242035. Harter, R. D., and Ahlrichs, J. L. 1967. Soil Sci. Soc. Amer., Proc. 31,30-33. Howard, G .J., and McConnell, P. 1967.5. Phys. Chem. 71,2974. Hunt, J. P. 1963. “Metal Ions in Aqueous Solution.” Benjamin, New York. Inoue, T., and Wada, K. 1968. Trans. 9th Inf. Congr. Soil Sci., 1968 Vol. 3 , pp. 289-298. James, D. W., and Harward, M. E. 1962. Clays Clay Miner. 11,301-320. Jenny, H., and Grossenbacher, K. 1963. SoilSci. SOC.Amer., Proc. 27,273-277. Johns, W. D., and Sen Gupta, P. K. 1967.Amer. Mineral. 52,1706-1724. Keay, J . , and Wild, A. I96 1. Clay Miner. Bull. 4,22 1. Kijne, J. W. 1969. Soil Sci.SOC.Amer., Proc. 33,539-542. Kim, J. T., and Weed, S. B. 1968.Agronomy Absrr. p. 90. Knight, B. A. G . , and Tomlinson, T. E. 1967. J. Soil Sci. 18,233-243. Kodama, H., and Schnitzer, M. 1968. Soil Sci. 106,73-74. Kodama, H., and Schnitzer, M. 1969. Proc. Inf. Clay Conf., 1969 (in press). Kononova, M. M. 1961. “Soil Organic Matter.” Pergamon Press, Oxford. Lailach, G. E., and Brindley, G. W. 1969. Clays Clay Miner. 17,95-100. Lailach, G . E., Thompson, T. D., and Brindley, G . W. 1968a. Clays Clay Miner. 16,285293. Lailach, G. E., Thompson, T. D., and Brindley, G. W. 1968b. Clays Clay Miner. 16,295301. Law, J. P., Jr., and Kunze, G. W. 1966. Soil. Sci. SOC.Amer., Proc. 30,32 1-327. Ledoux, R. L., and White, J. L. 1966. J. Colloid Interface Sci. 21,127. McAuliffe, C., and Coleman, N. T. 1955. Soil Sci. SOC.Amer., Proc. 19,156-160. McLaren, A. D., and Peterson, G. H. 1965. ASA (Amer. SOC.Agron.) Monogr. 10,259284. McLaren, A. D., and Seaman, G. V. F. 1968. Soil Sci. SOC.Amer., Proc. 32,127. McNeal, B. L. 1964. Soil Sci. 97,96-102. Marshall, K. C. 1968. Trans. 9fh Inr. Congr. Soil Sci., 1968 Vol. 3, pp. 275-280. Mortland, M. M. 1966. Clay Miner. 6,143-156. Mortland, M. M. 1968a. Trans. 9th I n f . Congr. SoilSci., 1968 Vol. I , pp. 691-699. Mortland, M. M. I968b. J . Agr. FoodChem. 16,706-707. Mortland, M. M., and Barake, N. 1964. Trans. 8th Inf. Congr. Soil Sci., 1964 Vol. 3, pp. 433-443. Mortland. M. M., and Meggitt, W. F. 1966. J. Agr. Food Chem. 14,126-129.

116

M . M . MORTLAND

Mortland, M. M., and Raman, K. V. 1967. J.Agr. Food Chem. 15,163- 167. Mortland, M. M., and Ranan, K. V. 1968. Clays Clay Miner. 16,393-398. Mortland, M. M., Fripiat, J. J., Chaussidon, J., and Uytterhoeven, J. 1963. J. Phys. Chem. 67,248-258. Murphy, R. C., Djuricic, M. V., Markey, S. P., and Biemann, K. 1969. Science 165,695697. Nahin, P.G. 1966.ClaysClayMiner. 13,317-330. Ovcharenko, F. D., Tarasevich, Yu. I., and Radul, N. M. 1967. Dopov. Akad. Nauk Ukr. RSR, Ser. B 29.63 I . Pafitt, R. L. 1969. Ph.D. Thesis, University of Adelaide, Adelaide, Australia. Pafitt, R. L., and Mortland, M. M. (1968). Soil Sci. Soc. Amer. Proc. 32,355-363. Peterson, G. H. 1957. Ph.D. Thesis, University of California, Berkeley, California. Raman, K. V., and Mortland, M. M. 1969. Soil Sci. Soc. Amer., Proc. 33,3 13-3 17. Rios, E. G., and Rodrigues, A. 1961. An. Real SOC. Espan. Fis. Quim., Ser. B 57, 117130. Russell, J. D. 1965. Trans. Faraday SOC.61,2284-2294. Russell, J. D., and Farmer, V. C. 1964. Clay Miner. Bull. 5,443-464. Russell, J. D., Cruz, M.,and White, J. L. 1968a.J.Agr. FoodChem. 16,21-24. Russell, J . D., Cmz, M., White, J. L., Bailey, G., Payne, W. R., Pope, J. D., and Teasley, J. 1. 1968b.Science 160,1340-1342. Schnitzer, M., and Kodama, H. 1966. Science 153,70-7 I . Schnitzer, M., and Kodama, H. 1967. SoilSci. SOC.Amer., Proc. 31,632-636. Schott, H. 1964. Kolloid-2.2. Polym. 199,158. Serratosa, J . M. 1966. Clays Clay Miner. 14,385-391. Sieskind, 0 . 1963. Ph.D. Thesis, Faculty of Science, University of Strasbourg. Skipper, H. D. 1970. Ph.D. Thesis, Oregon State University, Corvallis, Oregon. Solomon, D. H. 1968. Clays Clay Miner. 16,3 1-39. Stevenson, F. J . 1969. SoilSci. 107,470-479. Stotzky, G., Dawson, J. E., Martin, R. T., and Kuile, C. H. H. 196 1. Science 33,1483. Swincer, G. D. 1968. Ph.D. Thesis, University of Adelaide, Adelaide, Australia. Swoboda, A. R., and Kunze, G. W. 1968. Soil Sci. SOC.Amer., Proc. 32,806-8 I 1. Tahoun, S., and Mortland, M. M. 1966a. Soil Sci. 102,248-254. Tahoun, S., and Mortland, M. M. 1 966b. Soil Sci. 102,3 14-32 1. Tarasevich, Yu. I., Radul, N. M., and Ovcharenko, F. D. 1967. Dokl. Akad. Nauk SSSR 173,6 15. Theng, B. K. G., Greenland, D. J., and Quirk, J. P. 1967. Clay Miner. 7,l-17. Uytterhoeven, J. 1962. Bull. Groupe Fr. Argiles 14,69-76. Uytterhoeven, J., and Fripiat, J. J. 1960. Int. Geol. Congr., Rep. Sess., Norden, 21st. 1960 Part 24, pp. 80-87. Valoras, N., Letey, J., and Osborn, J. F. 1969. Soil Sci. Soc. Amer., Proc. 33,345-348. Van Olphen, H. 1967. Cfays CfayMiner. 15,423-435. Wada, K., and Inoue, T . 1967. Soil Sci. Plant Nutr. (Tokyo) 13,9- 16. Walker, G. F. 1967a. Clay Miner. 7,111-1 12. Walker, G. F. 1967b. Clay Miner. 7,129-143. Walker, G . F., and Garrett, W. G. 1967. Science 156,385-387. Weber, J. B. 1966.Amer. Mineral. 51,1657-1670. Weber, J. B.,and Scott, D. C. 1966. Science 152,1400-1402.

CLAY-ORGANIC COMPLEXES A N D INTERACTIONS

I17

Weber, J. B., Perry. P. W., and Upchurch, R . P. 1965. Soil Sci. SOC.Amer., Proc. 29, 678-688. Weber, J. B.. Meek, R . C., and Weed, S. B. 1969. Soil Sci. Soc. Amer., Proc. 33, 382385. Weed, S. B., and Weber, J. B. 1968.Amer. Mineral. 53,478. Weed, S. B., and Weber, J. B. 1969. Soil Sci. Soc. Amer., Proc. 33,379-382. Weiss, A. 1963. ClaysCIay Miner. 10,191-224. Yariv, S., Russell, J. D., and Farmer, V. C . 1966. IsraelJ. Chem. 4.20 1-2 13.

This Page Intentionally Left Blank

BI RDSFOOT TREFOIL Robert R. Seaney a n d Paul R. Henson Cornell University, Ithaca, New York and U. S. Department of Agriculture, Beltsville, Maryland

Page 1.

Introduction ............................. ...... . ..... ........................................... Origin and Distribution ............................................................... B. Agricultural History ,.................................. .............................. ... C . Economic Importance ................................ ... ... ........................... Morphology . . .. .. ........... . .................................................................. A. Root ...................... ..................... B. Stem and Leaf ........ A.

II.

111.

.................... ................................... C. Seedling Growth ........................................................................ D. Vegetative Propagation .............................. ....................... .......... IV. Adaptation.............................................. B. Soils and Soil Fertility C . Inoculation A.

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

................................................................... H . Insects ...................................................................................... Utilization ............................................................. .......................... A. Hay and Silage .......................................................................... B. Pasture .................................................................................... C. Feeding Value ........ .................................. VI. Genetics and Cytology .... V.

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

B.

VII.

Inheritance of Characters ........................... .. . . ........ .....................

Breeding ....................,

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

B. Variability and Methods .............................................................. C . Pod Dehiscence ........................................................................ I I9

120 120 120 121 i22 122 122 123 23 25 25 25 26 27 127 127 128 129 129 131 131 133 135 136 139 139 140 142 143 143 144 145 146 147 147 148 148

120

ROBERT R. SEANEY A N D PAUL R. HENSON

D. Seed Yield ...... ..................................... E. Seedling Vigor ................................ F. Disease Resistance .......................... G. Winterhardiness ................................. H. Clone Crosses ................................. I . Inbreeding and Hybridization J. Varieties ....................... References ...............................................

I.

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

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

I49 149 I50 I50 I50 151 I51 153

Introduction

A. ORIGINAND DISTRIBUTION The genus Lotus consists of a diverse group of annual and perennial species widely distributed throughout the world. Depending upon the system of classification, there are approximately 80 or up to 200 different species in the genus (Callen, 1959; Isely, 1951; Zandstra and Grant, 1967). The greatest diversity of species is found in the Mediterranean basin, an indication that this area was probably the center of origin for the Old World species. Species endemic to North America extend along the West coast, from British Columbia to Mexico and Lower California. California contains more of the New World species than any other state (Ottley, 1923, 1944). Three perennial trefoil species are used for forage production in the United States. The most important of the three, birdsfoot trefoil, Lotus corniculafus L., is extensively grown for pasture and hay in North central and Northeastern United States and Eastern Canada. Narrowleaf trefoil, L. tenuis Wald et Kit., is an important pasture legume on heavy, imperfectly drained soils in New York, California, and Oregon. Big. trefoil, L. pedunculatus Cav., because of susceptibility to certain diseases and lack of tolerance to drought, is used only on the moist coastal soils of the Northwest and, to a limited extent, on low-lying coastal soils of the Southeast. B. AGRICULTURAL HISTORY Robinson’s review (1 934) of the early agricultural history of trefoils in Europe indicates that the value of trefoil as a forage crop has been recognized for more than 200 years. Ellis (l774), in England, wrote of the desirability of trefoil as a feed for cattle. Published reports on birdsfoot trefoil in Europe, as reviewed by Robinson (1934) and MacDonald (1946), indicate a continuing interest in the species but apparently no

BIRDSFOOT TREFOIL

121

real expansion of trefoil culture until 1900 or later. Difficulty of harvesting substantial quantities of seed, as reported by J. Anderson (1777), undoubtedly hindered greater use of the species. Vianne ( 1870) reported that birdsfoot trefoil was common in pastures in all parts of France, but it was not actually cultivated. Reports by de Rothschild ( 1 920) and Schribaux ( 1 922) show that some 50 years later the species was being cultivated in various areas of the country. In Italy, Orsi (1 953) reported that birdsfoot trefoil had been cultivated for a very long time but that the literature on the species was negligible. In 1949 Panikkar stated that birdsfoot trefoil was a valuable hay and pasture crop in certain areas of India. Bridsfoot and big trefoil have been grown in Australia and New Zealand. However, only big trefoil is commercially important (Levy, 1918; C. A. Gardnerand Elliot, 1945). Trefoil species introduced into South America from Europe have proved to be well adapted from southern Brazil across to Chile. A. L. Gardner and Alburquerque ( 1965) reported birdsfoot trefoil to be widely distributed in Uruguay; and Ortiz et ul. (1961) have reported trefoil to be well adapted to Southern Chile. In tropical environments, the trefoils, particularly big trefoil, grow well and are productive legumes for pasturage. Hasaka (1957) noted trefoil growing in Hawaii at elevations above 900 feet. Quiros (1946) reported that birdsfoot trefoil, probably imported in herbage seed from Europe a great many years ago, grows wild in the Highlands of Costa Rica. It is not known when or how trefoils were introduced into the United States. It is quite probable that seed entered as impurities in ballast deposits, and in shipments of other seed from Europe, and became established on the East and West Coasts. Trefoils first became naturalized in eastern New York, western Washington, western Oregon, and northwestern California. According to MacDonald ( 1946), two species of trefoil occurred in New York, a broadleaved species, L. corniculutus, found on upland soils, and a slender or narrowleaf species, L. tenuis, usually found on heavier bottomland soils. In the Northwest, Nelson (1917) reported the narrow-leaved species near Portland, Oregon. C. ECONOMICIMPORTANCE

Most of the acreage of birdsfoot trefoil for pasture and hay is located in the north central and northeastern states. Since 1957, extension agronomists in the United States have been making annual estimates of the acreage of the various varieties of birdsfoot trefoil. These estimates were

122

ROBERT R. SEANEY A N D PAUL R. HENSON

compiled by Z. M. Saunders, J. E. Jernigan, and J. R. Paulling, and published in Trends in Forage Crop Varieties. According to these reports, the total acreage in birdsfoot trefoil increased from 770,000 acres in 1957 to slightly over 2 million acres in 1967. Wedin et al. (1 967) estimate that there are 30 million acres of unproductive pasture in the midwestern states and 10 million in the northeast. On much of this acreage, birdsfoot trefoil would be well adapted as a permanent pasture legume. Research to reduce establishment hazards, to improve animal gains through management, and to develop improved varieties of birdsfoot trefoil should result in the continued rapid expansion of this important legume in permanent pastures. II. Morphology

A. ROOT Birdsfoot trefoil, L. corniculatus, has a long tap root with numerous lateral branches. Branches from the primary root are quite large in diameter, but secondary branches become smaller and form a thick fibrous root system, especially in the upper 1 to 2 feet of soil (MacDonald, 1946). The root system of trefoil is not as deep as that of alfalfa, but is more extensive in distribution in the upper soil. Comparisons by MacDonald (1946) have shown trefoil rooting to a depth of 3.5 feet, and alfalfa to 5.5 feet. Differences in rooting depth and distribution have been used to explain why trefoil is more persistent than alfalfa on shallow, poorly drained soils.

B. STEMA N D LEAF There is considerable variation in leaf and stem morphology within L. corniculatus. Size, shape, color, and pubescence of stems and leaves vary greatly among different genotypes. Growth habit of stems may be prostrate, ascending, or erect. I n general, stems of the erect types are smaller in diameter and less rigid than stems of alfalfa. Base of the stem is rounded while the vigorously growing portions are square (MacDonald, 1946). Branching always occurs at the leaf axis of main and secondary stems, and the amount and symmetry of branching varies. Under optimal environmental conditions, stems may reach 3-4 feet in length. Each leaf consists of 5 leaflets, 3 attached to the terminal end of the petiole and 2 at the base. Leaflets are typically obovate, although shape may vary from rounded to oblanceolate. Leaves are attached alternately on opposite sides of the stem. During darkness the leaflets close around

BIRDSFOOT TREFOIL

123

the petiole and stem, similar to the night closing of leaves in white clover (MacDonald, 1946).

C. FLOWER A N D POD The inflorescence of trefoil is a typical umbel consisting of 4 to 8 florets attached by short pedicels to a long peduncle. Each floret consists of a calyx with 5 united sepals and a typical legume corolla with 5 petals. Two petals are fused together to form the keel, which is enclosed by two wing petals and the standard. Petal color varies from a light to dark yellow, and may be tinted with faint orange or red stripes. Color at the keel tip can be yellow, brown, or red (Buzzell and Wilsie, 1963). Reproductive parts of the flower consist of 10 stamens and a simple pistil. One stamen is attached individually to the base of the flower, while the other 9 stamens have fused filaments which form a tube surrounding the ovary. The fused stamens are of different lengths, five long stamens alternating with four shorter ones (MacDonald, 1946). An average of five to six legumes or pods are borne at right angles to the tip of the peduncle, giving the appearance of a birds’s foot-thus, the common name birdsfoot trefoil. Pods are long, cylindrical, and brown to almost black. Winch (1958) found average pod dimensions of Viking variety to be 25 mm in length and 3 mm in diameter. Pods contain 15-20 seeds attached to the ventral suture, and at maturity, they split along both sutures and twist spirally to discharge seed. D. POLLINATION A N D SEED DEVELOPMENT Pollen is shed from the anthers of trefoil before the flower is fully opened. About the time the standard petal begins to expand and pull away from the wing petals, anthers dehisce and pollen appears in the tip of the keel. At the full flower stage, when standard is fully erect and expanded, the filaments have elongated and pushed pollen into the keel tip, covering the surface of the stigma (Giles, 1949). At this stage the presence of stigmatic membrane prevents pollen germination (Tome and Johnson, 1945; Giles, 1949). Manipulation of the pollinating mechanism must take place in order to break the stigmatic membrane. Once the film is ruptered the papillosse cells of the stigma secrete a fluid in which the pollen then germinates. Pollination takes place when insects force their way toward the base of the flowers. Continued pressure by the insect causes extrusion of pollen and stigma through the tip of the keel, thus transferring pollen from flower to insect and from insect to stigma. Pollen germination and pollen tube ingress into the stigma begin about one-half hour after

I24

ROBERT R. SEANEY A N D PAUL R. HENSON

pollination. Fertilization usually takes place within 24-48 hours after cross-pollination (Giles, 1949; Wojciechowska, 1963). L. corniculatus has 20-70 ovules per ovary, the average number being about 45. Usually only 40% of the ovules, or an average of about 20 ovules per ovary, develop into mature seed (Giles, 1949; Hansen, 1949; Bubar, 1958). Bubar (1958) found that ovules within an individual ovary vary considerably in rate of development. Because of this, ovaries contain some fertilizable ovules for 8-10 days, although individual ovules are fertilizable for only 2 or 3 days. After pollination, pods develop rapidly; they reach maximum length in about 3 weeks. The color of the pods changes from green to light brown, and finally to black. Seeds become physiologically mature slightly before, or at the time pods turn light brown. Wiggans et al. (1956) noted that seeds mature 7-10 days before seed pod dehiscence. Rate of pod development is influenced by weather conditions. s. R. Anderson ( I 9 5 3 , working in Iowa, found that mature pods and seeds formed 24-47 days after pollination, while in New York, Winch (1958) found that the same stage of development required 26-38 days. In a study of the morphological factors associated with pod dehiscence, Buckovic ( 1 952), suggested that the differential loss of moisture from exocarp and mesocarp tissue resulted in tension between fiber layers which causes separation and twisting of the two valves of the pod. Pods dehisced after losing 40-60% of their original moisture content. Buckovic (1952) found that the rate of moisture loss from the pod influenced the amount of dehiscence. Rapid loss of moisture caused a high incidence of dehiscence, while pods which dried slowly did not dehisce even though percent pod moisture was the same. Relative humidity indirectly influences the rate of pod shattering by changing the moisture content of the pod (Metcalfe et al., 1957). In the Midwest and Northeast, seed losses from pod shattering can be high when relative humidity drops below 40%. There is considerable variation in color and size of seed of L. corniculatus. Seed color can be olive green, brown, and sometimes almost black (J. B. Smith, 1956). The seed coat is frequently mottled with black spots, which appear as small dots on some seeds to large patches on others. Seeds of birdsfoot trefoil are very small. Measurements by J. B. Smith ( 1 956) show that the average dimensions of seed of Viking variety to be I .4 mm long, 1.2 mm wide, and 0.9 mm thick. R. McKee (1949) and MacDonald (1946) have counted from 375,000 to 420,000 seeds per pound. The seed coats of mature trefoil seeds are often impermeable to water.

BIRDSFOOT TREFOIL

125

Hard seeds will not germinate unless the seed coat is scarified or otherwise treated to allow rapid imbibition of water. In New York, Brown (1955) found that hand harvested seed contained an average of 90% hard seed. Amount of hard seed dropped to about 40% when the seed crop was harvested by combining. Ill. Physiology

A. PHOTOPERIOD

L. corniculatus is a long-day plant requiring a 16-hour day length for full flowering (Joffe, 1958). As day length is shortened to 15 hours, blooming is slightly retarded, and at 14-14.5 hours the development of floral premordia is so restricted that blooming is very sparse. Plants grown under short days have a prostrate rosette type of growth compared to the erect habit of plants grown under long days (G. W. McKee, 1963). Leaf:stem ratios are lower when plants are grown under low light intensities or short days (Rhykerd, 1959). Occasionally abortive buds appear under optimum photoperiod. Other factors such as temperature, nutrient level, and insects and diseases may also restrict development of floral premordia or cause abortion of flower buds. B. CYANOCENESIS Many species of Lotus contain cyanogenetic glucoside which can be hydrolyzed by enzymatic action to produce free hydrocyanic acid (Grant and Sidhu, 1967; Phillips, 1968). Most collections of L. cornicularus contain cyanogenetic glucoside and corresponding enzyme. Concentrations of hydrocyanic acid in the forage can range from 0.017 to 0.005% of green weight (MacDonald, 1946). In some populations of birdsfoot trefoil, a small percentage of plants are acyanogenetic. The authors (unpublished) found about 4% acyanogenetic plants in the variety Viking. We attempted to isolate an acyanogenetic strain of birdsfoot trefoil b i recurrent selection of HCN negative plants selected from the variety Viking. A population was obtained which contained about 95% acyanogenetic plants. The concentration of hydrocyanic acid in some species of Lotus is poisonous to animals (Gurney and White, I94 1; Henry, 1938). In studies of H C N toxicity of L. corniculatus, it was shown that sheep could be killed if large quantities of plant juice (200-300 ml) were introduced into the rumen (Dougherty and Christensen, 1953). However, under normal conditions of pasturing animals on birdsfoot trefoil, there have

126

ROBERT R. SEANEY A N D PAUL R. HENSON

been no authenticated reports or clinical evidence of hydrocyanic acid poisoning. C. SEEDLING GROWTH There is considerable difference in seedling vigor among varieties of L. corniculatus. Seeding plants of Empire type trefoils are, in general, less vigorous than the European types. In studies of seedling vigor of Empire and Viking, Shibles and MacDonald ( 1962) concluded that the greater seedling vigor of Viking was due to a greater rate of photosynthetic area production. Net photosynthetic rate per unit area of cotyledon or leaf surface was the same for both varieties. However, Cooper and Qualls ( 1 968) have shown that the explanation for differential growth rate between Viking and Empire depends on the temperature at which seedlings are grown and studied. At a temperature of 27"C, Viking had a greater leaf area ratio (LAR) than Empire, but a similar net assimilation rate (NAR). These results are similar to those of Shibles and MacDonald ( 1 962). However, at a temperature of 2 1"C, the superiority of Viking over Empire was related to a greater net assimilation rate. Several studies have shown that variations in light intensities affect the amount of dry matter which is partitioned into leaves, stems, and roots. Rhykerd et al. (1959) noted that the leaf-stem ratio of birdsfoot trefoil was low at low light intensities but increased with increasing light intensities. This was in contrast to alfalfa, Medicago sativa L., and red clover, Trifolium pratense L., which had high 1eaf:stem ratios at lower light intensity and lower leafistem ratios at high light intensity. G. W. McKee ( 1962) also found that leaf area per plant of birdsfoot trefoil decreased under conditions of moderate shading. Both top and root growth of birdsfoot trefoil, as well as alfalfa and red clover, are inhibited by low light intensity (Gist and Mott, 1957). Cooper ( 1 966,1967) pointed out that trefoil is not less tolerant to shading than other legumes. He found that shading by companion crops reduced growth of birdsfoot trefoil and alfalfa proportionately. However, the effects of shading make trefoil more susceptible than alfalfa to competition and stress within the environment. The fact that seedlings of birdsfoot trefoil are smaller than those of alfalfa, may allow early competition to affect survival of trefoil more than alfalfa. Cooper and Qualls ( 1968) suggest that seedling vigor of different varieties or strains of trefoil can be predicted by measurement of speed of germination and speed of elongation of seedling plants. Such measurements were made during the early nonphotosynthetic stage of develop-

BIRDSFOOT TREFOIL

I27

ment. In comparisons of the varieties Leo, Tanna, Viking, and Empire, speed of germination and elongation values were related to the seedling vigor and yield of the variety when grown in the field. Speed of germination and elongation, and seedling vigor were significantly greater for Leo than for Tanna, Viking, or Empire.

D. VEGETATIVE PROPAGATION Trefoil plants can be propagated vegetatively by either stem or root cuttings. Stem cuttings which include a single node develop shoots from the axillary bud, and roots from the callus tissue of the internode area. Ostazeski and Henson ( I 965) found that type of cutting can influence the number of stems and vigor of the plant developing from the cutting. They compared cuttings made with one node with cuttings having two nodes, the lower node was stripped of leaves and placed in the rooting medium. Cuttings with two nodes developed shoots from the upper and underground nodes and produced plants that were more vigorous than those developing from single node cuttings. These studies indicate the importance of using cuttings made in a similar manner when evaluating performance of individual plants. Midgley and Gershoy (1946) used root cuttings to propagate trefoil plants. Segments of roots taken 3-6 inches below the crown developed multiple shoots and roots from callus tissue at both ends of the root segment. IV. Culture

A. ADAPTATION

In Europe, birdsfoot trefoil is more productive than most legumes on soils that are imperfectly drained, infertile, or droughty (Robinson, 1934). Birdsfoot trefoil in Uruguay is widely distributed and well adapted on soils too poor for alfalfa (A. L. Gardner and Alburquerque, 1965). Ortiz et al. (1961) reported that in Chile birdsfoot trefoil tolerates acid soils better than alfalfa and red clover. In North America, birdsfoot trefoil is adapted from the Atlantic Coast west to eastern Kansas, Nebraska, and the Dakotas in the United States, and the southern part of the Province of Ontario in Canada (Kingsbury and Winch, 1967). The southern limit of adaptation includes the higher lands in western North Carolina (Chamblee, 1960) and Tennessee (Fribourg, 1963), and northern Arkansas (Offutt, 1967). Narrowleaf birdsfoot trefoil, L. tenuis, is an important constituent in pastures on the heavy soils in the Hudson River Valley of New York

128

ROBERT R. SEANEY A N D PAUL R. HENSON

(MacDonald, 1946). In California, Peterson et al. ( 1 953) reported that narrow-leaf trefoil is grown throughout the state, but that it is used most extensively on clay soils difficult to drain after irrigation. Big trefoil, L. peduncularus, according to Howell ( 1 948), is well adapted to the acid coastal soils of the Pacific Northwest. I t grows well on low-lying soils that are frequently flooded during the winter months. Attempts to grow big trefoil in the Southeast have not been completely successful, because of its susceptibility to diseases, primarily Rhizoctonia. In Florida, Wallace and Killenger ( 1 952) reported big trefoil to be of value for forage in the low-lying flatwood areas of the State.

B. SOILSAND SOILFERTILITY Early reports on birdsfoot trefoil, based largely on observations, indicated that the species would grow well on infertile, droughty, or poorly drained soils. Later reports based on fertilizer trials indicate that trefoil does respond to improvements in soil fertility (MacDonald, 1946). In New York, MacDonald ( 1 946) studied the effect of various fertilizers and lime on the growth of birdsfoot trefoil. He concluded that, while birdsfoot trefoil tends to produce better than other legumes on normally poor soils, the soil-fertility requirements of trefoil did not differ from those of other commonly used legumes. In Vermont, Varney (1958) reported a good response to phosphorus and potash on clay and sandy loam soils. However, after 4 years there was considerable loss of stand on the fine, sandy loam soils of Vermont. The 10-year-old stand on heavy clay was as good, if not better, than the initial stands. In Indiana, Foy‘ and Barber (1961) conducted studies to isolate soil properties responsible for yield differences observed between 20 soil types. The yields obtained in the above studies, conducted in the greenhouse at a high fertility level, agreed in general with field response of trefoil on the different soil types. The multiple correlation of phosphorus, organic matter, and aggregation index on yield accounted for 58% of the variation in yield between the 20 soils. In Iowa, Hughes ( 1 962) reported good production of birdsfoot trefoil on eroded (pH 5.6) Shelby soil with adequate phosphate fertilization. He concluded, however, that “liming and fertilization in keeping with soil tests is advised.” In Maryland, Hunt and Wagoner (1963) reported a significant increase in percentage of birdsfoot trefoil, in association with orchardgrass, from potash in combination with low and high lime applications on an acid (pH 4.9) Codorus silt loam. The percentage of birdsfoot trefoil increased with applications of 83 lb/A of potash at both lime levels. Stands were

BIRDSFOOT TREFOIL

129

reduced on high lime and high potash (249 Ib/A) applications, possibly due to greater competition from orchardgrass. It is apparent that birdsfoot trefoil, where adapted, needs adequate amounts of phosphate, potash, and lime for satisfactory growth. Stands are generally longer-lived and more productive on clay soils. C. INOCULATION

Seed of all trefoil species should be inoculated before planting. In many areas, the proper rhizobium bacteria are not endemic to the soil or present in sufficient number to give adequate inoculation. A special group of rhizobium bacteria are required to give effective inoculation of trefoil. Bacteria that inoculate other legumes, such as alfalfa or clover, will not inoculate trefoil species. Within the genus Lotus, strains of bacteria may be infective and effective on one species but not another. Erdman and Means (1949) isolated strains of bacteria that were both infective and effective on L. corniculutus and L. tenuis, but were only infective on L. uliginosus (L. pedunculatus). Several factors have been found to influence inoculation. G. W. McKee ( 1962) noted that all degrees of shading and plant competition depressed seeding growth and inoculation of birdsfoot trefoil. Seedlings of trefoil were not adequately inoculated when grown in association with oats, weeds, or S-37 orchardgrass. Functional inoculation in the above studies was achieved several weeks after the competing crop was removed. G. W. McKee ( 1 962) suggested that Empire and Viking need 50% of natural daylight to be nodulated adequately. Trefoil seedlings can nodulate in soils with pH values of 4.5-7.9 (G. W. McKee, 1961). However, best nodulation has been obtained at pH 6.06.5 (J. H. Smith, 1955). Maximum effectiveness of rhizobium bacteria also depends on an adequate supply of soil nutrients (Lynch and Sears, 195 1). Alexander and Chamblee (1965) have studied the effects of sunlight and drying on the nodulation of trefoil. Exposure of inoculated seed to sunlight for 86 hours resulted in a less effective nodulation. A similar effect was noted after the seed was held for two to three weeks in a dry seed bed.

D. ESTABLISHMENT Seedling plants of birdsfoot trefoil are generally lacking in vigor when compared to alfalfa and red clover, and stands may be lost due to shading or competition from other species and weeds. However, good stands of trefoil can be obtained if special care is taken with respect to seedbed

130

ROBERT R. SEANEY A N D PAUL R. HENSON

preparation, date, rate, and depth of seeding, grass associations, and weed control. In the Northern States, seedings should be made in early spring (Midgley and Stone, 1958; MacDonald, 1946). At more Southern locations, August seedings have been successful, as reported in New Jersey by Ahlgren et al. (1953, in southern Illinois by Pierre and Jackobs (1954), and in western North Carolina by Chamblee ( I 960). Fall or February seedings are recommended in California (Peterson et al., 1953). When trefoil is sown in the spring, the presence of significant numbers of hard seed can result in poor stands. In Vermont, Wood ( 1 958) reported better stands of trefoil when using scarified seed. After fall seedings in Vermont, hard seed germinated the following spring, resulting in satisfactory stands (Midgley and Stone, 1958). In studies of spring seedings, Brown (1955), found only that only a small fraction of hard seed germinated during the seeding year. The next year, only 25% of the hard seed germinated and only 0-25% of these seedlings survived. Tests on rate of seeding birdsfoot trefoil have shown that, while greater numbers of seedlings will establish at high seeding rates, yields of herbage have not been significantly increased by seeding more than 5 or 6 Ib/A (MacDonald, 1946; Pierre and Jackobs, 1954; Wakefield and Skaland, 1965). Drilling seed over banded fertilizer placed 1 M to 1 W inches below the seed has increased the percent emergence and vigor of seedlings (Tesar et al., 1954; Briggs, 1953). However, in studies in New Jersey, banding fertilizer under the seed did not significantly increase establishment of trefoil seedlings (Duell, 1964). Response to band seeding often depends on nutrient level in soil, and environmental conditions during establishment. Since trefoil seed is small and growth rate of seedling is slow, depth of seeding is extremely important. Under ideal greenhouse conditions, MacDonald (1946) noted the percent emergence when planted at %, W ,and 1 inch was 90, 53, and 1 1 , respectively. Under field conditions on a loam soil in New Jersey, Ahlgren et al. ( 1955) reported that at depths of M , W ,and 1 inch, the emergence was 38,3 I , and 20% respectively. A field study in Kansas indicated generally good emergence at ?4-inch planting depth with little emergence at the I-inch depth of planting (Stickler and Wassom, 1963). Similar results were obtained by the authors in a study at Beltsville, Maryland (unpublished), where seedlings were made in 1963 on a silt loam soil. The average percentage of seedling emergence from seed of Empire and Viking, planted at M - , M-,N-, and I-inch depths, was 3 1 , 18, 8, and 3%, respectively. At all depths except the last,

BIRDSFOOT TREFOIL

131

the emergence of Viking was significantly greater than that of Empire. Birdsfoot trefoil should be seeded on a firm, smooth seedbed and no deeper than f-i inch. E. COMPANION GRASSES Birdsfoot trefoil may be seeded alone or in association with a grass. When used for grazing, trefoil is generally seeded with a grass. Simple mixtures of grass with trefoil is preferred to more complex mixtures. Moderate seeding rates of a single grass species result in less competition to trefoil during establishment. Good stands of birdsfoot trefoil are generally secured in mixtures with timothy (Chevrette et al., 1960). In a comparison of birdsfoot in mixtures with timothy, bromegrass, and orchardgrass in the Northeastern states, Ronningen et al. ( 1 954) reported better establishment with timothy and bromegrass. In Ohio (Parsons and Davis, 1964), good birdsfoot-grass mixtures were maintained through four harvest years, 1955- 1958, with Kentucky bluegrass, timothy, bromegrass, and orchardgrass. Low-producing pastures of Kentucky bluegrass can be greatly improved by renovation and establishment of birdsfoot trefoil. Williams (1953) found that establishment of satisfactory stands of trefoil on old pasture sites depended on the destruction of the sod and the preparation of a good seedbed. Plowing followed by harrowing resulted in the best stands of birdsfoot. In Maryland, Decker et al. (1 964) obtained excellent stands of birdsfoot trefoil in an old Kentucky bluegrass sward with specially designed sod-seeder openers equipped with fertilizer and seeding attachments.

F. WEEDCONTROL Effective control of weed competition during seedling growth is one of the most important factors in assuring successful stand establishment. Under most environments, birdsfoot trefoil can not effectively compete with fast growing weed species. Control of weeds during establishment often results in larger trefoil plants, more plants per unit area, and higher forage yields (Wakefield and Skaland, 1965; Scholl and Brunk, 1962). Companion crops, such as oats, have been used to suppress weed competition in trefoil seedings. However, competitive effects of the oats on trefoil are often detrimental to seedling growth and persistence (Scholl and Brunk, 1962). The desirability of using a companion crop for establishing trefoil depends on the particular environment under question and the economics of the crops being grown. Clipping or mowing has also been used to control excessive weed growth. Effectiveness of this

132

ROBERT R. SEANEY AND PAUL R. HENSON

treatment depends on the weed species involved and the relative stage of growth of the trefoil and weeds. In some cases, suppression of broadleaved weed species, by mowing, allows weed grasses, such as foxtail or fall panicum, to become more detrimental than the broad-leaved weeds (Kerr and Klingman, 1960). Use of chemicals offers the best method of removing weed competition from new seedings of trefoil. Several herbicides are available for control of both broad-leaved weeds and weed grasses. The application of the amine salt of 4-(2,4-dichlorophenoxy)butyric acid (2,4-DB) at the time weed seedlings are 1 to 2 inches in height will control many of the annual broad-leaved weeds such as lambsquarter, mustard, and pigweed. Under certain conditions, particularly when growth rate is slow, the trefoil seedlings will show some symptoms of injury. However, seedlings soon outgrow these effects. Postemergence applications of 2-sec-butyl-4,6-dinitrophenol (dinoseb) can also be used for control of annual broadleaved weeds in birdsfoot trefoil. Since trefoil seedlings are more susceptible to dinoseb injury than alfalfa, lower rates of dinoseb are recommended for use in trefoil (Linscott and Hagin, 1968). S-Ethyl dipropylthicarbamate (EPTC) incorporated into the surface 1 or 2 inches of soil before planting, controls some of the annual weeds and nearly all annual grasses (Linscott and Hagin, 1967). EPTC is very effective for control of yellow nutsedge. Use of this chemical means that trefoil must be seeded without companion grasses. Properly used, EPTC has little or no effect on emerging trefoil seedlings. Postemergence application of 2,2-dichloropropionic acid (dalapon) to new seedings gives good control of annual grasses such as foxtail, barnyard grass, and witchgrass (Scholl and Staniforth, 1957). Higher rates of dalapon have been used to suppress perennial grass species in fields used for seed production. In New York and Vermont, early spring applications of dalapon have increased Viking and Empire seed yields (Flanagan and MacCollom, 1964). The above chemicals are examples of selective herbicides which control grasses or broad-leaved weeds but do not significantly injure trefoil seedlings. These herbicides can be used singly or in combination. Combination treatments such as 2,4-DB plus dalapon, or EPTC plus 2,4-DB, or EPTC plus dinoseb, offer opportunity for control of both grasses and broad-leaved weeds in new seedling of trefoil (Peters, 1964; Linscott et al., 1967; Linscott and Hagin, 1968). Compared to other legumes such as red clover, white clover, or alfalfa, birdsfoot trefoil is more tolerant to herbicides such as 2,4-DB and dalapon. Studies by Fertig ( 1961) indicated that trefoil recovers more

BIRDSFOOT TREFOIL

133

rapdily than alfalfa after treatments of 2,4-DB where some stunting of seedling growth occurred. When dalapon is applied to trefoil containing volunteer red, alsike, or white clover, the trefoil is little affected but the clovers are severely injured. Varieties of trefoil differ in tolerance to herbicides. Empire has been found to be more tolerant than Viking to either fall or spring applications of 2,4-DB, (2,4-dichlorophenoxy) acetic acid (2,4-D), or 2-(2,4,5-trichlorophenoxy)propionic acid (silvex) (Fertig et al., 1960). These comparisons were made on established stands of trefoil being used for seed production. Tolerance or susceptibility of 2,4-D is genetically controlled. By using a program of phenotypic recurrent selection, we have been able to obtain a strain of L. corniculatus which is significantly more tolerant to 2,4-D than the variety Viking (unpublished). Breeding for tolerance to a specific herbicide may be a way of increasing herbicide selectivity.

G. DISEASES' Crown and root rot diseases cause significant economic losses of birdsfoot trefoil. Surveys indicate that crown and root rot fungi are prevalent in most areas of use and that these pathogens can lower forage yields and quality and also reduce stands (Drake, 1958; Miller et al., 1964). Losses from root rots are greater in the South than in the Northeast or North Central regions of the United States. N o single pathogen is primarily responsible for root rot. Rather the disease is associated with a parasite-saprophyte complex which varies widely in different environments. Fungi which have been associated with the crown and root rot complex are species of Fusarium, Verticillium spp., Macrophomina phaseoli, Mycoleptodiscus terrestris, and M . sphaericus (Kainski, 1959; Ostazeski, 1967). Though Rhizoctonia solani and Sclerotinia trifoliorum are themselves causes of specific trefoil diseases, they, too, are sometimes isolated as part of the complex. First symptoms of severe infection include the failure of plants to resume growth after harvest. In some cases, a slow regrowth may begin, but permanent wilting occurs before shoots are 3-4 inches high. Upper tap roots of such plants usually have an extensive central decay, leaving only a thin shell of functional tissue around the perimeter. The color of the infected tissue varies from cork-colored tan to a reddish-brown or

' Prepared by S. A. Ostazeski, Crops Research Division, ARS, USDA.

134

ROBERT R. SEANEY A N D PAUL R. HENSON

black. Lateral roots of these plants are easily broken or shredded, although signs of fungal colonization may not be apparent. When Mycoleptodiscus terrestris is part of the complex, the advancing margin of decay is frequently a dark reddish-brown to black. M. terrestris is an aggressive fungus with the ability to compete with other species to colonize root tissues (Pettit et al., 1966). Experiments have shown that spores of M. terrestris can germinate and infect leaves and stems of birdsfoot trefoil. Spots on leaves are small, reddish brown with an indistinct chlorotic halo. The leaf and stem infections associated with M. terrestris, if they occur naturally, could be important in disseminating the disease (McVey and Gerdemann, 1960). A related fungus, M. sphaericus also attacks the roots of birdsfoot trefoil. This fungus produces smaller uniformly round black sclerotia in invaded tissues (Ostazaeski, 1967). Sclerotinia trifoliorum causes a watery, soft rot of the lower stems, crown, and upper tap root. Damage usually occurs in later winter and early spring and is often associated with a heavy snow cover. When infected, the plant is usually killed and tufts of white cottony mycelium can be found attached to the lower stems or near.the crown (Kreitlow, 1962). Sclerotium rolfsii, the cause of southern blight, attacks the roots, crown, and stems near the soil surface, causing rapid death of the aboveground parts. During warm rainy weather, a cottony mycelium frequently covers the lower stems of infected plants. Since this disease occurs infrequently, it is of minor importance on birdsfoot trefoil. Although Rhizoctonia solani has been found in diseased crowns and roots of birdsfoot trefoil, it is not considered a primary agent of root rot. However, it does cause a foliar blight and can be very destructive in dense stands, particularly during hot, humid weather. The fungus attacks lower leaves first and rapidly moves throughout the crown and foliage. Blighted leaves turn gray and then brown, and are often plastered to the stems by a weblike growth of tan to brown mycelium. Damage occurs in patches, and sometimes up to 90% of the affected plants are killed. The disease may be partially controlled by harvesting at the first sign of infection (Kreitlow, 1962). Leaf spot and stem canker caused by Stemphylium loti is probably the most widespread foliage disease of birdsfoot trefoil. The fungus attacks the aerial portions of the plant, causing premature leaf drop. Usually one lesion is sufficient to cause the leaflet to drop prematurely. Leaf lesions are round, reddish brown, and sunken. They enlarge to a diameter of about 5 mm, darkening and forming a concentric zonate pattern. Stem

BIRDSFOOT. TREFOIL

135

cankers are copper-colored, ranging from dot size to elongate lesions about 25 mm long. Young stems may be girdled and killed. The fungus can also attack immature seed pods and cause shriveling and discoloration of seed (Graham, 1953). Phomopsis blight caused by Phomopsis loti is a recently described disease of birdsfoot trefoil (Upadhyay, 1966). It has been most important in the Midwestern states, but also has been observed in Virginia and New York. Stem lesions are small with a light to dark brown border and gray to tan center. An individual lesion or a group of lesions can kill stems by partial or complete girdling. Fusarium roseum has been reported in New York to attack stems and leaves under conditions of high relative humidity and high temperature (Ford, 1959). Infection causes the bleaching and drooping of the stem tips of both young seedlings and older plants. This disease occurs only sporadically, and it is of minor importance. Virus infections in birdsfoot trefoil are not known to cause serious damage. However, curly top virus and tobacco ring spot virus have been found in species of Lotus (Klostermeyer and Menzies, 195 1 ; Ostazeski, 1965). These viruses may play a role in stand persistence and possibly could predispose plants to attack by other pathogens. Many species and varieties of trefoil have been tested for resistence or tolerance to specific disease organisms. Certain introductions of L. corniculatus have some resistance to Stemphylium loti (Ford, 1960; Drake, 1962), or Sclerotinia frifoliorum (Barr and Callen, 1963). In the case of foliar blight caused by Phomopsis loti certain introductions have very significant differences in susceptibility (Upadhyay, 1965). Studies such as these suggest that existing genetic variation for resistance to specific diseases could be used in efforts to develop disease-resistant varieties.

H. INSECTS Several insects attack birdsfoot trefoil and cause losses of both forage and seed (Neunzig and Gyrisco, 1955). In central and eastern United States meadow spittlebug, Philaenus spumarius, is an important insect pest. The nymph produces a characteristic spittle or white foamy mass on stems and leaves as it feeds by sucking plant sap. Feeding by this insect causes general stunting of the plant and abortion of flower buds. The alfalfa plantbug, Adelphocoris lineolatus, and potato leafhopper, Empoasca fabae, are also sucking insects that can cause severe injury to birdsfoot trefoil. The alfalfa plantbug is particularly destructive because of its abundance and ability to destroy stem terminals and

136

ROBERT R. SEANEY A N D PAUL R. HENSON

flowers. Infestations of potato leafhopper can reduce stands of new seedings as well as lower forage yields and quality (Mathur and Pienkowski, 1967). Injury is characterized by a general stunting of the plant and change of leaf color to yellow or shades of red and purple. The trefoil seed chalcid, Bruchophagus gibbus, a small black wasplike insect, lays its eggs in developing seed pods 5-10 days after pollination (Neunzig, 1957). The larvae are the destructive stage .of this insect for they feed within the maturing ovule, leaving only a hollow inviable seed. Seed samples from various areas indicate that up to 50% of the seed can be destroyed by the chalcid larvae. In fields where considerable seed is left on the ground, infestation the following year can be quite severe. Thus, proper management of seed harvest is one way of preventing build up of this insect. Present insecticides do not offer adequate control. Recommendations for control of insect pests of trefoil vary from region to region. In most areas, proper applications of insecticides have been profitable in preventing loss of forage and seed production (Bader and Anderson, 1962a; Ridgeway and Gyrisco, 1959). I. SEEDPRODUCTION In 1969, certified seed of birdsfoot trefoil was produced in New York, Vermont, California, Minnesota, Iowa, and Missouri. About 85% of the total certified acres in the United States were located in New York and Vermont, and the varieties Empire and Viking were grown on 85% of the total acreage. Although exact statistics are not available on amounts of seed produced in each state, it is clear that, at present, major production of seed in the United States is centered in New York and Vermont. Canada also produces large quantities of trefoil seed. About 4 000 acres were certified in 1969. This was about equal to the total acreage certified in the United States. Empire has been the main variety grown for certification in Canada. Surveys of wholesale seed shipped to retailers in 12 Northeastern states indicate trends in use of forage species and varieties (Pardee, 1968). In 1960, these surveys indicated 900,000 pounds of trefoil seed moved to retail outlets in the Northeast. This amount gradually declined to 608,000 pounds in 1968. In 1960,24% of the trefoil seed supplied to retailers was Empire and 33% Viking. By 1968, these proportions had shifted to 5 1 % Viking and 19% Empire. Since the early 1950s, a significant proportion of the seed used in the United States has been imported from Europe, principally France and

BIRDSFOOT TREFOIL

137

Italy. In the period 1964 to 1968, an average of 500,000 pounds has been imported annually. Under optimal environmental conditions, trefoil plants have the capacity to produce large quantities of seed. Estimates of potential yields range from 600 to 1000 pounds of seed per acre. However, because of problems in seed harvest caused by pod dehiscence and indeterminant flowering, the amounts of seed harvested usually represent only a small portion of the amounts of seed produced. Seed yields in the North Central and Northeastern United States have been as high as 200-500 pounds per acre, with an average production in the range of 50-150 pounds per acre (Winch, 1958). Highest yields in California have reached 450 pounds per acre (Peterson et al., 1953). Year to year differences in seed set, and difficulty in harvest, often cause wide differences in average annual seed yields. Seed set in trefoil depends on pollination of flowers by insects, primarily by various species of Hymenoprera. Both pollen and nectar collecting honey bees are capable of tripping the pollinating mechanism (Bader and Anderson, 1962a). Adequate numbers of bees are necessary for pollination of all flowers. Morse (1955) has shown that, in New York, honey bee populations averaging one bee per square yard are sufficient to pollinate all flowers. As a general practice, seed growers import bees into seed production areas to ensure maximum pollination and seed set. Studies comparing seed yields of clear stands of trefoil and trefoilgrass mixtures have been conducted in several areas of the United States. In New York, Winch (1958) found that seed yields were reduced when trefoil was grown in mixtures with bromegrass, orchardgrass, tall fescue, or timothy. Lower seed yields were noted as density and height of the associated grass increased. Clear stands of trefoil seeded without grasses usually gave bette; seed yields and seed of higher purity. MacDonald ( 1946) found that lodging of trefoil plant before or during flowering significantly reduced seed yields. In Iowa, S. R. Anderson and Metcalfe ( 1 957) noted reduced lodging and increased seed yields when trefoil was grown with Kentucky bluegrass, orchardgrass, or timothy. Highest yields were obtained in the trefoil-Kentucky bluegrass mixture. Such studies comparing clear stands and trefoil-grass mixtures indicate that clear stands have the potential for highest seed yields. However, trefoil-grass mixtures often have higher seed yields because of less lodging. All varieties of L. corniculatus flower and set seed over an extended period of time, usually 2 to 3 weeks or longer. Individual plants will have both ripe pods and new flowers, and some pods will dehisce before

138

ROBERT R. SEANEY A N D PAUL R. HENSON

flowering and pod set are completed. Because of this indeterminate type of flowering and seed development, very critical timing of harvest is required in order to obtain maximum seed yields. Cutting too early results in seed loss because of the harvest of many immature and inviable seed. Cutting too late results in seed loss from pod dehiscence or shattering. Winch and MacDonald (1 96 1 ) recommended that harvest should begin when 70-80% of the pods are mature, i.e., when light brown to brown. S. R. Anderson (1955) suggested slightly earlier harvest, “when maximum number of pods are light green to light brown.” Several methods have been used to harvest seed. Probably most common in the Northeast is to mow, windrow, and then combine directly from the windrow. The crop is usually windrowed at the time of mowing with a windrower attached to the mower bar. The windrow is allowed to dry before combining. A second method involves harvesting and storing the crop like hay, and then threshing. Tests by MacDonald (1 957) have shown that this method results in excessive loss of seed. Direct combining of trefoil has been used to some extent, but in general, it is not an acceptable method of harvesting trefoil seed. This method has the disadvantage of running large quantities of green forage through the combine, which slows harvest and causes seed loss by clogging. Chemical defoliants and desiccants have been successfully used to defoliate and dry the plant before harvest (Jones, 1952; Cooper and Corms, 1952; Wiggans e? al., 1956). Such treatments make direct combining faster and significantly reduce the seed loss which occurs when combining green material. However, at present, desiccants and defoliants are not extensively used in the harvest of trefoil seed. In order to extend the period of seed harvest, flowering and seed set of trefoil can be delayed by clipping in the spring. In general, studies indicate that clipping at bud stage or at early flowering delays seed maturity from 5 to 10 days. Winch (1958) clipped trefoil at early bud and delayed seed harvest of Empire 4 days and Viking 7 days. Seed yields were not influenced by these early spring clipping treatments. However, later clipping treatments lowered seed yields and delayed harvest. Other studies also indicate that late clipping in June or during flowering reduces seed set (Bader and Anderson, 1962b). Insects can significantly reduce seed yields (Bader and Anderson, 1962b; Guppy, 1958). MacCollom (1 958), working in seed production fields in Vermont, found that early season application of insecticides increased seed yields of both Viking and Empire. In treated areas, there were more flowers and flowering occurred over a shorter period of time than in the untreated areas. In the Northeast, applications of insecticides

BIRDSFOOT TREFOIL

I39

have proved profitable in controlling insects in seed production fields. Seed yields can be increased by controlling weeds in established stands of trefoil. I n Vermont and New York, early spring application of dalapon has reduced competition from perennial grasses and resulted in significant increase in seed yield (Flanagan, 1961; Flanagan and MacCollom, 1964). Although several herbicides have been used to control broad-leaved weeds in established stands of trefoil, there is little evidence of increased seed production by use of such treatments (Fertig et al., 1960; Henson and Schoth, 1962). Winch et al. ( 1968) have suggested a preventive weed control program for seed production fields. The program starts with establishment, using a combination of 2,4-DB and dalapon, followed by yearly treatments with 2-chloro-4,6-bis(ethylamino)-s-triazine(simazine), starting in the late fall of the seeding year. Winch et al. found that fall applications of simazine for each of 3 years did not result in a completely grass free stand, but did reduce weed competition and increase seed yields. V.

A.

Utilization

HAYA N D SILAGE

Properly cured hay of birdsfoot trefoil is fine-stemmed, leafy, and nutritious, and is comparable in quality to other good legume hays. The more erect growing varieties of European origin, e.g., Viking, Mansfield, Cascade, and Granger, are more productive than Empire or Dawn, except in those regions where winterkilling is a problem. These latter varieties are semierect in growth habit and are winter hardy. Comparative yields of Viking and Empire in New York, as reported by MacDonald ( 1957), show consistently higher hay yields for Viking over Empire. Tests were conducted for four harvest years each at five locations in New York State. I n these tests alfalfa and Viking yielded the same, however, soils at two locations were unsuited to alfalfa. Birdsfoot trefoil is usually cut for hay as the plants come into bloom. I n New Jersey, Duel1 and Gausman (1957) reported maximum hay yields of imported European trefoil from cuttings made at the I/lO-bloom stage. The percentage of protein declined with maturity but maintained relatively high levels. Pierre and Jackobs (1953), in Illinois, reported higher 2-year average yields from cuttings made in the full-bloom stage for the Empire and Italian imported varieties. They found that percent of trefoil and total root weight of both varieties was equally good following harvests at pre-bloom, I / 10-bloom, and full-bloom. The percentage of protein of Empire hay remained high for all managements. The average

140

ROBERT R. SEANEY A N D PAUL R. HENSON

protein content of Italian declined from a high of 2 1.4% in hay cut at l/lO-bloom to 17.4% in hay cut at seed pod stage. In Ohio, Parsons and Davis (1964) concluded after extensive tests that trefoil should be managed to avoid very early or late fall harvests. They found that a mixture of imported European trefoil and orchardgrass persisted under a 4-cutting hay management system. Protein production tended to be constant for managements having a common time interval between cuttings. Managements having an 8-week interval between cuttings produced more protein than those cut at 6-week intervals. Unfavorable weather during harvest of trefoil hay can greatly increase dry matter losses. In a New York study comparing leaf losses of different legumes harvested for hay, MacDonald (1 946) reported losses of 17, 19, 27, and 32% for alfalfa, red clover, Empire birdsfoot trefoil, and European birdsfoot trefoil, respectively. When rain occurred 6 hours after cutting, leaf losses increased and a longer time was required for curing the hay. In another New York study, Trimberger et al. (1 962) reported dry matter loss of 31% for Viking and 40% for Empire because of rains after cutting. Birdsfoot trefoil in pure stands will frequently lodge as it approaches maturity. Associations of timothy and orchardgrass with birdsfoot trefoil will reduce lodging and field losses in harvesting and curing the hay. Good silage can be made of birdsfoot trefoil. Allred (1955) obtained satisfactory silage from a second cutting of birdsfoot trefoil containing 18% dry matter. This preliminary study was conducted in laboratory silos of 100-pound capacity. Wittwer et al. (1958) conducted silage studies to evaluate unwilted birdsfoot trefoil with unwilted and wilted red clover. All silage rated “very good” to “excellent,” and had an exceptionally good olive green color. The clover silages had a clean acid odor, while the odor of the birdsfoot trefoil silage was clean but very bland. All the silages were satisfactory with respect to milk production and maintenance of body weight. Results of the feeding trials indicated no significant differences in milk production, dry-matter consumption, and body weight changes.

B. PASTURE Birdsfoot trefoil is an excellent legume for increasing production of permanent pastures in the North Central and Northeastern states. In Indiana, Mott et al. (1 953) reported that beef production from Empire birdsfoot trefoil-bluegrass pasture under rotational grazing was 57 % greater than a similar pasture without trefoil. Annual applications of 120 pounds of N per acre to all grass pastures were not as productive as tre-

BIRDSFOOT TREFOIL

141

foil grass, but were significantly more so than the grass pastures without nitrogen. In southeast Iowa, Scholl and Hughes (1955) reported beef production on pastures renovated and seeded to Empire trefoil was increased 2 % times over that on unimproved pastures. These pasture soils were of low fertility and not suited for cropping. In a later study in southern Iowa, Wedin et al. (1967) reported similar beef gains on pastures renovated and seeded to Empire birdsfoot trefoil. The three pasture treatments evaluated in this experiment were: (1) unimproved; (2) fertilized with N and P; and (3) renovated trefoil-grass. The average beef gains for the 6 years (1 958- 1963) were 157,250, and 423 kg/ha for treatments 1,2, and 3, respectively. Differences in production were all highly significant. Pastures were grazed continuously using tester animals to determine average daily gains, and other steers were added or removed to provide uniform grazing pressure on all pastures. In Ohio, Davis and Klosterman (1959) reported a gain in TDN of 4 1 % for European birdsfoot trefoil-grass pastures over fertilized grass pasture under rotational grazing. In a later Ohio study, Van Keuren et al. 1969) determined the productivity in terms of animal returns of the Empire and Viking varieties under rotational and continuous grazing with sheep and cattle. During the 6-year period of study (1 958- 1963), animal liveweight gains were greater from the 3-paddock rotational grazing system than from continuous grazing. The animal gains from grazing the more prostrate Empire variety were greater than gains from the Viking variety for both sheep and cattle. Van Keuren and Davis (1 968) reported that both varieties persisted better under rotation than under continuous grazing. However, Empire persisted better than Viking under each grazing system by both sheep and cattle. Empire trefoil continued to be a significant component of the pasture for the entire 6 years, whereas the Viking became marginal in value after 3 years. In a clipping-grazing experiment in Minnesota, mixtures containing alfalfa were more productive than Empire trefoil alone or with grass, based on yields of harvested forage (Yawalker and Schmid, 1954). In general, the percentage of Empire birdsfoot trefoil in mixtures with the more aggressive legumes, alfalfa and red clover, was too low to greatly affect production. In general, the results of renovating and seeding birdsfoot trefoil in unimproved permanent pastures has greatly improved animal gains. A secondary advantage for the use of birdsfoot in permanent pastures has been the nonbloating characteristic of the species when grazed. The amount of stable foam produced from green legumes varies with species of legumes, and is correlated with bloat-inducing potential of the species

142

ROBERT R. SEANEY A N D PAUL R. HENSON

(Mangan, 1959). The foam production of samples of various legumes has been determined by Pressey et al. (1963), Kendall (1 964), and Cooper et al. (1966). Foam production of red and white clover was very high, but birdsfoot trefoil produced less than half as much as the other species. Kendall (1964) suggests that the foam production in nonbloating legumes may be limited by tannin in trefoil and pH of the rumen fluids. McArthur and Miltimore (1966) attribute bloat to foaming caused by a protein whose sedimentation velocity is 18 Svedberg units (S). Forages that induced bloat contained 4.5% of 18 S protein. They reported the content of 18 S protein in alfalfa, red clover, and birdsfoot trefoil to be 5.3,3.9, and O S % , respectively. All evidence indicates birdsfoot trefoil to be a nonbloating legume. C. FEEDING VALUE The feeding value of birdsfoot trefoil hay has equaled that of alfalfa and other good legume hays when fed to sheep (Ingalls et al., 1965) and to dairy cows (Loosli et al., 1950). I n addition, Loosli found that milk from cows fed trefoil hay contained more carotene, vitamin A, and tocopheral, and exhibited better keeping qualities than milk from cows receiving Ladino clover and timothy hays. The relatively higher carotene content of trefoil herbage was reported in Indiana by Burger ( 1 947). In New York, Trimberger et al. (1962) studied the feeding value of trefoil hay cut at different stages. They found that birdsfoot trefoil hay and a grass-legume hay were equally good in milk production. Late-cut trefoil hay was damaged by rain in 1958. In both years, late cutting (July 7 and 8) decreased the palatability of the hay, and reduced daily consumption. However, in both years the actual digestibility of these hays exceeded the calculated digestibility (Reid et al., 1959) by 2 and 8 percentage points for 1958 and 1959, respectively. They concluded from this research that “the higher digestibility coefficient over a longer period of time, particularly later in the season, extends the time for harvesting good hay from birdsfoot trefoil forage.” N o real differences in feeding value of Viking and Empire birdsfoot trefoil hay was apparent in these investigations. Directly related to feeding value are reports that birdsfoot trefoil can be stockpiled for later grazing or hay. In Pennsylvania, Mays and Washko (1960) reported that the grazing of pure stands of birdsfoot trefoil could be delayed until July 1 without sacrificing high yield of good quality forage. When the intitial growth is removed as hay or silage between June 1 and 5, the birdsfoot trefoil aftermath may be grazed anytime up to September 1. Comparative studies of plant development of

BIRDSFOOT TREFOIL

143

birdsfoot trefoil, alfalfa, and red clover by D. Smith (1962) revealed that Empire birdsfoot trefoil produced only one growth from crowns when left uncut, whereas alfalfa and red clover each produced three distinct growths when left uncut. Carbohydrate root reserves in birdsfoot trefoil were maintained at a very low level during the growing season under all cutting treatments until fall storage occurred. Smith concluded that Empire birdsfoot trefoil, after the initial vegetative expansion, is dependent for growth primarily on carbohydrates synthesized by existing top growth rather than on root reserves. In a later study by D. Smith and Nelson ( 1967), the height and frequency of cut of Empire birdsfoot trefoil and Vernal alfalfa were compared. They reported that a tall stubble of birdsfoot trefoil was needed for high yields under all cutting frequencies. A tall stubble was needed by alfalfa only for the most frequent cutting system. Regardless of the frequency of cut, the remaining stubble of birdsfoot trefoil must carry green leaves to produce the energy needed for regrowth. Thus, when Empire is stockpiled the plants continue to grow: the new growth comes from upper axillary buds of the stems. The stockpiled herbage may be grazed off over a period of time. If cut for hay, the stubble left must contain a sufficient number of green leaves for regrowth. VI.

Genetics and Cytology

A. CYTOLOGY Chromosome numbers have been determined for approximately 70 different species of Lotus. About one-third of the species have a basic chromosome number of 6, and two-thirds a basic chromosome number of 7. Diploid and tetraploid species occur within both groups. North American species of Lotus are all diploids, with 2 N = 12 or 14. A large number of European species have a basic chromosome number of 6, while North American species are usually x = 7 (Grant, 1965). L. corniculutus is a tetraploid with 2 N = 24 chromosomes. Chromosome pairing is usually bivalent with the occurrence of an occasional quadrivalent. Frequency of quadrivalent pairing is about 1 quadrivalent in every 4 microspore mother cells (Wernsman et ul., 1964). Dawson (1941) and others suggest that L. cornicufutus is an autotetraploid of L . tenuis. This conclusion is based on the chromosome number of these species ( L . tenuis, 2 N = 12), the morphological characteristics, the bivalent pairing in L . corniculutus and the tetrasomic inheritance of cyanogenesis. Stebbins ( 1950), believes that bivalent pairing

144

ROBERT R. SEANEY A N D PAUL R. HENSON

combined with tetrasomic inheritance is evidence that L. corniculatus is a segmental allopolyploid. From an analysis of the phenolic constituents in Lotus, Harney and Grant (1963, indicate that L. corniculatus is more likely to be an allotetraploid than autotetraploid. Grant and Sidhu ( 1967) interpret their data on HCN reaction of species in the L. corniculatus group to indicate that L. tenuis, as well as other species, could be ancestors of the tetraploid L. corniculatus. Other evidence that L. tenuis is a progenitor of L. corniculatus is found in the work of Wernsman et af. ( 1 964). Backcross progenies of the interspecific cross 4X L. tenuis X L. corniculatus showed bivalent pairing. This would indicate a high degree of homology between chromosomes of these two species. B. INHERITANCEOF CHARACTERS Dawson (1941) found that cyanogenesis in L. corniculatus is determined by a single dominant gene inherited tetrasomically. The acyanogenetic plants having the recessive gene lack the enzyme which is necessary for hydrolysis of the cyanogenetic glucoside. The concentration of hydrocyanic acid is probably quantitatively determined by a series of modifying genes. Other inheritance studies also show tetrasomic inheritance in birdsfoot trefoil. Donovan ( 1 959) and Donovan and McLennan ( 1 964), working with crosses between the large-leaved L. corniculatus var. vulgaris and small leaved L. corniculatus var. arvensis found large leaf to be dominant, with an autotetraploid type of inheritance. Leaf color, or actually chlorophyll content, in the variety Viking was found by Poostchi and MacDonald ( 196 I ) to be determined by a single dominant gene showing random four-chromosome type segregation. In certain accessions of L. corniculatus, keel tip color may be yellow, brown, or red. Both brown and red keel tip are dominant to yellow (Bubar and Miri, 1965; Buzzell and Wilsie, 1963). Brown keel tip is primarily determined by a single gene with probably two or more suppressor genes reducing the expected proportion of brown to yellow keel tip plants. A quantitative type of inheritance has been proposed for the intensity of the brown color of the keel tip. Buzzell and Wilsie (1963) also studied the inheritance of flowering time and length of flowering stem in crosses between Empire and Viking varieties. Dominance for early flowering was related to the flower stem lengths of the parent plants. Other characters in trefoil which show tetrasomic type of inheritance are pubescence, a chlorophyll deficiency, flower color, streaks on the corolla and self-incompatibility (Bubar and Miri, 1965).

BIRDSFOOT TREFOIL C.

145

CROSSAND SELF-COMPATIBILITIES

Many plants of L. corniculatus are self-incompatible. That is, they set little or no seed after self-pollination by insects or by hand (Silow, 193 1 ; Tome and Johnson, 1945). The degree of self-incompatibility ranges from plants which set no self seed to those which set a limited amount. In a population of some 360 plants which had been previously selected for desirable agronomic characters, Seaney (1964) found that 45% of the plants set no self seed, 46% set from 0.01 to 0.49 seed per flower pollinated, and 7% set from 0.5 to > 1.0 seed per flower. Occasional plants were found which set as high as 5 to 8 seeds per flower pollinated. There are several mechanisms which may be responsible for the selfincompatibility in trefoil. Giles ( 1949) has noted three abnormal processes, one or more of which could account for self-incompatibility. First, self pollen tubes grow at a slower rate through the ovarian tissue than do pollen tubes of cross pollen. Second, pollen tubes of self-pollinated plants approach the micropyle of unfertilized ovules, but frequently fail to enter and effect fertilization. And third, the frequency of aborting ovules is much greater in self-pollinated plants than in crosspollinated plants. Giles ( 1 949) suggested that the primary cause of selfincompatibility is the failure of ovules to be fertilized rather than abortions of ovules after fertilization. Self-incompatibility apparently is not related to differential growth rates of pollen tubes through the style. Both in vivo and semi-in vitro techniques have demonstrated that there is no difference between growth rates of cross- and self-pollen tubes (Giles, 1949; Miri, 1964; Spiss and Paolillo, 1969). Studies of the inheritance of self-incompatibility have been made in both L. corniculatus and L . tenuis (Elliott, 1946; Brandenburg, 1960; Bubar and Miri, 1965). Results of these studies are inconclusive and indicate that inheritance in both the diploid and tetraploid species is very complex. Under field conditions, L. corniculatus is largely cross-pollinated. All plants that are highly self-incompatible probably set 100% cross seed. However, plants which are self-fertile can set self-seed, and the amount of self-seed set is directly related to the self-fertility of the individual plant (Seaney, 1964; Miri and Bubar, 1966). Bansal ( 1965) studied the amount of cross and self-seed set occurring when plants were grown under cage isolation and pollinated by honey bees. The amount of self seed set for a specific clone was directly proportional to the relative self-fertility of that clone. The amount of seed set also varied when different pollen parents were used to make the cross. When crosses were made under cage isolation, it was found that different

146

ROBERT R. SEANEY A N D PAUL R. HENSON

arrangements of the clones and different row spacings influenced the foraging habits and visitations of the bees and resulted in different proportions of cross and self seed. When cross-pollinations are made by hand in the greenhouse, emasculation is not necessary to obtain complete crossing, if the plants are highly self-incompatible. However, plants that set as few as 0.2 seed per flower when self-pollinated may set 1-3% self seed when cross-pollinated without emasculation (Seaney, 1964). Both bud pollination and treatment of flowers with growth regulators have been unsuccessful in increasing self seed on self-incompatible plants (Giles, 1949). However, the authors (unpublished) have noted that postpollination treatment of plants with high temperatures sometimes increases self seed set. When making self-pollinations by hand, both Giles (1949) and R. McKee (1949) reported that greater seed set was obtained by tripping plus scarification of the stigmatic surface than by tripping alone. Seaney (1962) also noted that the highest number of self seed per flower was obtained when the stigma was extruded through the keel tip and lightly touched against a wooden toothpick. Another method which was equally effective in obtaining self seed set, was simply to roll the umbel or flower cluster between the fingers. D. INTERSPECIFIC HYBRIDIZATION

Attempts have been made to cross L. corniculatus with a number of diploid and tetraploid species of Lotus. Grant (1 965) has catalogued all interspecific crosses between Lotus species. Within the genus Lotus a relatively large number of successful interspecific crosses have been made. However, L. corniculatus has been successfully crossed with only four other species, L. tenuis, L. pedunculatus (both 2X and 4x), L. palustris, and L. coimbrensis. Bent (1958) attempted interspecific crosses in order to introduce into L. corniculatus the superior seedling vigor of L. tenuis and the rhizomatous and delayed shattering characteristics of L. pedunculatus. Plant characteristics of the interspecific hybrid, such as leaf size and shape, number of flowers per umbel were intermediate to the parent species. Bent, as well as Gershon (1961) were unsuccessful in isolating the delayed pod shattering or rhizomatous characteristics of L. pedunculatus in advanced generation populations of the interspecific cross. Phillips and Keim (1968) were successful in crossing L. corniculatus with L . coim-

BIRDSFOOT TREFOIL

147

brensis which has indehiscent seed pods, but did not observe indehiscence in F1or backcross progenies. Several workers have crossed L. corniculatus with both diploid and tetraploid plants of L. tenuis (Keim, 1952; Mears, 1955; Wernsman et al., 1965). The hybrids are vigorous, intermediate in appearance, and show a relatively high degree of fertility when intercrossed. From the backcross progenies of (L. corniculatus X L. tenuis) (L. corniculatus) the authors (unpublished) have been able to select lines which are significantly superior in seedling vigor to the variety Viking. Grant has made extensive studies of the inheritance and linkage relationships in interspecific hybrids between diploid species (2N = 12) closely related to L. corniculatus (Grant et al., 1962; de Nettancourt and Grant, 1964a,b; Harney and Grant, 1964). He has been successful in obtaining a number of different crosses involving diploid species such as L. tenuis, L. jlicaulis, L. japonicus, L. schoelleri, L. krylovii, L. Borbasii, and L. alpinus. These studies have indicated dominance of such characters as ascending growth habit, red stem color, striping of flower bud, reddish-brown keel tip color, seed coat mottling, and presence of HCN in leaves. Studies such as these are valuable in providing information to elucidate the feasibility of transferring certain traits into L. corniculatus. The numerous and divergent characteristics of this genus enhance the possibility of using interspecific hybridization for variety improvement. However, as yet, use of interspecific hybridization has not resulted in the development of superior agronomic types. VII.

Breeding

A. OBJECTIVES

Trefoil breeding programs in Canada and United States have been largely concerned with improvement and development of varieties within the species L. corniculatus. Although big trefoil, L. pedunculatus, is used for hay and pasture in the coastal regions of Northwestern and Southeastern United States, there has been little work on variety improvement since development of the varieties Beaver and Columbia. The primary objective of breeding programs in North America has been to increase forage yields of the crop when used for pasture or hay. Specific breeding objectives to accomplish this have involved improvement of seedling vigor, resistance to root rots, winterhardiness and rapid

148

ROBERT R. SEANEY A N D PAUL R. HENSON

recovery growth after cutting. Very little work has been done in the important areas of quality, or resistance to insects and foliar diseases.

B. VARIABILITY A N D METHODS There is a wide range of variability within L. corniculatus. Differences exist both within and between collections. J. B. Smith (1956) described the range of variation for certain characters in birdsfoot trefoil. The collections which were studied showed wide differences in seedling vigor, date of flowering, rate of early spring growth, rate of recovery growth, and winterhardiness. No differences were noted for rate of seed germination or amounts of hard seed produced. In studies of replicated clonal materials, the authors (unpublished) have found significant differences between individual plants for such additional characters as leaf size, shape, and color; flower size and shape; growth habit; flower color; seed production; number of stems per crown; and crown recovery. This latter character concerns the ability of some plants to produce regrowth from crown or basal stem buds after first vegetative growth has reached flower or seed stage of maturity. Several methods have been used to develop new varieties of trefoil. Some breeders have used phenotypic mass selection followed by random crossing of the selected clones. Others have used various types of progeny testing to evaluate clones for subsequent intercrossing and increase as a variety. Recurrent selection programs, combined with either progeny testing or phenotypic selection, have also been used to a limited extent.

c. POD DEHISCENCE In seed production fields, considerable loss of seed often occurs because of pod shattering (S. R. Anderson, 1955). Breeders have attempted to develop strains of L. corniculutus which do not shatter readily. Peacock and Wilsie (1 957) found differences between clones for time of dehiscence in relation to pod maturity. Some clones were found to be readily indehiscent. That is, shattering did not occur until sometime after pods were fully mature. Clones were selected for resistance to pod shattering, and by one cycle of recurrent selection pod dehiscence was reduced by 17%. Study of progeny from a series of diallel crosses indicated that the clones showing highest general combining ability produced the best FI combination. As yet, selection and breeding within L. corniculatus have not resulted in isolating strains which have significant shattering resistance under field conditions. Interspecific hybridization may offer opportunity for transferring pod indehiscence into L. corniculutus (Gershon, 1961; Phillips and Keim,

BIRDSFOOT TREFOIL

149

1968). Pods of L. coimbrensis are completely indehiscence while L. angustissimus, L. tetragonolobus, and L. ornithopodioides show considerable pods indehiscence. However, incrosses between L. corniculutus and L. climbrensis, Phillips and Keim (1968) have not observed indehiscence in F, or backcross progenies. D. SEEDYIELD Within most varieties and strains of trefoil, there is considerable variation for seed set between individual plants. Peacock and Wilsie (1960) concluded that observed differences in seed production among clones of L. corniculatus were probably due to differences in genotype. Although Peacock and Wilsie ( 1960) found no increase in seed production in either the first or second cycle of selection in a recurrent selection breeding program, there were individual second cycle crosses which were outstanding in seed production. They suggested that selection and crossing of superior second cycle parents should give increases in seed set in the third cycle generation. Other investigators have also found genetic variation for seed yield (Buzzell and Wilsie, 1964). Estimates of hereditability suggest that a large part of phenotypic variance for seed set is genetically controlled (Albrechtsen et al., 1966). Although selections for high seed yields have been incorporated into experimental synthetics, no varieties have yet been developed specifically for higher seed yields. E. SEEDLING VIGOR Trefoil seedlings are, in general, less vigorous than those of alfalfa or red clover. Because of the relatively slow seedling growth, successful establishment of trefoil is often difficult under conditions of competition from other crops and weeds. The objective of several breeding studies has been to increase seedling vigor. Henson and Taymen (196 I ) found that the rate of seedling growth was related to seed size. Large seeds produced more vigorous seedling plants than small seeds. Selection and intercrossing of plants having large seeds resulted in lines having greater seedling vigor. Draper and Wilsie (1 965) have shown that seed size can be increased by recurrent selection. Three cycles of recurrent selection in the varieties Viking and Empire resulted in increases in seed size of 20% per cycle in Viking and 25% per cycle in Empire. Twamley ( 1967) tested open-pollinated progeny lines of European type trefoils and also concluded that seedling vigor was directly related

150

ROBERT R. SEANEY A N D PAUL R. HENSON

to seed size. By use of an intensive selection and breeding program, Twamley was able to incorporate increased seeding vigor into the variety Maitland.

F. DISEASERESISTANCE In Southern areas of the United States crown and root rot diseases cause serious economic loss. Henson ( 1962), and others have worked on the development of strains resistant to root rotting organisms such as Rhizoctonia spp., Fusarium spp., and Macrophomina phaseoli. There are clonal differences for susceptibility to the root rots, but completely resistant plants have not been found. However, estimates of hereditability and genetic advance do indicate opportunity for increasing the resistance of trefoil to root rot. Miller (1 968) estimated the combining ability of four clones of trefoil for root rot by studying diallel progenies grown as spaced plants. Miller suggested that significant specific combining ability effects indicate that use of single crosses might be more effective than use of clone synthetics in breeding for root rot resistance. The variety, Dawn, developed cooperatively by the USDA, Agricultural Research Service, and Missouri Agricultural Experiment Station, is the only variety presently in use which was specifically bred for resistance to root rot. Dawn has shown significant tolerance to root rots in Missouri when compared to the varieties Viking and Empire.

G. WINTERHARDINESS In the Northern United States and in Canada, winterhardiness has been one objective of all trefoil breeding programs. Bubar and Lawson ( 1959) found differences in winterhardiness between varieties, strains, and ecotypes. These differences are hereditable and can be transferred into first generation progeny. The variety Leo was developed from materials selected for increased winterhardiness.

H. CLONECROSSES Bansal (1965) studied the yield performance of single crosses, double crosses, and four-clone synthetics produced under cage isolation by random bee pollination. Progenies were evaluated as spaced plants. Fourclone synthetics and single crosses of the same parental clones wire higher yielding then the double cross combinations. Bansal attributed this to the higher proportion of self or sib seed set in the double crosses. The advanced Fz generations of the single crosses were lower yielding than

BIRDSFOOT TREFOIL

151

their parental single crosses. The use of specific clone combinations, such as single and double crosses, will not be practical until some system of controlling pollination makes possible obligate crossing under field conditions. 1.

INBREEDING A N D HYBRIDIZATION

The effects of inbreeding in birdsfoot trefoil are, in general, similar to those found in other cross-pollinated crops. Decreased vegetative vigor and self-fertility have been noted in successive inbred generations (Brandenburg, 1960; Seaney, 1967). Single crosses of inbreds are higher yielding than either parent. However, Seaney (1 967) found that average forage yields of F, crosses within groups of S4, Ss, and Ss inbreds were less than yields of the check variety Viking. Specific combinations of Ss inbreds were 35% higher in yield than Viking. In these studies progenies were evaluated as spaced plants.

J. VARIETIES Varieties within the species L. corniculatus are of two distinct types, commonly referred to as “European” or “Empire” types. The primary difference between these forms is their growth habit, maturity, seedling and recovery growth rates, and winterhardiness. When grown and compared under field conditions it is relatively easy to distinguish between Empire and the other European type varieties. However, growth habit, date of flowering, and the gross morphological characteristics of European type varieties are quite similar, and identification of particular varieties is extremely difficult. In a series of studies, Nittler and Kenny ( 1 965) used controlled environmental conditions for variety identifications. Plants of Viking and common European trefoil were easily distinguished from plants of Empire by the effect of photoperiod on stem length and number. Distinguishing Viking from common European types was more difficult. The influence of temperature on the percentage of plants in bloom can also be used to distinguish between Empire, Viking, and common European (Nittler and Kenny, 1965). Other studies show that foliar applications of N-dimethylaminosuccinamic acid (B-995) cause Empire plants to be shorter and more decumbent than Viking or common European (Nittler and Kenny, 1966). Empire is a selected ecotype which was found in fields in Albany County, New York. All seed increases and selections of Empire trace to this origin. Empire is semierect and fine stemmed and flowers 10- 14 days later than common European types. Flowering is indeterminate, for

152

ROBERT R. SEANEY AND PAUL R. HENSON

flowers are continually produced on new growth which originates from node buds along the stem. Compared to European-type trefoils, spring growth of Empire and recovery after harvest are slow. Empire is very winter-hardy in Northeastern United States and in Canada. Empire was developed by the Cornell University Agricultural Experiment Station, Ithaca, New York. Dawn, a four-clone synthetic, was developed from selections made in the variety Empire. Dawn closely resembles Empire, although it is slightly more erect and earlier flowering. Trials in Missouri show that Dawn has greater resistance to root rots than either Empire or Viking. Dawn was developed cooperatively by the Missouri College of Agriculture, and Agricultural Research Service, USDA, Columbia, Missouri. Leo is a variety originating from an accession imported from USSR. Selections for this variety were made from a group of plants which had survived several winters at Macdonald College, Quebec, Canada. Leo belongs to the same subspecific group within L. corniculutus, as does the variety Empire. Tests in Quebec, Canada, show this variety to have excellent early spring vigor, and greater winterhardiness than either Empire or Viking. Leo was developed at Macdonald College of McGill University, Quebec, Canada. Viking trefoil is a European type variety obtained by intercrossing selections from Denmark and ecotypes growing in Columbia and Schuyler counties, New York. Viking is erect, early maturing, and superior in winterhardiness to imported lots of trefoil from Italy or France, and to the varieties Granger and Cascade. In New York, Viking yields 15-20% more forage than Empire. This variety was developed by the Cornell University Agricultural Experiment Station, Ithaca, New York. Mansfield is similar to Viking in growth habit, winterhardiness, and range of adaptability. The variety was developed by combining selections from two imported seed lots and a seed stock from Columbia county, New York. Mansfield, like Viking, has greater seedling vigor than Empire and also recovers rapidly after cutting. Mansfield was developed and released by the Vermont Agricultural Experiment Station, Burlington, Vermont. Cascade is an early maturing variety developed by selecting and intercrossing a large number of plants from an accession collected in France. Cascade has good seedling vigor and recovery growth, but is not as winterhardy as either Viking or Mansfield. Cascade was developed cooperatively by the Washington Agricultural Experiment Station, and the U.S. Soil Conservation Service, Pullman, Washington. Granger was developed from a seed lot imported from France. It is

BIRDSFOOT TREFOIL

153

uniform in type, having upright growth, broad leaves, and good seedling vigor. Granger is similar to Cascade in forage production and winterhardiness. Granger was developed cooperatively by the Oregon Agricultural Experiment Station and Agricultural Research Service, USDA, at Corvallis, Oregon. Douglas is a variety selected from several European lots grown in Douglas County, Oregon. It is a vigorous fast growing selection similar in appearance to other European types. Tana is similar in appearance and performance to Cascade and Granger. In some tests in New York State, Tana appears to have more rapid recovery after cutting than other European type varieties. However, winterhardiness is not as good as either Empire or Viking. Tana was released by the Montana Agricultural Experiment Station, Huntley, Montana. Maitland is a synthetic of 1 I clones selected for seedling vigor, forage yield, and winterhardiness. It is equal to Viking variety in winterhardiness, but is superior in seedling vigor and spring growth. Maitland was developed by the University of Guelph, Guelph, Ontario, Canada. Two varieties of the species L. pedunculatus have been developed by the Oregon Agricultural Experiment Station and Agricultural Research Service, USDA, Astoria, Oregon. Beaver is a smooth or glabrous-leafed variety, and Columbia is a selection having pubescent leaves and stems. Foliage and general appearance of these varieties are similar to L. corniculatus. Both varieties are well adapted to the wet winter and cool summers of coastal Oregon and Washington. REFERENCES Ahlgren, G . H., Briggs, R. A., Sprague, M. A., and Wakefield, R. C. 1955. N e w J e r s e y A g r . Exp. Sro., Bull. 779. Albrechtsen, R. S., Davis, R. L., and Keim, W. F. 1966. C r o p . Sci. 6,355-358. Alexander, C . W., and Chamblee, D. S. 1965. Agron. J. 57,550-553. Allred, K. R. 1955. Ph.D. Thesis, Cornell University, Ithaca, New York. Anderson, J. 1777. “Essays Relating to Agricultural and Rural Affairs,” 2nd ed., pp. 228234. London. Anderson, S. R. 1955. Agron. J. 47,483-487. Anderson, S. R., and Metcalfe, D. S. 1957. Agron. J . 49,52-55. Bader, K. L., and Anderson, S. R. 1962a. Agron. J . 54,306-309. Bader, K. L., and Anderson, S . R. 1962b. C r o p Sci. 2, 148-149. Bansal, R. D. 1965. Ph.D. Thesis, Cornell University, Ithaca, New York. Barr, D . J. S., and Callen, E. 0. 1963. Phyroprorecrion 44, 18-24. Bent, F. C. 1958. Ph.D. Thesis, Cornell University, Ithaca, New York. Brandenburg, N. G. 1960. Ph.D. Thesis, University of Maryland, College Park, Maryland. Briggs, R. A. 1953. Ph.D. Thesis, Rutgers University, New Brunswick, New Jersey.

154

ROBERT R. SEANEY A N D PAUL R. HENSON

Brown, C. S. 1955. Ph.D. Thesis, Cornell University, Ithaca, New York. Bubar, J. S. 1958. Can. J . Bot. 36,65-72. Bubar, J. S., and Lawson, N. C. 1959. Can. J. Plant Sci. 39, 125-126. Bubar, J. S., and Miri, R. K. 1965. Nature 205, 1035-1036. Buckovic, R. G. 1952. M.S. Thesis, Oregon State College, Corvallis, Oregon. Burger, 0. J. 1947. M.S. Thesis, Purdue University, LaFayette, Indiana. Buzzell, R. I., and Wilsie, C. P. 1963. Crop Sci. 3, 128-130. Buzzell, R. I., and Wilsie, C. P. 1964. Crop Sci. 4,436-437. Callen, E. 0. 1959. Can. J . Bot. 37, 157-165. Chamblee, D. S. 1960. North Carolina, Agr. Exp. Sta., Bull. 411. Chevrette, J. E., Folkins, L. P., Gautheir, F. M., and Greenshields, J. E. R. 1960. Can. J . Plant Sci. 40,259-267. Cooper, C. S. 1966. Crop. Sci. 6,63-66. Cooper, C. S. 1967. Crop Sci. 7, 176-178. Cooper, C. S., and Corms, W. G. 1952. Sci. Agr. 32,281-284. Cooper, C. S., and Qualls, M. 1968. Crop Sci. 8,756-757. Cooper, C . S., Eslick, R. F., and McDonald, P. W. 1966. Crop Sci. 6,215-216. Davis, R. R., and Klosterman, E. W. 1959. Ohio, Agr. Exp. Sta., Res. Circ. 75. Dawson, C. D. R . 1941. J . Genet. 42,49-72. Decker, A. M., Retzer, H. J., and Swain, F. G. 1964.Agron. J . 56,2 1 1-2 14. de Nettancourt, D., and Grant, W. F. 1964a. Can. J . Genet. Cytol. 6,29-36. de Nettancourt, D., and Grant, W. F. 1964b. Can. J . Genet. Cytol. 6, 277-287. Donovan, L. S. 1959. Can. J . Plant Sci. 39, 141-157. Donovan, L. S., and MCLenndn, H. A. 1964. Can. J . Genet. Cytol. 6, 164-169. de Rothschild, H. 1920. C. R . Acad. Agr. France 6, 505-508. Dougherty, R. W., and Christensen, R. B. 1953. Cornell Vet. 43,481-486. Drake, C . R. 1958. Plant Dis. Rep. 42, 145-146. Drake, C. R. 1962. Plant Dis. Rep. 50,509-5 12. Draper, A. D., and Wilsie, C. P. 1965. Crop Sci. 5.3 13-3 15. Duell, R. W. 1964. Agron. J . 56, 503-505. Duell, R. W., and Gausman, H. W. 1957. Agron. J . 49.3 18-3 19. Elliott, F. C. 1946. M.S. Thesis, Iowa State College, Ames, Iowa. Ellis, W. 1774. "The Modern Husbandman," 1st ed., pp. 143-145, London. Erdman, L. W.,and Means, U. M. 1949. Soil Sci. Soc. Amer., Proc. 14,170-175. Fertig, S . N. 196 1. Proc. Northeast. Weed Contr. Conf.15,357-358. Fertig, S. N., Meadows, M. W., and Bayer, G. 1960. Proc. Northeast. WeedContr. Conf. 14,308-3 13. Flanagan, T. R. I96 1. Proc. Northeast. Weed Contr. Conf. 15,249-253. Flanagan, T. R., and MacCollom, G. B. 1964. Proc. Northeast. Weed Contr. Con$ 18, 3 15-3 18. Ford, R. E. 1959. Phyropathology 49,48 1-486. Ford, R. E. 1960. Plant Dis. Rep. 44,276-280. Foy, C. D., and Barber, S. A. 1961. Agron. J . 53, 109-1 10. Fribourg, H. A. 1963. Tennessee, Agr. Exp. Sta., Bull. 371. Gardner, A. L., and Alburquerque, H. 1965. Proc. 9th Int. Grassland Congr., 1965 VOI. 2, pp. 1053-1058. Gardner, C. A.. and Elliot, H. G. 1945. West.Australia Dep. Agr.J. 22,358-360. Gershon, D. 1961. Ph.D. Thesis, Cornell University, Ithaca, New York. Giles, W. L. 1949. Ph.D. Thesis, University of Missouri, Columbia, Missouri.

BIRDSFOOT TREFOIL

155

Gist, G. R., and Mott, G. 0. 1957. Agron. J . 49.33-36. Graham, J. H. 1953. Phytopathology 43,577-579. Grant, W. F. 1965. Can. J . Genet. Cytol. 7,457-471. Grant, W. F., and Sidhu, B. S. 1967. Can. J . Bot. 45,639-647. Grant, W. F., Bullen, M. R.,andde Nettancourt, D. 1962. Can. J . Genet. Cytol. 4,105-128. G u p p y , J . C. 1958. Entomology 90,523-53 I . Gurney, E. H., and White, C. H. 1941. Queensland Agr. J . 55,297-299. Hansen, H. W. 1949. Iowa Srare Coll. J . Sci. 24,57-59. Harney, P. M., and Grant, W. F. 1964. Can. J . Genet. Cyrol. 6, 140-146. Harney, P. M., and Grant, W. F. 1965. Can. J . Genet. Cytol. 7,40-51. Hasaka, E. Y. 1957. Hawaii Univ. Agr. Ext. Circ. 367. Henry, T. A. 1938. J . SOC.Chem. Ind., London 57,248-249. Henson, P. R. 1962. Crop Sci. 2,429-432. Henson, P. R., and Schoth, H. A. 1962. US.Dep. Agr., Agr. Handb. 223. Henson, P. R., and Taymen, L. A. 1961. Crop Sci. 1, 106. Howell, H. B. 1948. Oregon, Agr. Exp. Sra., Bull. 456. Hughes, H. D. 1962. In “Forages” (H. D. Hughes, M. E. Heat, and D. S. Metcalfe, eds.), 2nd ed., pp. 187-204. Iowa State Univ. Press, Ames, Iowa. Hunt, D. J., and Wagoner, R. E. 1963. Agron. J . 55, 16-19. Ingalls, J. R., Thomas, .I.W.,Benne, E. J., and Tesar, M. 1965. J. Anim. Sci. 24,1159-1 164. Isely, D. 195 I . Iowa Srare Coll. J . Sci. 25,439-482. Joffe, A. 1958. South African J . Agr. Sci. 1,435-450. Jones, L. G. 1952. Down Earrh 8,2-4. Kainski, J. M. 1959. Ph.D. Thesis, Cornell University, Ithaca, New York. Keim, W. F. 1952. Ph.D. Thesis, Cornell University, Ithaca, New York. Kendall, W. A. 1964. Crop Sci. 4,391-393. Kerr, H . D., and Klingman, D. L. 1960. Weeds 8, 157-167. Kingsbury, C. H., and Winch, J. E. 1967. Ontario Dep. Agr. Foodlnf. LeaJlerAGDEX 122. Klostermeyer, E. C., and Menzies, J. D. 1951. Phytopathology 41,456-458. Kreitlow, K . W. 1962. U.S.Dep. Agr., Agr. Handb. 223. Levy, E. B. 1918. New Zealand J . Agr. 17,347-35 1. Linscott, D. L., and Hagin, R. D. 1967. Weeds 15,264-267. Linscott, D. L., and Hagin, R. D. 1968. Weeds 16, 182-184. Linscott, D. L., Seaney, R. R., and Hagin, R. D. 1967. Weeds 15,259-264. Loosli, J . K.. Krukovsky. V. N.. Lofgreen. G. P. and Musgrave, R . B. 1950. J . Dairy Sci. 33,228-236. Lynch, D. L., and Sears, 0. H. 195 I. Soil Sci. SOC.Amer., Proc. 15,176-180. MacCollom, G. B. 1958. J . Econ. Enromol. 51,492-494. MacDonald, H. A. 1946. Cornell Univ. Agr. Exp. Sra., Memoir. 261. MacDonald, H. A. 1957. Farm Res. 23, No. 4, 8. McArthur, J. M., and Miltimore, J. E. 1966. Proc. loth Inr. Grassland Congr., I966 Sect. 2, pp. 5 18-52 1. McKee. G. W. 1961. Agron. J . 53,237-240. McKee, G . W. 1962. Crop Sci. 2,315-317. McKee, G . W. 1963. Crop Sci. 3,205-208. McKee, R. 1949. Agron. J . 41,313-316. McVey, D. V., and Gerdemann, J. W. 1960. Phyropathology 50,416-421. Mangan, J. L. 1959. New Zealand J . Agr. Res. 2,47-61. Mathur, R. B., and Pienkowski, R. L. 1967. J . Econ. Entomal. 60,207-209.

156

ROBERT R. SEANEY A N D PAUL R. HENSON

Mays, D. S., and Washko, J. B. 1960. Pennsylvania Agr. Exp. Sra., Sci. Farmer 8,2. Mears, K. A. 1955. M.S. Thesis, University of Vermont, Burlington, Vermont. Metcalfe, D. S., Johnson, I. J., and Shaw, R. H. 1957. Agron. J . 49, 130-133. Midgley, A. R., and Gershoy, A. 1946. J . Amer. SOC.Agr. 38, 197-199. Midgley, A. R., and Stone, E. 1958. Vermont, Agr. Exp. Sta., Bull. 607. Miller, J. D. 1968. Crop Sci. 8,41-43. Miller, J. D., Kreitlow, K. W., Drake, C. R., and Henson, P. R. 1964. Agron. J . 56,137- 139. Miri, R. K. 1964. Ph.D. Thesis, Macdonald College, McGiil University, Montreal, Canada. Miri, R. K., and Bubar, J. S. 1966. Can. J . Plant Sci. 46,411-418. Morse, R. A. 1955. Diss. Absrr. 15,946. Mott, G . O., Smith, R. E., McVey, W. M., and Beeson, W. M. 1953. Indiana, Agr. Exp. Sta., Bull. 581. Nelson, J. C. 1917. Torreya 17, 151-160. Neunzig, H. H. 1957. Ph.D. Thesis, Cornell University, Ithaca, New York. Neunzig, H. H., and Gyrisco, G. G. 1955. J . Econ. Entomol. 48,447-450. Nittler, L. W., and Kenny, T. J. 1965. Crop Sci. 5,457-459. Nittler, L. W., and Kenny, T. J. 1966. Crop Sci. 6,601-604. Offitt, M. S. 1967.Arkansas Farm Res. 16, No. 5 , 8 . Orsi, S. 1953. Ann. Ente Consor Inrerprov. Tosc. Sementa IV, 30. Ortiz, F. S., McCune, D. L., and Grove, H. V. 1961. Agr. Canaderia 6,26-27. Ostazeski, S . A. 1965. Plant Dis. Rep. 49,855-856. Ostazeski, S . A. 1967. Mycologia 59,970-975. Ostazeski, S. A., and Henson, P. R. 1965. Crop Sci. 5,253-254. Ottley, A. M. 1923. Univ. California Berkeley, Publ. Bot. 10, 189-305. Ottley, A. M. 1944. Briffonia 5.81-123, Panikkar, M. R. 1949. Indian Farming 10,444-447. Pardee, W . D. 1968. Cornell Univ., Dep. Plant Breed. Mimeo 68-1. Parsons, J. L., and Davis, R. R. 1964. Ohio, Agr. Exp. Sra., Res. Bull. 967. Peacock, H. A., and Wilsie, C. P. 1957. Agron. J. 49,429-431. Peacock, H. A., and Wilsie, C. P. 1960. Agron. J. 52,321-324. Peters, E. J. 1964. Agron. J. 56,4 15-4 19. Peterson, M. L., Jones, L. G., and Osterli, V. P. 1953. California, Agr. Exp. Sra., Circ. 421. Pettit, R. E., Calvert, 0. H., and Baldridge, J. D. 1966. Plant Dis. Rep. 50, 753-755. Phillips, R. L. 1968. Crop. Sci. 8, 123-124. Phillips, R. L., and Keim, W. F. 1968. Crop Sci. 8, 18-21. Pierre, J. J., and Jackobs, J. A. 1953. Agron. J. 45,463-468. Pierre, J. J., and Jackobs, J. A. 1954. Illinois, Agr. Exp. Sta., Circ. 725. Poostchi, I., and MacDonald, H. A. 1961. Crop Sci. 1,327-328. Pressey, R., Synhorst, S. H., Bertram, J., Allen, R. S., and Jacobson, N. L. 1963. J. Anim. Sci. 22,970-978. Quiros, C. M. 1946. Rev. Agr., Sun Jose 18,91-92. Reid, J. T., Kennedy, W. K., Turk, K. L., Slack, S. T., Trimberger, G. W., and Murphy, R. P. 1959. J. Dairy Sci. 42,567-571. Rhykerd, C. L. 1959. Agron. J . 51,7-9. Rhykerd, C. L., Langston, R., and Mott, G. 0. 1959. Agron. J . 51, 199-201. Ridgeway, R. L., and Gyrisco, G. G. 1959. J . Econ. Entomol. 52,836-838. Robinson, D. H. 1934. EmpireJ. Exp. Agr. 2,274-283.

BIRDSFOOT TREFOIL

157

Ronningen, T. S., Wakefield, R. C., Atwood, S. S., Lowe, C. C., Murphy, R. P., Fortman, H. R., Stuckey, I. H., Varney, K. E., and Burger, 0. J. 1954. New Jersey, Agr. Exp. Sta.. Bull. 777. Scholl, J. M., and Brunk, R. E. 1962. Agron. J. 54, 142- 144. Scholl, J. M., and Hughes, H. D. 1955. Iowa Farm Sci. 9,715-716. Scholl, J. M., and Staniforth, D. W. 1957. Agron. J. 49,432-435. Schribaux, E. 1922. C. R . Acad. Agr. France 8, 124- 13 1. Seaney, R. R. 1962. Crop Sci. 2.81. Seaney, R. R. 1964. Crop Sci. 4,440-441. Seaney, R. R. 1967. Agron. Abstr. 59, 18. Shibles, R. M., and MacDonald, H. A. 1962. Crop Sci. 2,299-302. Silow, R. A. 1931. Welsh Plant Breed. Sta. Bull. 12,234-240. Smith, D. 1962. Crop Sci. 2,75-78. Smith, D., and Nelson, C. J. 1967. Crop Sci. 7, 130- 133. Smith, J. B. 1956. M.S. Thesis, Cornell University, Ithaca, New York Smith, J. H. 1955. Ph.D. Thesis, Cornell University, Ithaca, New York. Spiss, L., and Paolillo, J., Jr. 1969. Crop Sci. 9, 173-176. Stebbins, G. L. 1950. “Variation and Evolution in Plants.” Columbia Univ. Press, New York. Stickler, F. C., and Wassom, C. E. 1963. Agron. J . 55,78. Tesar, M. B., Lawton, K., and Kawin, B. 1954. Agron. J . 46,189-194. Tome, G . A., and Johnson, 1. J. 1945. J . Amer. SOC. Agron. 37, 101 1-1023. Trimberger, G . W., Kennedy, W. K., Reid, J. T., Loosli, J. K., Turk, K. L., and Krukovsky, V. N . 1962. New York, Agr. Exp. Sta., Ithaca, Bull. 974. Twamley, B. E. 1967. Can. J . Plant Sci. 47,603-609. Upadhyay. J. 1965. Plant Dis Rep. 49,724-728. Upadhyay, J. 1966. Phytopathology 56,760-765. Van Keuren, R. W., and Davis, R. R. 1968. Agron. J. 60,92-95. Van Keuren, R. W., Davis, R. R., Bell, D. S., and Klosterman, E. W. 1969. Agron. J. 61, 422-425. Varney, K. E. 1958. Vermont, Agr. Exp. Sta., Bull. 608. Vianne, E. 1870. “Prairies et Plantes Fourrages.” Paris. Wakefield, R. C., and Skaland, N. 1965. Agron. J. 57,547-550. Wallace, A. T., and Killinger, G. B. 1952. Florida, Agr. Exp. Sta., Circ. S-49. Wedin, W. F., Vetter, R. L., Scholl, J. M., and Woods, K. W. 1967. Agron. J. 59,525-528. Wernsman, E. A., Keim, W. F., and Davis, R. L. 1964. Crop Sci. 4,483-486. Wernsman, E. A., Davis, R. L., and Keim, W. F. 1965. Crop Sci. 5,452-454. Wiggans, S. C., Metcalfe, D. S., and Thompson, H. E. 1956. Agron. J . 48, 281-284. Williams, W. A. 1953. Agron. J . 45, 331-333. Winch, J. E. 1958. Ph.D. Thesis, Cornell University, Ithaca, New York. Winch, J. E., Anderson, G. W., and Jones, G. F. 1968. Can. J . Plant Sci. 48, 393-399. Winch, J . E., and MacDonald, H. A. 1961. Can.J. Plant Sci. 41,523-532. Wittwer, L. S., Kennedy, W. K., Trimberger, G. W., and Turk, K. L. 1958. New York, Agr. Exp. Sta., Bull. 931. Wojciechowska, B. 1963. Genet. Pol. 4, 53-63. Wood, G . M. 1958. Vermont Farm & Home Sci. 3,4. Yawalker, K. S., and Schmid, A. R. 1954. Agron. J. 46,407-41 I . Zandstra, I. I., and Grant, W. F. 1967. Can. J. Bor. 46,557-583.

This Page Intentionally Left Blank

THE CONFIGURATION OF THE ROOT SYSTEM IN RELATION TO NUTRIENT UPTAKE K. P. Barley Waite Agricultural Research Institute,

1.

II.

111. IV.

V.

VI.

Glen Osmond, South Australia

Introduction ....... ........................................................... Geometrical ption A. Methods ........................................................ ..................... B. Principal Geometrical Features ....................................................... Nutrient Transference in the Soil ................. .................... Physiological Conditions Governing Uptake ......... .............. A. Uptake Parameters........................................................................ B. Variation between and along Roots .................................................. The Influence of Configuration on Uptake ...................................... A. Effective Radius ............................................................................ B. Abundance and Density ................................................................. C. Distribution .... ............................................................... D. Root Elongatio ient Uptake .............................................. Conclusions ........... ........... ............... References ..... ...............................................................

I.

Page I59 16 1 161 164

I67 171 171

175 177 177 181 186 194 197 198

Introduction

After Bray (1954) aroused interest in the idea that the “mobility” of a nutrient in the soil might have a strong influence on its supply to plants, attempts were made to define just what was meant by “mobility,” and to measure the properties concerned. Tepe and Leidenfrost (1958) measured the extent to which an ion exchanger depleted a column of saturated soil, and found that the length of the depleted zone varied from 2 mm for phosphate to 2 cm for nitrate and chloride. Klute and Letey (1958) introduced a suitable method, and values of the apparent diffnsivity for given ion species in the soil were measured for the first time by Schofield and Graham-Bryce ( 1960) and by Porter et al. (1960). Because the values of the apparent difisivity found for ions in soils were one or more orders of magnitude less than those known for dilute aqueous solutions, soil scientists were led to believe that soils offer a I59

160

K. P. BARLEY

large resistance to the transfer of nutrients to the plant root. For example, in their review “Movement of Nutrients to Plant Roots” S. R. Olsen and Kemper ( 1968) conclude that “although nutritional adequacy has historically been characterized most often by the amount of nutrient in the soil, the rate at which the nutrient can move . . . is an equally important factor.” This may be so, but it needs to be emphasized that the resistance to nutrient transfer cannot be inferred from knowledge of soil properties alone; nor is it sufficient to know in addition how well the roots can absorb. Just as the resistance to the conduction of heat offered by an insulating body depends not only on the thermal properties of the materials of which it is composed, but also on its size and shape, so the resistance offered by the soil to the transfer of nutrients to the root depends upon the size and shape of the paths along which nutrients must travel. The paths are determined chiefly by the configuration of the root system. Although they emphasized the influence of the properties of the soil itself, S. R. Olsen and Kemper recognized that the form of the root system was important also, and referred briefly to the latter in Section V of their review. The present account pursues the topic further. Qualitatively it is apparent that the shape and extent of root systems influence the rate and pattern of nutrient uptake from the soil. Although numerous empirical studies of this topic have been reported, the facts gathered have too often contributed little to understanding. Clearly a theoretical framework is needed to structure our knowledge of the subject and to guide intelligent management of nutrient uptake. In soil-plant systems the number of states that exist is so vast in relation to the number that can be examined in experiments, that blind empiricism is unlikely to get us very far. Apart from its didactic value, the role of theory is to motivate and guide the design of experiments. To the extent that theory can be quantified, it can be used to predict the values of variables and parameters that may most profitably be examined. Experiments test the value of hypothetical solutions, and, similarly, the results of management test the value of predictions. At the outset we need to remember that the configuration assumed by the root system is influenced markedly by nutrient supply. Since 1892 when Nobbe conducted his classical experiments, it has been known that, when plants suffer from an overall deficiency of nitrogen, their roots branch more in regions where the soil is locally enriched with nitrogen fertilizer. Numerous other examples of this kind of growth response have since been reported (see, for example, Sayre, 1934; Duncanand Ohlrogge, 1958). The influence of mineral nutrition on root growth is too large a subject to be dealt with here; the reader is referred to Troughton (1957)

THE CONFIGURATION OF THE ROOT SYSTEM A N D NUTRIENT UPTAKE

161

and Viets (1965) for surveys of evidence from the field, and to recent papers by Brouwer and Loen (1962), May et al. (1965, 1967), and Hacket ( 1968a,b) for physiological interpretations. II.

Geometrical Description of the Root System

A. METHODS The form and extent of the root system can be described by mapping the roots as they appear on the vertical and horizontal faces of a trench. The classical investigations of Weaver (1926) were conducted in this way. The method records the distribution of the main roots and the thicker laterals only, as it is not practicable to dissect fine laterals from the soil in the field. A more complete picture of the root system may be obtained by washing the soil away from excavated prisms to expose the roots. To help retain the roots in position, sets of pins are usually pressed into the soil before the prism is excavated (Blaser, 1937). Such methods are too laborious and crude for adequate quantitative data to be obtained, but they provide a useful qualitative picture of the root system. Roots can be separated from loose soils by soaking and washing away the soil, preferably by agitating the samples in screens in a suitable washing machine (Gates, 1951). The roots can be separated from coarse soil particles retained on a screen by elutriation in a moving current. Many soils are difficult to wash away, but good results can be obtained with nearly all soils if fragmented samples are first soaked in sodium hexametaphosphate solution, and then dispersed with a high speed stirrer for a short time. After this treatment root sections, together with other macroscopic pieces of organic matter, are eluted and retained on a fine screen (Barley, 1955). The separate is floated in a dish, and the total root length is determined by counting the intercepts made by the short sections of root on randomly distributed line transects. The transects are conveniently formed by a hairline in the eyepiece of a low power microscope. The diameter of the roots can also be determined with an eyepiece graticule. Following Newman (1966), the total length of root per separate L is given by

where N (cm-') is the number of intercepts per unit length of transect and A (cm2) is the plane area on which the root lengths and transects are distributed. As fragmented screened soil is employed in the above method,

162

K. P. BARLEY

composite samples can be prepared from sample cores in the conventional manner. There is no need to remove foreign material from the separate unless root weight or composition is also required. Total root length and the length of hair-bearing root may be recorded separately using a twochannel counter, the latter quantity being of particular interest when lengths are correlated with nutrient uptake. The local distribution of roots can be described by locating root intercepts on polished sections of the soil, the sections being made at randomly chosen loci in the direction of the principal planes of symmetry of the root system. In an early study, Fitzpatrick and Rose ( 1 936) counted the ends of roots protruding from the faces of soil cores. More recently the method has been refined by the use of polished sections and low power microscopy (Barley and Sedgley, 1961; Melhuish, 1968). The nature of the distribution -random, under-, or overdispersed -can be ascertained by measuring distances to the nearest and successive neighbors, the distribution of neighbors of all orders being related to the xz distribution (Thompson, 1956). When polished sections are prepared for examination of the root distribution, root length can be estimated from counts of the number of roots intercepted. The theory of random lines in three-dimensional space shows that, when numerous straight lines intersect unit cube, the mean length of the secants is 2/3 (Kendall and Moran, 1963, p. 76). If we represent roots by straight segments meeting at bends, then, provided the individual segments within a given volume of soil can be considered to be located and oriented at random, so that the intercepts which they make with each principal plane are distributed at random, it follows easily from the above result that Lv = 2m (2) where Lv(cm-2), the rooting density, is the length of root per unit volume of the soil, and m (cm-2), is the arithmetic mean of the number of root axes intercepted per unit area of plane for the three principal planes. In practice (2) provides a useful estimate of Lv for well established'root systems in uniform soils (Melhuish and Lang, 1968). Where the root disstribution is markedly anisotropic, as in the early seedling stage when the main roots and first-order laterals are growing geotropically, or in coarsely structured soils, Lv can be found from the anisotropy and the sum mI mII mrrIfor the three principal planes (Lang and Melhuish, 1970). As noted in Section I a key item of information needed in uptake studies is the apparent path-length for the transfer of nutrients to a set of roots. Usually we wish to know the volume fraction of the region containing a

m.

+ +

THE CONFIGURATION OF THE ROOT SYSTEM A N D NUTRIENT UPTAKE

163

nutrient that lies within any given distance of the root network in that region. The probability distribution dPld6 of the distance 6 from random points of origin to the nearest point of contact with slender, straight rods distributed randomly in three-dimensional space has been derived by Ogston ( 1958), Eq. ( 1 2). Provided that the root segments are randomly located, we can employ Ogston’s solution, and, by integration,

where E6O

(1 1)

where V,, is the maximum rate of uptake analogous to that defined by Eq. (lo), but expressed per unit area of the surface of a root zone of radius q. They show also that, provided the initial concentration c, < p, a good approximation may be obtained by substituting the linear boundary condition f=-ac,

r = q, t > 0

(12)

where a (cm sec-I) is the apparent surface conductance of the root (analogous to the surface conductance in the theory of heat conduction), and is defined by T,,= aC,

when C

= c,

(13)

Strictly, neither condition ( 1 1 ) nor the linear approximation (12) hold at extremely low concentrations, since there is a limit below which the plant cannot achieve any net uptake. The limits are of the order lo-' M for nitrate (C. Olsen, 1950), and M for phosphate and potassium (Asher and Loneragan, 1967; Williams, 196 1).

THE CONFIGURATION OF THE ROOT SYSTEM A N D NUTRIENT UPTAKE

173

The values of V , p , a, etc., given in the physiological literature have usually been found by measuring the uptake of ions by sets of excised roots, when the uptake time is limited to a few hours, or by the root systems of whole seedlings or young plants, when the uptake time may be several days. We need to know whether such values apply over longer periods of time. Care needs to be taken when averaging uptake values for a set of roots. From transfer theory we know that components of the ion flux are influenced by the product aq. For reasons given below the values of a and q for the set of root zones comprising a root system are likely to be correlated, so that, statistically (Yq # (1! (for zero correlationFy i = 0). Yet physiologists conventionally find the value of a (or more sophisticated alternatives) by expressing the uptake as uptake per unit root abundance and relating this to solution concentration. For present purposes we can avoid this difficulty and define 7,by

where Q (moles cm-’) is the uptake per unit length of root. Over a period of weeks the continuation of uptake clearly depends upon the growth of the plant, and, as shown by Nye and Tinker (1969),

where W (8) is the weight of the plant, L (cm) is the length of root per plant, and u (moles g-’) is the gravimetric concentration of the absorbed element in the plant. Note that when u is constant Cur, depends upon dWldt. Equations ( 14) and ( 15) suggest one possible approach by which nutrient uptake models of the kind described in this review may be linked with models of photosynthesis. Values of 7,derived from Eq. (14) do not depend upon hypotheses about uptake mechanisms. If 7) is assumed to be constant, Eq. (14) may be used to derive 6 from data of Loneragan and Asher (1967) for the steady uptake of phosphate by a number of species from continuous flow cultures over 4 weeks. For a constant radius of 0.02 cm the estimated values of CU are of the order of cm sec-’, and this is within the range found in short-term experiments with low-salt plants. So far we have considered the physiological conditions governing nutrient uptake without reference to effects of the concomitant uptake of water by the root. As indicated in Section 111 the water flux density Y,, at the root surface r = q depends not only on the transpiration E , but also on the abundance of the roots. If by analogy with the well known Leaf Area Index (LAI) we term the product 27r$ LAthe “Root Area Index” (RAI),

174

K. P. BARLEY

then, as LA is of the order of lo2cm-' under crops, and may be as high as lo3 cm-' under grasses (Newman, 1969), and as is commonly about 0.02 cm, the RAI is likely to be of the order 10' or lo2. Since E rarely exceeds 1 cm day-' (1 x loe5cm sec-I), provided a reasonably high proportion of the root surface is operative, v,, is unlikely to exceed cm sec-', and will more commonly be of the order of lo-' cm sec-'. Assuming for the present that at ordinary values of v,, effects of water uptake on the conductance a are small, then, when v,, > a , ions accumulate at the root surface and diffusion is away from the root; when, as is usual, v,, 1-10 mM. As the rate of transpiration rises, the selectivity of the root tends to decline (Pitman, 19651, and passive uptake to the shoot tends to increase, the most straightforward example being the positive effect of transpiration on the uptake of Si by oats (Jones and Handreck, 1965). In experiments with well stirred solutions when u is expressed per unit area of root surface, we can generally disregard the area of the root hairs. This is because ions from the ambient solution pervade the free space of the root, uptake through the plasmalemma can occur throughout the cortex, and the hairs add little to the area of the interface between free space and plasmalemma. Provided c, = co this is likely to hold in the soil

THE CONFIGURATION OF THE ROOT SYSTEM A N D NUTRIENT UPTAKE

175

also. When this is so the uptake per unit length Q is correlated with q P (volume per unit length) (see, for example, Fig. 4 of Russell and Newbould, 1968). But in the soil, when c,,/co+ 1 , and particularly if the soil is relatively dry, most of the uptake is likely to occur via the root hairs (Section V). Since ions tend to accumulate in the hairs (Lauchli, 1967), the chief regulating barrier then resides in the plasmalemma lining the inner tangential wall of the epiderm and/or in the plasmalemma of the outermost file of the cortex. When this is so Q will be correlated with r ) (surface area per unit length). As we shall see below this description is too simple, but it serves to show that the microscopic flow path followed by ions within the root may not be the same when the root is in a stirred solution as when it is in the soil. This raises doubts about the utility of determining plant parameters such as a by measuring uptake from stirred solutions; at least we need to compare conventional values with those found by measuring Q in soils of known properties, and fitting transfer equations such as Eq. (8) (Clarke and Barley, 1968). B.

VARIATION BETWEEN A N D ALONG

ROOTS

Not all roots are concerned primarily with the absorption of water or nutrients, and root form often shows adaptation for the performance of functions other than absorption. The fleshy roots of many perennial and biennial dictotyledons, for example, act as organs of storage, as does the cortex of the main roots of many monocotyledons; even laterals may be modified to form storage organs in certain species. The prop-roots of tall grasses show mechanical adaptation; for example the proproots of maize have thick double rings of fibers. The roots of plants adapted to wet places frequently have abundant aerenchyma; and, while aerenchyma is most conspicuous in the marsh plants, it is by no means confined to them. The various adaptations may give rise to obvious dior trimorphism within a root system (Kokkonen, 1931; Jacques, 1937; Barley, 1953). Dimorphism may also result from the presence of mycorrhizal roots, and mycorrhizal associations are common in grasses including the cereals, clovers, and horticultural plants. It has been shown that mycorrhizal roots can sustain their ability to absorb phosphate for much longer periods than uninfected roots (Bowen, 1968), and the possible role of mycorrhizas in the nutrition of crop and pasture plants deserves more attention. Along the length of a root characteristic differences in form and structure are found corresponding with the various stages in development and

176

K. P. BARLEY

degeneration. Close behind the elongating tip root hairs arise, and in the hair zone the root is “glued” to the soil by its mucilage (Section V); elsewhere it separates easily from the soil; in older zones, owing to exfoliation of the outer cells, the central cylinder is often left within a wider air-filled channel (see Fig. 2b of Head, 1968). When the outer cells degenerate, the outermost intact cortical layer generally becomes suberized, when it is termed an exodermis. Although water and solutes penetrate the exodermis when the root is immersed (Kramer and Bullock, 1966), this is less likely when the root is in the soil, as the suberized zones are poorly wetted. Given the known differences in form and structure between and along the roots of a plant, it is obvious that simple relations between Q and r) of the kind mentioned in Section IV, A are unlikely to account for more than part of the variation in physiological uptake ability within a root system. Using a series of potometers, Grasmanis and Barley (1969) found, for example, that in stirred solution QNon/r)or QNo3/q2varied by a factor of 4 or 5 along the length of the pea radicle, most of the variation being associated with differences in protein content between the zones. Recently the study of uptake and translocation along the root has been expedited by the technique of scanning the root after uptake from radioactive solutions (Bowen and Rovira, 1967). The technique also lends itself to the study of uptake by different members of the root system (Bowen and Rovira, 1969). In either case careful account must be taken of the influence of isotopic exchange. An example of the results is given for seminal roots of wheat in Fig. 2. After the roots had been treated with 5X M p h o ~ p h a t e - ~ in ~ , calcium ~ ~ P sulfate solution for 15 minutes, some of the plants were removed for scanning, and the remainder were transferred to phosphate-:”P in calcium sulfate for a further 210 minutes to allow translocation of the 32P.The scan at 15 minutes (Fig. 2) shows the usual subapical peak (see also Brown and Cartwright, 1953; Grasmanis and Barley, 1969) and a second peak in the zone where laterals were developing. Much of the absorbed %&P was retained in the growing tip (see also Kramer and Wiebe, 1952), but it was translocated readily to the tops from zones proximate to the tip and from the laterals. Finally it is known that the various members of the root system differ in their relative rates of absorption of different nutrients. For example, Russell and Sanderson (1967) showed with small potometers that the ratio QP/Qsrfor first-order laterals of barley was twice that for the main root axes. We conclude that, while form may be important, the pattern of nutrient uptake depends also on physiological differences and gradients in the root system.

THE CONFIGURATIONOF THE ROOT SYSTEM A N D NUTRIENT UPTAKE

0

SCALE

5

177

10cm

FIG.2. Distribution of 32Pin a wheat seedling (Bowen and Rovira, 1969). Top: After 15 minutes in 5 x 10-RMph~sphate-~l.~*P. Bottorn:After a further 210 minutes in 5 x M p h ~ s p h a t e - ~ ~Only P . one of five seminal roots is shown. The radioactivity of laterals is included, and the position of the most distal lateral is shown by the arrow. V.

The Influence of Configuration on Uptake

A.

EFFECTIVE RADIUS

1. The Axial Part of the Root As noted in Section IV, when cq co, uptake occurs throughout the cortex and we can disregard the root hairs; also, for purposes of illustration we can treat the problem deterministically and assume that T and hence a are likely to be related to q2.Provided the roots do not interact, the influence of r) on uptake per unit surface area can be found by making the appropriate adjustment to a, and referring to solutions of Eq. (8) subject to Eq. (12). Carslaw and Jaeger [( 1959, p. 337, Eq. (IS)] provide a solution for the case v = 0, from which it can be seen that, when t is of the order of days, as 7 decreases, the influence of the parameter qalk,8 outweighs that of KLt/r)*,so that the rate of uptake per unit surface area increases. For the ordinary range of values of other variables, q has an appreciable influence when (Y =s cm sec-l. Furthermore, the surface area of a given weight of roots varies inversely with r). Nye ( I 966) borrows likewise from Carslaw and Jaeger, but he underestimates the effect of r) through assuming that a is independent of r). This , that Q is likely to be directly prois appropriate only when c , , ~ ~ , ,so portional to q. But when this is so the root hairs are influential, and r) no longer determines the effective radius of the root. Steady state solutions of Eq. (5) for v > 0 and subject to various boundary conditions are given by S. R . Olsen and Kemper ( 1968, p. 13 l ) , from which it can be seen that r) appears in a logarithmic term. 2 :

178

K. P. BARLEY

2. The Irlfluence of Root Hairs As noted in Section IV, when c,, 4 co, most of the uptake is likely to occur via the root hairs. Before evaluating the role of the hairs in uptake, we need to know the time for which they continue to absorb nutrients. Unfortunately such information is meager. Hairs may persist for long periods in many species, but they do not necessarily continue to absorb ions, as the walls of persistent hairs may thicken or even become lignified (McDougall, 1921). Root hairs generally collapse after a few days or weeks, but in the Gramineae, and in the cereals in particular, the hairs tend to persist, thin-walled hairs being found on all parts of the root system. A recent report that the root hairs of barley, Hordeum vulgare L., live for only 2 days (McElgunn and Harrison, 1969) should be discounted, as the roots were illuminated brightly and stained at hourly intervals with neutral red, and this treatment kills cells in the barley root (Patterson, 1941). In dicotyledons having roots that show extensive secondary thickening -and these include the common pasture legumes (Soper, 1959)-the root hair cells are lost together with the rest of the epiderm as thickening proceeds, but in other dicotyledons the hairs may persist for weeks or even months (Whitaker, 1923). a . Qualitative Efects. The most important function of root hairs may well be the maintenance of liquid continuity between water in the cell wall and pore water in the soil. Main roots and laterals are far too wide to occupy the narrow voids into which water menisci retreat as the soil dries. Even those voids that can just be entered by hairs (radius = 5 p ) drain at a suction of only 0.3 bar. However, the hairs are more effective than this value would indicate, as their walls secrete mucilage, and the mucilage infiltrates into finer pores. The author has observed that more mucilage is secreted when the soil is dry than when it is moist. Although the existence of 3 “junction resistance” has not yet been established for hairless roots grown in situ, a resistance of this kind has to be invoked to explain the low ion uptake observed when roots are disturbed and pressed back onto the soil (Clarke and Barley, 1968; see particularly their Fig. 7). In addition to any effect that root hairs may have on junction resistance, the significance of local alteration of the soil around the hairs needs to be considered. Local changes in sorption characteristics resulting, for example, from pH shifts, or complexing with diffusible exudates or rhizosphere products, can set up diffusion gradients either toward or away from the root. Wilkinson et al. (1 968) show how such effects can lead to local jepletion of calcium independently of uptake. b. Geometrical Efects. The geometrical effects of root hairs on nutri-

THE CONFIGURATION OF THE ROOT SYSTEM AND NUTRIENT UPTAKE

179

ent transfer lend themselves to mathematical description. The choice of boundary conditions in published investigations has been highly arbitrary. Bouldin ( I96 I ) treats the problem as though the hairs acted independently. But the hairs are so closely spaced that, even in highly buffered soils, the zones of depletion around individual hairs begin to overlap after about a day. Further difficulties arise when the hairs are commensurate in width with the soil granules, as bulk values of the transfer coefficients are then unlikely to apply in the vicinity of the hairs. One line of approach adopted first by Passioura ( 1 963) is to replace the real root with a suitably defined equivalent cylinder of radius a , and to consider transference within the composite cylindrical region bounded internally by the epidermal ring of radius r), and defined by r) -= r s a s r < @J. In certain circumstances we can predict the influence of hairs on uptake with reasonable confidence. In particular, if the region occupied by the hairs is exhausted in a time that is negligible compared with the period of uptake under consideration, we can adopt the condition

-

where c' is the volumetric concentration of extractable forms of the element considered, and a = r) 1, where 1 is the length of the root hairs: and define the uptake per unit length of hairless root as

+

and that of a root with hairs as

The above condition will hold for a lightly buffered ion, when the root has a high propensity for uptake, and the flux through the soil is small. Nye ( 1 966) considers the nonconvective case Y = 0, for which an explicit solution is available [Carslaw and Jaeger, 1959, p. 336, Eq. @)I. As one illustration of his results we find that when r) = 0.02 cm, a = 0.1 cm, k, = 1 X lop8cm2sec-l, andj= 1, the uptake over 1 week is doubled by the hairs. A steady-state solution is readily obtained for Y > 0 [Gardner, 1965, p. 567, Eq. (49)], from which it can be seen that the influence of the hairs diminishes as v increases.

180

K. P. BARLEY

In the cases that we have been considering, apart from the initial uptake, the flow of ions occurred solely from the bulk of the soil to the region occupied by the hairs. At the other extreme we consider conditions in which depletion occurs only within a thin shell of soil in close proximity to the epidermal cells, and in which repletion is so slow as to be unimportant. This may hold, for example, for the uptake of manganese from those soils, in which uptake is thought to depend upon the “constant reduction” of insoluble MnOr (Passioura and Leeper, 1963). As hair growth commonly increases the surface area of the outer epidermal walls from 2 to 10 times, the hairs are potentially influential in the uptake of such nutrients. Their actual effect depends, however, on the extent to which they grow through voids rather than on or in the soil matrix, so that

where a is now the radius describing the effective surface area of the root, and S = g(t) is the rate at which the element is transferred to the root per unit area of the surface so defined. c . Experimental Studies. The presence of absorbed P and Sr in root hairs has been shown directly with the X-ray microanalyzer (Lauchli, 1967). Similar resolution does not appear to have yet been achieved in studies of the depletion of soil around roots. Conventional autoradiography can show the macroscopic pattern of depletion around roots, and, following the observations of Walker and Barber (1961) on the depletion of Rb, a variety of such patterns has been observed. As noted above, great care needs to be taken in inferring conclusions about uptake from observations of local depletion around roots. When there is independent evidence of uptake commensurate with the depletion and ion convection is small, the interpretation of depletion patterns is relatively straightforward. Observations particularly pertinent to the root hair problem have been made by Lewis and Quirk (1967),who worked with an acid soil known for its high phosphate sorption. They showed that, after 5 days uptake, the zone in which P was depleted around wheat roots had a clearly delineated boundary that coincided with the tips of the root hairs. Moreover the width of the zone changed little in the subsequent 26 days, and was little altered by a 3-fold increase in the added phosphate: these observations suggest that the width of the zone was governed chiefly by the length of the hairs. The chief difficulty in measuring the effects of the hairs on uptake is to obtain a suitable control. Clearly what is required for comparison is a

THE CONFIGURATION OF THE ROOT SYSTEM A N D NUTRIENT UPTAKE

18 1

hairless root in an equivalent physiological state; comparisons with hairless roots obtained by altering the conditions of the culture are of little value as the roots may differ anatomically and physiologically in many ways besides the presence or absence of hairs. Champion and Barley ( 1969) have shown that it is possible to grow roots along clay slopes with and without penetration of the clay by hairs by controlling the mechanical state of the clay. This approach may prove profitable in uptake studies. In horizons where peds are highly developed, roots tend to be clustered in the macrovoids. The peds are often coated with cutans; as certain cutans are known to impede ion diffusion (Soileau et al., 1964), the question of whether or not root hairs are able to penetrate the surface of peds has some significance for nutrition. Champion and Barley show that root hairs are capable of penetrating moderately resistant, remolded clays, and that they do not grow only in existing voids. In describing the effects of root hairs on the uptake of exchangeable ions, we may expect that in general k,, K , , will have one set of values in the region 7 s r s a and a second set of values in the region r > a. Perhaps the most profitable approach to adopt, since we cannot measure the values in the inner region directly, is to measure the bulk values of the transfer coefficients together with the surface conductance a at r = 7 and the uptake, Q. The required value of k, = K , may be found using a nonreactive ion, and the case k, # K , may then be examined using a reactive ion having the same charge.

B. ABUNDANCE A N D DENSITY 1 . Abundance

Under this heading we consider the problem of the minimum length of root needed to meet plant demand for absorbed nutrients. For spaced plants the relevant measure of abundance is length per plant; for closed communities it is often more convenient to refer to LA,the length of root under unit area of ground surface. We deal first with the case in which there is no external resistance to nutrient transfer. This is exemplified by uptake from well stirred solution, when the rate of uptake depends only on the conductance a and the solution concentration C. In general, a increases with whole plant demand, and decreases as the plant approaches salt-saturation. The demand per unit length of root can be varied either by amputating part of the plant, or, with less physiological disturbance, by dividing the roots between a complete solution and one lacking a specified nutrient. Following the latter approach, Gile and Carrero (1917) found that the rate of uptake of

182

K. P. BARLEY

N , P, K, or Fe per unit weight of root by young maize plants rose to a maximum as the proportion of roots being supplied was reduced. A maximum of this kind pertains to a specified value of C, and should be distinguished clearly from the maximum I/ found when supply is nonlimiting (see Section IV). The 50-year-old experiment of Gile and Carrero needs to be repeated using shorter uptake periods and a wider range of concentrations; but in the absence of more recent data we use their results to obtain a crude estimate of the minimum length of roots required to meet a given demand. If we assume that their maize roots had a specific length of 5 x lo3 cm g-I (author's data for maize), the maximum rate of uptake of nitrate measured by Gile and Carrero is equivalent to 4 peq cm-' sec-I. The rate of uptake of N03-N by cereal crops in the tillering phase is generally about 1 kg ha-' day-' (8 peq cmT2sec-'), so that, if there were no external resistance, this supply could be maintained when LA > 2 cm-l. This value is exceeded after 1 or 2 weeks growth of cereal crops, and by the late tillering stage LA usually exceeds 200 cm-*. Greater lengths of root will be needed to meet the plant nutrient requirement in the soil than in stirred solution, because the soil offers a resistance to ion transfer, and because roots compete in depleting the soil. Moreover for most nutrients c is buffered at low vdlues. (We note, without considering further here, that when convection brings ions to the root faster than they can be absorbed, the resistance of the soil to back diffusion serves to raise c at the root surface above the initial concentration.) Subject to the physiological maxima described above, the maximum rate of uptake from the soil that can be achieved by a root system can be predicted by treating the roots as a perfect sink (condition of Eq. 16). When the roots are spaced widely so that they act independently, their steady rate of uptake subject to condition (16) is given by Gardner [ 1965, p. 567, Eq. (49)].* To take one example: if c, = 1 me 1-', 7 = 0.02 cm, v,, = 1 X lo-' cm sec-I, k , = 1 X 10-6cm2sec-', O=O.l,andj= 1, then Q = 0.17 peq cm-' sec-', and an uptake of N 03-N at the rate of 1 kg ha-' day-' could be maintained when LA9 47 cm-'. In general, however, the roots may not act independently, and in the next section we consider what happens when they compete during uptake.

2. Densiry The resistance offered by the soil to the transfer of ions depends not only on soil transfer coefficients k,, K, etc., but also on the rooting density Lv.Where the roots have a reasonably uniform propensity for uptake, we *Incorrectly; the exponent in Gardner's Eq. (49) should be w/2?rDO.

THE CONFIGURATIONOF THE ROOT SYSTEM A N D NUTRIENT UPTAKE

183

may profitably define a radius of influence b (cm) as

The value of b in topsoils under dense crops or pastures is usually less than 2 mm even when the root hairs are neglected: and, as mobile ions can move rapidly through distances of a centimeter a day in moist soils, roots are certain to compete strongly in the top soil. a. Theoretical. Although the problem of transference to sets of competing roots may at first appear intractable, it can be solved if, following Philip ( I957), we consider radial transfer in the hollow cylinder 1) s r s b for the present we ignore any effect of root hairs- subject to the condition

aclOr= 0,

r = b,

t >0

For v = 0 explicit solutions of Eq. (8) are available subject to condition (21) above and to certain specified conditions operating at the root surface, r = 17 (Carslaw and Jaeger, 1959, pp. 334-339). For v > 0, Eq. (8) may be solved numerically (Passioura and Frere, 1967). To provide a concrete illustration of the effects of rooting density on ion uptake the solution of Eq. (8) subject to condition (21) and the condition of constant conductance a defined by Eq. (12) has been evaluated for u = 0, for specified values of k,, 8 , and 7 and t = 4 days. Remembering that the value of LITin top soils under dense crops or pastures varies from 2 to 50 cm-2, it is apparent from Figs. 3a and 3b that when a> 1 X lops cm sec-’ and j = 1, more than sufficient roots are likely to be present to deplete the topsoil thoroughly and rapidly. When a < lopscm sec-I and j 3 100 uptake is linearly related to L,. over all of the above range. This suggests that plants with a low rooting density are likely to be susceptible to deficiency of the less mobile nutrients and there is some evidence to support this view (Cornforth, 1968). Transpiration (v > 0) tends to increase the rate of uptake at any given value of Lv,but this hardly affects the generality of our conclusion for ions that are strongly buffered by adsorption, as the solution concentration, and hence the convective flux, of such ions is always small. Root hairs also tend to increase the rate of uptake at given values of LIT.When effects of hairs are mainly geometrical they may be expected to displace the curves shown in Fig. 3 so as to increase the rate of depletion. But the effects of hairs on competition between neighboring roots are far less certain when local qualitative changes in the soil occur. A given uptake may be satisfied, for example, by the depletion of a relatively narrow shell of soil when the sorption

184

K. P. BARLEY ..

0

2 ROOTING

4 6 DENSITY (cm-2)

'""I

-%

-

8

I

80-

60-

a

2 ROOTING

4 6 DENSITY (cm-*)

8

FIG.3 . The influence of rooting density L I .on the depletion of the labile pool. ( T o p ) for cm sec-' and different buffering capacitiesj, (borrorn) for j = 10 and roots of a =' 1 X cm2 sec-'; t = 4 days; different surface conductance a. v = 0 ; 7 = 5 x lo-' cm; k, = I x 6, = 0.2.

capacity is decreased by hair exudates. We can now see the importance of the root hair problem outlined in Section V; its solution is needed before we can describe the effects of rooting density on uptake with greater realism. b. Experimental. Although it is difficult to control the rooting density precisely, a range of densities can be obtained with a variety of methods. The main difficulty is to avoid confounding the density with changes in whole plant demand or with the total supply.

THE CONFIGURATION OF THE ROOT SYSTEM AND NUTRIENT UPTAKE

185

If density is varied by amputating some of the roots, there may well be an unknown compensatory increase in the conductance a of the remaining roots in response to the greater plant demand per unit length of root. Insofar as variation in a is likely to accompany actual variation in LIr,this is acceptable to the experimenter. Of more concern is the likelihood of compensatory growth, particularly in long-term experiments. Provided not too many roots have been amputated, the relative growth rate of the remainder increases so that there is little change in the absolute rate of increase of root dry weight per plant (Humphries, 1958). When amputation has been more severe the rate of increase in weight may decline, but, whether this happens or not, the remaining roots branch more frequently and lateral elongation is enhanced. Because of the compensatory production of laterals rooting density in the amputated treatment rapidly overhauls that in the control and may even surpass it (Brouwer, 1966). An alternative approach adopted by several investigators is to vary the plant density itself, or, in pot culture, the volume of soil per pot. Careful interpretation is needed as the rooting density is now confounded with the total supply per plant. Moreover, the rooting density near the base of the plant remains relatively high on all treatments, even though the spaceaverage value differs between treatments. Following this approach, Cornforth ( 1968) obtained results, which, despite the difficulties of interpretation mentioned, are indicative of the effects predicted theoretically in the preceding section. Cornforth’s method was to vary the depth of a nitrogen- and phosphorus-deficient soil in pots. His results for oats, TABLE I I I Root Concentration and Nutrient Uptake from Soil Columns of Different Depth (Cornforth, 1968) Soil depth (cm)

Parameter Oats Root dry weight (g/l) 4,.(meq/l soil) 4,.(meq/l soil) Kale Root dry weight (g/I) qS (medl soil) q p (meq/l soil)

7.6

1.2 2.0 0.16 -

-

15.2

30.4

60.8

L.S.D. ( P = 0.05)

-

-

2.0 0. I4

2. I 0.09

0.2 2.1 0.06

0.09 0.002

1.2 3 .O 0.2 I

0.8 3.1 0.15

0.4 2.9 0.1 I

0.15 0.6 0.03

186

K. P. BARLEY

Avena sativa, L., and kale, Brassica oleracea L., given in Table 111, show that the uptake of nitrogen per unit volume of soil ( q Nwas ) independent of rooting density for both plants, and this suggests that the deficient soil was depleted thoroughly at all densities. In contrast the uptake of phosphorus per unit volume of soil (qr) decreased as the rooting density decreased, the relative reduction in qf. being less than the relative reduction in density, probably because of the associated increase in the total supply. A better method of controlling density, free from the above effects of demand, supply, or compensatory growth, would be to divide the roots in differing proportions between given volumes of labeled and unlabeled soil having a common nutrient status. The author has been unable to find published data, which include a record of rooting density, obtained by such means. C. DISTRIBUTION

I. Local Distribution In Section V, B we treated the radius of influence as though it could be specified by a single value. In fact b represents a mean of a set of values, and we now consider effects that the nature of the local distribution may have on uptake at any given rooting density. Though quantitative data are lacking, local aggregation (clumping) of roots is likely to be common, particularly if soil peds are strong, when the roots tend to follow the larger voids (Edwards er al., 1964). Underdispersion might be expected due to local depletion of the less mobile nutrients in the zones around the earlier formed roots. However, the pattern resulting from competition is not simple and depends on the scale of observation. Competition may even give rise to fine scale aggregation, owing to younger roots growing preferentially in the gaps left vacant between the regions depleted by the older roots [compare with the whole plant distributions described by Pielou (1960)l. As noted in Section 11, A, the intercepts made by roots of any particular component in a mixture can be identified with autoradiography after labeling the tops of the component with a readily translocated radioisotope. In a two-component mixture, departures from random association can be detected by finding the freunquency of nearest neighbor pairs in the possible classes (labeled labeled -: --) and applying a x2 test (Pielou, 196 1). In this way Litav and Harper ( 1 967) examined the local mixing of the roots of two component mixtures of cereals and of a mixture of oats and peas. The null hypothesis of random mixing was tenable except when nitrogen

++, +-, -+,

+,

THE CONFIGURATIONOF THE ROOT SYSTEM A N D NUTRIENT UPTAKE

187

was applied to the leaves of one component. This led to undermixing. Undermixing also occurs when substances exuded from the roots of one plant inhibit those of its neighbors; this is known to occur in guayule (Parthenium argenratum A.) (Bonner, 1946). Overmixing might be expected if the roots of one plant make the local environment more favorable for those of another. This may occur, for example, when a nodulated legume and a grass are grown together, although, as noted above, Litav and Harper found no evidence of this in a mixture of peas and oats. In the absence of any published account, a highly simplified theoretical outline of the influence of the local distribution of roots on ion uptake is given below. We treat the problem as two-dimensional and consider radial transfer to a root of radius r). For any given set of roots a region of influence may be associated with any particular root by describing the area containing all points of space nearer to that root than to any other. When the centers of the roots are distributed randomly so that they constitute a Poisson field of points, the regions so defined are “Voronoi” polygons with a mean number of six sides (Miles, 1970). The area variance of the Voronoi polygons is given by Gilbert (1962) as 0.280 m-2, where m is the number of points (root axes) in unit area. For each polygon so defined we substitute a circle of equal area having radius Bi.For a random distribution of roots the variance of Bi is obtained immediately from the area variance given above. Higher order moments were not known to the author, but direct measurement of the areas associated with a sample of 300 points showed that the distribution of Bi fitted a gamma distribution. In addition, a slightly skewed distribution of Bi was obtained, without altering the mean and with only a negligible change in variance, by applying the transformation

where Bi = skewed value, Bi = gamma value, 4 = constant, s2 = variance of the gamma distribution, a n d 3 = mean of the gamma distribution. As in the density problem we consider the region r) 6 r c Bi and solve Eq. (8) subject to boundary conditions (12) and (21) for a suitable range of values of Bi. The uptake Qi for each value of Bi is then multiplied by the appropriate frequency, and finally the total uptake is found by summation. The nature of the distribution might be expected to exert its greatest influence at intermediate values of the density m and buffering capacity j. A range of parameter values corresponding to 0.2 6 m C 2, 1 S j 6 1000

K. P. BARLEY

188

was examined in terms of the model outlined above. The rate of depletion did not differ by more than 5% between the most efficient (regular) and the least efficient (4 = 1.5) distribution. It would be interesting to examine uptake by more highly aggregated sets of roots, but the model is unsatisfactory when the centers of equivalent circles deviate widely from those of the corresponding polygons. Moreover, real roots have hairs, and, when the roots are aggregated, overlap between the regions penetrated by the hairs of neighboring roots becomes important at relatively low densities. For the random distribution the probability of overlap has been derived by Roach, 1968, Eq. (4.1) and is given by P

=

1 - exp [-47rm(r)

+ fj2]

(23)

If r ) + f = 1 mm the hairs of one fifth of the total number of roots overlap when m = 2.0 cm-2. Several investigators have varied the local distribution of roots by growing plants in pots filled with resistant peds of coarse or fine size. Results are difficult to interpret as local distribution has invariably been confounded with rooting density, owing to differences in growth in the two grades of peds. As expected, the uptake of the most mobile ions (NOs-, C1-) is little altered by ped size, the soil being depleted rapidly whatever the ped size (Wiersum, 1962).

2. General (Macroscopic) Distribution

Gross genotypical differences in the distribution of roots are often of considerable ecological or agronomic significance. One example is pcovided by the history of pasture development on the “Ninety Mile Desert” of south eastern South Australia. In this region, which owes its name not to low rainfall but to dearth of nutrients in the sandy soils, depauperate scrubland has been converted to productive pasture by applying superphosphate together with trace amounts of copper and zinc salts and introducing exotic species. The pastures were at first based on subterranean clover, Trifolium subterraneum L. The residual value of the applied phosphate is high, but, when the superphosphate is withheld after several annual applications, the clover soon exhibits sulfur deficiency. If lucerne, Medicago sativa L., is grown instead, sulfur deficiency does not arise (J. K. Powrie, unpublished data). Most of the applied sulfate is leached from the topsoil during the winter, and it accumulates at 4 m in the deeper sands. This is well below the clover roots, few of which penetrate below

THE CONFIGURATION OF THE ROOT SYSTEM A N D NUTRIENT UPTAKE

189

1 m, but is within the depth explored by the lucerne. More of the water stored in depth is also recovered by the lucerne, and, in recent years, there has been a marked changeover to the deeper rooted legume. a. Impedance to Nutrient Transfer. Below a certain value of L,, the uptake of a given nutrient per unit volume of soil, q,, from a given layer or compartment in the root zone is likely to be directly related to L17(see Fig. 3). On a macroscopic scale it is sometimes convenient to consider the depletion of a layer or compartment of soil penetrated by roots rather than radial transfer to individual roots, and to introduce the concept of impedance to ion transfer I,, analogous to the impedance to water flow (Iw) defined by Gardner and Ehlig ( 1 962) and modified by Cowan ( 1 965). When convection predominates, the theory used by these authors to predict the pattern of water uptake from successive layers of the soil may also be used to predict the convective transfer of ions to the root surface. However it is difficult to extend this kind of analysis to provide a theory of ion uptake, because a variable surface resistance (lla) is present, and too little is known about (i) the variation of a along or between roots (Section IV, B), (ii) the influence of v or plant water suction on the value of the surface resistance, or (iii) the way in which ion uptake from one layer may influence the uptake from other layers. b. Correlative Studies. As so many uncertainties exist, it is best to proceed empirically at present and find the extent to which LVand qLare correlated. Overgeneralized notions of “uptake pattern” are too indefinite to be useful when dealing with nutrients, because the different nutrients are absorbed in amounts and proportions that alter with the stage of growth, and because their relative availability may vary widely with depth. Close correlation is unlikely even for particular nutrients, as uptake depends not only on the form of the roots but also on their physiological state (Section IV, B). The correlation of root system and uptake patterns has been expedited greatly by the advent of tracer methods. Indeed without such methods it would be difficult to describe the pattern of nutrient uptake. (We note here that tracer methods are also useful for studying the absorption of nutrients placed in particular compartments of the rooting zone. However, the topic of fertilizer placement is outside the scope of this review, as it requires proper consideration of local and general growth responses of the roots to added nutrients.) The simplest application of tracer methods is to find the rate at which root systems spread through the soil. This may be done by placing a readily absorbed and translocated isotope at chosen depths, and finding the time at which it can first be detected in the tops of plants immedi-

190

K. P. BARLEY

ately above and at various lateral distances from each site of placement. Phosphate labeled with 32Pis convenient for this purpose, as it does not move readily in most soils, and it can be detected easily with a portable rate-meter. The estimation of the rate of uptake from the various notional compartments into which the root zone may be divided is more difficult. Consider a compartment of the root zone of (arbitrary) unit volume within which the soil reacts uniformly with a given nutrient, and within which the concentration of the nutrient c,‘ and the proportion of roots of each kind or quality is uniform. The uptake within the compartment ql = g(Lv,c,’).Two cases will now be compared: First, let a tracer be mixed uniformly with the soil in the compartment so that its concentration is ci’. Provided that there is no effect of radiation on uptake (condition i) and that local equilibrium has been attained in the exchange between the added isotope and the labile ions in the soil (condition ii),

Also, provided that cl‘*4 cl’ (condition iii) and that the addition of the tracer does not alter LV (condition iv),

where H is a coefficient for the particular nutrient and compartment. Second, let the same amount of tracer be localized in a region occupying a volume fraction w of the compartment so that its concentration in the labeled region is cl;/o. Then, provided the above conditions remain satisfied,

as before, and, in either case,

The above conditions are not generally appreciated, but they need to be considered carefhlly in devising a working procedure. Practical difficulties such as contamination of nonlabeled compartments and physical disturbance of the soil during injection must be considered also. Condition ( i ) can usually be satisfied without sacrificing sensitivity by injecting small (10 ml) portions of solution at a number of locations within a chosen compartment. The usual advice is to inject the isotope near the middle of the compartment. As LV does not in general vary linearly with

THE CONFIGURATION OF THE ROOT SYSTEM A N D NUTRIENT UPTAKE

191

depth or distance from the base of the plant this will introduce bias, and the safest procedure is to inject at random within the compartment. Conditions (iii) and ( i v ) require that the amount of isotope together with any carrier added be negligible compared with the amount of labile nutrient present in the labeled region. Russell and Newbould ( 1 968) state that the added tracer has to equilibrate with a similar volume of soil in all compartments, but this is not necessary. They state also that lack of knowledge about the extent to which the tracer equilibrates with the pool of labile nutrient before absorption prevents the estimation of absolute values of qL.But this is not so except in short term studies, as diffusionequilibrium is not required, and local equilibrium in the exchange between the tracer and labile forms is approached closely within several days even in slowly reacting soils. Where slow reaction is a problem, departures from equilibrium can, moreover, be minimized by adjusting the concentration of carrier so that the added solution has approximately the same concentration as the equilibrium solution. The usual strategy is to inject the tracer into a single compartment of the root zone of each of a number of randomly chosen test plants. The alternative of sampling plants at a number of lateral distances from a common labeled compartment is more economic when the roots are well distributed, and either procedure is valid statistically provided that the depths of labelling are assigned randomly. Whichever strategy is adopted, the sampling grid has to be sufficiently wide to prevent interference, and this matter often requires preliminary investigation. The choice of isotope is governed by the element under study; but many investigators have begun with 32Pfor reasons given above, and because phosphate in the soil solution undergoes little isotopic exchange with root phosphate (Overstreet and Jacobson, 1946). With isotopes having a long half-life it is convenient to label the soil once at the beginning of an experiment, and sample and count over extended periods, rates of uptake being found by sampling at short time intervals. This procedure is often adopted with 32P,when the investigator has to accept the possibility that radiation may interfere with uptake at least in the early stages of the experiment, as initial doses of 100 FC or more per compartment are likely to be required. There is scope for the use of in place of 32P as the former has a longer half-life (25 days). The accuracy ofany method of measuring uptake pattern can be checked by comparing the total uptake per plant with the sum of the uptakes from the various compartments into which the root zone is divided. When the labeled regions are representative of their compartments, good agreement can be obtained in the field with tracer methods (Nye and Foster, 196 la).

192

K. P. BARLEY

If no suitable isotope of a particular element exists, a guide to the relative uptake to be expected from different compartments may sometimes be obtained by substituting an isotope of a closely related element. For example, 89Srhas frequently been substituted for 45Ca,as the latter is inconvenient to count because of its soft beta radiation. It is worth noting that this particular substitution overestimates the proportion of Ca absorbed from compartments of low Ca status, because the ratio qRSSr/qCa increases exponentially as the level of exchangeable Ca in the soil decreases (Andersen, 1967). The literature on nutrient uptake patterns is extensive but mostly unrewarding. Investigators seldom attempt to correct their data for differences in isotopic dilution between compartments. Very few accompany their data on distribution of nutrient uptake with data on the distribution of roots. Tracer methods have made it possible to examine critically older beliefs that the subsoil is an important source of nutrients. The older claims were based chiefly on excavation studies which showed that the roots of many crops penetrated to considerable depths. Weaver et al. (1 922) attempted to support such arguments experimentally, but, while the observations they made on the extent of root systems have lasting value, their uptake experiments lacked suitable controls. In retrospect their experiments seem to have been designed to demonstrate the correctness of convictions about the importance of deep roots rather than to test hypotheses. Tracer studies have repeatedly shown that deep roots can absorb and translocate nutrients readily, provided the nutrients are present in available forms. Nevertheless the subsoil is usually a poor source of nutrients, even though important exceptions occur, as when plants growing on sandy top soils absorb most of their potassium from underlying clays, or when nitrate, sulfate or other ions are leached into the subsoil. Low uptake of phosphate from subsoils is common, largely due to low availability, and to the presence of more abundant supplies of P in the topsoil. Nye and Foster ( I96 1b) provide the data given in Tables IV and V on the uptake patterns found under two annual crops-the cereal maize, Zea mays L., and the tap-rooted legume pigeon pea, Cajanus indicus, Spreng-and under a natural savannah of perennial grasses and deep-rooted shrubs. The observations were made on a phosphorus- and nitrogen-deficient soil, the crops being top dressed with superphosphate and sulfate of ammonia, each at 200 kg/ha. In the earlier stages of growth the pigeon pea absorbed most of its phosphorus from zones around the tap root, but the maize roots spread and absorbed more widely. Later in the season lateral roots of pigeon pea spread horizontally, so that by 80 days the pea

THE CONFIGURATION OF THE ROOT SYSTEM A N D NUTRIENT UPTAKE

193

TABLE IV Uptake of Phosphorus from Different Compartments of the Root Zone of Annual Cropsaeb

Crop Maize

Pigeon pea

Age (days)

Depth (cm)

34

80

Radial distance from base of plant (cm) 0-20

20-40

0-12.5 12.5-25.0 25.0-37.5

43 16 0

18 15 1

3 4 0

0-12.5 12.5-25.0 25.0-37.5 37.5-50.0

16 12 I 0

29 12

14

1

2

2

1

34

0-12.5 12.5-25.0 25.0-31.5

59 34 2

3 2 0

0 0 0

80

0-12.5 12.5-25.0 25.0-37.5 37.5-50.0

19 13

18 10

12 17

2

2 I

5

0

40-60

10

I

"Data from Nye and Foster (1961b). "Values are expressed as percentage of total uptake., TABLE V The Distribution of Phosphorus Uptake and Root Dry Weight with Depth under Bush Fallow",*

Plant Grasses

Herbs and shrubs

Depth (cm)

Uptake (%)

Root concentration (%)

0-25 25-50 50-75

70

81

19 11

14 6

0-25 25-50 50-75

64 23 13

66 27 7

"Data from Nye and Foster ( 196 1 b). *Values are expressed as percentage of totals for 0-75 cm.

had depleted just a s wide a zone as the maize. The top-dressed crops derived only 7-1 1% of their phosphorus from below 25 cm, whereas the plants in the unfertilized bush fallow derived one-third of their phosphorus from below that depth. Approximately 9 kg/ha of phosphorus was added

194

K. P. BARLEY

to the soil surface each year by the bush fallow, so that 3 kg/ha of phosphorus was retrieved from below 25 cm. Under the system of subsistence farming studied by Nye and Foster, this slow retrieval and surface accumulation of phosphorus by the natural bush fallow is essential for the maintenance of fertility. Data on root concentration were not given for the annual crops; under the bush fallow the root distribution corresponded roughly with the distribution of phosphorus uptake (see Table V). Tracer techniques are also informative in revealing where competition for nutrients occurs. In a paper on competition between pairs of the grasses, perennial ryegrass (Loliurn perenne L.), meadow fescue (Festuca pratensis Huds.), and their triploid hybrid, O’Brien et al. ( 1967) show that the hybrid had an advantage in mixtures because of its greater root development in the deeper subsoil; the ryegrass and fescue absorbed less P from 60 cm depth when mixed with their hybrid than when by themselves or with one another. Differences in the specific abundance or activity (abundance or activity of tracer per unit mass of element) of tracers present in different tissues or different plant organs indicate the compartments of the root zone from which the nutrient originated. Such information is of considerable interest when we wish to consider the management of crop nutrition in relation to different stages of development. Following earlier indications obtained in dietary surveys, Ellis (1966) showed that the higher specific activity of wSr in the peel than in the flesh of the potato was due to the strontium in the peel having been derived from shallower depths in the soil than the strontium in the flesh. Probably this reflects differences in the times of absorption, most of the strontium in the flesh being absorbed later than that in the peel. Observations of this kind, and indeed the whole subject of the pattern of nutrient uptake, have received more urgent attention in recent years because of the need to know the fate of nuclear fission products added in fallout to the soil. The interested reader is referred to the informative Annual Reports of the Letcombe (formerly Radiobiological) Laboratory of the U. K. Agricultural Research Council. A N D NUTRIENT UPTAKE D. ROOT ELONGATION

So far we have treated the root system as though it were stationary, but in fact the system extends continuously during the growth of a crop. As the rates at which roots elongate are commensurate with those at which the most mobile ions are transferred, we cannot neglect the possible influence of root elongation on uptake.

THE CONFIGURATION OF THE ROOT SYSTEM AND NUTRIENT UPTAKE

195

Whenever convection fails to meet demand, the growth of the root system into fresh, undepleted regions of the soil will increase the rate of nutrient supply. An aspect that is likely to be very important in relation to crop nutrition is the stage of crop development at which the roots elongate most rapidly, and that at which they cease to elongate. Interesting differences are known to exist among cultivars; for example, McClure and Harvey (1962) found that, in contrast to its parent lines, a highyielding sorghum hybrid had a root system that spread widely and for a long time after flowering. When the front of the root system extends through the soil, its progress increases the volume of the root zone more rapidly when the root system is “open”-volume proportional to third power of radial extentthan when it is “closed”-volume directly proportional to depth of the root zone. The influence of elongation on uptake can be described most simply for the one-dimensional case. When the front of a dense uniformly deep root system moves downward through a homogeneous soil, a steady state will be achieved provided that the uptake rate is not too high. The rate of uptake F of a nutrient that can be sustained per unit area of the front is given by Gardner (1968) as

where c, is the initial concentration of the nutrient in the soil solution, cs is the concentration at the front, v (cm sec-’) is the apparent velocity of

water through the soil (positive upward), and U (cm sec-l) is the velocity with which the front extends downward. When there is no reaction with the soil (j= I ) , v and U are additive, but, when nutrients are buffered by adsorption, the downward growth of the root system is j times more influential than the upward flow of the soil water. The actual uptake of less mobile nutrients is unlikely to be influenced by elongation growth to the extent predicted by Eq. (27), because, as shown in Table I, root systems become attenuated toward the extremities of the rooting zone, so that there we need to consider the transfer of nutrients to single roots rather than to a front. When single roots are considered, we again expect elongation of the root to be more important in the uptake of less mobile nutrients. If, for example, we make appropriate substitutions in the heat conduction equations for an infinitesimal sink moving in a straight line with velocity U [Carslaw and Jaeger, 1959, p. 267, Eq. (2)], we find the productjU appearing as a term, not U alone. For slender bodies the analysis becomes very involved. The most

196

K. P. BARLEY

direct approach might be to assume that the absorbing region has a simple shape and size, and then seek solutions of a transfer equation in three-dimensional form. Miles ( 1 965) provided an elaborate analysis of diffusion to slender moving bodies, but the boundary conditions adopted were inappropriate for plant roots. A more flexible procedure is to use a source-sink model, initially without regard to shape, and to seek the surface on which a chosen boundary condition is satisfied. By relaxing constraints such as successive sink strength or spacing, a solution may finally be obtained fitting a surface having the desired shape (Anderssen et al., 1969). When diffusion predominates, the radius of the diffusion field around the plant root is often very small in relation to the length of the field, and, when this is so, a rough indication of the influence of root elongation on uptake may be obtained by neglecting axial components of the flux and treating the problem as if the root traversed a series of discrete thin layers of soil (Passioura, 1963). Single roots can then be represented by one of the following sinks: ( i ) Stationary infinite cylinder of fixed radius r). At t > 0 the uptake is given by Q=-27~?

$kft

r =

q

dt

(28)

(ii) Cylinder of fixed radius r] and length 1 S r ) undergoing translation in the axial direction at constant velocity U.This represents an elongating root in which uptake is confined to a zone of given length. The flux density varies over the surface, but the total flux is constant so that

where T = 1/U. (iii) Cylinder elongating at constant velocity U (change of shape, no translation). The root elongates from 1 = 0 at t = 0 to U t at time t and absorbs throughout its length so that $ff,r = dl = $;f,r = , dt and as in (i)

,

Sink (ii) is of particular interest as it applies if the root has a hairbearing zone of fixed length, and, as seems likely in relatively dry soils, uptake from other zones is negligible (see Section V, A, 2). Note that in this case the constant rate of uptake depends not on U,but on the time T = l/U.To give one example pertinent to the uptake of a mildly buffered

THE CONFIGURATION OF THE ROOT SYSTEM A N D NUTRIENT UPTAKE

197

nutrient by main roots from a relatively dry soil, when k, = 1 x cm2 sec-l,j = 10, d = 0.1, a = 1 X cm sec-l, and 7 = 5 x cm, the constant rate of uptake Q by a root zone for which T = I day is predicted to be twice the near steady rate for a stationary root at large values oft. As noted in Section V, A, 2 there are soils in which, even if the root is wholly capable of uptake, depletion of certain nutrients is limited to a thin shell of soil adjacent to the epidermal wall. If the uptake is rapid then the rate of uptake per root varies directly with U. Insofar as the history of branching and elongation determine 15, and LA,root elongation is widely recognized as being important in relation to uptake. That the motion of the root itself may influence uptake is less obvious. In the absence of experiments, the topic remains highly speculative. VI.

Conclusions

Recent advances in methods of examination enable the rapid, quantitative determination of rooting density and distribution. As a wide range of tracers, chiefly radioisotopes, has become available, it is possible also to describe the pattern of nutrient uptake from the soil. Studies of the transfer of ions to single or widely spaced roots have shown that resistance to ion transfer within the soil can reduce the rate of uptake. When convection is negligible, this is true for even the most mobile ions. When root systems (sets of roots) are studied, as contrasted with single roots, the resistance to nutrient transfer depends not only on soil properties (transfer coefficients), but also on the configuration of the root system itself. The rooting density tends to have an overriding influence on the rate at which the soil is depleted. Indeed, the rooting density within the topsoil under well established crops or pastures is so great that the resistance to transfer is small except for the most immobile nutrients, at least when the soil is moist. Under seedlings and generally in subsoils, where the rooting density is relatively low, the resistance to nutrient transfer needs to be considered carefully. Diffusion theory facilitates prediction of the hypothetical effects of the configuration of the root system on nutrient uptake. However the processes operating are not nearly so simple as has to be assumed, and the main use of theory is in the design of experiments of a kind that are amenable to analysis. Our present ability to interpret uptake patterns is limited chiefly by lack of knowledge of functional differences between roots of different age and form.

198

K. P. BARLEY ACKNOWLEDGMENT

The author wishes to thank colleagues at the Waite Institute and at the F. C. Pye Field Environment Laboratory, CSIRO, Canberra for helpful comments. A grant from the Nuffield Foundation enabled the author to complete the manuscript in Canberra.

REFERENCES Andersen, A. J. 1967. Riso Rep. 170, 32. Anderssen, R. A., Hale, R. P., and Radok, J. R. M. 1969. Plant Soil 30,271-290. Asher, C. J., and Loneragan, J. F. 1967. Soil Sci. 103,225-233. Barber, S. A. 1962. Soil Sci. 93,34-39. Barley, K . P. 1953. Australian J . Agr. Res. 4,283-291. Barley, K . P. 1955. Agron. J. 47, 145-147. Barley, K. P., and Sedgley, R. H. 1961. Australian J . Sci. 23, 297-298. Barrow, N. J. 1966. Ausfralian J . Agr. Res. 17,849-863. Bama, D. N., and Dutta, K. N. 1961. Empire J. Exp. Agr. 29,287-298. Blaser, R. A. 1937. J. Amer. SOC.Agron. 29,421-423. Bonner, J . 1946. Bot. Gaz. (Chicago) 107,343-35 1. Bouldin, D. R. 196 I . Soil Sci. SOC.Amer., Proc. 25,476-480. Bowen, G. D. 1964. Australia, Commonw. Sci. Ind. Res. Organ., Div. Soils, Div. Rep. No. 1/64. Bowen, G. D. 1968. Trans. 9th Int. Congr. Soil Sci., 1968 Vol. 2, pp. 219-228. Bowen, G. D., and Rovira, A. D. 1967. Australian J . Biol. Sci. 20, 369-378. Bowen, G. D., and Rovira, A. D. 1969. At. Energy Australia 12, 2-7. Bray, R. H. 1954. Soil Sci. 78,9-22. Brouwer, R. 1954. Acta Bot. Neer. 3,264-309. Brouwer, R. 1966. Proc. Easter Sch. Agr. Sci., Univ. Notringham 12, 153-166. Brouwer, R., and Loen, E. A. 1962. Jaarb. Inst. Biol. Scheik. Onderz. LandbGewass pp. 19-25. Brown, R., and Cartwright, P. M. 1953. J . Exp. Bot. 4, 197-221. Carslaw, H. S., and Jaeger, J. C. 1959. “Conduction of Heat in Solids,” 2nd ed. Oxford Univ. Press, London and New York. Champion, R. A., and Barley, K. P. 1969. Soil Sci. 108,402-407. Clarke, A. L., and Barley, K. P. 1968. Australian J . Soil Res. 6, 75-92. Cockroft, B., Barley, K. P., and Creacen, E. L. 1969. Australian J . Soil Res. 7,333-348. Cormack, R. G. H. (1944). Amer. J . Bot. 31,443-449. Cornforth, 1. S. 1968. J . Soil Sci. 19, 291-301. Cowan, 1. R. 1965. J . Appl. Ecol. 2,221-239. Dittmer, H . J. 1938. Amer. J . Bor. 25,654-657. Dittmer, H. J. 1940. Soil Conserv. 6, 33-34. Dittmer, H. J. 1949. Amer. J . Bor. 36, 152-155. Duncan, W. G., and Ohlrogge, A. J. 1958. Agron. J. 50, 605-608. Edwards, W. M., Fehrenbacker, J. B., and Vavra, J. P. 1964. Soil Sci. SOC.Amer., Proc. 28, 560-564. Ellis, F. B. 1966. Ann. Rep. Agr. Res. Council Radiobiol. Lab. (Letcombe Regis) 16,47-48. Ellis, F. B., and Barnes, B. T. 1968.Ann. Rep. Agr. Res. Council Letcombe Lab. (Letcombe Regis) 19,24-26.

THE CONFIGURATION OF THE ROOT SYSTEM A N D NUTRIENT UPTAKE

199

Epstein, E., and Hagen, C. E. 1952. Plant Physiol. 40,620-624. Evans, H. 1938. Ann. Bor. (London) [N.S.] 2, 159-182. Fitzpatrick, E. G., and Rose, L. E. 1936. Amer. Soil Survey Assoc. Bull. 17, 136-145. Foy, C. D., Burns, G . R., Brown, J. C., and Fleming, A. L. 1965. Soil Sci. Soc. Amer., Proc. 29964-66. Frere, M. H., and Axley, J. H. 1964. Soil Sci. 97,209-2 13. Gardner, W. R. 1965. I n “Soil Nitrogen” Bartholomew, W. V., and Clark, F. E. (eds.), pp. 550-572. Amer. SOC.Agron., Madison, Wisconsin. Gardner, W. R. 1968. Trans. 9th Int. Congr. Soil Sci., 1968 Vol. I , pp. 135- 142. Gardner, W. R., and Ehlig, C. F. 1962. Agron. J. 54,453-456. Gates, C. T. 1951. J . Australian Insr. Agr. Sci. 17, 152-154. Giddings, J. C., Kucera, E., Russell, C. P., and Myers, M. N. 1968. J. Phys. Chem. 72, 4397-4408. Gilbert, E. N. 1962. Ann. Math. Stat. 33,958-972. Gile, P. L., and Carrero, J. 0. 1917. J. Agr. Res. 9,73-95. Grasmanis, V. 0. G., and Barley, K. P. 1969. Australian J. Biol. Sci. 22, I3 13- 1320. Greacen, E. L., Barley, K. P., and Farrell, D. A. 1968. Proc. Easrer Sch. Agr. Sci., Univ. Nottingham 15,256-269. Hackett, C. 1968a. Proc. Easrer Sch. Agr. Sci., Univ. Nottingham 15, 134-147. Hackett, C. 1968b. New Phytol. 67, 287-299. Head, G. C. 1968. J . Hort. Sci. 43,275-282. Humphries, E. C. 1958. Ann. Bot. (London) [N.S.] 22.25 1-257. Jacques, W. A. 1937. New Zealand J . Sci. Technol. 19A, 441-450. Jenny, H. 1966. Plant Soil 25,265-289. Jenny, H., and Overstreet, R. 1939. J. Phys. Chem. 43, 1185-1 196. Jones, L. H. P., and Handreck, K. A. 1965. Plant Soil 23,79-96. Kafkafi, U., Posner, A. M.,and Quirk, J. P. 1967. SoilSci. SOC.Amer., Proc. 31,348-353. Kautsky, J., Barley, K. P., and Fiddaman, D. K. 1968. Australian J. Soil Res. 6, 159-167. Kendall, M. G., and Moran, P. A. P. 1963. “Geometrical Probability.” GriRn, London. Kiesselbach, J. A., and Weihing, R. M. 1938. J. Amer. SOC.Agron. 27, 538-541. Klute, A., and Letey, J. 1958. Soil Sci: SOC.Amer., Proc. 22,213-215 Kokkonen, P. 1931. Acta For. Fenn. 37, 1-123. Kramer, P. J., and Bullock, H. C. 1966. Amer, J. B o f . 53,200-204. Kramer, P . J., and Wiebe, H. H. 1952. Plant Physiol. 27, 661-674. Kutschera, L. 1960. “Wurzelatlas Mitteleuropaischer Ackerunkrauter und Kulturpflanzen.” DLG-Verl.-Ges. Frankfurt. Lang, A. R. G., and Melhuish, F. M. 1970. Biomerrics (in press). Lauchli, A. 1967. Planta 75, 185-236. Lavy, T. C., and Barber, S. A. 1964. Soil Sci. SOC.Amer., Proc. 28,93-97. Lewis, D. G., and Quirk, J. P. 1967. Planr Soil 26,445-453. Litav, M., and Harper, J. L. 1967. Plant Soil 26,389-392. Loneragan, J. F., and Asher, C. J. 1967. Soil Sci. 103,311-3 18. McClure, J. W., and Harvey, C. 1962. Agron. J. 54,457-459. McDougall, W. B. 1921. Amer. J . Eot. 8, 171-176. McElgunn, J. D., and Harrison, C. M. 1969. Agron. J. 61,79-80. Marriott, F. H.C., and Nye, P. H. 1968. Trans. 9th Inr. Congr. Soil Sci., 1968 Vol. 1, pp. 127-34. May, L. H., Chapman, F. M., and Aspinall, D. 1965. Australian J . Biol. Sci. 18,25-35, May, L. H., Randles, F. M., Aspinall, D., and Paleg, L. G. 1967. Australian J . Biol. Sci. 20, 273-83.

200

K. P. BARLEY

Melhuish, F. M. 1968. Ann. Bor. (London) [N.S.] 32, 15-22. Melhuish, F. M., and Lang, A. R. G. 1968. Soil Sci. 106, 16-22. Miles, J. W. 1965. Proc. Roy. Soc., Ser. A. 284, 137-145. Miles, R. E. 1970. Marh. Biosci. 6,85-127. Newman, E. I. 1966.J. Appl. Ecol. 3, 139-145. Newman, E. I. 1969. J. Appl. Ecol. 6, 1-12. Nielson, J. A. 1964. Ecology 45,643-646. Nobbe, F. 1892. Landwirt. Vers. Sra. 4,212-224. Nye, P. H. 1966. Plant Soil 25, 81-105. Nye, P. H., and Foster, W. N. M. 1961a. Trans. 7th lnt. Congr. Soil Sci. 1960 Vol. 2, pp. 2 15-222. Nye, P. H., and Foster, W. N. M. 1961b. J . Agr. Sci. 56,299-306. Nye, P. H., and Tinker, P. B. 1969. J. Appl. Ecol. 6,293-300. O’Brien, T. A., Moorby, J., and Whittington, W. J. 1967. J . Appl. Ecol. 4,5 13-520. Ogston, A. G. 1958. Trans. Faraday SOC. 54,1754-1757. Olsen, C. 1950. Physiol. Plant. 3, 142-164. Olsen, S. R., and Kemper, W. D. 1968. Advan. Agron. 20,91-149. Overstreet, R., and Jacobson, L. 1946. Amer. J. Bor. 33, 107-1 12. Passioura, J. B. 1963. Plant Soil 18, 225-238. Passioura, J. B., and Frere, M. H. 1967. Australian J. Soil Res. 5, 149-159. Passioura, J. B., and Leeper, G. W. 1963. Narure 200,29-30. Patil, A. S., King, K. M., and Miller, M. H. 1963. Canadian J. Soil Sci 43,44-5 1. Patterson, E. K. 1941. Amer. J. Bor. 28,628-638. Pavlychenko, T. K. 1942. Nut. Res. Council Canada Publ. 1088,46. Philip, J. R. 1957. Proc. 3rd Congr. lnt. Comm. lrrig. Drainage 1957 Question 8, pp. 125- 154. Phillips, R. E., and Brown, D. A. 1964. Soil Sci. SOC.Amer., Proc. 28,758-763. Pielou, E.C. 1960. J. Ecol. 48,575-584. Pielou, E. C. 1961. J. Ecol. 49,255-269. Pitman, M. G. 1965. Australian J. Bioi. Sci. 18,987-998. Pittman, U . J. 1964. Canadian J. Plant Sci. 44,283-287. Porter, L. K., Kemper, W. D., Jackson, R. D., and Stewart, B. C. 1960. Soil Sci. SOC. Amer., Proc. 24,460-463. Racz, G. J., Rennie, D. A., and Hutcheson, W. L. 1964. Canadian J. Plant Sci. 42,430436. Roach, S. A. 1968. “The Theory of Random Clumping.” Methuen, London. Russell, R. S., and Ellis, F. B. 1968. Nature 217,582-583. Russell, R. S., and Newbould, P. 1968. Proc. Easter Sch. Agr. Sci., Univ. Notringham 15, 148- 167. Russell, R. S., and Sanderson, J. 1967. J. Exp. Bor. 18,491-508. Sayre, C. B. 1934. Proc. Amer. SOC. Horf. Sci. 32,564-471. Schofield, R. K., and Graham-Bryce, 1. J. 1960. Nature 188, 1048-1049. Soileau, J. M., Jackson, W. A., and McCracken, R. J. 1964. Soil Sci. 15, 1 17- 123. Soper, K. 1959. New Zealand J. Agr. Res. 2, 329-341. Stickler, F. C., and Pauli, A. W. 1961. Trans. Kansas Acad. Sci. 64,81-90. Tepe, W., and Leidenfrost, E. 1958. Landwirt. Forsch. 11, 217-230. Thompson, H. R. 1956. Ecology 37,391-394. Torssell. B. W. R., Begg, J. E., Rose, C. W., and Byrne, G. F. 1968. Ausrrafian J. Exp. Agr. Anim. Husb. 8,533-543.

THE CONFIGURATION OF THE ROOT SYSTEM AND NUTRIENT UPTAKE

201

Troughton, A. 1957. Bull. Commonw. Past. Fld. Crops No. 44pp. 163. Ueno, M.,Yoshihara, K., and Okada, T. 1967. Nature 213, 530-532. Viets, F. G . 1965. I n “Soil Nitrogen” Bartholomew, W. F., and Clark, F. E. (eds.), pp. 508-548. Arner. SOC.Agron., Madison, Wisconsin. Walker, J. M., and Barber, S. A. 1961. Science 133, 881-882. Weaver, J. E. 1920. Carnegie Inst. Washington Publ. 292. Weaver, J. E. 1926. “Root Development of Field Crops.” McGraw-Hill, New York. Weaver, J. E., and Bruner, W. E. 1927. “Root Development of Vegetable Crops.” McGrawHill, New York. Weaver, J. E., Jean, F. C., and Crist, J. W. 1922. Carnegie Inst. Washington Publ. 316. Whitaker, E. S. 1923. Bot. Gaz. (Chicago) 76,30-58. Wiersum, L . K. 1962. Plant Soil 16,42-61. Wilkinson, H. F., Loneragan, J. F., and Quirk, J. P. 1968. Science 161, 1245-1246. Williams, D. E. 1961. Plant Soil 15,387-399. Wilson, B. F. 1967. Bot. Gaz. (Chicago) 128,79-82.

This Page Intentionally Left Blank

FROST AND CHILLING INJURY TO GROWING PLANTS H. F. Mayland and J. W. Cary Snake River Conservation Research Center,

Soil and Water conservation Research Division, Agricultural Research Service,

U S . Department of Agriculture, Kimberly, Idaho

Page 203 ....................... 206 ........................ 206 206 209 212 D. "Bound" Water ............................................................................ Cold Lability of Enzymes ..................................................................... 21s A. In V i m Evidence ........................................................................ 21s 216 B. In Vivo Evidence ....................................................... 217 Membrane Composition and Permeability ................................................ 217 A. Description ..................................................... 219 B. Composition.. .............................................................. 220 C. Permeability ................................................................................. 220 Protection from Freezing ..................................................................... 220 A. Evidence of Chemical Effects ............................. 222 B. Mechanisms of Freeze-Injury Protec 226 C. Undercooling and Nucleation ........ D. Chilling Injury ......................... .............. 228 230 23 1

I . Introduction ............................... 11. Physicochemical Principles of Prote A. Structural Requirements .........

111.

IV.

V.

VI.

I. Introduction

Freezing injury in plants represents a major economic loss to agriculture. Reingold (1960) reports crop losses in the United States resulting from cold weather as follows: Crop

Percentage loss

~

Almonds Apples Citrus Stone fruits Cereals Strawberries Grapes

10 8 8 10

3-4 30-40 10

203

204

H. F. M A Y L A N D A N D J. W. CARY

Another survey of the crop-freeze problem was conducted in the contiguous United States for the period 1963 to 1968 (Prestwich, L., “Freeze Damage to Crops,” unpublished research work, U.S.S. Agr. Chem., 1969). He found that production of an estimated 3.6 million acres of cropland was destroyed annually by freezing, and that lost production was valued at 341 million dollars per year (Table I). TABLE I Average Annual Crop Freezing Losses for Years 1963-1968, Continental United States“ Freezing losses Loss relative to all crop losses Crop

Acres (millions)

Dollars (millions)

% of acreage

% of value

Fruits Vegetables Field crops

0.45 ( 1 5 ) h 0.30 (6) 2.90 ( I )

215 (12.0) 58 (2.5) 68 (0.4)

12 8 79

63 17 20

Prestwick, L., unpublished research work, U.S.S. Agr.-Chem. (1969). bData in parentheses are percentage of total crop acres or dollar value lost.

While the above data represent losses resulting from ice-induced injuries, there may also be crop production losses caused by low temperature which go unnoticed and are thus unaccounted for. For example, Kuraishi et al. ( 1 968) reported that unhardened pea plants were killed at -3°C without ice formation. In addition to losses that are directly attributable to ice formation in plants, there are other yield-reducing factors that may be attributed indirectly to cold temperatures. Plants such as cotton, peanuts, and other tropical species may be permanently injured by cool temperatures of 0 to +lO°C (Sellschop and Salmon, 1928). Majumder and Leopold ( 1967) have reported that callose plugs form in or along the phloem sieve tubes and that this contributes to the low temperature responses of some species. Xylem elements of fruit trees may be permanently occluded by exposure to freezing temperatures (Daniel1 and Crosby, 1968). Restricted water movement resulting from xylem vessel occlusions limits tree growth and fruit yield and plays a role in peach tree decline. Early research on freezing phenomena in plants centered on plant selection and classification according to their ability to become cold

FROST AND CHILLING INJURY TO GROWING PLANTS

205

hardy and survive freezing temperatures. Recent work has centered on the differentiation between plants with or without the ability to harden. Perhaps the most fascinating problems are yet to be encountered in the study of cold stress and freezing in nonhardy plants. This includes the varying ability of plants to survive cold temperatures, as during the seedling establishment of corn, beans, and sorghum or during vegetative growth of legumes and pollination and flowering of horticultural plants and small grains. These so-called “nonhardy” plants have, therefore, been subdivided into tender and resistant types in various geographical areas. For example, beans, corn, and peach blossoms in temperate climates may be considered as “tender” crops, while peas, lettuce, and sugarbeets are more cold resistant, although none of these plants are thought of as having the ability to become cold hardy, as do winter wheat and many perennials. Research on the conditions associated with plant adaptation to cold temperatures has been carried on for nearly a century, and excellent discussions of cold hardiness may be found elsewhere (Levitt, 1956, 1966b, 1967). Smith (1 968) summarized the inability of past cold-hardiness studies to satisfactorily associate changes in plant constituents with frost tolerance. He reported that, “. . . although differences in chemical changes during cold-hardening exist among species, there is still a question as to whether these alterations in plant metabolism are intimately involved in the development of frost hardiness or whether they are merely associated changes.” Recent approaches using biochemical techniques are providing definitive evidence of an enzyme system (peroxidase isozyme components) showing major response to cold temperature stresses by plant tissues capable of cold-hardening (McCown et al., 1969a,b). When hardened plant material is cooled slowly, ice first forms in the extracellular space (Levitt, 1956). The equilibrium vapor pressure of ice is less than that of pure liquid water at any given temperature below 0°C. Thus as the water in the extracellular space freezes, the chemical potential falls below that of the cell sap, and water diffuses from the cells through the semipermeable membrane. The cells become freely permeable, perhaps because of rupture of the plasma membrane by ice crystals when intracellularly frozen, or simply from disruption of the normal structure of the plasma membrane. Protoplasm may be injured by freezing in two ways -dehydration and mechanical strain. Within certain limits, dehydration is injurious only in conjunction with mechanical strain because dehydration increases the consistency of the protoplasm. The protoplasm is thus more brittle and more liable to rupture under the action of the deforming stress. Super-

H. F. MAYLAND A N D J. W. CARY

206

imposed upon the above types of injury is the action of concentrated electrolytes within the cell. When dehydration exceeds a certain limit, the increased consistency becomes irreversible and must be regarded as a form of coagulation. Such coagulation is frequently an irreversible colloidal change. Cell death may not result immediately upon coagulation, but eventually membranes rupture and other macromolecules are irreversibly denatured and freezing injury results. This picture of the freezing process, however, does not explain the mechanics of injury nor does it describe cold stress phenomena. Ice formation and freezing injury in plants have been previously reviewed (Luyet and Gehenio, 1940). Levitt (1956, 1967) has published extensive reviews of factors associated with cold hardiness of plants. Redistribution of water in winter cereals and the subsequent effect of freezing stresses on plant survival were reviewed by Olien ( 1 967a). Idle and Hudson (1 968) and Scarth (1 944) presented a limited discussion on chilling injury and the physical effects occurring during ice formation in plants. Mazur ( 1969) discussed concepts, experimental approaches and results of tissue preservation by freezing and relates these to botanical oriented freezing studies. The discussion presented here concentrates on the effects of low-termperature stress on cell membranes and other macromolecules in the cell and relates these to the overall plant response to chilling or frost injury. II.

Physicochemical Principles of Protein Structure

A.

STRUCTURAL REQUIREMENTS

Proteins must be flexible to accomplish their biochemical functions associated with conformational changes. Protein flexibility is provided by weakening or strengthening of intramolecular bonds that maintain secondary and tertiary structure. When temperatures decrease, macromolecules become excessively rigid or brittle, and thus inactive. The primary structure of proteins is chemical valence bonding in a sequence of amino acids and disulfide bonds. The secondary structure is the polypeptide-chain configuration (series of amino acids) yielding H-bonding between peptide, N -H, and C =0 groups. Tertiary structure is the pattern of packing of the secondary structures. B. BONDING I . Types Kauzmann ( 1959) lists seven types of intramolecular bonds that might

FROST A N D CHILLING INJURY TO GROWING PLANTS

207

be expected to influence the polypeptide chain configuration. These are: a. Hydrogen bonds between peptide linkages b. Hydrophobic bonds c. Salt linkages (ion pair bonds) and other electrostatic forces d. Hydrogen bonds other than those between peptide links e. Stabilization by electron delocalization f. Dispersion forces (London forces), protein chemist’s term of secondary bonding g. Disulfide groups and other cross linkages H-bonds and hydrophobic bonds are likely to have the most important functions because of the relatively large number of peptide and hydrophobic groups in nearly all proteins. The H-bond is suited to play an important role in physiological processes because of the small bond energy (Table 11) and small activation energy involved in its formation and rupture (Pauling, 1960). Many protein properties depend on configurations present in localized regions of the molecule, and these configurations might be determined by some less abundant types of bonds. It is not really safe to say that any of the bonds are “less important than others” except that salt linkages are not prominent contributors to the stability of proteins (Kauzmann, 1959; Matsubara, 1967). TABLE I I Bonding Energies in Kilocalories per Molen Bond

c-c C-N

c-s s-s

H-bond Hydrophobic

Experimental

Calculated

68 65 52 48.5 -

64 53 57 50 Generally 2-10 Less than 3

“The value of hydrophobic bonding energy is a function of the nonpolar groups involved and also temperature, decreasing with a reduction in temperature (Nemethy and Scheraga, 1962b). Other data are from Levitt (1962 Copyright C 1962 Academic Press, New York).

2. Hydrogen Bonding between Peptide Links Recent research on protein hydration has demonstrated the close interaction between the hydration shell surrounding protein molecules and the physicochemical properties of the proteins themselves (Bernal, 1965). The hydration shell consists of several layers of water molecules

208

H. F. MAYLAND A N D J. W. CARY

in an icelike sheath surrounding and linking the protein molecules (Bernal, 1965; Nemethy and Scheraga, 1962a,b). Structure is considered essential for maintaining protein properties and functions. Any alteration of this water structure would result in changes in both the secondary and tertiary protein structures and would be defined as denaturation (Kauzmann, 1959). Such changes prevent proteins from functioning properly because of steric incompatibilities with coenzymes. 3. Hydrogen Bonding Other Than Those between Peptide Linkages

Examples of H-bonding apart from peptide linkages in proteins include carboxylate ion to the phenolic hydroxyl of tyrosine, carboxylate and hydroxyl of threonine or serine and the carboxylate ion and the thiol group of cysteine (Kauzmann, 1959). The energy of this H-bond type is much less than that of the H-bond between two peptide groups. Nonpeptide H-bonds may modify properties of dissociable groups. However, it does not seem likely that nonpeptide H-bonds make a major contribution to the stability of native proteins. 4 . Hydrophobic Bonding

The role of the hydrophobic bonds or hydrophobic regions of protein molecules (Fig. 1) has received increasing attention in recent years. Nonpolar side-chain groups of protein molecules modify the water structure in their neighborhood in the direction of greater “crystallinity” (Shikama, 1965b). Nemethy and Scheraga (1962b) consider the hydrophobic bond formation in a protein to consist of two processes: (1) two or more nonpolar side chains which are surrounded by water come into contact, and (2) thereby decrease the total number of the water molecules around them. Hydrophobic bonds play a unique role in stabilizing

FIG. 1. Schematic representation of a protein molecule, especially showing interactions between side-chain R groups in an aqueous solution. The R, and R. represent polar sidechain R groups and nonpolar side-chain groups, respectively. In this model the hydrophobic bonds are pictured with a lattice-ordered layer of water around them, as shown by broken lines.

FROST A N D CHILLING INJURY TO GROWING PLANTS

209

native protein conformation since these bonds are a function of the water structure around the nonpolar group (Shikama, 1965b; Nethey and Scheraga, 1962b). Nonpolar amino acids constitute 35 to 45% of proteins (Shikama, 1965b). Examples of these nonpolar side chains are: the methyl group of alanine, the isopropyl group of valine, the isobutyl group of leucine, the sec-butyl group of isoleucine, the benzyl group of phenylalanine, and the methyl mercaptan group of methionine (Shikama, 1965b). These nonpolar side chains have a low affinity for water. The polypeptide chain configuration in proteins, which brings large numbers of these groups into contact with each other, removes them from the aqueous phase. This configuration is more stable than others, all other things being equal (Kauzmann, 1959). 5. Disulfide Bonds

Disulfide bonds ( S S ) consist of the intramolecular cross linkages by cystine or phosphodiester links. When this type of bond is located in the macromolecular chain, it is impossible for the chain to fold into less stable configurations (Kauzmann, 1959). 6. Other Bonding Types

The effect of electrolytes and nonelectrolytes will probably depend on the degree to which they cause reorientation of the structured water surrounding the macromolecule. Small, strongly polar molecules, having strong hydrogen bonding characteristics, may break down the highly structured water envelope. Binding of small organic molecules may have strong binding affinity on the inside of the protein helix. Urea molecules, for example, are bound to peptide bonds which normally would be buried within the protein molecule, but protein becomes denatured following the bonding change resulting from the action of the urea molecule (Kauzmann, 1959). Some ions may help to stabilize the protein structure and protect it against denaturation caused by other agents (Boyer et al., 1946a,b).

C. INACTIVATIONAND DENATURATION The overall integrity of protein structure depends on both apolar (hydrophobic) and polar (H-bonding) interactions. Changes in the bonding may induce changes in the protein molecule which result in denaturation and loss of activity. Denaturation, although having a number of definitions, will be used here as “a process(es) in which the spacial

2 10

H. F. MAYLAND A N D J . W. CARY

arrangement of the polypeptide chains within the molecule change from that typical of the native protein to a more disordered arrangement” (Kauzmann, 1959). Denaturation may occur when H-bonding is broken, or when hydrophobic bonds are displaced. Bello (1966) has shown that hydrophobic denaturants are effective in disrupting deoxyribonucleic acid (DNA) structure. The hydrophobic bond is of prime importance in the stabilization of the native protein conformation at normal physiological temperatures. As the temperature is lowered, however, hydrophobic effects become weaker and hydrogen bonds more stable. The effects expected may be: (1) denaturation resulting from disruption of hydrophobic regions, (2) structure stabilization resulting from hydrogen bond stabilization, or (3) denaturation and formation of a new hydrogen-bonded conformation (Bello, 1966) or disulfide bridge (Levitt, 1966b). An example of the latter is Kavanau’s hypothesis (see Langridge, 1963). He proposes that some enzyme inactivation, such as phosphatase and peroxidase at low temperatures (ca. - 10°C), is attributable to an increase in intramolecular H-bonding so that active centers lose their specific configuration. Stability may also result from disulfide bonds or cystine bridges which are found in some heat-stable enzymes. The heatstable enzyme thermolysin does not have cystine bridges but must obtain its stability from hydrophobic interaction and perhaps, in addition, metal chelation (Matsubara, 1967). Sulfhydryl (SH) and disulfide (SS) groups help maintain the primary structure of proteins and control of the enzyme activity. Since changes in the steric conformation of proteins may be affected by freezing and thawing (Levitt, 1966a), it follows that these groups may also be involved in the physiological processes that accompany the changes in water activity (Tappel, 1966). Measurements of the SH and SS contents of plants before and after freezing have indicated a conversion of protein SH and SS when the freezing resulted in killing, but not when the plants survived uninjured (Levitt, 1962). Similar results were obtained with injury by heating. On the other hand, when plants of different hardiness were compared, a positive correlation was found between SH content and resistance to freezing injury. Plants incapable of hardening at low temperatures also showed a marked increase in SH at hardening temperatures, but only if permitted to wilt (Levitt et al., 1961). Levitt (1962) therefore proposed a hypothesis which assumes that ice forms extracellularly when a plant is frozen and the water that separates the protoplasmic proteins moves to these extracellular ice loci, thus causing the cell to dehydrate. At a certain degree of dehydration, which

FROST A N D CHILLING INJURY TO GROWING PLANTS

21 1

varies with the plant resistance to freezing injury, the SH and SS groups of adjacent protein molecules would approach one another closely enough to permit chemical reactions to occur (see Levitt, 1962). The reaction could be of two kinds: an oxidation of two SH groups to SS, or an SH SS interchange reaction. In each case the result would be an intermolecular SS bond. Since the SS bond is covalent, it is far stronger than the hydrogen of hydrophobic bonds (Table 11) which are responsible for much of the tertiary structure of the protein. Consequently, when thawing occurs and water reenters the protoplasm, pushing the proteins apart, the newly formed SS bonds remain intact, whereas many of the weaker hydrogen and hydrophobic bonds are broken by the stresses, and protein molecules then unfold or denature. If the intermolecular SS bond forms by SH SS interchange, the unfolding could occur during the freezing process since an intramolecular SS bond would be broken. If a sufficient number of intramolecular SS bonds are formed, the unfolding would lead to protein denaturation and cell death. The above hypothesis seems to fit many natural conditions and provides a useful explanation of injury (Levitt, 1967). A study of desiccation injury in cabbage leaves supports Levitt’s sulfhydryl-disulfide hypothesis (Gaff, 1966). Structural protein extracted from cabbage leaves displayed an apparent unfolding at water potentials less than -40 bars. The degree of unfolding increased with increasing disiccation until cell death occurred at -94 bars water potential. Direct evidence is still lacking to support the sulfhydryl hypothesis of freezing injury. Trials to visualize tissue bound SH groups by electron microscopy have given only equivocal results (Pihl and Falkmer, 1968). Addition of SH-containing compounds (i.e., cysteine and glutathione) to chloroplast systems has failed to provide protection against freezing (Heber and Santarius, 1964). Krull ( 1967), however, reports conclusive evidence that frost resistance in epidermal cells of red cabbage is increased by mercaptoethanol, which alters disulfide content of proteins. Addition of nonpenetrating sugars protected epidermal cells of red cabbage, but no evidence was obtained for the protection of surface SH groups on cell wall membranes by the sugars (Levitt and Haseman, 1964). It was concluded that the protection must, therefore, be internal to the cytoplasmic proteins. The SH groups of proteins are of considerable chemical interest since they are the most highly reactive of the amino acid side chains. The SH groups have a varying reactivity, which is as yet unexplained except for some broad steric possibilities (Battell et al., 1968). Some proteins do not contain disulfide bridges. One such protein is glycogen phosphorylase,

*

212

H. F. MAYLAND A N D J . W. CARY

which can have two sulfhydryl groups per mole of enzyme bound without loss of enzymatic activity. A second class of sulfhydryl groups in the same protein when bound by amperometric titration results in complete loss of enzymatic activity and denaturation. The first two SH groups must be fully exposed on the enzyme surface, allowing the possible disulfide bond formation between phosphorylase monomers, which then results in intermolecular disulfides connecting enzyme molecules into large aggregates. Upon protein denaturation, another class of sulfhydryl groups will be exposed; the number depends upon conditions, but will include as many as 12 more SH groups per mole (Battell et al., 1968). D. “BOUND” WATER

1 . De3nition

Current usage in cryobiology loosely defines “bound” water as that which does not freeze (Meryman, 1966). The energy status of this water is shown in Table 111. There is little doubt that biochemical systems contain liquid water at subfreezing temperatures, and that the amount of this bound water (Fig. 2) decreases with temperature (Levitt, 1956; Toledo et al., 1968) and/or with molecular denaturation (Pichel, 1965). TABLE I11 Vapor Pressure versus Temperature for Water and Ice and the Corresponding Vapor Pressure Potential of the Water“ Aqueous vapor pressure Temperature (“C) 0 -1

-2 -3 -4 -5 -6 -7 -8 -9

- 10 - 15

Potential

Ice (mmHg)

Water (mmHg)

Joules kg-’

4.579 4.217 3.880 3.568 3.280 3.013 2.765 2.537 2.326 2.131 I .950 1.241

4.579 4.258 3.956 3.673 3.410 3.163 2.93 1 2.7 I5 2.5 I4 2.326 2.149 1.436

0 -1213 -2426 -3620 -4827 -60 12 -7188 -8326 -9465 -10,675 -I 1.807 -2036 1

-Bars 0 12 24 36 48 60 72 83 95 107 I I8 209

“Assumptions are: atmospheric pressure and ice and water at vapor pressure equilibrium.

FROST A N D CHILLING INJURY TO GROWING PLANTS

213

20

c

15

r 0

.-

B

10

5

0 Temp 0 Tension

-5’ 60

-loo

-15’

118

209

-20”

-25O

-30°

FIG. 2. Progressive ice formation with decreased temperature in the lichen Cetraria richardsonii. Temperature in degrees centigrade and tension in bars (from Table 111) at the ice-water interface under equilibrium conditions. (From Levitt, 1956 Copyright 0 1956 Academic Press, New York.)

2 . Experimental Evaluation of “Bound” Water

Microorganisms maintain about 10% of their total water in a nonfrozen state at -20°C (Mazur, 1966). This 10% residual water in cells is not normal supercooled water, but is water bound to cellular solids by forces of varying strength. Even at nonfreezing temperatures, sharp distinctions cannot be made between wholly “free” water or liquid water which at one extreme is totally unengaged in relationships other than with itself, and the other extreme to totally “bound” water which is active in determining secondary or tertiary macromolecular structure. Some progress in measurement of bound water appears to be possible, utilizing nuclear magnetic resonance (NMR) spectroscopy. Toledo et al. (1968) were able to measure the bound water content of wheat flour dough with good precision, for a given temperature, such as - 18”C, regardless of total water content. Considerable progress has already been made in defining protein hydration characteristics at freezing temperatures. Kuntz et al. (1 969) reported the hydration of proteins and nucleic acid solutions at -35°C to be 0.3-0.5 g of water per gram of protein. Nucleic acids were three to five times more hydrated than proteins. It is well to point out that high-resolution NMR spectra analysis shows that the “bound” water is not “icelike” in any literal sense, although it is clearly less mobile than liquid water at the same temperature. There is a remote possibility

214

H. F. MAYLAND AND J. W. CARY

that this “bound” water may be related to “anomalous” or “poly water,” which is receiving much current attention (Lippincott et al., 1969). Attempts have been made to differentiate between the physical properties of cytoplasmic protein-water extracts of cold-hardy and nonhardy plants (Brown, J. H.,Bula, R. J., and Low, P. F., unpublished information, Purdue University). Essentially no differences were found in the apparent specific heat capacities, ice nucleating abilities, or the amount of water absorbed to the dry protein. Partial specific volumes were similar, but showed increases as plants were exposed to decreasing temperatures. 3. Chemical Potentials All the water in plants first supercools and then begins to freeze, generally in the extracellular space, as the temperature is lowered under “equilibrium” conditions (rate s 1°C per minute). The liquid water remaining within the cell is subjected to a lesser change in chemical potential than that surrounding the ice crystal outside the cell (Table 111). Dehydration of cellular protoplasm occurs during freezing in response to gradients in water energy. The vapor pressure gradient caused by extracellular freezing may be used to estimate the driving potential for water flow only if temperature and electrical gradients are negligible. As ice crystals grow in an aqueous solution, the solutes tend to be largely excluded from the crystal, and thus they concentrate in the solution. If specific ions are present in the solution, particularly F- and NHt, (Fand N H t are highly toxic and generally not present in plants) a preferential trapping of ions in the crystal can occur, resulting in potentials of 20-30 V or more between the crystal and the solution (LeFebre, 1967). While this has not been measured in plants, it could conceivably enter into the reactions that take place in the bound water and membrane regions during freezing. Since freezing releases heat, it is also possible that signifficant thermal gradients develop across cell walls and membranes. The technique of atomizing microorganism cells in 02-freeatmospheres of known relative humidity has been used to study “bound” water. Organisms thus exposed rapidly lose 90-95% of their total water content, but the remainder is less easily lost. Webb (19 6 3 , using this aerosolization method, reported that the death rate was directly related to the amount of “bound” water removed from these cells (Fig. 3). Thermodynamic analysis of the death rates obtained during two periods (0 to 1 hour and 1 to 5 hours) and a wide range of temperatures indicated that death results from a tightening of molecular structures and is associated with

FROST AND CHILLING INJURY TO GROWING PLANTS

215

relatively small activation energies (Webb, 1965). Very few deaths occur at above 70% relative humidity (RH),(corresponding to water potential of - 130 bars at 20°C or a temperature effect of - lO"C), but a sudden increase in the cell's sensitivity occurs as the R H is lowered further.

-

0.05

u1

-30 2 P 0

Y

E 0.03-

-20

e

f0

z 0,

0.02-

-

8

-10

RH 10 Tension

30

50

70

90

>IOOO

goo

4ao

130

0

0

2 I

2

o g

FIG. 3. The effect of relative humidity (RH)on the water content and death rates of Serratia rnarcescens. Death rate K = In N J N , with K , representing the time interval between 0 and 1 hour, while Kr represents interval of I to 5 hours. S.rnarcescens ordinarily has 400 g of water per 100 g of solids. Data were taken at 25°C. Tension (water potential) is in bars, as taken from Table 111. (From S. Webb, 1965, "Bound Water in Biological Integrity," Thomas Springfield, 111. with permission.)

Ill.

Cold Lability of Enzymes

A. In Vitro EVIDENCE The main factor contributing to protein denaturation by freezing and thawing is the change in water structure around the native protein molecule during freezing and thawing. Shikama (1 965b) has shown that there is a critical temperature region in which catalase and myosin are denatured during freezing and thawing. Denaturation begins at - 12°C for catalase and -20°C for myosin. The double-stranded helical structure of DNA is not broken down by freezing (0 to - 192°C) and thawing (Shikama, 1965a). Infrared spectroscopy of DNA, however, showed that structural changes occurred in the molecule which corresponded to the water activity where microorganism viability was lost (Webb, 1965). X-ray analysis of the water remaining on the macromolecule suggested that water reorientation also occurred (Webb, 1965). Although there may be several different DNA enzyme to water interactions, Cox (1968) has suggested that loss of the water layers from the DNA molecule produces a biologically inactive moiety by semireversible formation of a hydrate. Some enzymes are not inactivated by freezing and thawing. Two of these

216

H. F. MAYLAND AND J. W. CARY

enzymes are invertase and sucrose phosphorylase (Barskaya and Vichurina, 1966).Glycogen phosphorylase b, in contrast to phosphorylase a , loses its enzymatic activity at 0°C (Graves et ul., 1965). Pyruvate carboxylase is rapidly inactivated by exposure to low temperature, but the enzyme inactivation is at least partially reversible by rewarming (Scrutton and Utter, 1964). Heber (1967)and Heber and Santarius (1964)considered dehydration of the adenosine triphosphate (ATP) synthesis system by freezing as responsible for its inactivation. This may occur above -8°C (Borzhkovskaya and Khrabrova, 1966),but dehydration may be more complete at -18" to -25°C (Ivanova and Semikhatova, 1966). Removal of functional water from the membrane system to the growing ice crystals apparently leads to the uncoupling of the phosphorylatory system from electron transport in the case of photosynthetic phosphorylation and, in other cases, to related effects (Heber and Santarius, 1964). B. In Vivo EVIDENCE The in vitro evidence for cold lability of enzymes is further supported by Ng (1969),who concluded that the decrease in cell yield of Escherichia coli with decreasing growth temperature resulted from the uncoupling of energy production from energy utilization. Stewart and Guinn (1969) observed a decrease in ATP with chilling of cotton seedlings at 5°C and concluded that oxidative and photophosphorylation were more sensitive to low temperature inhibition than systems that use ATP. The close association of both enzyme and membrane sensitivity to low temperature is reinforced here. The inner mitochondria1 membrane 'contains the entire electron transfer chain as well as the enzymes of oxidative phosphorylation (Green and Tzagoloff, 1966). Kuiper (1969a)postulated that membrane ATPase is sensitive to denaturation by freezing. He (Kuiper, 1967) reported that potato ATPase was cold labile except when treated M 1,5-difluoro-2,4-dinitrobenzene,which with compounds such as was found to increase water permeability of bean root cell membranes and to afford considerable protection of bean plants against freezing damage. Pullman et al., (1960)also reported ATPase to be cold labile and inactivated at temperatures of 4°C.McCarty and Racker (1966),in searching for coupling factors for photophosphorylation, reported cold lability at ATPase activity. This loss of activity at 0°C was accelerated by salts and was pH dependent. Cyclic photophosphorylation of intact and broken chloroplasts isolated from frozen and unfrozen leaves of winter wheat and spinach was examined by Heber and Santarius (1967).Living and frost-killed leaves were supplied with radioactive sucrose, and in both

FROST AND CHILLING INJURY TO GROWING PLANTS

217

cases this sucrose was converted into a number of organic compounds including organic acids. It was concluded that the destruction resulting from freezing of the phosphorylation reactants (which provide the energy necessary to maintain life) takes place in vitro and in vivo. Freezing and/or freeze-drying is a common method for the long-term preservation of animal viruses and may be a satisfactory method for such plant viruses as tobacco mosaic, southern bean mosaic, tomato bushy stunt, and others (Kaper and Siberg, 1969). Turnip yellow mosaic virus, however, is structurally injured by in vitro freezing of its water solutions and is completely degraded into its KNA, which remains intact, and its protein component, which becomes predominatly fragmented (Kaper and Siberg, 1969). The temperature at which an enzyme is denatured by heat can be significantly increased for certain enzymes if they are preconditioned by exposure to increasingly higher temperatures. Similarly, conditioning of bean plants (Phaseolus acutifolius, var. Tepary Buff) to cool temperatures tended to increase the heat stability of the extracted malic dehydrogenase (Kinbacher, E. J., unpublished, University of Nebraska). Very strong contrasts to enzyme denaturation at subzero temperatures may be found in nonequilibrium freezing (lowering of temperature at rates in excess of 10- 100 centigrade degrees per minute) studies of single cells as well as of higher plants (Doebbler et al., 1966). IV.

Membrane Composition and Permeability

A. DESCRIPTION

Cellular membranes must also be considered in any discussion of freezing injury in plants. Nearly all cells killed by freezing and thawing show membrane damage (Mazur, 1966). Water movement from the cell to the extracellular space during slow freezing was previously discussed. This freezing process does not always kill the plant. Figure 4 illustrates the rate at which the supercooled water in plant cells (yeast) would be expected to equilibrate with the external frozen water by dehydration of the cellular constituents (Mazur, 1966). In addition to water movement across cell walls, we must consider water movement across organelle membranes, such as for mitochondria and chromosomes. Some correlation has been found between structural alteration of certain organelle membranes and associated enzyme activity as a function of freezing rate (Sherman and Kim, 1967). These authors pointed out that there are differences in the reaction and resistance of various organelles to ice formation and dissolution in and around them.

218

H. F. MAYLAND A N D J. W. CARY

-

0

0 -2 -4 -6 -8 -10 -12 -14 -16 -18 -20 Temperature, ("C)

FIG.4. Calculated percentages of supercooled intracellular water remaining at various temperatures in yeast cells cooled at indicated rates. Where V = volume of supercooled water in cell, and V , = initial water in cell. The dashed line represents equilibrium conditions. (From Mazur, 1966 Copyright 0 1966 Academic Press, New York.)

Thus, the damage at the cell membrane surface may cause a decrease in the capacity of the living protoplasmic membrane to serve as a barrier against ice inoculation into the cell. Freezing injury in nonhardy plants has been observed to result from a disruption of the diffusion barrier by intracellular ice formation and subsequent mechanical rupture which exposes cellular contents to the freezing site (Olien, 1961). Sakai and Yoshida (1968) concluded that freezing injury in cabbage cells resulted from disruption of the plasma membrane which is a very important structural component of the cell. Electron microscopy shows the cell membrane to be highly ordered. In almost all plant cells this membrane consists of a layered material approximately 75 A thick. Two dark electron-dense areas, each being about 25 A,are separated by a light layer. Stein (1967) represents the membrane schematically as a sandwich containing a bimolecular lipid center with polar groups on the exterior side. The paraffinic lipid sectors are bonded

FROST A N D CHILLING INJURY TO GROWING PLANTS

219

primarily by hydrophobic bonding (Green and Tzagoloff, 1966) to polypeptide chains or mucoproteins which make direct contact with the aqueous exterior or interior of the membrane system. B. COMPOSITION Chemical analyses show high concentrations of phospholipids, cholesterol, and protein. The lipid composition is species-dependent. The protein mass may be two to three times that of the lipids. Many of these proteins are enzymes, such as ATPase and acetylcholinesterase. The phospholipids are composed of a large hydrophilic phosphate ester grouping. It is expected that the hydrophilic groups of the phospholipid will be preferentially situated in the aqueous interface and the hydrophobic fatty acid chains will interlock with one another. The lipid composition may determine the membrane permeability (Christophersen, 1967). This is supported not only by an increase in the fatty acid content in hardened plants, but also by a preferential accumulation of polyunsaturated fatty acids, especially linoleic and linolenic fatty acids (Gerloff et al., 1966). Insects and microorganisms, in addition to higher plants, contain increased proportions of unsaturated fatty acids, or more highly unsaturated fatty acids, if they are grown at low temperatures (Chapman, 1967). This is further supported by Kuiper ( 1 969b), who reported that applying galactolipids to fruit flower buds increased resistance of flowers to freezing as tested 2-3 days later. Application of other lipid types resulted in decreased resistance. This relationship demonstrates the importance of lipids for water transport across membranes and for membrane stability against freezing. Siminovitch et al. ( 1 968) reported increases in polar lipids (principally phospholipids) and lipoproteins without changes in total lipids in living bark cells of the black locust tree during the development of extreme freezing resistance. Damage to the lipoproteins occurs when the last traces of water are removed as ice so that the lipoprotein complexes are brought into actual contact with one another (Keltz and Lovelock, 1955; Lovelock, 1957). Such disruption of the membranes may allow nucleation of the supercooled water within the cell. Heber ( 1 967) attributed the uncoupling of phosphorylation from electron transport (Section 111, B) to damage of chloroplast membranes resulting from freezing. Within the cell protoplasm are numerous bodies which are also enclosed within membranes. Chloroplast membranes are frost sensitive (Heber and Ernst, 1967). Increase in activity of some mitochondria1 enzymes and all those of lysosomes is found when these organelles in

220

H. F. MAYLAND A N D J . W. CARY

animal cells are disrupted as in freezing and thawing (Tappel, 1966). Plant cell microbodies, if similar to animal cell lysosomes (Frederick et al., 1968), are cell organelles containing families of hydrolytic enzymes in a nonreactive state. The lysosome membrane is a complex one consisting of a unit phospholipid-protein and associated protein. The membrane complex can be made permeable or can be disrupted by freezing and thawing. After release, lysosomal enzymes initiate catabolic reactions which could rapidly lead to considerable disorganization within the cell (Tappel, 1966). Because of their high latency and content of hydrolytic enzymes of broad specificity, the lysosomes appear to be the most important cell structure involved in the freezing injury (Tappel, 1966). C. PERMEABILITY Olien (1965, 1967b) extracted water-soluble, cell wall carbohydrate polymers from tissues of winter cereals. He reported that polymers isolated from cold-hardy tissue interact with the ice-liquid interface, resulting in less perfectly structured ice. The polymers had little effect on the freezing temperature, but interfered with the liquid solid reaction as a competitive inhibitor. Similar findings have been reported by Trumanov and Krasavtsev (1 966). It has been observed (Cary and Mayland, unpublished) that ice may form and melt in the leaves of such plants as peas (Pisurn sativurn), lettuce (Lactuca sativa), and sugarbeets (Beta vulgaris) without causing visible damage if temperatures do not drop below -5°C and the freezing time is not longer than 5 or 6 hours. Increases in membrane permeability accompany the cold-hardening process (Levitt, 1956). Plants like sugarbeets, peas, and lettuce may be protected from ice injury by highly waterpermeable membranes, or by some polysaccharide ice-interface reaction as suggested by Olien (1965, 1967b). It is possible that permeable membranes allow particular polysaccharides to move onto the extracellular surfaces where they can interact with growing ice crystals. Hassid (1 969), in his review of polysaccharide biosynthesis in plants, emphasizes the further importance of the plasma membrane as a source of cell wall building materials. V.

Protection from Freezing

A. EVIDENCE OF CHEMICAL EFFECTS

Protection against freezing damage has been obtained by microclimate modification. Adding water via surface or sprinkler systems has

FROST AND CHILLING INJURY TO GROWING PLANTS

22 1

been successful in some cases because of the great heat capacity of water. High-expansion foams which blanket plants, providing insulation against temperature changes, are being developed for use on low growing crops. Some data show that protection against freezing damage may be achieved by use of chemicals. Within the group of compounds generally classified as growth retardants are several which may protect against chilling (Tolbert, 1961) and freezing injury. One such compound, 2chloroethyl trimethylammonium chloride (CCC, also Cycocel) has provided at least limited protection against freezing damage, as well as protection against drought (Shafer and Wayne, 1967). Treatment with CCC increased freezing resistance in cabbage, one-year-old pear trees, tomatoes, and wheat (Shafer and Wayne, 1967; Michniewicz and Kentzer, 1965; Wunsche, 1966). Similar treatment with CCC increased winter hardiness of cabbage (Marth, 1965), and wheat (Toman and Mitchell, 1968). The compound 1,5-difluoro-2,4-dinitrobenzenegave protection to young bean plants against an 8-hour freezing period at -3°C (Kuiper, 1967). Similar responses have been reported for N,N-dimethylamino succinamic acid (B-nine, B995, and Alar). Significant increases in cold temperature tolerance were not observed after spraying tomato transplants with Alar (Hillyer and Brunaugh, 1969). Using this chemical resulted in more flowers on apple and cherry trees, and greater number of sweet corn ears (Cathey, 1964). CCC and B-nine, however, are relatively long lived (months to one year) and so may be undesirable for short-term protection of tender plants. The chemicals discussed here are generally classed as growth regulators. Their effect on flowering and final crop yield has not been fully evaluated. Some preliminary work suggests that snap bean yield (Sanders and Nylund, 1969)can be reduced and pea yield (Maurer et al., 1969) can be increased in some cases by applying B995 or CCC. Another chemical, N-decenylsuccinic acid, applied 4 hours before initiation of freezing temperatures, has been shown to prevent apple blossom injury when exposed to -6°C for 2 hours (Hilborn, 1967). Applying this compound at any time before the 4-hour prefreeze interval was ineffective. Earlier studies with this same fatty acid (Kuiper, 1964b) showed that the compound induces freezing resistance in young bean plants. When the fatty acid was sprayed on flowering peach, apple, and pear trees, most of the flowers resisted freezing injury at -6°C. Inducing frost resistance in strawberry flowers by application of decenylsuccinic acid and a few of its monoamides was reported by Kuiper (1967). Flower survival was: control, 8% ; decenylsuccinic acid, 10%; decenyl-N,N-dimethylsuccinamic acid, 30%; and decenyl-N,N-di-

222

H. F. MAYLAND A N D J . W. CARY

methylsuccinichydrazide, 40%. There is experimental evidence that decenylsuccinic acid is incorporated into the lipid layers of the cytoplasmic membrane, where it raises the membrane permeability to water where only the viscosity effect is observed (Kuiper, 1964b).The beneficial effects of decenylsuccinic acid found by Kuiper have been challenged. Newman and Kramer (1966), attempting to duplicate Kuiper’s findings (1964a,b), found that roots of intact bean plants are killed by exposure to M decenylsuccinic acid. They concluded that this chemical acted as a metabolic inhibitor and that increases in water permeability resulted from root injury. Heber and Ernst (1 967) isolated a high-molecular protein (possibly a nucleoprotein) from chloroplasts of hardy spinach leaves which was effective in protecting chloroplast membranes from frost injury. This isolated protein was also heat stable against 90°C for 2 minutes. Dycus (1969) observed less injury by high and low temperatures after spraying plants with zinc-containing compounds, but not copper or iron. He also isolated a subcellular particle from the tomato plant which seemed to be associated with zinc content and low temperature tolerance. Zinc ions are powerful inhibitors of ribonuclease (RNase destroys RNA) and could therefore influence protein synthesis (Hanson and Fairley, 1967). Since zinc concentrations in the plant are inversely related to RNase activity (Kessler, 1961), additional zinc would be helpful in controlling the activity of this hydrolytic enzyme, which might be released from plant cell microbodies (Section IV, B) during cold temperature stress. Zinc as well as boron and manganese may increase protoplasmic viscosity (Shkol’nik and Natanson, 1953) and, therefore, increase freezing resistance. DeVries and Wohlschlag ( 1969) have isolated a glycoprotein from an Antarctic fish which was responsible for 30% of the freezing-point depression of the fish’s serum. A more critical look at these substances and related compounds might provide opportunities for control of frost susceptibility in plants. B. MECHANISMS OF FREEZE-INJURY PROTECTION 1 . Bond Protection Although little direct evidence is available on the nature of freezing processes, it is suspected that, aside from direct mechanical rupture of ice crystals, lipoprotein alteration of membranes may be a primary cause of freezing injury (Heber and Santarius, 1964). Water removal during freezing may lead to lipoprotein injury. Structure alteration (denaturation)

FROST AND CHILLING INJURY TO GROWING PLANTS

223

could then be caused by hydrogen bond breakage allowing lipoprotein interaction. This is supported by the fact that hydroxyl-containing compounds such as sugars act as protective substances against the inactivation of the phosphorylation process. Heber and Santarius ( 1 964) explain the protective action of sugars and other compounds by their ability to retain or substitute for water via hydrogen bonding in proteins sensitive to dehydration. Sugars provide protective action against freezing injury in cabbage cells (Sakai, 1962). Apparently this protection is a surface phenomenon which prevents removal of the surface water layer from protoplasts by the dehydrating action associated with freezing and thawing. In simple cells, inositol (benzene ring surrounded by 6-OH) gives some degree of protection from stress such as freezing or radiation. It is suggested that the protection afforded by this chemical results from its ability to protect H-bonding (Webb, 1965) or possibly in substituting for the water structure. Sokolowski et al. ( 1 969), however, discounted the suggestion that inositol takes the place of water in maintaining the stability of desiccated cells. They suggested that the observed inositol effect may result from a conformational change in the protein brought about by inositol binding at positions adjacent to the reaction site. A large number of cryoprotective compounds (chemicals which preserve cellular integrity at subzero temperatures) have been evaluated for their effectiveness in protecting simple cell systems. Although these cyroprotective compounds may be diverse, some generalizations may be made even though the mechanisms of protection may not be completely explained. These generalizations go far toward correlating cryoprotective activity with-molecular structure (Doebbler, 1966). After examining the molecular structure of known cryoprotective solutes, it is apparent that all are capable of some degree of hydrogen bonding (Doebbler e f al., 1966). Some association of the cryoprotective agents with the cell membrane appears to also take place. Steric and electrostatic properties of protective additives perhaps act via effects on adsorption which can also influence the recoverability of frozen cells (Rowe, 1966). An interesting further generalization with regard to hydrogen bonding is the similarity in types of compounds that afford cryoprotection and those that protect microorganisms against drying or radiation, or which protect proteins against thermal denaturation (Doebbler, 1966; Webb, 1965).

2. Membrane

meets

Rowe ( 1966) has suggested that cryoprotective compounds interact directly or indirectly with the cell membranes to stabilize the water-

224

H. F. MAYLAND A N D J . W. CARY

lipid-protein complex tertiary structure. It is at the cellular membrane level that biological integrity appears to be insulted by freezing (Livne, 1969), and it is at the membrane level that biochemical understanding of cryoprotection must be sought. A correlation between mole equivalents of potential hydrogen bonding sites provided by a solute and protection of some simple cell systems during freezing has been reported (Doebbler, 1966). There is a question, though, as to how quantitative this method would really be because of the variation in hydrogen bonding energies. Jung et al. ( 1 967) found that applying certain purines and pyrimidines enhanced the development or maintenance of cold hardiness. Hardy plant varities contained greater amounts of DNA and ribonucleic acid (RNA) in the water-soluble trichloroacetic acid (TCA)-precipitable protein fraction than those of less hardy varieties during the development and maintenance of cold hardiness. In addition, the content of these constituents was increased by exposing the plants to low temperatures at a short photoperiod. The metabolic processes were altered by the chemical treatments in a manner that made the TCA-fraction of the nonhardy plants more nearly like that of untreated plants of a hardy variety. This supports the conclusions of others (Siminovitch et al., 1962) that watersoluble protein content is related to development and maintenance of cold hardiness. Thermostability of human and bovine serum albumin has been increased when the protein was combined with fatty acids and related compounds (Boyer et al., I946b). The protective action of the fatty acid ion increased with chain length up to C12, but maximum stabilization at high concentrations was obtained with C, and Cs. Native proteins were protected against heat denaturation by fatty acids which prevented viscosity increases in heated solutions. In another paper, Boyer et al. (1 946a) reported that low fatty acid concentrations prevented an increase in viscosity due to denaturation by urea or guanidine. The action of the fatty acid anions appears to result from their combination with certain groups or areas of the molecule and is probably the result of the combination of the anion with both the positive and the nonpolar portions of the protein. Protective action against freezing damage in higher plants has been evaluated from a standpoint of membrane permeability. Kuiper ( 1967) studied the effect of surface active chemicals as regulators of plant growth and membrane permeability. Several compounds, including the decenylsuccinic acid groups, were tested for their effects on water permeability of bean roots and growth retardation of young bean plants. In each group the effectiveness increased by increasing the number of carbon atoms.

FROST AND CHILLING INJURY TO GROWING PLANTS

225

There appears to be a definite effect of the hydrocarbon chain length of the surface active chemical on both permeability and resistance to freezing. These surface active chemicals probably affect permeability of the plasma membrane and its resistance to freezing by incorporating the molecules into the lipid layers of the plasma membrane. Charged lipids occurring in the plasma membrane may contribute in the same way to the permeability characteristics and the freezing resistance of the membrane. Kuiper (1 969b) also reported that when galactolipids were added to the root environment, an increase in water transport through the plant was observed. Applying this lipid to fruit flower buds increased resistance to flower freezing as tested 2 or 3 days later. The results demonstrate the relation of lipids to water transport across membranes and to membrane stability against freezing. Dimethyl sulfoxide (DMSO), which is a dipolar aprotic solvent with a high dielectric constant and a tendency to accept rather than donate protans, has been used as a carrier for many compounds used in cryoprotective studies. DMSO has been found to prevent loss of respiratory control and to decrease inefficiency of oxidative phosphorolation of plant mitochondria stored in liquid nitrogen (Dickinson et al., 1967). The mechanism by which DMSO protects some biological membranes against freezing damage is not known and, in fact, its beneficial effect of altering the permeability characteristics has been disputed in some studies (Chang and Simon, 1968). They, instead, attribute the in vivo effects of DMSO primarily to its ability to alter enzyme reaction rates. The native form of biopolymers is surrounded by the ordered arrays of water molecules. Substitution or removal of the biopolymer’s hydration sheath by DMSO would be expected to alter the protein configuration (Chang and Simon, 1968). It is possible, therefore, that at low concentrations DMSO permits a protein molecule, such as RNA, to assume a more open, less hydrogenbonded configuration. DMSO is an excellent fat solvent and has been shown to remove some fatty acids from the bacterial membrane (Adams, 1967), and membrane porosity may, therefore, increase (Ghajar and Harmon, 1968). This increase in permeability in cell membranes by DMSO may be similar to natural changes occurring during the cold-hardening processes. The protection afforded to plasma lipoproteins by polyhydroxyl compounds against damage by freezing or drying has a parallel in the case of some simple cells (Keltz and Lovelock, 1955). The mechanism of the damage caused by freezing and drying has some points of similarity to the picture of temporary collision complexes occurring in the exchange of lipids between lipoprotein complexes. The difference lies in the avail-

226

H. F. MAYLAND A N D J . W. CARY

ability of water molecules to interact with an exposed group in the parent lipoprotein complexes in solution. In the frozen system, the ice lattice may draw water molecues away from the lipoprotein complexes, disrupting the structure of the complexes. It is noteworthy that all the molecule species that protect against damage by freezing or drying are themselves rich in hydroxy groups (Keltz and Lovelock, 1955). Possibly their presence offers the lipoprotein complexes some alternative molecules to associate with in the place of water molecules that have become unavailable. In the presence of either water or some other molecule-containing hydroxyl groups, the lipoprotein complex may rearrange to allow some internal compensation and assume a configuration which returns to the original structure when water is readmitted to the system. C. UNDERCOOLING AND NUCLEATION Even though the temperature is below the freezing point of plant water, ice crystals may not form. The solution must first be nucleated. The nucleation of undercooled liquid water is not well understood. Pure water may cool below -30°C without forming ice. Elaborate preparations are required to demonstrate this, since even the slightest foreign particle may cause nucleation (Dorsey, 1948). The initiation of ice crystal formation is evidently a surface interface reaction. Davis and Blair ( 1969) have presented data suggesting that the presence of strain energy in suspended particles may enhance their ability to cause nucleation. It is not known whether or not lattice strain energies could be important in ice nucleation in plant tissue. This may be involved in the increased undercooling which occurs with faster cooling rates in some plants (Cary and Mayland, unpublished). Mechanical shock does not cause nucleation in either bulk solution or plant tissue (Dorsey, 1948; Kitaura, 1967). It has been established that ice forms preferentially in the extracellular spaces. The pressure of the water in the extracellular spaces may be important. Dorsey ( 1948) has stated that the spontaneous (undercooled) freezing temperature of solutions tends to parallel decreases in the true freezing point induced by the addition of salt. The freezing point may also be decreased by raising the pressure (Evans, 1967). If pressure affects the spontaneous undercooling in a similar way, nucleation should occur first in the extracellular spaces where liquid pressures are generally less than those inside the cells. An ice crystal is also an excellent nucleator. This nucleation may occur as ice from the atmosphere settles on the plant surface, contacting a continuous liquid film leading to the extracellular spaces and thus causing nucleation of the water in these areas.

FROST A N D CHILLING INJURY TO GROWING PLANTS

227

Single (1 964) found that wheat plants could be stored in a dry (low humidity) chamber at -3 to -5°C almost indefinitely without the formation of ice crystals. Yet he felt this would not occur in the field since crystals are present in the air which could initiate freezing of undercooled water in the plant upon contact with the leaf. When following the progression of ice formation in his wheat plants, Single noted varying resistance to freezing in different plant parts, the rate of advance of ice being sharply reduced at internodes compared to leaves. It is known that capillaries and type of solute affect both the rate of growth and the shape of the ice crystals (Pruppacher, 1967a,b). Beans (Phaseolus vulgaris, var. Pinto and Sanilac), corn (Zea mays), and tomatoes (Lycopersicon esculentum) were exposed to temperatures ranging from -2 to -3°C for various lengths of time (Mayland and Cary, 1969). When the relative humidity was less than loo%, the plants undercooled and ice crystals did not form for several hours. Eventually some plants did begin to freeze at random. When ice crystals were allowed to come in contact with the leaves, undercooling stopped and ice began to form in the tissues resulting in death of the tissue from mechanical cell rupture. In these experiments, it appeared that nucleation occurs on the surface if the atmospheric dew point is reached. Otherwise, nucleation occurs inside the plant tissue and factors within the plant may exert some influence on the spontaneous nucleation temperature. Results supporting this hypothesis have also been reported by Kitaura (1 967) and Modlibowska ( 1 962). Once nucleation has occurred, the ice phase spreads rapidly (i.e., with velocities of up to 1 cmlsec) through the conductive tissue, so long as the temperature of the tissue is below the freezing temperature of the solution it contains. The rate of ice nucleation from the conductive elements into the surrounding cellular tissue depends on the initial energy of the water in the plant, at least in the case of beans (Cary and Mayland, unpublished). When the energy of the plant water is high (-6 to -8 bars), the spread of ice is rapid throughout the leaves and results in death. If the plant-water energy is lower (- 12 to - 15 bars in the case of beans), the ice spreading rate is less by at least an order of magnitude, resulting in bean leaf damage of the type shown by Young and Peynado ( 1 967) for citrus leaves. The energy level of plant water is also related to the anatomical characteristic of leaf surfaces. It has been shown (Cary and Mayland, unpublished) that undercooled water in corn seedlings with a water potential of -18 bars is not nucleated by ice crystals on the leaf surface. Corn seedlings with higher potentials (-8 bars) are easily nucleated by exterior ice crystals. While the nature of this barrier to nucleation is not under-

228

H. F. MAYLAND AND J. W. CARY

stood, some aspects of the problem concerning nucleation sites have been discussed by Salt (1 963). Kaku and Salt (1 968) concluded that the freezing temperature of conifer needles increased ultimately as the number and quality of favorable nucleation sites increased. Hudson and Brustkern (1 965) observed permeability differences in young moss leaves of different ages and that the permeability increased with hardening and with age. Upon exposure to freezing temperatures, cells supercooled until a wave of intracellular freezing was initiated at -8°C in some leaves. These authors observed the freezing wave progressing from one cell to another and suggested that this probably occurred via the plasmodesmata. They further observed in very young leaves that the freezing did not start spontaneously, but was initiated by inoculation through the imperfectly developed cell walls at the apices of the leaves at approximately -4°C. As previously noted, some compounds provide freeze-injury protection to tender plants. Protection may result from changes in the membrane which render it less susceptible to rupture by ice crystals. The benefit may come about from permeability changes allowing rapid water transmission out of the cell to external ice crystals, thus reducing the chance of nucleation inside the cell. An alternative explanation, which has received little attention, is that these compounds may increase the stability of undercooled water (1) by changing the surface properties of the leaf so that ice crystals on the surface are not able to initiate nucleation in the extracellular spaces, or (2) by increasing membrane permeability allowing solutes from the cell to leak into the extracellular spaces, thus causing the fluid to be more stable to undercooling, or (3) by directly affecting the nucleation temperature of water in the plant. The reports on urea effects on frost tolerance fit into this possibility. Occasionally, spraying with urea has been credited with decreasing damage during mild freezes (van der Boon and Tanczos, 1964). However, as pointed out earlier, urea is very undesirable as far as protein denaturation in an ice crystal system is concerned. Urea strongly affects the molecular structure of water, and so it is possible that it could lower the nucleation temperatures and prevent ice crystal formation during a mild freeze, yet increase damage to the cell constituents if ice crystals do appear. D. CHILLING INJURY

It is known that temperatures well above freezing cause injury to many tropical plants. From the preceding sections it is not difficult to see how

FROST AND CHILLING INJURY TO GROWING PLANTS

229

a shift in hydrogen and hydrophobic bonds caused by cool temperatures could result in an irreversible change in specific enzymes. Moreover, it is possible that such changes occur in temperate climate plants during cool periods causing changes in growth that go unnoticed or unexplained. Protoplasmic streaming used as an index in cellular heat injury in plants may also be used to differentiate between cold-sensitiveand cold-insensitive plants. The differentiation is made on the ability of protoplasmic streaming to continue in cells exposed to cold temperatures. In some chill-sensitive plants, protoplasmic streaming stops suddenly when the tissue temperature drops to 1 WC, although streaming continues in cells from chill-resistant plants almost to 0°C (Langridge and McWilliam, 1967). Lyons et al. (1964) studied the physiochemical nature of mitochondria1 membranes of chilling-sensitive and chilling-resistant plants. The mitochrondria of chilling-resistant species had the greatest capacity for swelling and the greatest degree of unsaturation of the membrane fatty acids. This degree of unsaturation is directly related to membrane flexibility. The swelling and contraction of mitochondria are closely correlated to oxidative phosphorylation. The phosphorylative system, which is dependent on membrane associated enzymes, could be disrupted by membrane inflexibility, and the available ATP supply could be reduced. Low temperatures, 0-lWC, are known to promote callose plug formation in the conductive tissue of beans (Majumder and Leopold, 1967). Another example of chilling effects is the marked change which occurs in corn growth as a result of cool root temperatures (Walker, 1967). Stewart and Guinn (1969) have reported an extensive decrease in the ATP levels of cotton seedlings resulting from 2 days of chilling at 5°C. Kuraishi et al. ( 1 968) have shown that even peas (Pisum sativum), commonly thought to tolerate some cold stress, may show biochemical changes as a result of chilling. Chilling of germinating cotton seed reduced plant height, delayed fruiting, and reduced fiber quality in direct relation to cold exposure time (Christiansenand Thomas, 1969). Garden bean (Phaseolus vulgaris, var. Sanilac) seedlings exposed to -1 "C for 4 hours had delayed fruiting and maturity dates (Mayland, unpublished). Potato leaves may be injured in varying degrees of severity depending upon leaf maturity, relative maturity of leaf sections, position of leaf, temperature, and other factors (Hooker, 1968). Rainbow flint corn seedlings are sensitive to chilling injury up to 5.5 C (Bramlage, W. J., unpublished, University of Massachusetts). Chilling stress effects on plants appears to offer new and challenging frontiers for plant physiologists.

230

H. F. MAYLAND A N D J . W. CARY

VI. Conclusions

The freezing temperature of plant material is first of all determined by the chemical potential of the plant water. The freezing point drops approximately 1°C for each 12 bars equivalent negative pressure (in the range of 0 to - 10°C). Ice does not form spontaneously as the temperature drops to the freezing point of the plant water. Rather, the solution tends to undercool to variable and somewhat unpredictable temperatures. When nucleation does occur, it will generally be in the extracellular space. The chemical potential of ice is less than that of liquid water at the same temperature. Water will consequently move from the cell to the ice lens on the outside so that the cell dehydrates as the lens grows. If the cooling rate is fast or if the cell has a low permeability, the dehydration of the cell will be too slow to maintain a stable supercooled solution and ice crystals will appear inside the cell. Ice crystals inside the cell increase the chances of injury. As soon as ice appears, the plant begins to undergo desiccation due to the decreasing volume of the liquid phase. The resulting stress increases by approximately 12 bars effective negative pressure for each degree below 0°C. Thus, as the temperatures decrease, the water stress may become great enough to cause significant chemical damage through the disruption of bonds. This can lead to membrane changes and protein denaturation. Stress from desiccation does not develop with decreasing temperature when the solution undercools without forming the ice phase, though some bonding in large molecules may still be rearranged. When tender plants show immediate frost injury from temperatures not lower than -3 or -4"C, the damage is caused primarily by ruptured cell membranes. Some nonhardy plants such as peas and lettuce can survive temperatures of -3 or -4°C for a few hours, even though ice is present in the plant tissue. Other plants, such as beans and corn, may survive similar freezing conditions only if the plant water undercools without the spread of ice through the tissue. Under these conditions the dew point of the atmosphere and the water content of the plants are important factors. As the temperature decreases below -8 or -1o"C, the chances of the ice phase being absent are small and plants which survive these conditions with the accompanying stress of -80 to -100 bars become those in the cold-hardy group. These are known to tolerate the growth of ice crystals in the intercellular cavities. When injury occurs in these plants, it is more apt to be from chemical bonding changes resulting from desiccation. Freezing and survival of cold-hardy plants has received much attention. Further progress in this area rests mainly in the realm of biochemistry,

FROST A N D CHILLING INJURY TO GROWING PLANTS

23 1

particularly with respect to composition and bonding in lipids and proteins. Causes of cold injury in some plants, aside from the mechanical forms of injury induced by ice formation, may not be separable from some of the effects of enzyme inactivation observed during freezing conditions. Here, the in vitro studies of Heber and Santarius (1964) and Young ( 1 969) on the sensitivity to freezing by enzymes of the election transport system of photosynthesis may provide further clues to chilling and freeze injury in plants. The in vivo studies of Stewart and Guinn ( 1 969), which showed a decrease in ATP of seedlings chilled at 5”C, certainly give support to the biochemical approach of studying low temperature stresses in plants. Future studies of the temperature sensitivity of the oxidative- and photophosphorylation system of plants should help describe injuries resulting from exposure to either cold or freezing temperatures. Freezing of nonhardy plants has received little study in spite of great economic importance. It is possible that significant advances may now be made in this area. Five areas in particular need to be studied: a. The chemical control of cell membrane permeability b. The identification and control of polysaccharides or other molecules which interact with the ice-water interface c. The internal control of nucleation of undercooled plant solutions d. The surface properties that prevent ice particles in the atmosphere from nucleating water in the plant e. Identification of low temperature-sensitive links in the electron transport system and evaluation of such links to determine opportunities for genetic alteration. REFERENCES Adams, G . A. 1967. Can. J . Biochem. 45,422-436. Barskaya,T. A.,and Vichurina,G. A. 1966. Biol. Nauki 2,153-155. Battell, M. L., Smillie, L. B., and Madsen, N. B. 1968. Can.J. Biochem. 46,609-615. Bello, J . 1966. Cryobiology 3,27-3 I . Bernal, J . D. 1965. Symp. SOC.Exp. Biol. 19,17-3 1. Borzhkovskaya, G . D., and Khrabrova, M. A. 1966. Fiziol. Rasr. 13,720-724. Boyer, P . D., Ballou, G. A., and Luck, J. M. 1946a.J . Biol. Chem. 162,199-208. Boyer, P. D., Luin, F. G., Ballou, G . A., Luck, J . M., and Rice, R. G. 1946b. J . Biol. Chem. 162,181-198. Cathey, N. N . 1964. Annu. Rev. PlantPhysioL 15,27 1-302. Chang, C.-Y., and Simon, E. 1968. Proc. SOC.Exp. Biol. Med. 128,60-66. Chapman, D. 1967. In “Thermobiology” (A. H. Rose, ed.). pp. 123-146. Academic Press, New York. Christiansen, M.N., and Thomas, R. 0. 1969. Crop Sci. 9,672-673.

232

H. F. MAYLAND AND J. W. CARY

Christophersen, J. 1967. I n ‘‘Molecular Mechanisms of Temperature Adaptation,” Publ. No. 84, pp. 327-350. Am. Assoc. Advance. Sci., Washington, D. C. Cox,C. S. 1968.5. Gen. Microbiol. 50, 139-147. Daniel], J. W., and Crosby, F. L. 1968. Proc. Amer. SOC.Hort. Sci. 93,128- 134. Davis, B. L., and Blair, D. N. 1969.5. Geophys. Res. 74,4571-4580. DeVries, A. L., and Wohlschlag, D. E. 1969. Science 163,1073-1075. Dickinson, D. B., Misch, M. J., and Drury, R. E. 1967. Science 156,1738-1739. Doebbler, G . F. 1966. Cryobiology 3,2- I 1. Doebbler, G. F., Rowe, A. W., and Rinfret, A. P. 1966. I n “Cryobiology” (H. T. Meryman, ed.), pp. 407-450. Academic Press, New York. Dorsey, N. E. 1948. Trans. Amer. Phil. SOC. [N.S.] 38,247-328. Dycus, A. M. 1969. HortScience 4,153. Evans, L. F. 1967. Nature (London) 213,384-385. Frederick, S.E., Newcomb, E. H., Vigil, E. L., and Wergin, W. P. 1968. Planta 81,229-252. Gaff, D. F. 1966. Ausr. 5. Bid. Sci. 19,291-299. Gerloff, E. D., Richardson, T., and Stahmann, M. A. 1966. Plant Physiol. 41,1280-1284. Ghajar, B. M., and Harmon, S. A. 1968. Eiochem. Eiophys. Res. Commun. 32,940-944. Graves, D. J., Sealock, R. W., and Wang, J. H. 1965. Biochemistry 4,290-296. Green, D. E., and Tzagoloff, A. 1966.5. Lipid Res. 7,587-602. Hanson, D. M., and Fairley, J. L. 1967. 5. Biol. Chem. 244,2440-2448. Hassid, W. Z. 1969. Science 165,137- 144. Heber, U. 1967.PlantPhysiol. 42,1343-1350. Heber, U., and Ernst, R. 1967. In “Cellular Injury and Resistance in Freezing Organisms” (E. Asahina, ed.), Proc. 11, pp. 63-77. Inst. Low Temp. Sci., Hokkaido Univ., Sapporo, Japan. Heber, U. W., and Santarius, K. A. 1964. Plant Physiol. 39,7 12-7 19. Heber, U. W., and Santarius, K. A. 1967. I n “The Cell and Environmental Temperature” (A. S. Troshin, ed.) pp. 27-34. Pergamon Press, Oxford. Hilborn, M. T. 1967. Phytoparhology 57,34 1. Hillyer, 1. G., and Brunaugh, H. L. 1969. HortScience 4,153. Hooker, W. J. 1968.Amer. PotafoJ. 45,250-254. Hudson, M. A.,and Brustkern, P. 1965. Planta 66,135-155. Idle, D. B., and Hudson, M. A. 1968. In “Low Temperature Biology of Foodstuffs” (J. Hawthorn, ed.), pp. 153- 175. Pergamon Press, Oxford. Ivanova, T. L a n d Semikhatova, 0.A. 1966. Eot. Zh. (Leningrad) 51,1266-1275. Jung,G. A., Shih, S.C.,and Shelton, D. C. 1967. PlantPhysiol. 42,1653-1657. Kaku,S.,andSalt,R.W. 1968.Can.J.Bor.46,1211-1213. Kaper, J. M., and Siberg, R. A. 1969. Cryobiology 5,366-374. Kauzmann, W. 1959. Advan. Protein Chem. 14,l-63. Keltz, A., and Lovelock, J. E. 1955. Fed. Proc., Fed. Amer. SOC.Exp. Biol. 14,84. Kessler, B. 1961. In “Plant Analysis and Fertilizer Problems” (B. Kessler and W. Reuther, ed.), pp. 314-322. Am. SOC.Hort. Sci. St. Joseph, Michigan. Kitaura, K. 1967. I n ‘‘Cellular Injury and Resistance in Freezing Organisms” (E. Asahina, ed.), Proc. 11, pp. 143-156. Inst. Low Temp. Sci., Hokkaido Univ., Sapporo, Japan. Krull, E. J. 1967. Ph.D. Thesis, University of Missouri (University Microfilms No. 67-917, Ann Arbor, Michigan). Kuiper, P. J. C. 1964a. Science 143,690-691. Kuiper, P. J. C. 1964b. Science 146,544-546. Kuiper, P. J. C. 1967. Meded. Landbouwhogesch. Wageningen 67-3,1-23.

FROST AND CHILLING INJURY TO GROWING PLANTS

233

Kuiper, P. J. C. 1969a. Proc. 1st Intern. Citrus Symp., 1968, Vol. 2, pp. 593-595. Kuiper, P. J. C. 1969b. Proc. 11th Int. Bor. Cong.. 1969, p. 118. Kuntz, I. D., Jr., Brassfield, T. S., Law, G. D., and Purcell, G. V. 1969. Science 163,13291331. Kuraishi, S., Arai, N., Ushijima, T., and Tazaki, T. 1968. Planrfhysiol. 43,238-242. Langridge, J. 1963.Annu. Rev. Plantfhysiol. 14,441-462. Langridge, J., and McWilliam, J. R. 1967. I n “Thermobiology” (A. H. Rose, ed.), pp. 231292. Academic Press, New York. LeFebre, V. 1967. J. Colloid Inreiface Sci. 25,263-269. Levitt, J. 1956. “The Hardiness of Plants.” Academic Press, New York. Levitt, J. 1962.J. Theor. Biol. 3,355-391. Levitt, J. I966a. Cryobiology 3,243-25 1. Levitt, J. 1966b. I n “Cryobiology” (H. T. Meryman, ed.), pp. 495-563. Academic Press, New York. Levitt, J. 1967. I n “Molecular Mechanisms of Temperature Adaptation,” Publ. No. 84, pp. 4 1-5 1. Am. Assoc. Advance. Sci., Washington, D. C. Levitt, J., and Haseman, M. 1964. Plant Physiol. 39,409-412. Levitt, J., Sullivan, C. Y., Johansson, N., and Pettit, R. M. 1961. Plant Physiol. 36,611616. Lippincott, E. R., Stromberg, R. R., Grant, W. H., and Cessac, G. L. 1969. Science 164, 1482- 1487. Livne, A. 1969. Isra. J. Chem.,Proc. p. 152. Love1ock.J. E. 1957. Proc. Roy.Soc., Ser. E 147,427-433. Luyet, B. J., and Gehenio, P. M. 1940. Eiodyanmica 3,33-99. Lyons, J. M., Wheaton, T. A., and Pratt, H. K. 1964. Plant Physiol. 39,262-268. McCarty, R. E., and Racker, E. 1966. Erookhaven Symp. Biol. 19,202-214. McCown, B. H., Hall,T. C.,and Beck,G. E. 1969a. Plantfhysiol. 44,210-216. McCown, B. H., McLeester, R. C., Beck, G . E., and Hall, T. C. 1969b. Cryobiology 5, 410-412. Majumder, S. K., and Leopold, A. C. 1967. Plant Cellfhysiol. 8,775-778. Marth, P. C. 1965. J. Agr. FoodChem. 13,331-333. Matsubara, H. 1967. I n “Molecular Mechanisms of Temperature Adaptation,” Publ. No. 84, pp. 283-294. Am. Assoc. Advance. Sci., Washington, D. C. Maurer, A. R., Ormrod, D. P., and Adedipe, N. 0. 1969. HortScience 4,182. Mayland, H. F., and Cary, J. W. 1969.Agron. Abstr. p. 33. Mazur, P. 1966. In “Cryobiology” (H.T. Meryman, ed.), pp. 213-315. Academic Press, New York. Mazur, P. 1969. Annu. Rev. Plant Physiol. 20,419-448. Meryman, H. T. 1966. I n “Cryobiology” (H.T. Meryman, ed.), pp. 1- 114. Academic Press, New York. Michniewicz, M., and Kentzer, T. 1965. Experientia 21,230-23 I. Modlibowska, 1. 1962.J. Hort. Sci. 37,249-261. Nemethy,G.,and Scheraga, H. A. 1962a.J. Chem. Phys. 36,3401-3417. Nemethy,G.,and Scheraga, H. A. 1962b. J. Phys. Chem. 66,1773-1789. Newman, E. L a n d Kramer, P. J. 1966. Plantfhysiol. 41,606-609. Ng, H. 1969. J. Bacteriol. 98, 232-237. Olien, C. R. I96 1. Crop Sci. 1,26-28. Olien, C. R. 1965. Cryobiology 2,47-54. Olien, C. R. 1967a. Annu. Rev. Plant Physiol. 18,387-408.

234

H. F. M A Y L A N D A N D J . W. CARY

Olien, C. R. 1967b. Crop Sci. 7, 156-157. Pauling, L. 1960. “The Nature of the Chemical Bond,” 3rd ed. Cornell Univ. Press, Ithaca, New York. Pichel, W. 1965. Amer. SOC.Heating, Refrig. Aircond. Engrs. J . 7,68-71. Pihl, E., and Falkmer, S. 1968. Histochemie 13, 289-295. Pruppacher, H. R. I967a. J. Glaciol. 6,65 1-662. Pruppacher, H. R. 1967b. 2.Naturforsch. A 22,895-901. Pullman, M. E., Penefsky, H. S., Datta, A., and Racker, E. 1960.J. Biol. Chem. 235,33223329. Reingold, N. 1960. “Cold Weather Agriculture,” Senate Doc. 91. U S . Govt. Printing Office, Washington, D.C. Rowe, A. W. 1966. Cryobiology 3, 12-18. Sakai, A. 1962. Nature (London) 193,89-90. Sdkai, A., and Yoshida, S. 1968. Cryobiology 5, 160- 174. Salt, R. W. 1963. Can. Entomol. 95, 1190-1202. Sanders, D. C., and Nylund, R. E. 1969. HortScience 4, 182. Scarth, G. W. 1944. New Phytol. 43, 1-12. Scrutton, M. C., and Utter, M. F. 1964. Fed. Proc., Fed. Amer. SOC.Exp. Biol. 26, 162. Sellschop, J. P. F., and Salmon, S. C. 1928. J. Agr. Res. 37,3 15-338. Shafer, N. E., and Wayne, N. J. 1967. Proc. 6th lnt. Congr. Plant Protect., 1967 pp. 463-465. Sherman, J. K., and Kim, K. S. 1967. Cryobiology 4,61-74. Shikama, K. 1965a. Nature (London) 207,529-530. Shikama, K. 1965b. Sci. Rep. Tohoku Univ., Ser. 4 31,67-73. Shkol’nik, M. Y., and Natanson, N. E. 1953. Dokl. Akad. Nauk S S S R 88, 1067-1070: Chem. Abstr. 47, 11359i (1953). Siminovitch, D., Therrien, H., Wilner, J., and Gfeller, F. 1962. Can.J. Bot. 40,1267-1269. Siminovitch, D., Rheaume, B., Pomeroy, K., and Lepage, M. 1968. Cryobiology 5,202225. Single, W. V. 1964. Aust. J . Agr. Res. 15, 869-875. Smith, D. 1968. Cryobiology 5, 148-159. Sokolowski, M. B., Weneck, E. J., Trkula, D., and Bateman, J. B. 1969. Biophys. J . 9, 950-953. Stein, W. D. 1967. “The Movement of Molecules Across Cell Membranes.” Academic Press, New York. Stewart, J. McD., and Guinn, C. 1969. Plant Physiol. 44,605-608. Tappel, A. L. 1966. In “Cryobiology” (H. T. Meryman, ed.), pp. 163-177. Academic Press, New York. Tolbert, N. E. 1961. I n “Plant Growth Regulation” (R. M. Klein, ed.), pp. 779-786. Iowa State Univ. Press, Ames, Iowa. Toledo, R., Steinberg, M. P., and Nelson, A. I. 1968. J. Food Sci. 33,3 15-3 17. Toman, F. R., and Mitchell, H. L. 1968.J.Agr. Food Chem. 16,771-772. Trumanov, 1. I., and Krasavtsev, 0. A. 1966. Soviet Plant Physiol. 13,439-444. van der Boon, J., and Tanczos, 0. G. 1964. Mede. Dir. Tuinbouw (Neth.) 27, 201-209. Walker, J . W. 1967. Soil Sci. SOC.Amer. Proc. 31,400-403. Webb, S . J. 1965. “Bound Water in Biolokical Integrity.” Thomas, Springfield, Illinois. Wunsche, U. 1966. Natunvissenschafren 53,386-387. Young, R. 1969. Proc. 1st Intern. Citrus Symp., 1968 Vol. 2, pp. 553-558. Young, R., and Peynado, A. 1967. Proc. Amer. SOC.Hort. Sci. 91, 157-162.

THE PLATINUM MICROELECTRODE METHOD FOR SOIL AERATION MEASUREMENT D. S. Mclntyre Commonwealth Scientific a n d Industrial Research Organization, Canberra, Austral io

I. The Method ....................................................................................... A. Introduction .............................................................. B. Basis of the Method and Application .......................... C. Review Outline ............................... 11. Electrochemistry ..... A. Reaction .......... B. Reaction Rate . .............................. C. Effect of p H . .................. E. Electrode Kinetics ....

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

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

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

Current-Moisture Relations ........ F. Current-Time Relations ...... G . Voltage-Time Relations ...... V. Conclusions ........................... VI. 0 2 Flux and Plant Response .......... E.

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

References .......

Page 235 235 236 24 I 24 1 243 245 248 252 253 266 266 267 268 268 268 269 27 1 272 275 275 276 278 28 I 28 I

I. The Method

A. Introduction

Although it has usually been considered that poor soil aeration is detrimental to plant growth, a definite relationship between plant growth and some measure of soil aeration has been sought with little success; reviews of soil aeration and plant growth studies have been published by Vilain ( 1 963) and Grable ( 1 966). Effort was concentrated initially on 235

236

D. S. MCINTYRE

the gas phase, but measurements of air-filled porosity, O2 and C 0 2 gas concentrations and diffusion rates, as well as air permeability, gave only very limited success. On the grounds that “evaluation of conditions at the interface between the root surface and the soil system presents the greatest possibility of ascertaining the influence of soil aeration on plant growth” and that “active root surfaces are covered with water films” so that “some method of measuring oxygen diffusion . . . through the liquid phase to a reducing surface similar to that of a plant root, would be of hndamental importance,” Lemon and Erickson (1 952, 1955) introduced the platinum microelectrode method. It was said to measure in situ the rate of oxygen diffusion through soil solution to a simulated root in the form of a thin platinum wire, at which O2was electrochemically reduced.

B. BASIS OF THE METHOD A N D APPLICATION Adapted from biological techniques for tissue and blood O2 content, in which it is referred to as the “oxygen cathode” (Davies, 1962), the method involves measurement of the electric current caused by the reduction of O2 at the surface of the cylindrical wire electrode, which is usually 4-5 mm long, with diameter in the range 0.4-1.2 mm, but most commonly about 0.5 mm. The microelectrode, suitably mounted, is pushed into the soil and cathodized with respect to a standard reference electrode. The ensuing current decreases with time as shown in Fig. la. 14

(a)

12

10

4

5

+

8

C

!i

3

0

6

4

2

0

0

2

3 Time (min)

4

5

6

I a - 21% b - 157% c - 10 2% d - 4.1%

0

1

0

" 01

"

'

02

" " ' 03 04

(b)

'

05

~

' ~ ' 06 07

~

00

'

~

0.9

Applied voltage (volts-ve vs SCE)

25

-

20

-

I

4

a

1

; 15 ?!

-

3

0

I0

-

5-

3 Cathode voltage (volts v e v s S C E )

FIG. 1. Microelectrode current relations for reduction of O2 in porous media. (a) Current-time relations for various unsaturated media. (b) Current-voltage relations for watersaturated media: 0 , sand: X. clay suspension: 0, glass beads 18p median diameter: letters u-e refer to concentration of Or in equilibrium with saturating solution; SCE, saturated calomel electrode. (c) Current-voltage relations for an unsaturated sandy loam; I, current as a function of applied voltage; 11, current as a function of effective voltage.

238

D. S. MCINTYRE

By plotting the current at a certain time against the applied voltage, current-voltage relationships can be established. For a clay suspension, soil paste, or water-saturated porous medium, a relation of the type shown in Fig. l b is obtained. Such a relation implies that if a voltage within the plateau region is applied to the microelectrode, the current will be independent of voltage, and limited only by the rate of diffusion of O2 to the electrode surface. Current at time t is given by

where & = Oo flux at time r at the surface of an electrode of radius a. Oxygen JIux (traditionally but incorrectly called oxygen diffusion rate, or ODR) may therefore be determined by measuring the current, usually after attainment of a quasistationary state at 4 or 5 minutes, and by use of the working equation

where i, = current in microamps, M = molecular weight of 0 2 (= 32 glmole), n = number of equivalents/mole of O2 (= 4), F = the Faraday (= 96,500 coulombslequiv),A =area of the electrode in cm2. Since introduction of the method, it has been used extensively in studies of the effect of aeration on plant and root growth and the behavior of soil organisms (see Stolzy and Letey, 1964a,b). Interpretation of results has been based on the model depicted in Fig. 2, which is considered to represent soil conditions near a root, or the microelectrode, in unsaturated soil (Wiegand and Lemon, 1958, 1963; Lemon and Kristensen, 1961; Kristensen and Lemon, 1964; Letey and Stolzy, 1967). For this model it is assumed that the flux is limited by the diffusion rate of O2 in solution to give current-voltage relations of the type in Fig. 1 b. The oxygen flux for such a coaxial cylindrical model is given by

where D = the effective diffusion coefficient for O2 in the soil solution; Cb = O2 concentration at the liquid-gas interface at mean radius b; C , = O2concentration at the electrode surface, at radius a. The mean thickness of the assumed liquid film separating the electrode from the gas phase is given by (b - a ) ; it will obviously depend on the moisture content of the soil. Although there must be some liquid con-

SOIL AERATION BY PLATINUM MICROELECTRODE

Platinum surface

239

- Water

\ sol I particles

Water film

(bl

Platinum electrode

FIG.2. The model which is assumed in order to explain microelectrode behavior. (a) Particles and solution separating the electrode from gas-filled pores (after Kristensen and Lemon, 1964). (b) Coaxial cylindrical model with water film of mean thickness 6r separating the electrode from gas-filled pores.

tinuity between the working electrode and the reference electrode, this model was considered to represent the physical conditions sufficiently well. In operation of the method, it is assumed that cb is the equilibrium concentration in solution for the particular partial pressure of O2 in the gas phase, and C, = 0. Putting ( b - a ) = 6 r , Eq. (3) can then be written

and 6 r has been called the “apparent mean liquid path length” (Kristensen and Lemon, 1964) and “mean filmthickness” (Letey and Stolzy, 1967). In general the method has been applied to soil-plant studies in two ways. First, using Eq. (2) there have been attempts to correlate plant and organism behavior with O2flux, or to determine a flux that is critical for

240

D. S. MCINTYRE

root growth.’ Second, by application of Eqs. (2) and (4), values of 6 r critical to root growth have been determined (Wiegand and Lemon, 1958, 1963; Kristensen and Lemon, 1964; Stolzy and Letey, 1964a,b; Letey and Stolzy, 1967). The last authors have measured a value of 20 x I 0-8 g ern+ min-’ as the minimum for root extension of many plants, and subsequently determined equivalent critical values of 6r for specified porosity and O2conditions. It should be noted that as fa,, is dependent on the radius of the electrode, and on the time of reading the current, these must be standardized for comparison of results. Lemon and Erickson (1 955) realized that the “boundary conditions around an electrode are not strictly similar to those around an exploring root” but stated that “electrode measurements have suggested that boundary conditions around an active root are more important to the rate of oxygen supply than was previously realized.” The apparent dependence, from this model, of oxygen flux on the effective diffusion coefficient in the solution, and on the concentration gradient across the water film, meant that the flux should be dependent on three main soil properties. They are moisture content, structure of the soil next to the electrode, and the O2 concentration in the gas-filled pores. As the last would be related to gaseous diffusion in the soil, the method gave promise of integrating the effects of those properties which control soil aeration and the rate of 0 2 uptake by roots. However, recent investigations of the method have shown that in unsaturated soils the current-voltage relations of the microelectrode do not show a plateau. In general the relations are of the form shown in Fig. lc. Such a relationship suggests that the process governing the rate of O2 removal by the microelectrode is not diffusion of O2through a water film. It throws considerable doubt on the diffusionmodel of Fig. 2, and has important implications in the interpretation of measurements made with the diffusion model in mind. Covey and Lemon (1962) in an analysis based on the coaxial cylindrical model, showed divergence of theoretical and experimental results for periods greater than 3 minutes. They concluded that the electrode (and plant root) geometry does not strictly meet the boundary conditions of the model. However, it can be replaced by another model (shown in Fig. 6) based on consideration of the electrode kinetics of a microelectrode inserted in unsaturated porous media (three-phase system). Although the microelectrode method is basically an electrochemical one, it was used widely for soil-plant studies without a thorough appraisal of the electrochemical principles which govern its response in three-phase It has been tacitly assumed that the rate-controlling process for O2 uptake by a root is also diffusion through an encompassing film of water.

SOIL AERATION BY PLATINUM MICROELECTRODE

24 I

media. In two-phase media, it has often been demonstrated experimentally that transport rate of 0 2 has an overwhelming influence on current limitation. For three-phase media it has been assumed that the response is similarly governed. Hence kinetics of the reaction, and the effect of transport rate of the product (OH-) from the electrode, were not initially investigated in three-phase media, while soil properties such as pH and electrical resistance were thought to have little influence on electrode response. Subsequent more critical investigations and analyses (Kristensen, 1966; Mclntyre, 1966b, 1967),have suggested that through lack of knowledge of the fundamental electrochemicalprinciples involved, the results of many workers are not universally applicable.

C. REVIEWOUTLINE Because the method is being widely used in agronomic, ecological, and sometimes plant physiological experimentation, any limitations that are inherent in the method because of the electrochemical and physical principles governing its behavior, should be examined. Therefore the remainder of this review will examine (1) the electrochemistry from the point of view of the reactions at the electrode, electrode kinetics, and transport processes; (2) the effect of the environment on the behavior of the microelectrode; (3) physical effects of insertion of the electrode on this environment. The proposed models will be reviewed and related to experimental results. Finally, conclusions will be made regarding the usefulness of the platinum microelectrode method and the reliability of measurements made with it. Details of the experimental method have been published several times previously (Birkle et al., 1964; Letey and Stolzy, 1564; Stolzy and Letey, 1964a; Lemon and Erickson, 1952, 1955; van Doren and Erickson, 1966; van Doren, 1958) and will be mentioned here only when results are dependent on the particular technique adopted. Neither will the problems of experimental use, such as extreme variability occurring in field conditions, be discussed; they have been treated by the workers mentioned above. This review is concerned rather with the fundamental principles underlying operation of the electrode in soil, as a system of controlled potential electrolysis of oxygen, and how those principles affect application of the method to soil-plant studies. II.

Electrochemistry

Electrochemical analysis is important in determining which process controls the current resulting from O2 reduction at the working electrode.

242

D. S. MCINTYRE

It is necessary to know the nature of the reaction, its products, and their subsequent behavior, the influence of transport processes and of pH in determining electrode response. One can then deduce, using experimental evidence, likely rate-controlling processes in soils where a variety of conditions may be met. The object is to determine in particular whether transport of 0 2 ,or some other process, is dominant in controlling the rate of reduction of 0 2and , hence the current. Previously it has been assumed that transport rate of O2 is the dominant rate-controlling process in both root uptake and in the electrochemical reduction of 0 2 . Evidence regarding the apposite reactions, and their rate in relation to transport processes, comes mainly from solution electrochemistry. This evidence indicates that when a platinum electrode having surface properties similar to those expected with the microelectrode, is used in solution, enhancement of transport of 02 by forced or even by natural convection, may alter the current controlling process from one of O2 supply rate to that of reaction rate. Similarly in porous media, it is known that for water-saturated conditions in which the nearest gas-liquid interface is the soil surface, the current is controlled by the rate of diffusion of O2 along very small concentration gradients. However as moisture is lost and air enters pores near the electrode, the opportunity exists for concentration gradients between the nearest gas-liquid interface and the electrode to be orders of magnitude greater than in saturated soils. This could then make possible a very high transport rate of O2 which may make the reaction rate the current-controlling process over parts of theelectrode. It is important also to recognize the influence of pH, for the voltage at which H+ ion reduction begins is dependent on the pH of the solution surrounding the electrode. As well, pH may affect the type of reaction occurring, and the reaction rate. Solution electrochemistry indicates that the pH has little effect on decomposition voltage2 of H+ until the pH decreases to about 4 (see later). This value is not often found in measurement on soils, but it must be remembered that soil pH is measured on an extract, and that the measured value decreases as the soil-water ratio increases. Thus the in situ pH value will be less than that measured on an extract. It will decrease with water content, and also with an increase in the salt concentration. Moreover the decomposition voltage for H+ zDecomposition voltage is defined by Potter (1961); here it is taken to mean that electrode voltage at which measurable current first occurs. Characteristic voltages for reduction or oxidation are usually expressed in terms of half-wave potential (EI/z),but the decomposition voltage is more pertinent in the present treatment.

SOIL AERATION BY PLATINUM MICROELECTRODE

243

depends as well on 02 concentration (Oden, 1962) and on current density (Potter, 196 1).

A.

REACTION

The net reactions commonly accepted as occurring at the platinum microelectrode have been (Lemon and Erickson, 1955; Oden, 1962; Stolzy and Letey, 1964a): (a) for acid media (Kolthoff and Lingane, 1952) or at pH < 5 (Charlot ef al., 1962) O2 + 4 H + + 4e--

2H20

(5)

(b) for alkaline media (Kolthoff and Lingane, 1952) or at 5 < pH < 12 (Charlot et al., 1962) Op

-

+ 2 H 2 0 + 4e-

40H-

(6)

From the theoretical and experimental investigations, Oden (1 962) considers that reaction (2) is relevant to a lower pH limit of about 3.5-4.0 for air-saturated solutions. Therefore we will consider only this reaction, as soil pH normally falls within the limits specified. A good deal of controversy exists among electrochemists as to the exact reaction. Apart from specific details, two substantially different courses have been proposed (Williams, 1966). The first involves production of H 2 0 2and its subsequent reduction, so that the reaction in more detail becomes

+ 2eH 2 0 2+ 2eO2+ 2 H 2 0 + 4e-

O2 + 2H,O

--

20H-

+ H202

20H-

-

40H-

(7)

The alternative hypothesis suggests that electroreduction of oxygen proceeds at a metal electrode by the formation and subsequent decomposition of surface oxides. In this case the overall reaction is of the form:

+ 2M + 2 H 2 0 + 2e2MOH + 2eO2+ 2 H 2 0 + 4e0 2

2M0

--

+2 M 0

+

2MOH 2 0 H 2M 2 0 H -

+

40H-

244

D. S. MCINTYRE

The third step in the reaction is rate-controlling when platinum electrodes are used. Both reactions require 4 electrons for the reduction of a molecule of 0 2 and , reaction (8) shows formation and destruction of an oxide and hydroxide of the metal. Support for the presence of H 2 0 2 comes from qualitative measurements by Laitinen and Kolthoff (1941), and quantitative measurement by Delahay ( I 950), during controlled electrolysis for long periods. Sawyer and Interrante (1961) considered that H Z 0 2was the primary reduction product at a prereduced electrode and proposed a mechanism invoking formation of platinum hydroxides during the reaction. Kozawa ( 1964) concluded that the proportion of H d 0 2formed electrochemically depended on the surface conditions of the platinum electrode, such as presence of oxide and/or adsorption of ions present in solution, and that the H 2 0 2 was decomposed catalytically. Further evidence of H 2 0 z formation was presented by Muller and Nekrassow (1965), who made measurements using a rotating disc electrode with a guard ring. They found that the presence of oxides of platinum retarded the reduction of O2to H 2 0 2(but not reduction to OH-), in alkaline solution, and that this reaction occurred rapidly only on oxide-free surfaces. The surface oxide, however, accelerated reduction of H202 by catalytically decomposing it. Other authors, mainly from purely electrochemical analysis, are of the opinion that the reaction proceeds without the production of H 2 0 z . Lingane ( 196 1) from chronopotentiometric measurements concluded that HeOr was not produced, and Riddiford (1961) on the basis of analysis of various data was of the same opinion. In a recently published book Hoare ( I 968) states that H 2 0 2is always formed by reduction of 0 2 at a cathode, but that at a current density less than A/cm2 (an approximate current of 6 microamps for the electrodes in common usage) catalytic activity of platinum is sufficient to reduce the H 2 0 2at the same rate as it is formed. For a greater current density HEOzaccumulates. Impurities can lower the catalytic activity and hence the required minimum current density for HZO2accumulation. Further reviews and references can be found in Williams (1966), Rickman et al. (1968), and Rickman (1966). The results have been obtained by a variety of techniques with variations in electrode pretreatment (and hence electrode surface properties), electrolyte nature and pH, temperature, current density, quantity of electricity passed, time of treatment, and intensity of stirring. It appears that the reaction may take different routes dependent on different combinations of these parameters. However, whatever the route taken, it appears necessary that for the reaction to proceed platinum oxides or

SOIL AERATION BY PLATINUM MICROELECTRODE

245

hydroxides must be present or formed during the reaction. Many of the differences would be resolved if H 2 0 2 is catalytically decomposed, for some arguments regarding its absence are based on interpretation of current-voltage-time relationships where small quantities may not be detected. Rickman et al. ( 1968) conclude that, in measurements with the platinum microelectrode, H 2 0 2is formed regardless of the state of oxidation of the electrode, but that the rate of its catulyric decomposition depends on the presence of an “active” oxide. An irreversibly formed oxide can exist on platinum from aging of a clean electrode, and Rickman et ul. (1968) consider that this inhibits the rate of decomposition of H 2 0 2 which can then be lost from the surface into the bulk solution, as suggested by Oden ( 1962). The details of the reaction are of interest from two points of view, rate and efficiency of the reaction. Some authors (e.g., Charlot et al., 1962) consider that reduction of H 2 0 2is a slow reaction, particularly at low pH values ( < 10). Therefore its presence may make the reaction rate low, particularly if its decomposition rate is diminished by the lack of sufficient oxides on the electrode (which applies to most microelectrode measurements in soils). Efficiency refers to the relation between current and amount of O2 reduced. Oden (1962) found from electrolysis of O2 in solution, that on the basis of a 4-electron reaction, total charge passed could account for the loss of 02,making the reaction 100% efficient; similar conclusions were reached by Willey and Tanner (1961). Slow decomposition of H 2 0 2could, however, reduce efficiency by allowing loss from the electrode surface due to diffusion. In soils, adsorption by, and reaction with, soil constituents could further reduce efficiency (Oden, 1962). As both variation in pH and state of the electrode surface apparently affect the decomposition rate of H 2 0 z , there could be variability in current from one soil to another, when measurement is made with the platinum microelectrode. B.

REACTION RATE

The reaction rate should be considered in relation to the rate of supply of the reactant, and whether the former is likely to be the currentlimiting process. On a “bright” platinum surface (as distinct from a platinized surface) the reaction rate is apparently highly dependent on the surface conditions. For such an electrode four initial conditions are recognized in general: ( 1) preoxidized, anodized (or chemical treatment, e.g., HNO& (2) prereduced, cathodized (or chemical treatment, e.g., H 2 S 0 4 ) ;(3) abraded (clean); (4) aged-oxidized (any of the above three

246

D. S. MCINTYRE

surfaces allowed to stand in water or air). Another condition has been introduced by Black and Buchanan ( I966), viz. (5) cathodized in phenolphthalein solution. Surfaces (2), (4), and ( 5 ) have been shown by these authors to behave similarly in air-saturated two-phase systems. Preoxidized electrodes apparently allow the greatest reaction rate (Lingane, 196 1; Sawyer and Interrante, 196 1 ; Rickman, 1966; Rickman et al., 1968). Lingane (1961) considers that at least one-tenth of the surface area of the electrode should be covered with freshly formed oxide in order to maintain the reaction rate at a maximum in air-saturated solution. The current strength will then depend on the diffusion rate of 02,but diffusion control is lost if the coverage is less than one-tenth of the surface area, and reaction rate becomes the current-controlling process. However, the utilization of preoxidized electrodes for soil measurements is both awkward and time consuming. Moreoever, in soils the electrode surface is abraded by insertion and then cathodized for 4 or 5 minutes before the current is read. From Lingane’s investigations, the deduction can be made that such treatment is sufficient to remove any oxides previously present. The actual initial condition of the surface is hard to determine for, as well as the treatment of the electrode mentioned above due to its normal use, mild abrasion may be used before each measurement. Sometimes the electrode may also be “aged oxidized” to some extent. Sawyer and Interrante (1961) showed that a reduced electrode in solution without voltage applied, is oxidized in a few seconds, but that the oxide is removed by cathodization for 2 to 3 minutes at a current density of 0.02 mA/cm2 (1-2 p A for the platinum microelectrode normally used). Presumably the same changes would occur with a freshly abraded electrode also and give rise to a standard surface. Rickman et al. ( 1 968) and Rickman ( 1 966) quoted evidence to show that an aged-oxidized surface inhibits 0 2 reduction and is hard to remove except by abrasion or strong chemical treatment. Black and Buchanan (1966) have found from electrolysis in sand saturated with 0.5 M KCI solution, that the aged-oxidized surface gives similar results to a prereduced electrode. Rickman et al. (1968) and Rickman (1966) also demonstrated by controlled potential electrolysis and chronopotentiometric measurements in a clay suspension, that formation and removal of oxides occurred on a platinum microelectrode. The presence of oxides increased the reaction rate (and hence current) as well as affecting the decomposition potential of 02.They were unable to demonstrate that the state of oxidation of the electrode significantly affected current, and reached the conclusion that diffusion rate of O2 in a suspension was so

SOIL AERATION BY PLATINUM MICROELECTRODE

247

low that any of the surfaces could maintain the resulting current. The effective diffusion coefficient in a saturated soil would be less than that in a suspension. Thus the reaction rate would need to be extremely low to be found limiting, considering the small concentration gradients existing in a saturated soil, for which the nearest gas-liquid interface is at the soil surface. In fact, it has been found (McIntyre, 1966b) that diffusion control is very strong in a 15% suspension, water saturated sand and glass beads, the solution in each case being air-saturated (Fig. lb). Black and Buchanan’s results (1966) lead to similar conclusions, for their plateau current was substantially the same for electrode surfaces that were preoxidized, prereduced, aged-oxidized, and prereduced in phenolphthalein. These determinations and comparisons do not necessarily apply to unsaturated porous media in which the concentration gradients may be considerably greater at parts of the electrode so that the transport rate of O2 should be correspondingly greater. The fact that current is continuously dependent on voltage (Kristensen, 1966; McIntyre, 1966b) shows that diffusion control is lost. There are even indications of the disappearance of diffusion control in suspensions when transport rate of O2 has been increased by saturating the suspension with 100% 0 2 as against 2 1% O2 (Lemon and Erickson, 1955). In a three-phase system, solid and liquid geometry will be such that over a significant proportion of the electrode surface very thin films would separate the electrode from the gas phase. Consequently O2 concentration gradients could be greater than those in saturated soils by possibly several orders of magnitude. If Lingane’s postulate ( 1 96 1) that a reduced electrode allows reaction rate control in solution is correct, its occurrence in three-phase systems must be highly probable. Lemon and Erickson ( 1952) calculated that in 5 minutes the concentration gradient could “reach out” 4-5 mm from the electrode. If parts of the electrode are separated from gas-filled pores by a film thickness of say only 10 p , the concentration gradient at those parts has the potential to be 400-500 times greater than when the thickness of the film is 4-5 mm. It does not require a very large proportion of the electrode area in this condition to materially increase the current. Theoretically, in a soil with say 10% air-filled porosity, the same percentage of the electrode area should be in contact with gas-filled pores, provided no change results from electrode insertion. Thus a significant area of the electrode is potentially able to receive O2 at a very high rate, which may make the current over this area controlled by the reaction rate. It has been pointed out by Stolzy and Letey (1 964a) that for comparison of results, standardization

248

D. S. M C I N T Y R E

is required in the radius of the microelectrode and in the time at which the current is read. If the reaction rate is likely to be a current-limiting process, surface conditions at time zero should also be made uniform to ensure that time is not having a further effect through its control of the state of the surface. It is generally found in determination of current-time relations that variability of current is greatest at short times, and becomes less as time increases beyond 3-4 minutes. This behavior suggests that cathodization is establishing a more uniform surface. For uniformity at time zero then, it is probably best (and simplest) to begin each time with a mildly abraded surface, produced with wet sand or with the medium under examination, if it is suitable. Such a surface may be maintained as a result of the abrasion caused by removal and insertion after each 5 minutes operation, but some added abrasion will also ensure removal of any contamination that may have occurred. Rickman et af. ( 1 968) state that reproducibility is not good with abraded electrodes; this is contrary to the author’s experience, and may depend on the manner of abrasion. Rickman et al. also advocate preoxidation of the microelectrode for 72 hours with concentrated nitric acid. By increasing the rate of reaction, preoxidation would give greater likelihood of diffusion control of current. However, the half-wave potential (and decomposition potential) for O2 at a preoxidized electrode, is sensitive to change in pH (Sawyer and Interrante, 1961), and soil properties will control the degree of abrasion and cathodization. Thus, unless the electrode is freshly oxidized before each measurement, the initial surface conditions for any determination may differ from the previous ones. Moreover, the main (perhaps the only) advantage of the platinum microelectrode method is its simplicity; acid pretreatment introduces a further complication to affect a variable which may not be of great consequence in the final result. O2 transport rate in unsaturated soils may be so great that the reaction rate will be the current-limiting process over a significant proportion of the active electrode area. Until it has been demonstrated that preoxidation allows diffusion control of current, it seems better to begin with a mildly abraded electrode -a condition that can be repeated easily. C. EFFECT OF PH

The pH of the adjacent medium may influence the electrode behavior by altering: (a) the exact course of the reaction, (b) the decomposition potential (or half-wave potential) of 02,(c) the decomposition potential of H+ ions, (d) the effective area of the electrode by precipitation of insoluble compounds on its surface. The pH can also affect the rate of re-

SOIL AERATION BY PLATINUM MICROELECTRODE

249

duction, for H 2 0 2is more stable in acid than alkaline solutions so that it is easier to exchange the equivalent of four electrons per 0, molecule in alkaline than in acid solutions, (Hoare 1968). Case (a) has been mentioned previously, and case (b) is of only minor consequence here (see Sawyer and Interrante, 1961; Black and Buchanan, 1966). Case (d) will be discussed under poisoning of the electrode. A knowledge of the decomposition potential of H+ ions is important in determining the voltage to be applied to the platinum microelectrode, particularly in anaerobic or near-anaerobic media. The voltage depends not only on H + ion concentration, but also on O2 concentration (Oden, 1962) and current density (Potter, 196 I ) . Current-voltage curves with varying pH have been presented by Kolthoff and Lingane (1952) and Oden ( I 962), for solutions, and by van Doren (1958) and Black and Buchanan (1966) for two-phase media. In such a measurement the decomposition potential of H+ is fairly obvious although the apparent applied voltage at which it occurs will depend on the extent of ohmic losses within the medium; in unsaturated soils the decomposition potential is not so obvious (see, e.g., Fig. 4).The results show that for air-saturated soil solutions pH has little apparent effect on the decomposition potential of H + for values between about 3.5 and 1 1 [and probably between 3.5 and 14 (Oden, 1962)], but that below a critical pH of about 3.5 the decomposition potential for H decreases negatively rather rapidly. Oden (1962) has pointed out also that as 0 2 concentration decreases, the critical pH for significant H + ion decomposition rises above 3.5. Under near anaerobic conditions, application of the voltages often employed (e.g., -0.65 V) can give rise to a current that will be wrongly interpreted as O2 reduction. Armstrong ( 1 967) has shown this effect well for a two-phase system with a measured pH of 6.3. The H + decomposition voltage changes from about - 0.8 V for air-saturated conditions to -0.5 V for near-anaerobic conditions. The dependence of the H+ ion decomposition voltage on the O2concentration has significance in two contexts: (1) determination of zero (residual current) in acid conditions; (2) determination of O2 flux in acid soils, for which, it is recognized, the pH of an extract overestimates the in situ pH. Oden (1962) has treated the second case theoretically by considering the points at which H ions neutralize the O H - ions formed in the reaction. For the surface of neutralization to just reach the electrode surface the requirement is +

+

250

D. S. MCINTYRE

where C = concentration and D = difisivity. In air-saturated solutions mole/liter, making at 20°C the critical H+ ion concentration is 3.3 X the pH = 3.5. Figure 3a shows the experimental results of Oden (1962) with current as a function of pH at constant applied voltage. The current increases suddenly at a pH of 3.5, indicating the reduction of H+ ions to HP. Acidity has little effect on current in the pH range 3.5-11.0. The platinum electrode (0.6 mm in diameter) was coated with a thin layer of collodion, which undoubtedly had a high electrical resistance. Thus with an applied voltage of -0.8, the efective voltage would be much smaller (see below; see also McIntyre, 1966b, 1967). The formation of OH- ions at the cathode from the reduction of O2 prevents simultaneous reduction of H+ ions. It follows from Eq. (9) that the critical value of pH (or CH+) is solely a function of the O2 concentration of the solution for, in porous media proportional changes in D should be the same for both H+ and O2in solution. Oden (1962) has found that the critical pH value for the external solution remains unchanged over the cathode (applied) potential range of - 0.7 to - 0.9 V. The H+ ion movement is therefore not caused by the electric field, but by diffusion. The critical pH value for a number of 0 2 concentrations in solution at 20°C has been calculated, and the results are shown in Table I. For a solution in equilibrium with less than 1.0% gaseous 0 2 the critical pH rises above 5 . Under completely anaerobic conditions the correlation TABLE I Critical pH Values for H Ion Reduction in Relation to the Concentration of OI in Solution When Equilibrium Exists Between the Solution and the Given Gaseous Or Concentration (at 20°C) +

Equilibrium gas concentration (%) 0.1 0.7 I .O 2.0 4.0 5.0 10.0 15.0 21.0 50

Cox in solution mole liter-') 0.0 14 0.094 0.14 0.27 0.54 0.68 1.35 2.03 2.84 6.15

pH, critical

value 5.8 5 .O 4.8 4.5 4.2 4.1 3.8 3.6 3.5 3.1

25 1

SOIL AERATION BY PLATINUM MICROELECTRODE

between the pH of the bulk solution and the cathode potential at the start of the reaction (incipient evolution of Hz) is shown by Oden (1962) to follow the Nernst theory, that is, with a change of 58 mV per pH unit. The relationship is shown in Fig. 3b. (0)

(b)

“t

Overvoltage

08V

0

4

2

6

8

10

12

14 p K

Relation between current,I. and pH at constant voltage

I

0

2

4

6

8

1 0 1 2 1 4

pH

Relation between pH and voltage at the stort of the reaction H f e - - c l / 2 HZ

FIG.3. (a) Experimental relationship between current (at constant voltage) and pH of an unbuffered solution saturated with air. (b) relation between voltage at the beginning of the reaction 2H+ 2e- 4 Hs,and the pH of an oxygen-free solution (both curves from Oden, 1962).

+

In suspensions and soils the true pH is difficult to specify although it is usually considered to be somewhat lower than the pH of an extract and to decrease with water content. Recent discussion (Harter and Ahlrichs, 1967a,b; Mortland, 1967) suggests that “acidity” can increase up to 100 times from the bulk solution to a clay surface when bulk solution pH is about 7.0. Thus the pH “seen” by the electrode may be up to 2 units lower than that determined from an extract. Furthermore in buffered systems, such as soil colloids, H+ion donors can occur within the “diffusion layer,’’ which means that there is less demand for H+ ion gradients for neutralization of the OH- ions formed. Critical pH values for H+ ion decomposition should therefore be correspondingly greater. There is little experimental evidence on this point from controlled measurement in soils. In Fig. 7a, slight change in the slope of the Tafel line for sandy loam of pH 6.3 and in equilibrium with 4.1 % O2 probably indicates that H+ ion reduction begins at - 0.5 V. In Fig. 7b a more marked change in slope occurs at - 0.7 V for one soil. Black and West (1969) using slurries and saturated sand obtained similar results to those found earlier by Oden ( 1 962). For their media when applied voltage was - 0.4 V or less, current did not change over the pH range of 5.4-3.2;

252

D. S. MCINTYRE

more negative voltages gave much higher currents at 3.2. Solutions were air-saturated; at near-anaerobic conditions even - 0.4 V may be sufficiently negative to reduce H+ ions in some soils. Oden’s curve was obtained at - 0.8 V (Fig. 3a), but his electrode was coated with collodion. Discrepancies between workers, in the apparent (applied) voltage at which certain reactions occur, can be explained in terms of different ohmic losses between the working and reference electrodes (see Section ILE, 1 ,a: Resistance Polarization). The applied voltage used and advocated by Stolzy and Letey ( 1 964a,b) is - 0.65 V versus Ag-AgC1 (about - 0.70 V versus SCE). Even allowing for substantial ohmic losses, it is highly probable that, in the light of Table I and the above discussion, such a voltage will be reducing H+ ions in many poorly aerated soils. It can be postulated that a significant proportion of the 20 X lo-* g cm+ min-’ found so often by these authors to be a critical flux for root growth is partly a result of H+ion reduction in a nearly anaerobic medium. There is evidence of this occurring in measurements by Letey et d. (1962) in which zero 0 2 gives an ODR of 10-20 X g ern+ min-I; however, this could also be due to transport of O2downward through the plant and roots.

D. POISONING OF THE ELECTRODE SURFACE Poisoning is usually taken to mean any alteration of the electrode surface which affects its efficiency to reduce oxygen, other than by substances involved in the main process, such as platinum oxides. Oden (1962) pointed out the possibility that at a certain pH value iron and aluminum oxides and/or colloids may reach the electrode surface, and that carbonates may be precipitated at the high pH values near the electrode. Rickman et al. ( 1968)found that some of these processes occurred, but only after electrodes were left in place for long periods. The effect was much greater in a loamy sand than in a clay and should not be significant in either for short-term measurements. Other substances which are known to “poison” platinum occur in soils in small or trace amounts-for exmple, PO4, S, Cu, As, and organic molecules- but under conditions where the electrode is inserted every 5 minutes their influence does not appear to be significant. However, Kozawa (1964) found some evidence of the effect of ions in solution, notably Ca2+,on the efficiency of reduction of O2 at a platinum electrode, while Rickman (1968) showed that the current at constant electrical conductivity of the soil was less for exclusively Ca2+ions or a mixture of Ca2+and Na+ ions, than for exclusively Na+ ions in solution. Bockris and Conway (1949) examined the effect of As, but found

SOIL AERATION BY PLATINUM MICROELECTRODE

253

that relatively large concentrations were necessary to induce significant poisoning. To the author’s knowledge no measurements have been reported which check poisoning in organic or highly reduced soils, where it may reasonably be expected to be greatest. In mineral soils poisoning does not seem to be a problem. E. ELECTRODE KINETICS When significant current flows because of a reaction occurring at an electrode in some medium, several nonequilibrium (rate) processes contribute to the reaction. In the present context they are: (1) the O2reduction rate, which is a function of voltage and state of the electrode surface; (2) the rate of reduction of Hz02 if formed, which depends on the same properties; (3) transport of O2 to the electrode; (4) transport of OHions from the electrode. Between them they control the kinetics of the electrode processes. The last two are dependent on properties of the medium, and one or more of the four processes may be dominant in control of the current strength. The simple model adopted to explain operation of the platinum microelectrode requires that transport of O2 by diffusion is the dominant process in current control, giving rise to current-voltage relations of the type in Fig. 1b. Thus, properties of the medium, such as electrical resistance, salinity, and pH, which may affect other processes, have been assumed to have no influence on current. Work by Kristensen (1966) and McIntyre (1966bJ967) showing for a number of three-phase media that current is continuously dependent on voltage (as in Fig. 4), makes application of these assumptions to such media doubtful. Similar currentvoltage relations have also been found by several other workers: Birkle er al. (1964), Gradwell (1965), Tackett and Pearson (1964), Sims and Folkes ( 1964), and Martin (1968). We will therefore examine possible theoretical relationships due to the kinetics of electrode processes, and how they are affected by the medium. The next step is to then relate theory to experimental results obtained with a microelectrode. Two distinct physical systems occur in porous media: (1) water saturated, with water at positive pressure (two-phase system); (2) unsaturated porous media, containing water and air (three-phase system). Their current-voltage relations are given in Figs. 1 and 4. The intermediate, and more important system in soil aeration, is that in which the porous medium is saturated, or nearly so, but the water is held under suction. There is a scarcity of good data for these conditions, but some evidence, presented below, supports the hypothesis that this system is modified locally by the electrode penetration in such a way that it reverts

254

D. S. MCINTYRE

50

IWW

45

80

70

I

I

I IP

a

I

40 35 30

25 20 15 10

5 0 3 Effective voltage (volts-ve vs SCE)

FIG.4. Some current-voltage relations obtained with unsaturated porous media (McIntyre, 1966). A, Loam of pH 4.5 in equilibrium with 2 I % Or;0,18p glass beads, pH 10.5 in equilibrium with 10% Or; 0,sandy loam pH 6.3 in equilibrium with 10% Oa.

to system (2). In two-phase systems, diffusion control occurs without exception. In three-phase systems, in which the transport rate of Onmust be considerably greater, so that the reaction rate may be limiting, other factors which alter markedly with decreasing moisture should be considered. The consequences of desaturation have been discussed in part by Kristensen ( 1966) and McIntyre ( I 966b, I967), and hence the aspects covered by these authors will be treated only briefly. Factors such as electrolyte transport, pH, and activation polarization, will be covered in more detail.

SOIL AERATION BY PLATINUM MICROELECTRODE

255

1 . Polarization of the Electrode If the reaction at the electrode (Eq. 6) is written as an equilibrium reaction it becomes

where kf and kb are the rate constants for the forward and backward reactions. The theoretical reversible potential (E,) of the reaction is given by the Nernst equation (see e.g. Laitinen, 1960; Potter, 1961) as Em

=

Eo

+ 2.3nFRT ~

log

(P ) (OH-)'

Reaction (10) is in practice highly irreversible at a platinum electrode (Charlot et al., 1962), so that a relatively large negative voltage must be applied to overcome energy barriers to the reduction reaction. In this state the electrode is said to be polarized, the degree of polarization being given in terms of the overvoltage (or overpotential) (q) by the relation q=V-Ee4

(12)

where V is the voltage applied externally to the electrode with respect to a reference electrode. Polarization may result from several processes which are controlled by properties of the electrode and the medium. Types of polarization relevant here are: (a) resistance polarization, which occurs when ohmic losses in the medium make the effective electrode voltage (V,) different from the voltage applied externally (CIA); (b) electrolyte transport polarization resulting from slow ion transport (by diffusion and electrical migration) from the vicinity of the electrode; (c) activation polarization, due to a slow reaction rate at the electrode; (d) concentration polarization, due to slow transport rates of reactants producing concentration gradients in the medium; and (e) reaction polarization, due to a slow secondary chemical reaction. Resistance, activation, and concentration polarization are probably the most important ones in relation to electrode response in three-phase media. If all types of polarization contribute to the total overvoltage (qT),then 1)T = 1)Res

+ 1 ) E l + 1)A + T ) C + T)Rea

(13)

256

D. S. MCINTYRE

Usually one type of polarization is dominant under the imposed conditions; in solutions it will be one of the last three. In solution electrochemistry (vRes)is eliminated by special methods (see Lingane, 1958) and (qEl)occurs very rarely, if ever. The latter does however occur in porous electrodes used for fuel cells (Austin et al., 1965; Rockett and Brown, 1966) both liquid- and gas-filled pores are present in these electrodes. a. Resistance Polarization. In a porous medium the electrical resistance to current flow will generally be quite appreciable. Kristensen ( 1966) and McIntyre ( 1 967) have measured some values which show that resistance can be large in unsaturated soils. McIntyre found that the resistance (R,) conforms to the relationship

which really expresses the resistance between spherical conductors with radii rl and rz, a distance d apart. The way in which the microelectrode is used gives rise to boundary conditions more of the form for spherical rather than cylindrical conductors (McIntyre, 1967). F o r d 9 rl or r 2 , the resistance becomes independent of the distance apart of the electrodes and a function of rl, r2, and CT only. Measurements made by McIntyre ( 1 967) are given in Fig. 5 and show that R , becomes independent of d at about 10 cm. (0)

(b)

25

2ot

~ o moist t cont 0 026 0 0.28

-0-0,

0

Vol. moist. cont. 0

x 0

0

5

15 Distance (cm)

10

20

25

0

5

0.25 0.27 0.30 10

15

20

Distance (cm)

FIG.5 . Electrical resistance of (a) a loam and (b) a sandy loam as a function of the distance between the salt bridge tip and platinum microelectrodes of diameters 0.56 mm (solid line) and 1.22 mm (dashed line). (From McIntyre, 1967.)

25

SOIL AERATION BY PLATINUM MICROELECTRODE

257

Kristensen (1 966) and McIntyre ( 1966b, 1967) have demonstrated the importance of the electrical resistance in determining the value of the microelectrode voltage effective in 0 2 reduction. If V , is the voltage applied between the working electrode and a reference electrode in a porous medium, and R, is the resistance between the same electrodes, then the effective voltage V , is given by

where i R , = qRes the resistance overvoltage. In two-phase systems resistance polarization is not important provided it is not great enough to lower VE beyond the minimum plateau voltage (see Fig. Ib). Van Doren ( 1958) and van Doren and Erickson (1 966) have shown on the basis of measurement in suspensions that using an applied voltage of -0.65 V, a resistance of 65,000 ohms can be tolerated. Their plateau current at 5 minutes was 3.0 PA, making the ohmic loss about 0.2 V. When the applied voltage was -0.8 V van Doren found that the current decreased as resistance was increased to 15,700 ohms. These results mean that the lower voltage limit of the plateau was about -0.5 V and the upper about -0.75 V, and that at -0.8 V H + ions were being reduced. Reduction of H continued with increasing resistance until with 15,700 ohms the effective voltage was below the decomposition voltage for H + ions in the particular suspension used. Van Doren and Erickson (1966) made recommendations for unsaturated soils on the basis of their measurements in suspensions. Their results demonstrate the dangers of extrapolation from two-phase to three-phase systems. It is obvious that the variation in effective voltage, and hence current, would be very large for systems represented in Fig. 4, if the applied voltage were kept constant and the resistance varied over the range 0-65,000 ohms. Kristensen (1966) and McIntyre (1966b, 1967) have discussed this aspect, and McIntyre has shown that this effect alone invalidates comparison of values of 0 2 flux determined in three-phase systems at constant applied voltage (refer to Fig. Ic). Results of Birkle e f al. ( 1 964), which showed that spacings of 10 cm to 250 cm between working and reference electrodes in a three-phase system had no influence on current, could be interpreted as implying no effect of increasing electrical resistance. The reason for this result is obvious from Fig. 5. Although electrical resistance has no influence on current if the applied voltage lies on a plateau in the current-voltage relations, it does affect the apparent applied voltage at which certain reactions occur. Thus +

258

D. S . MCINTYRE

differences in decomposition and half-wave potentials, and the voltage at which pH affects the current, have no analytical significance, even in saturated media, unless ohmic loss in the medium is accounted for. For example, Black and Buchanan (1966) found a slightly more negative apparent half-wave potential for O2 reduction than did Sawyer and Interrante (1961), but the former authors made measurements in sand, the latter in solution. Effect of elecrrolyte. Soil conductivity ( (T) is a function of electrolyte (moisture) content, nature, concentration, continuity, tortuosity, and temperature. It can be shown from the dependence of the electrical resistance on the electrode radius, and consideration of the electrical field, that soil conditions in the vicinity of the electrode (the “diffusion layer”) will be the most important in determining the electrical resistance. Here the electrolyte may be dominated by OH- ions, other anions such as CIbeing repelled, with cations such as Ca2+,Mg2+,K+, Na+, H+ being present in varying quantities, the amount of the last depending on pH. Increase in electrolyte concentration due to the discharge of OH- ions can increase or decrease (T depending on whether its value is below or ab0v.e a maximum which can occur in electrolyte solutions (Andrew and Jones, 1966). However, as the maximum appears to occur at concentrations of 3-6 M , an increase in (T rather than a decrease is to be expected in soils. Rickman ( 1 968) has shown an effect of electrolyte constitution on current by measurements made with only the tip of the platinum microelectrode in contact with various solutions. Under these conditions a thin film of electrolyte exists on the electrode above the meniscus (Maget and Rothlein, 1965; Will, 1963). It is considered that such thin films will occur on the electrode in three-phase media. Rickman found that if the electrical conductivity of the solution were kept constant but the electrolyte constitution varied, CaClp solutions and mixed CaCI2-NaCI solutions gave a significantly lower current than solutions of NaCl or Na2S04 alone. His current-voltage curves for this condition were similar to those for three-phase media, and he concluded that the reason for his results was “evidently the migration of ions parallel to the electrode surface.” This latter was suggested as one possibility for voltage sensitivity of current in three-phase media by McIntyre (1966b). On the other hand, Kozawa ( 1964) showed specific effects of certain cations, particularly Ca2+,in solutions, in which efficiency of the reaction (and hence current) was affected by their presence. In Rickman’s experimental technique ( I 968), the actual voltage of the

SOIL AERATION BY PLATINUM MICROELECTRODE

259

platinum electrode was measured potentiometrically with respect to a second reference electrode. Such a couple should also be measuring the redox potential of the system, so that the measured potential will be a combination of the redox potential and the effective voltage. Hence it appears that the effective voltage for Or reduction between Ca’+ and Na+ systems, at constant measured voltage, should differ by the differences between redox potentials of Ca++and Na+ systems. b. Electrolyte Transport Polarization. In porous media, desaturation enhances O2 transport rates but at the same time impedes removal of OH- ions from the electrode. This could allow a buildup of the concentration of OH- ions near the electrode. Slow transport of OH- ions was suggested by Lemon and Kristensen (1961) as a current-limiting process in unsaturated soils. Mclntyre (1 967) propounded it as a possible reason for voltage sensitivity of current, either by interaction of OHions with the electric field, or by enforced movement along the electrode in very thin moisture films. Rockett and Brown (1966) consider that the latter occurs in the porous electrodes of fuel cells and is important in determining output. Each ion would have a velocity component parallel to the electrode surface, causing ohmic loss and local changes of effective voltage. Ion movement within the film above a meniscus on an electrode decreases the local effective voltage, sometimes to zero (Will, 1963; Maget and Rothlein, 1965). Occurrence of a similar phenomenon would explain the lack of a plateau in the current-voltage relations found for unsaturated porous media, as the effective voltage would vary from point to point over the electrode surface. When the applied voltage is increased, the variability would be maintained so that on parts of the electrode the effective voltage could reach the decomposition potential for H + ions, while at other parts the effective voltage is still very low. This explanation is favored by Rickman ( 1 968). Theory and principles of transport processes in the “diffusion layer” of solutions is not well established (Newman, 1968); in three-phase media, principles of such processes will be even more complex. In soils with a pH less than 6, where the surface of neutralization of OH- ions by H ions approaches the electrode, the electrical potential gradient should be large. This means that anions other than OH- should be excluded from the “diffusion layer” whatever the concentration of the soil electrolyte. Therefore OH- ions will be the only carriers of current in this region making electrolyte transport polarization likely in acid soils. If few H + ions are present (pH > 8) the distance over which the electrical potential is active will be much greater, so that potential gradients +

260

D. S. MCINTYRE

will be small. This may allow the presence in the “diffusion layer” of other anions as well as OH- ions, and they will also carry current. If salinity is high, the conditions are similar to those occurring in solutions in which an excess of indifferent electrolyte masks any influence of transport of the active ion on current (Lingane, 1958). Thus electrolyte transport polarization could be expected for conditions of alkaline pH and low salinity, but not when the salt content of an alkaline medium is high; it could be expected at acid pH values irrespective of the salinity. In fuel cell literature (Austin et af., 1965; Rockett and Brown, 1966), it has been postulated that limitation of current by electrolyte transport makes current proportional to C O T .This relationship has been found also by Maget and Rothlein (19 6 3 , referred to by Evans (1968), and in corrosion investigations (see Evans, 1968), but Evans considers it to be a result of limitations to reaction rate in the breaking up of the O2molecules. In measurements with a microelectrode in cellulose sponge where electrolyte content would be low -van Gundy and Stolzy ( 1963) showed an approximate square root relation for O2gas concentrations up to 2 1 %, and McIntyre ( 1 966b) from measurements in a loam with high electrolyte concentration (1.33 M KCl) but low pH (4.5 in 1 :5 extract of 1.33 M KCI), found that current was related to Cox by a square root relation for O2 gas concentrations up to 50% (see Fig. 8). In both cases electrolyte transport could be a significant factor, and limitation of current by electrolyte transport polarization will make current continuously dependent on voltage. c. Activation Polarization. Activation polarization results from the reaction rate at the electrode being the current-limiting process, because of a slow step in the electrochemical reaction. lt is indicated if current is continuously dependent on voltage, as in Fig. 4.Generally the slow process is electron transfer, but in the reduction of 0 2 Evans ( 1 968) states that “the sluggishness is usually attributed to the difficulty of splitting the oxygen molecule,” the reaction 0 2 -+ 0 0 requiring high energy. Lingane (196 1 ) has stated that activation polarization can occur at a bright platinum electrode in stirred solution, and Maget and Rothlein ( 1 965) have found it to occur in a meniscus on a platinum electrode above a solution. These have been discussed previously. Conditions in the vicinity of the electrode suggest the probability of activation polarization occurring, and Fig. 6 shows a proposed model to explain how this could occur. The assumption is that liquid films are thin enough over part of the electrode to maintain a high rate of supply of O2 and that at such points O2 concentration is constant and in equi-

+

SOIL AERATION BY PLATINUM MICROELECTRODE

26 1

librium with that in the gas phase. Under these circumstances current is given by (Laitinen, 1960): i = nFAC,, kr exp

- u q A nF

RT

(16)

where Cox is the 02 concentration both at the electrode and in bulk solution, a and kf are constants of the reaction and qA is the activation overvoltage which, provided qEl,q c , and qReaare negligible, is given by

E , being obtained from Eq. ( 1 1). From Eq. (16) taking logarithms of both sides we obtain: log i/A = log nFCox k ,

-

that is, using Eq. (17), log ilA

=

!!!d!h + log nFCox kf b

(19)

which is one form of the Tafel equation with b = --2.3 RT a nF

If Cox and pH are constant at the electrode surface, the latter condition occurring for bulk solution pH between about 4 and 11, the last term of Eq. (19) and E , are constant, giving rise to the relation l o g i =VA + K

b

The current-voltage relations of McIntyre ( 1966b)and Kristensen ( 1966) in all of which VE could be determined, have been found to conform to (2 I ) . Some are plotted in Fig. 7. With activation polarization, current is dependent on 0 2 concentration as well as on voltage, and at constant effective voltage one can examine this relationship. Equation (16) can be written, using (17), i = n F A Cox k f exp

unF [-7 (V, - Em)]

D. S. MCINTYRE

262

Wcter- filled pores to infinity

..:.. .,..'. .,...:. b e o i l water

/

Platinum electrode surface


z

U

‘Compiled from data of Martin e l al. (1967). *Molecular weight estimated by gel filtration (Sephadex).

412

FRANCIS E. CLARK A N D ELDOR A. PAUL

and Mayaudon, 1966; Jenkinson, 1968). Following incubation of 14Cglucose in soil, Wagner and Mutatkar (1968) found the highest specific activity to occur in amino compounds normally occurring in the cell walls of bacteria and fungi, and thus presumably, microbial cell walls would accumulate in soil organic matter to a greater extent than would cell cytoplasmic constituents. Ford and Greenland (1 968) found that densimetric fractionation of soil organic matter separated a light fraction consisting primarily of small-sized fragments, together with phytoliths. Although the light fraction contained some humic material, its composition approximated that of plant rather than humic material, and its turnover rate was relatively rapid. It was found that 25-60% of the mineral nitrogen released during incubation came from the nonhumic light fraction even though this fraction accounted for a much smaller percentage of the total soil organic matter. When viewed collectively, recent work indicates that the organic components of soil are composed of at least three fractions when considered on a dynamic basis: (1) decomposing plant residues and the associated biomass which turn over at least once every few years; (2) microbial metabolites and cell wall constituents that become stabilized in soil and possess a half-life of 5-25 years; and (3) the resistant fractions, which in grassland soils are composed of humic components ranging in age from 250 up to 2500 years (Campbell et al., 1967; Scharpenseel et al., 1968). It is pertinent to ask why much of the soil humus is stable and to inquire into the time factors involved in humus turnover. In this inquiry, attention will be focused largely on the nitrogenous moiety of soil organic matter. The resistance of proteinaceous constituents of soil humic materials to degradation by a wide array of microbial proteases can probably be attributed to a lack of flexibility of the substrate and to mechanical as well as chemical shielding of localized areas of the substrate surface. Colloidal organic substances are rendered less susceptible to biodegradation by physical sorption on surfaces on or within the layers of clay minerals (Estermann et al., 1959; Greenland, 1965), by chemical combinations with calcium, iron, and aluminum (Brydon and Sowden, 1959; Mortensen and Himes, 1964), or by lodging in micropores inaccessible to microbial cells (Rovira and Graecen, 1965). Much of the organic matter in soil is associated with the silt and coarse clay fractions (Arshad and Lowe, 1966; Grant, 1967). The C:N ratio of the adsorbed materials generally decreases with decreasing particle size (Chichester, 1969). This decrease in C:N ratios was found by Chichester to coincide with a much greater release of nitrogen upon incubation of the finer-sized fractions, indicating a preferential adsorption of nitrogen-rich and more easily mineralizable components by the clay particles.

413

THE MICROFLORA OF GRASSLAND

The interaction of clays with organic matter is shown in Table XVII. The addition of 5 % montmorillonite to the soil increased the amounts of added carbohydrate carbon which could be recovered as amino acids TABLE X V l l Hemicellulase Activity of Soil in Which Hemicellulose or Glucose Was Decomposed“ ~

Carbohydrate added

Hemicellulose

Glucose

~~

~~~~

Montmorillonite added (%)

Period of decomposition (days)

0 0 0 0 0

6 12 30 90 700

7.8 8.7 8.1 6.0 3.1

5 5 5 5 5

6 12 30 90 700

11.6

0 0 0 0 0

6 12 30 90 300

6.1 7.2 6.0 3.5

0.06 0.06 0.08 0.08 0.07

5 5 5 5 5

6 12 30 90 300

15.5 14.2 12.6 10.6 9.2

0.16 0.16 0.12 0.07 0.08

Carbohydrate C recovered in amino acids (% of added C)

11.6 10.0 10.0 7.1

5.0

Hemicellulase activity

1.03 1.oo

0.92 0.85 0.18 2.22 2.49 1.93 1.40 0.58

“Compiled from data of L. H. Sorensen (1969).

after incubation. The content of enzymes capable of degrading hemicellulose increased with increasing amounts of hemicellulose carbon originally added, was reasonably stable for 90 days, but decreased to a low level by 700 days. Hemicellulase concentration was low in soil to which only glucose had been added, even though a fairly large proportion of the radioactive glucose carbon was retained as amino acids in the presence of montmorillonite. These observations indicate that the amino acid metabolites originating from the added hemicellulose and fixed by montmorillonite must have been at least partially enzyme protein (L. H.

FRANCIS E. CLARK A N D ELDOR A. PAUL

414

Sorensen, 1969; Galstyan er al., 1968). It also indicates a reasonable stability of the adsorbed enzyme to degradation by microorganisms. A similar stabilization of I4C-labeled extracellular polysaccharides has been postulated by Cheshire et al. (1969).

c.

HUMICMATERIALS The growth of soil organisms with humic acids as sole sources of carbon is slow (Flaig and Schmidt, 1957; Tepper, 1963). Decomposition is enhanced when humic materials are enriched with supplemental sources of carbon and nitrogen (Mrysha, 1966; Szegi, 1967). In addition, mixed cultures usually degrade humic materials to a greater degree than individual pure cultures (Pontovich, 1938; Schonwalder, 1958). It is likely that this mixed culture stimulation of degradation also occurs in the natural soil environment and that it is at least partly responsible for the priming effect of green manures on the mineralization of soil organic matter (Lohnis, 1926; Broadbent, 1947; Paul et al., 1967). In attempting to explain the generally low degree of microbial utilization of humic carbon, several workers have proposed that the humic molecule consists of a highly aromatic core with some nitrogen-containing heterocyclic rings and more labile, nitrogen-rich, peripheral side chains (Ovchinnikova, 1965; Chesire et al., 1967; Szegi, 1967). The suspected presence of stable free radicals in humic acids has been confirmed by electron paramagnetic resonance spectrometry (Steelink and Tollin, 1967; Scheffer and Ziechmann, 1967). These free radicals are assumed to occur as quinone structures, and it has been shown that the more aromatic gray humic acids contain more radicals than brown humic acids (Kleist and Mucke, 1966). Many soil microorganisms utilize humic materials primarily as a source of nitrogen. The indication of the amino acid nature of much of the soil organic nitrogen (Jenkinson and Tinsley, 1960; Bremner, 1965, 1967; Scharpenseel and Krausse, 1962; Mayaudon, 197 1 ) emphasizes the importance of studies on extracellular proteases. Microorganisms produce a battery of extracellular proteolytic enzymes which can hydrolyze a variety of peptide and ester bonds in molecules of varying size under a wide variety of physiological conditions. The multiplicity of different proteases synthesized by microbial cells and the low substrate specificity of these enzymes enable the microflora, in general, to be very active in the dissimilation of numerous types of proteinaceous materials (Vlassak, 1966). Although most microbial proteases exhibit maximum activity at a certain pH, they remain stable and active over a much wider range of pH than does any of the animal proteases (McConnell, 1950): An extracellular protease of Cephalosporium, for example, was found to MICROBIALUTILIZATION

OF

THE MICROFLORA OF GRASSLAND

415

be active on three different proteins within a pH range of 5.0-9.5 (Oleniacz and Pisano, 1968). Another characteristic of microbial proteolysis is the ability of microorganisms, particularly the fungi, to synthesize several different types of extracellular proteases simultaneously (Wang and Hesseltine, 1965; Hagihara, 1960). Most proteases follow a similar pathway of catalysis. The spatial configuration and arrangements of individual amino acids in a substrate molecule are the important criteria in determining the specificity of proteolytic enzymes, while the size of the protein molecule is only of minor importance. The analysis of the action of proteolytic enzymes assumes a rigid enzyme and a flexible substrate. Therefore, the peptide chains of a protein molecule will only be split at those bonds where the surrounding segment of the peptide chain meets the specificity requirements of the enzyme and has sufficient flexibility to fit into the active site. Folding of peptide chains into a compact, three-dimensional structure prevents some or all of the peptide bonds from having the correct contact with the active sites of the enzyme and imparts a high degree of stabilization against proteolysis to some globular proteins, such as ovalbumin. The association with clays and the aromatic moiety of humic materials would make proteolysis even more difficult in the soil system. Several microorganisms that occur commonly in soil have been reported capable of decomposing humic acids at least to some extent. Fungi which have been found the most active are also capable of degrading lignin. Among fungi reported to be agents of humus decomposition are the following: Penicilliurn (Kudrina, I95 1 ; Mathur and Paul, 1967); Aspergillus (Kudrina, 195 1); Polystictus (Burges and Latter, 1960; Hurst, 1963); and Trichoderma (Schonwalder, 1958). Bacterial genera reported as agents of humus decomposition are the following: Arthrobacter (I. L. Stevenson, 1967); Bacillus, Corynebacterium, Nocardia, and Streptomyces (Schonwalder, 1958); Proactinomyces and Actinomyces (Ochilova, 1961); and Pseudomonas (Nikitin, 1960; Mishustin et al., 1968). I t is interesting to note that the above genera are almost without exception among those listed as the prominently proteolytic members of the soil microflora. Genera commonly described as actively proteolytic are, among the bacteria, Pseudomonas, Bacillus, Clostridium, Proteus, Arthrobacter, Flavobacterium, Serratia, Streptococcus, and Micrococcus (Pelczar and Reid, 1958); among the actinomycetes, Streptomyces, Nocardia and Actinomyces (Cochrane, 196 1 ; Waksman, 1967); and among the fungi, Penicillium, Aspergillus, Cephalosporium, Gliocladium, Trichoderma, Rhizopus, Mucor, and Mortierella (Cochrane, 1958; Hagihara, 1960). Attempts to define the biochemical pathways in humic decomposition

416

FRANCIS E. CLARK A N D ELDOR A. PAUL

have met with little success. There is evidence that the first step in decomposition is the reduction of carboxylic groups to primary alcohol groups (Hurst, 1963). Fungi capable of reducing the carboxyl groups of aromatic acids are also able to decolorize humic acid (Hurst et al., 1962). Mathur and Paul ( 1967) tested the ability of Penicilliurn frequentans to attack humic materials of different molecular weights. The larger moieties of humic acid were reduced to smaller-sized molecules, suggesting that surface groupings were being peeled off. Hydrolysis of the ether bond in humic acid was believed to be the major method of degradation. Salicyl alcohol and salicylaldehyde were measured after fungal cleavage under conditions of restricted aeration. VII. Nitrogen Transformations in Grassland Soils

A. NITROGEN FIXATION Many grassland soils contain large concentrations of organic nitrogen. Both the initial accretion of this nitrogen and its continued maintenance under land-use systems in which little fertilizer nitrogen is applied are largely microbiological phenomena which currently are receiving intensive study. Recent advances in the biochemistry of nitrogen fixation have been adequately reviewed by Burris (1969) and Hardy and Knight ( 1967). Among these advances have been the isolation of ferredoxin and the elaboration of a scheme for nitrogenase reactions based on an electron-activation, two-site hypothesis, as diagrammed by Hardy and Burns (1968). The development of the acetylene assay for nitrogen fixation has also been a major contribution of the study of cell-free nitrogen-fixing systems. This technique measures nitrogenase activity as shown by reduction of acetylene to ethylene and detection of the ethylene in the hydrogen flame of a gas chromatograph. Although it is an indirect method of analysis, it is an inexpensive, fast, and sensitive means of measuring nitrogen fixation. However, it must be standardized for the conditions of incubation and proper precautions, such as adequate controls, must be taken during analysis (Hardy e t a / . , 1968; W. D. P. Stewart, 1969).

1 . Symbiotic Nitrogen Fixation Various legumes can readily fix 55-225 kg of nitrogen per hectare annually. Under specialized conditions, there have been claims for as much as 840 kg of nitrogen fixed annually (W. D. P. Stewart, 1966). The environmental factors involved in the fixation of nitrogen by legumes have been reviewed by Vincent (1965). Natural grasslands contain a broad range of legumes (Budd and Best,

THE MICROFLORA O F GRASSLAND

417

1964; Whitman and Stevens, 1952). The latter workers found 27 species of legumes in western North Dakota grassland. Half of the total legume productivity, which ranged from 25 to 90 kg/ha, was accounted for by four species of legumes. Legume productivity usually accounted for less than 10% of the total range productivity and at times, for no more than I %. In quadrat counts, legumes averaged about 30 stalks/m2,whereas the grasses showed from one thousand to several thousand stems/m2. Although of minor importance as structural components in grassland, the legumes present therein are generally well nodulated, have deep, extensive root systems, and show an active period of growth during the moist season. Although the amount of nitrogen fixed annually in the leguminous nodules may be quite low, nevertheless their cumulative role in nitrogen accretion over a number of years can be substantial. Hopefully, the recent advances in methodology will shortly make available more informative data on the amount of nitrogen fixed in leguminous symbioses in various types of grassland. Plants other than legumes may also serve as hosts for nodulating organisms (Bond, 1967; W. D. P. Stewart, 1966, 1967). Although the microbial endophytes involved in nonleguminous symbioses are not as yet adequately defined, a wide variety of associations is postulated (Table XVIII). For plants such as Alnus and Ceanothus, the nodulating bacteria are actinomycetes rather than rhizobia. Eleagnus and Shepherdia are rootnodulating shrubs of this type and both are widely distributed throughout many areas of grassland habitat. Recently, Farnsworth and Hammond ( 1 968) have reported that sagebrush (Arternisia lodviciunu) and prickly pear cactus (Opuntiu frugilis) bear root nodules that appear capable of fixing atmospheric nitrogen. If their observation is substantiated, it will extend the range of nonleguminous symbioses and should encourage the search for other associations. Species of Artemisia and Opuntia are widely distributed throughout the western Great Plains. Mayland et al. (1 966) noted an increase of 15N in Artemisia during studies of fixation by algae in Arizona desert but attributed the enrichment of isotopic nitrogen to utilization of nitrogen which has been fixed and excreted by the algal crust. The leaf nodules and wet leaf surfaces of vegetation in humid tropical regions provide suitable sites for the development of nitrogen-fixing bacteria such as Azotobacter, Beijerinckia, Clostridium, Klebsiella, and Xanthomonas (Ruinen, 196 1 , 1965; Silver el ul., 1963: Vasantharajan and Bhatt, 1968). It is doubtful that phyllosphere floras contribute significantly to nitrogen accretion in semiarid grassland. Mycorrhizal nitrogen fixation has been postulated from field evidence

FRANCIS E. CLARK A N D ELDOR A. PAUL

418

concerning the ability of Pinus species that bear mycorrhizae to grow vigorously in nitrogen-poor soil (G. Stevenson, 1959). In his reviews on TABLE XVlIl Known or Postulated Nitrogen-Fixing Symbioses or Associations Involving Nonleguminous Plants

M icrobiont

Host genera involved Symbioses or associations involving Alnus, Casuarina, Ceanothus, Compronia, Discaria, Eleagnus, Hippophae, Myrica, Shepherdia; possibly Arremisia and Opuntia Dryas, A rctosraphylos, Cercocarpus, Purschia Podocarpus

Cycas, Encephalartos, Siangeria, Ceratozamia, Macrozamia, Zamia

involved spermatophytes Actinomycetes

Unknown Fungus? (Saxton, 1930); bacterium? (McLuckie, 1922) Blue-green algae

Gunnera

Blue-green algae

Numerous and diverse genera of both graminous and nongraminous plants

Fungi

N-fixing organisms in the phyIlosphere* Associations involving lower plants Pteridophytes Azolla Blue-green algae Bryophytes Blue-green algae Blasia, Curvicularia, A nthoceros Lichens Collema, Leptogium, Blue-green algae Pelrigera

Site of microbiont and host interaction Root nodules

Root nodules are present' Root nodules

Coralloid root hypertrophies Wartlike leafbase hypertrophies Mycorrhizal roots'

Leaf surfaces

Leaf pores Thallus cavities

The lichen thallus, which is a fungusalga symbiosis

'Nitrogen fixation is assumed on the basis of ecological observations, but definitive proof is lacking. *The microbionts involved (Azotobacrer, Beuerinckia, Closiridium, blue-green algae) are known to fix nitrogen even if not in association with higher plants, but association with higher plants is believed to augment their ecological role as fixers of nitrogen.

THE MICROFLORA O F GRASSLAND

419

nitrogen fixation by higher plants other than legumes, Bond (1967, 1968) has cited evidence that podocarp nodules, containing a phycomycetous nonseptate endophyte, fix nitrogen by means of a mycorrhizal association. Mycorrhizae are known to occur widely among the grasses (Nicolson, 1959: Dorokhova, 1967), but thus far there is no good evidence that they are capable of fixing atmospheric nitrogen. Lawrence et al. ( 1 967) have considered the role of Dryas in vegetation development on newly exposed mineral soil following glacier ice recession. Nitrogen turnover measurements in marine and brackish environments (W. D. P. Stewart, 1967) and studies of fixation by Hippophae in dune sand (W. D. P. Stewart and Pearson, 1967) have shown fixation ranging from a few to 100 kg of nitrogen per hectare to occur in these nonleguminous communities. Nevertheless, at the present state of knowledge, leguminous symbioses must be considered as more important for nitrogen accretion in grassland than are the various nonleguminous symbioses involving higher plants.

2 . Asymbiotic Nitrogen Fixation Recognition of the paucity of legumes in many grassland areas has led to a continued interest in the role of asymbiotic bacteria and other simple plants in the nitrogen economy of grassland soils. Moore (1966) has summarized extensive data concerning asymbiotic nitrogen fixation in a wide variety of soils. Many of the earlier experiments were based on Kjeldahl measurements of long-term field plots. They tend to show a nitrogen imput averaging about 22 kg/ha annually. In this total is included the nitrogen accruing from rainfall, an imput which in itself can be considerable (Eriksson, 1952; Junge, 1958; Mackenthun, 1965). Allison (1953, although not doubting that asymbiotic fixation does occur, expressed skepticism on the validity of many of the older measurements showing gains of nitrogen ranging up to 100 kg/ha per year. It is now generally agreed that the use of isotopic nitrogen or acetylene methodology is necessary in order to obtain meaningful measurements of nitrogen fixation. Delwiche and Wijler (1956) conducted 15Nz fixation experiments on grass sod and associated cultivated soils and found no nitrogen fixation to occur asymbiotically unless a source of energy, such as glucose, was added. More recently, W. A. Rice et al. (1967), also using isotopic nitrogen, measured 19 pg of N per gram of soil fixed in 28 days in laboratory samples of a prairie soil containing straw residues of the most recent crop. The addition of higher concentrations of straw or waterlogging greatly increased the rate of fixation (Table XIX). R. J. Myers et al. (1970) measured nitrogen fixation in undisturbed soil cores 15 cm in diameter

TABLE XIX Measurements of Asymbiotic Nitrogen Fixation Using I5N or C2H, Methodology

P h)

Reference W. A. Rice et al. ( I 967)

R. J. Myers et al. ( 1 970)

Mayland et al. ( 1966) Shtina et al. (1968)

Methodology

Experimental conditions

N

Soil aliquots given laboratory incubation (a) At field capacity, plus 1% straw (b) At field capacity, plus 5% straw (c) Waterlogged, plus 1% straw (d) Waterlogged, plus 5% straw 15-cm diameter undisturbed soil cores (a) Nostoc surface crust only (b) 0-2.5 cm deep (c) 2.5 cm deep (d) Total 7.5 cm core, and using C Z H Z 15-cm diameter core of cultivated soil plus 2% wheat straw (a) 0-2.5 cm deep (b) 2.5-7.5 cm deep (c) 7 5 cm deep Algal crusts of Arizona desert Algal crusts on chestnut soil, USSR Algal crusts on rendzina soil, USSR Mountain meadow (Colorado, U.S.A.) soil block containing algae in moss habitat Savanna (Nigeria) soil incubated moist and with 35% oxygen Sand dune (U.K.) marine environment dominated by Nostoc, 0-1 cm depth Jordan Fertility Plots (Pennsylvania, U.S.A.) (a) No fertilizer added, 0-15 cm depth (b) 27 kg/ha of N added, plus PK (c) 81 kglha of N added, plus PK

I5

l5

l5

N

N N

Porter and Grable ( I 969) Odu and Vine ( 197 I )

W. D. P. Stewart (1967) Hardy et al. (1968)

l5

N N

C,H,

N fixed (pglgram of soill28 days)

0

19 150

67 240

I” 239 4.3 0 7.3

0.6 0.9 0 43 46.5 62.2 5.0

I .o

47.6

33.9 22.4 11.8

?

3

s

THE MICROFLORA OF GRASSLAND

42 1

and found (Table XIX) lower nitrogen gains than were found by W. A. Rice et al. According to Myers and co-workers, the top 7.5 cm of cultivated soil amended with straw and incubated in a normal, controlled atmosphere fixed nitrogen equivalent to 0.6 kg/ha during 28 days of incubation. The same clay soil contained Nostoc crusts which fixed 239 pg of N per gram of Nostoc crust. In the natural field soil, no fixation occurred below 2.5 cm. The accumulated value of nitrogen fixation, as measured by 15N, in the Nostoc crust and in the top 2.5 cm of soil, was 1.52 kg/ha. Measurement of fixation in the same soil by the acetylene technique indicated 1.8 kg of N per hectare was fixed during equivalent time. techniques Data from other recent experiments utilizing I5N and C2H2 to measure fixation are also summarized in Table XIX. To present the information in a comparable form, data from the various authors have been recalculated on a micrograms per gram basis. Also further to facilitate comparison, data for incubation intervals longer or shorter than 28 days have been recalculated to 28 days. The microflora involved in asymbiotic nitrogen fixation is diverse. At least 15 genera of bacteria and many blue-green algae are now known to fix nitrogen. The classical nitrogen-fixing organism Azotobacter and the more recently isolated Beijerinckia occur sporadically and in low numbers in grassland soils (Ross, 1960; Di Menna, 1966; W. D. P. Stewart, 1969) and in soils generally. Exceptions are soils of the Nile Delta and sugarcane soils of Brazil in which high populations of Azotobacter and/or Beijerinckia occur. In semiarid grassland, neither genus is believed to be of importance in nitrogen fixation. Members of the genus Clostridium are widely distributed in grassland soil (Di Menna, 1966; Campbell et al., 1967; Ross, 1958, 1960). They can utilize the breakdown products of natural substrates, such as straw (W. A. Rice et al., 1967), to fix large concentrations of nitrogen. The clostridia are doubtless the most prolific nitrogen-fixers of the asymbiotic bacteria. Members of the genera Pseudomonas, Achromobacter, Klebsiella, and Bacillus can readily be isolated on nitrogen-free agar (MeiMejohn, 1968; Paul and Newton, 1961; Moore, 1966; Anderson, 1955), but they appear of doubtful effectiveness as nitrogen-fixers in field soil. The significance of such organisms as Methanobacterium and Desulfovibrio probably is not large. The photoautotrophic bacteria are widely distributed in aquatic environments but their functional occurrence in semiarid grasslands has not been demonstrated. Metcalfe and Brown (1957) described two Nocardia species from grassland that were able to fix atmospheric nitrogen. Fedorov and Ilina ( 1 960) also noted evidence of nitrogen fixation by actinomycetes, but Jensen ( 1965) concluded that

422

FRANCIS E. CLARK A N D ELDOR A. PAUL

evidence for nitrogen fixation by nonsymbiotic actinomycetes is mostly negative or unconvincing. Various species of blue-green algae are known to fix atmospheric nitrogen (Burris, 1956; Moore, 1966; Shields and Durrell, 1964; Shields et al., 1957; Russell, 1950; Hernandez, 1956). Cameron, Mayland and associates have calculated that algal surface crusts collected from Arizona desert rangelands could fix up to 10 kg/ha of nitrogen annually if the soil surface was fully covered with such crusts (Cameron and Fuller, 1960; Fuller et al., 1961; Mayland and McIntosh, 1966; Mayland et al., 1966). Since native rangeland is by no means fully covered by Nostoc or other crusts, their calculated value must be scaled down considerably for field conditions. Porter and Grable (1969) found algae contributed during 10 days approximately 4 kg of nitrogen per hectare in a wet mountain meadow in which conditions were very favorable for algal growth. Liverworts and mosses may variously contribute to nitrogen fixation. Bond and Scott (1955) found the liverwort Blasia pusilla capable of fixing lSNNz. Moore ( I 966) and W. D. P. Stewart ( I 966) have discussed the contribution that the photosynthetic lower plants may make to nitrogen fixation in various natural environments. Heterotrophic nitrogen-fixers appear to play their major role by being widely distributed and fixing small or very small quantities of nitrogen over prolonged periods, particularly in microhabitats containing available substrates and restricted oxygen (W. D. P. Stewart, 1969; W. A. Rice et al., 1967), possibly in the rhizosphere (Mishustin, 1967), and in the case of the photosynthetic lower plants, singly or in mixed associations, in microhabitats of favorable light intensity and with high moisture, at least intermittently.

B. NITRIFICATION Many workers have reported that there is an inhibition of nitrification in grassland soil (Lawes, 1889; Miller, 1906; Lyon and Bizzell, 1918; Richardson, 1938; Parberry, 1945; Theron, 195 1 : Theron and Haylett, 1953: Soulides and Clark, 1958; Woldendorp, 1962; Boughey et al., 1964; E. L. Rice, 1964, 1965; Munro, 1966; Neal, 1969). In such soil, the nitrate content is commonly observed to be negligible or absent while the ammonium content, although small, is usually of measurable quantity. This is in contrast to most cultivated soils, in which more of the mineral nitrogen is to be found in the nitrate than in the ammonium form. Various explanations have been offered for the relatively poor nitrification that occurs in grassland soil. T h e most commonly supported hypothesis is that there is direct suppression of the nitrifying bacteria by the

423

THE MICROFLORA O F GRASSLAND

living plant root. This was first suggested by Lyon and Bizzell(l9 18) and Lyon et al. (1923) and was later endorsed by Theron (195 1 ) and Theron and Haylett (1953). In recent years other workers have presented data showing that exposure to either intact plant roots or to aqueous extracts prepared therefrom influences the activity of the nitrifying bacteria (Novogrudskii, 1963: Molina and Rovira, 1964; E. L. Rice, 1964, 1965; Boughey et al., 1964; Munro, 1966; Neal, 1969). Varying degrees of inhibition of nitrification by grass roots have been reported. Munro (1966), working in Africa, found that all of the seven grass species tested inhibited nitrification. In short-term experiments involving exposure of nitrifiers to aqueous extracts of grass roots, oxidation of both ammonium and nitrite nitrogen was strongly to very strongly inhibited (Table XX). In contrast, Neal (1969) found that while the native dominant grasses in western Canada failed to depress nitrification, grasses and forbs that increase on or invade overgrazed land commonly produced nitrification inhibitors. TABLE XX Influence of Aqueous Extracts of Grass Roots on the Activity of Nitrifying Bacteria" Nitrite N produced Source of root extract

PPm

Percent inhibition

None, control Hyparrhenia jilipendula Cynadon dactylon Rhyncheletrum repens Sporobolus pyramidalis Eragrostis curvula Themeda triandra Pennisetum purpureum

7.9 0.8 2.7 2.1 1 .o 1.6 1 .o 3.4

90 66 66 88 80 88 57

-

Nitrite N oxidized

PPm

Percent inhibition

9.0 0.2 1.9 1.9 2.4 1.6 0.5 5.0

98 79 79 74 82 95 45

Data of Munro ( 1 966).

Other workers have expressed doubt that inhibitions of nitrification are caused by toxins of root origin and indeed some even doubt the occurrence of any such inhibition in grassland. Brar and Giddens (1968) undertook to extract possible inhibitors from grassland soil but following extraction, noted no improvement in the nitrifying capacity of the soil. They concluded that the low nitrifying capacity in their soil was caused by soil acidity and a low population of nitrifiers. Ross (1958) also believed

424

FRANCIS E. CLARK A N D ELDOR A. PAUL

that the low nitrifying potential of some grassland soils was due simply to their acidity. Parberry (1945) and Michniewicz (195 1 ) believed there might be insufficient aeration for good nitrification in grassland soils covered by a layer of decaying plant residues. Harmsen and van Schreven (1955) doubted the validity of any such explanation, since well-drained grasslands generally have good aeration, often far superior to that of arable land. O’Connor et al. (1962) and Robinson (1963) suggested that the low production of nitrate in grassland was linked to a low population of nitrifying bacteria caused by the scarcity of ammonium ions in grassland and the direct competition of grass roots for such ions. Chase et al. (1967) found that grasses grown in Ontario, Canada, did not compete for ammonium to the detriment of the nitrifiers nor did the grass roots inhibit nitrifiers by other means. Populations of nitrifiers in urea-treated grass and fallow plots were fully comparable. It would appear that the mechanisms by which and the extent to which different grassland soils inhibit nitrification are in need of further study. Presently the bulk of the evidence is that such inhibition does exist and, as E. L. Rice ( 1 965) and Neal (1969) have pointed out, possibly the inhibitory effect may be of importance in conserving the low amount of available nitrogen present in grassland soils or in determining the success of individual species to invade or to establish dominance in plant communities. C. DENITRIFICATION AND VOLATILIZATION The three general pathways for gaseous losses of nitrogen from soil are those of enzymatic denitrification, chemodenitrification, and volatilization of ammonia. Recent reviews concerning these pathways are available (Broadbent and Clark, 1965 : Gardner, 1965). 1 . Enzymatic Denitrifcation

In order for enzymatic denitrification, a process in which microorganisms turn to a nitrate respiration in lieu of an oxygen respiration, to occur, it is necessary that nitrate be present, that oxygen be absent or greatly limited in its availability, and that sufficient available substrate or energyyielding material be present to permit microbial activity. Additionally of course, such environmental factors as moisture, temperature, reaction, etc. must be favorable for microbial activity. Grassland soils, particularly those of semiarid rangelands, are characteristically low in nitrate, and they are usually well aerated. Accordingly, enzymatic denitrification in them can be expected to be negligible. However, should nitrate be present and should there be either regions or microsites of poor aeration, the de-

THE MICROFLORA OF GRASSLAND

425

nitrification process should not be greatly limited because of any shortage of microbial substrate (Greenland, 1958, 1965). Grassland soils usually contain considerable amounts of organic matter in the upper few centimeters of the profile. 2. Chemodenitrification

Chemodenitrification involves loss of gaseous nitrogen to the atmosphere due to nitrite instability (Clark, 1962). Inasmuch as the nitrite dismutation requires the presence of nitrite and is promoted by microbial metabolism (Clark and Beard, I96 I ) , chemodenitrification occurs primarily in soils in which the nitrification process is sufficiently retarded to permit the accumulation of nitrite as an intermediate product in the transformation of ammonium to nitrate and in which the organic matter content is sufficiently high to support an active microflora. The inhibition of nitrification that occurs in grassland soil, discussed above, is often characterized by nitrite accumulation (Clark et al., 1960). In the work of Soulides and Clark ( I 958), the gaseous nitrogen loss via chemodenitrification in some laboratory lots of grassland soils treated with urea nitrogen was more than twice that measured for nearby intertilled soil. Even though chemodenitrification losses of considerable magnitude can be measured in the laboratory (Reuss and Smith, 1965: Soulides and Clark, 1958; Wullstein and Gilmour, 1964), at present there appears to be no justification for attempting to extrapolate such laboratory data to finite values in the field. Most grassland soils do not receive ammoniacal fertilizers and therefore such soils seldom if ever contain appreciable nitrite. Only in grassland receiving ammonium nitrogen and in addition characterized by a poor nitrifying capacity would any serious nitrogen loss via chemodenitrification be expected. Even then, should warm temperature and drying conditions prevail, losses via ammonia volatilization doubtless would be of much greater magnitude. 3. Volatilization of Ammonia

The loss of gaseous ammonia from soil is basically a physicochemical process controlled by meteorological conditions and soil characteristics (Gardner, 1965; Mortland, 1958). Microorganisms are involved to the extent that they contribute to the supply of ammonium ions in the soil by producing enzymes that hydrolyze urea or by decomposing soil organic matter. Microbial release of ammonium nitrogen from soil organic matter occurs at a rate that is usually less than one or more of the following: the concurrent rate of nitrogen demand by growing plants; the ammonium retention capacity of the soil: and/or the ammonium-oxidizing

426

FRANCIS E. CLARK A N D ELDOR A. PAUL

potential of the chemotrophic nitrifying bacteria in the soil. Because of these demands, volatilization of ammonia microbially released from soil organic matter can rarely be expected to occur. Volatilization of ammonia is known to occur from grassland soils given surface applications of urea or aqua-ammonia (Allison, 1955, 1966; Volk, 1959, 1961). There is usually a high urease content in grassland litter: consequently, ammonia volatilization is usually higher from grassland soil than from bare soil following equivalent applications of urea fertilizer to the two soils. On unfertilized grassland in semiarid regions, some loss of ammonium nitrogen from the system is known to occur from urea nitrogen voided by grazing livestock. B. A. Stewart (1 970) found that following addition of steer urine to bare soil with surface drying conditions, volatilization loss could be as much as 90% of the applied nitrogen. However, urine-voided nitrogen that becomes volatilized is not necessarily lost from the soil-plant system. It may be returned to the soil in precipitation or it may be reabsorbed by moist surfaces (plant, soil) or free water surfaces within the ecosystem. ACKNOWLEDGMENT This review stems from participation by the authors in the Grasslands Biome segment of the International Biological Program. Its compilation was supported in part by National Science Foundation Grant No. GB7824 and in part by the Canadian National Research Council, being issued thereunder as CCIBP Report No. 41. The authors gratefully acknowledge helpful suggestions and discussion received from many colleagues at the Pawnee and Matador grassland sites in the United States and Canada. REFERENCES Alexander, M. 1961. “Introduction to Soil Microbiology.” Wiley, New York. Allen, L. A., Harrison, J., Watson, S. J., and Ferguson, W. S. 1937. J . Agr. Sci. 27,27 1-293. Allison, F. E. 1955. Advan. Agron. 7,2 13-250. Allison, F. E. 1966.Advan. Agron. 18,219-258. Anders0n.G. R. 1955. J . Bacteriol. 70,129-133. Arshad, M. A., and Lowe, L. E. 1966. Soil Sci. Soc. Amer.. Proc. 30,73 1-735. Babiuk, L. A., and Paul, E. A. 1970. Can. J . Microbiol. 16,57-62. Balloni, W., and Materassi, R. 1968. Trans. 9th Int. Congr. Soil Sci., 1968 Vol. 2, pp. 159162. Bayliss-Elliot, J. S. 1926. Ann. Appl. Biol. 13,277-288. Beetle, A. A. 1952. J . Range Manage. 5, 141-143. Beijerinck, M. W. 1888. Bot. Ztg. 46,742-750. Biederbeck, V. O., and Paul, E. A. 1968. Can. SOC.Soil Sci., Proc. 14,7. Billing, E.,and Baker, L. A. E. 1963.5.Appl. Bacteriol. 26,58-65. Bisby, G . D.,James, N., and Timonin, M. 1.1933. Can. J . Res. 8,253-275. Bisby, G. D., Timonin, M. I., and James, N. 1935. Can. J . Res. 13,47-65. Bond, G . 1967. Annu. Rev. Plant Physiol. 18,107- 126.

THE MICROFLORA OF GRASSLAND

427

Bond, G. 1968. In “Recent Aspects of Nitrogen Metabolism in Plants” (E. J. Hewitt and C. V. Cutting, eds.), pp. 15-25. Academic Press, New York. Bond, G., and Scott,G. B. 1955.Ann. Bot. (London) [N.S.] 19,67-77. Botner, P. 1967. C . R. Acad. Sci., Ser. D 265,1468- 1470. Boughey, A. S., Munro, P. E., and Meiklejohn, J. 1964. Nature 203,1302- 1303. Brar, S. S., and Giddens, J. 1968. Soil Sci. SOC.Amer., Proc. 32,82 1-823. Bremner, J. M. 1965. In “Soil Nitrogen” (W. V. Bartholomew and F. E. Clark, eds.), pp. 93-149. Amer. SOC.Agron., Madison, Wisconsin. Bremner. J. M. 1967. I n “Soil Biochemistry” (A. D. McLaren and G. H. Peterson, eds.), pp. 19-66. Marcel Dekker, New York. Brierley, W. B. 1923. I n “Microorganisms of the Soil” (E. J. Russell, ed.), p. I 18- 147. Longmans, Green, New York. Bristol Roach, B. M. 1926.Ann. Bot. (London) 40,149-201. Bristol Roach, B. M. 1927.Ann. Bot. (London) 41,509-5 17. Broadbent, F. E. 1947. SoilSci.SOC.Amer., Proc. 12,246-249. Broadbent, F. E., and Clark, F. E., 1965. In “Soil Nitrogen” (W. V. Bartholomew and F. E. Clark, eds.), pp. 344-349. Amer. SOC.Agron., Madison, Wisconsin. Brown,P.E. 1917.Science46,171-175. Brydon, J. E., and Sowden, F. J. 1959. Can. J . Soil Sci. 39,136- 143. Budd, A. C., and Best, K. F. 1964. “Wild Plants of the Canadian Prairies.” Queen’s Printer, Ottawa. Burges, A. 1967. I n “Soil Biology” (N. A. Burges and F. Raw, eds.), pp. 479-492. Academic Press, New York. Burges, A. 1968. Trans. 9th Int. Congr. SoilSci., 1968 Vol. 2, pp. 29-35. Burges, A., and Latter, P. 1960. Nature 186,404-405. Burri, R. 1903. Zentralbl. Bakteriol., Parasitenk., Infektionskr. Hyg., Abt. 2 10,756-763. Burris, R. H. 1956. In “Inorganic Nitrogen Metabolism” (W. D. McElroy and B. Glass, eds.), pp. 3 16-343. John Hopkins Press, Baltimore, Maryland. Burris, R. H. 1969. Proc. Roy. Soc., Ser. B 172,339-354. Burstrom H. G. 1965. In “Ecology of Soilborne Plant Pathogens” (K. F. Baker and W. C. Snyder, eds.), pp. 154- 166. Univ. of California Press, Berkeley, California. Cameron, R. E., and Fuller, W. H. 1960. SoilSci. SOC.Amer., Proc. 27,353-356. Camp, T. R. 1963. “Water and its Impurities.” Reinhold, New York. Campbell, C. A., Paul, E. A., Rennie, D. A., and McCallum, K. J. 1967. SoilSci. 104,8145 and 2 17-224. Carlisle, A., Brown, A. H. F., and White, E. J. 1966. J . Ecol. 54,65-98. Casida, L. E., Jr., Klein, D. A., and Santoro, T. 1964. Soil Sci. 98,37 1-376. Chase, F. E., Corke, C. T., and Robinson, J. B. 1967. I n “Ecology of Soil Bacteria” (T. R. G. Gray and D. Parkinson, eds.), pp. 593-61 I. Liverpool Univ. Press, Liverpool. Chesire, M. V., Cranwell, P. A., Falshaw, C. P., Floyd, A. J., and Haworth, R. D. 1967. Tetrahedron 23,1669- 1682. Chesire, M. V., Mundie, C. M., and Shepherd, H. 1969. Soil Biol. Biochem. 1, 117-130. Chichester, F. W. 1969. SoilSci. 107,356-363. Clark, F. E. 1940. Trans. Kansas Acad. Sci. 43,7544. Clark, F. E. 1949.Advan. Agron. 1,241-288. Clark, F. E. 1962. Trans. Int. Soil Conf. New Zealand 1962 pp. 173-176. Clark, F. E. 1967. In “Soil Biology” (N. A. Burges and F. Raw, eds.), pp. 15-49. Academic Press, New York. Clark, F. E., and Beard, W. E. 196 1. Trans. 7th Int. Cong. Soil Sci., I960 Vol. 3, pp. 50 1 508.

428

FRANCIS E. CLARK A N D ELDOR A. PAUL

Clark, F. E., Hervey, R. J., and Blank, L. M. 1947. US.,Dep. Agr., Tech. Bull. 935. Clark, F. E., Beard, W. E., and Smith, D. H. 1960. Soil Sci. SOC.Amer., Proc. 24,50-54. Cochrane, V .W. 1958. “Physiology of Fungi.” Wiley, New York. Cochrane, V. W. 1961.Annu. Rev. Microbiol. 16,l-26. Crosse, J . E. 1959.Ann. Appl. Eiol. 47,306-3 17. Crosse, J. E. I963.Ann. Appl. Eiol. 52,97- 1 1 1. Dahlman, R. C. 1968. In “Methods of Productivity Studies in Root Systems and Rhizosphere Organisms” (M. S. Ghilarov, ed.), pp. 11-21. Akad. Nauk. S. S. S. R., Leningrad. Dahlman, R. C., and Kucera, C. L. 1965. Ecology 46,84-89. Dalbro, S. 1956.14th Int. Horr. Congr., Rep., 1955 p. 770. Daubenmire, R. F. 1968. Advan. Ecol. Res. 5,209-266. Delwiche, C. C., and Wijler, J. 1956. Planr Soil 7, 113-129. Di Menna, M. E. 1957. J . Gen. Microbiol. 17,678-688. Di Menna, M. E. l958a. Nature 181,1287. Di Menna, M. E. l958b. J . Gen. Microbiol. 18,269-272. Di Menna, M. E. 1 9 5 8 ~J. . Gen. Microbiol. 19,58 1-583. Di Menna, M. E. 1959. New ZealandJ. Agr. Res. 2,394-485. Di Menna, M. E. 1966. New ZealandJ. Agr. Res. 9,218-226. Dix, R. L., and Butler, J. E. 1954. J . Range Manage. 7,265-268. Doeksen, J., and van der Drift, J. 1963. “Soil Organisms.” North-Holland Publ., Amsterdam. Domsch, K. H. 1962. Zentralbl. Bakteriol., Parasitenk., Infektionskr. Hyg., Abr. 2 116,3378. Dorokhova, N. A. 1967. I n “Mycotrophy in Plants” (A. A. Imshenetskii, ed.), pp. 221-23 1. Akad. Nauk S.S.S.R.. Moscow. Doxtader. K . G. 1969. Eacreriol. Proc., Abstr. 69th Annu. Meeting p. 14. Dubach, P., and Mehta, N. D. 1963. Soils Fert. 26,293-300. Duggeli, M. 1904. Zentralbl. Bakteriol., Parasitenk., Infektionskr. Hyg., Abt. 2 12,602-6 14. Durrell, L. W. 1959.Amer. Midl. Narur. 61,322-328. Dwivedi, R. S. 1966. Trop. Ecol. 7,84-89. Dye, D. W. 1964. New Zealand J . Sci. 7,26 1-269. England, L. M., and Rice, E. L. 1957. Eot. Gaz. (Chicago) 118,186-190. Eriksson, E. 1952. Tellus 4,215-232. Estermann, E. F., Peterson, G. H., and McLaren, A. D. 1959. Soil Sci. SOC.Amer., Proc 23, 3 1-36. Farnsworth, R. B., and Hammond, M. W. 1968.ASA (Amer. Soc.Agron.)Absrr. p. 5 3 . Fedorov, M. F., and Ilina, T. K. 1960. Mikrobiologiya 30,832-838. Flaig, W. 1964. Geochim. Cosmochim. Actu 28,1523- 1525. Flaig, W., and Schmidt, H. L. 1957. Arch. Mikrobiol. 27, 1-32. Flint, E. A. 1958. New ZealandJ. Agr. Res. 1,99 1-997. Ford, G . W., and Greenland, D. J. 1968. Trans. 9th Inr. Congr. Soil Sci., 1968 Vol. 2, pp. 403-410. Fred, E. B., and Waksman, S. A. 1928. “Laboratory Manual of General Microbiology.” McGraw-Hill, New York. Fuller, W. H., Cameron, R. E., and Raica, N. 1961. Trans. 7th Int. Congr. Soil Sci., 1960 VOI 2, pp. 6 17-624. Galstyan, A., Tatevosian, G., and Havoundjian, Z. S. 1968. Trans. 9th Inr. Congr. Soil Sci., 1968 Vol. 3, pp. 28 1-288.

THE MICROFLORA OF GRASSLAND

429

Gams, W. 1967. Mitt. Biol. Bundesans. Land-Forstwirt., Berlin-Dahlem 123,l-77. Gardner, W. R. 1965. I n “Soil Nitrogen” (W. V. Bartholomew and F. E. Clark, eds.), pp. 550-572. Amer. SOC.Agron., Madison, Wisconsin. Gibson, T. 1939. Proc. 3rdInt. Congr. Microbiol., 1939 Abstr. Commun. pp. 304-305. Golley, F. B. 1960. Ecol. Monogr. 30,187-206. Graff, O., and Satchell, J. E. 1967. “Progress in Soil Biology.” North-Holland Publ., Amsterdam. Graham, D. C., and Hodgkiss, W. 1967. J . Appl. Bacteriol. 30,179-189. Grant. P. M. 1967. Rhodesia, Zambia MalauiJ. Agr. Res. 5,211-213. Greenhill, A. W.,andChubnall, A. C. 1934.Eiochem. J . 28,1422-1427. Greenland, D. J. 1958. J . Agr. Sci. 50,82-91. Greenland, D. J. 1965. Soils Fert. 28,415-425 and 521-532. Greenland, D. J.. and Ford, G. W. 1964. Trans. 8th Int. Congr. Soil Sci.,1964 Vol. 3, pp. 137-148. Gruber, T. 1909. Zentralbl. Bakteriol., Parasitenk., Infektionskr. Hyg., A b t . 2 22,401-4 16. Gyllenberg, H. 1955. Physiol. Plant. 8,644-652. Hagihara, B. 1960. In “The Enzymes” (P. D. Boyer, H. Lardy, and K. Myrback, eds.), 2nd rev. ed., Vol. 4, pp. 193-2 13. Academic Press, New York. Hansen. E. H.. and Schnitzer, M . 1966. Trans. Int. Soil Con$, 1966 pp. 87-92. Hardisson, C., and Robert-Gero, M. 1966. Ann. Inst. Pasteur, Paris 111,486-496. Hardy, R. W. F.,and Burns, R. C. 1968. Annu. Rev. Eiochem. 37,331-358. Hardy, R. W. F., and Knight, E., Jr. 1967. In “Progress in Phytochemistry” (L. Reinhold, ed.), pp. 387-469. Wiley, New York. Hardy, R. W. F., Holsten, R. D., Jackson, E. K., and Burns, R. C. 1968. Plant Physiol. 43, I 185-1 207. Harmsen, G . W., and Jager, G. 1963. I n “Soil Organisms”(J. Doeksen and J. van der Drift, eds.), pp. 245-25 1. North-Holland Publ., Amsterdam. Harmsen, G. W., and van Schreven, D. A. 1955. Advan. Agron. 7,299-398. Hayes, M. H. B., Stacey, M., and Standley, J. 1968. Trans. 9th Int. Congr. Soil Sci., 1968 Vol. 3 , pp. 247-256. Heard, A. J. 1965. J . Agr. Sci. 64,329-334. Hellreigel. H . . and Wilfarth. H. 1888. Beilagrheft Z. Ver. Robenruc,Xer-Ind. D P U IReichs. . Hernandez, S . C . 1956. J . SoilSci. SOC.Philipp. 8,19-22. Hodgson, G . W., Hitchon, K., Takuchi, K., Baker, D. L., and Peake, E. 1968. Geochim. Cosmochim. Acta 32,737-772. Hofmann. E. 1963. Recent Progr. Microbiol., Symp. 8th Int. Congr. Microbiol., 1962, pp. 2 16-220. Hofmann, E., and Hoffmann, G. 1966. Advan. Enzymol. 28,365-390. Hogg, B. M., and Hudson, H. J. 1966. Trans. Brit. Mycol. SOC.49,185-192. Hopkins. H. 1954. J. Range Manage. 7,255-258. Hudson, H. J. 1962. Trans. Brit. Mycol. SOC.45,395-413. Hudson, H. J.,and Webster,J. 1958. Trans. Brit. Mycol. Soc. 41, 165-177. Hurst, H. M. 1963. I n “Enzyme Chemistry of Phenolic Compounds” (J. B. Pridham, ed.), pp. I2 1- 152. Macmillan, New York. Hurst, H. M., Burges, A.,and Latter, P. 1962. Phytochemistry 1,227-23 I . Huss, H. 1907. Zentralbl. Bakteriol., Parasitenk., Infektionskr. Hyg. Abt. 2 19,50-70. Iswaran, V . , and Harris, J. R. 1968. Trans. 9th I n t . Congr. Soil Sci., 1968 Vol. 3 , pp. 50151 I . Jackson, R. M. 196.5.N e w Zealand J. Agr. Res. 8,865-877.

430

FRANCIS E. CLARK A N D ELDOR A. PAUL

James, N. 1955. Can.J . Microbiol. 1,479-485. Jenkinson, D. S. 1968. Soil Sci. 105, 25-39. Jenkinson, D. S., and Tinsley, J. 1960. Roy. SOC.Dublin Sci. Proc., Ser. A 1, 141-147. Jensen, H. L. 1931.SoilSci. 31,123-158. Jensen, H. L. 1933. Proc. Linn. SOC.NewSouth Wales 58,181-185. Jensen, H. L. 1963. Sci. Horr. 16,I5-2 I . Jensen, H. L. 1965. In “Soil Nitrogen” (W. V. Bartholomew and F. E. Clark, eds.), pp. 436480. Amer. SOC.Agron., Madison, Wisconsin. John, R. P. 1942.Ann. Bor. (London) [N.S.] 6,323-349. Junge, C. E. 1958. Trans. Amer. Geophys. Union 39,241-248. Katznelson, H. 1965. In “Ecology of Soilborne Plant Pathogens” (K. F. Baker and W. C. Snyder, eds.), pp. 187-207. Univ. of California Press, Berkeley, California. Katznelson, H.,and Stevenson, I. L. 1956. Can.J . Microbiol. 2,611-622. Katznelson, H., Lochhead, A. G., and Timonin, M. I. 1948. Bor. Rev. 14,543-587. Keddie, R. M. 1959. J . Appl. Bacreriol. 22,403-416. Kelly, J. M., and Opstrup, P. A. 1968. In “Progress in Terrestrial and Freshwater Ecology,” Publ. No. 250, pp. 112-1 13. Oak Ridge Nat. Lab., Oak Ridge, Tennessee. Kendrick, W. B., and Burges, A. 1962. Nova Hedwigia 4,3 13-342. Kleist, H.,and Mucke, D. 1966. Experientia 22,136-137. Kononova, M. M. 1966. “Soil Organic Matter.” Pergamon Press, Oxford. Krasil’nikov, N. A. 1944. Mikrobiologiya 13,144- 146. Kreutzer, W. A. 1969..I. Colorado-Wyoming Acad. Sci. 6,17. Kroulik, J . T., Burkey, L. A., and Wiseman, H. G. 1955.J. Dairy Sci. 38,256-262 and 263272. Krzysch, G. 1965. Z. Acker.-Pjanzenbau 122, 108-140. Kudrina, E. A. 195 I . Tr. Pochv. Insr. V . V . Dokuchaeva, Akad. Nauk SSSR 38,185-253. Kurtz, W. G . W., Dawson, P. S. S., and Blakely, E. R. 1969. Can. J . Microbiol. 15,27-33. Kuster, E. 1967. In “Soil Biology” (N.A. Burges and F. Raw, eds.), pp. I 1 1-127. Academic Press, New Y ork. Kutzner, H. J. 1956. Ph.D. Dissertation, Hohenheim. Kutzner, H. J. 1968. Landwirt. Forsch. 21,48-62. Ladd, J. N., and Butler, J. A. 1966. In “Use of Isotopes in Soil Organic Matter Studies,” F.A.0.II.A.E.A. Tech. Meeting, 1963,pp. 143-152. Pergamon Press, Oxford. Lal, S. P., and Yadav, A. S. 1964. Indian Phyroparhol. 17,208-2 18. Last, F. T. 1955. Trans. Brit. Mycol. SOC.38,221-239. Latter, P. M.,and Cragg, J. B. 1967.J . Ecol. 55,465-482. Latter, P. M., Cragg, J. B., and Heal, 0. W. 1967. J . Ecol. 55,445-464. Lawes, J. B. 1889. J . Roy. Agr. SOC.Engl. 25,1-24. Lawrence, D. B., Schoenike, R. E., Quispel, A., and Bond, G. 1967. J . Ecol. 55,793-813. Leben, C. 1961. Phyroparhology 51,553-557. Leben, C . 1963. Phytoparhology 53,778-78 I . LeClerg, E. L., and Smith, F. B. 1928. SoilSci. 25,433-441. Lewin, R. A. 1962. “Physiology and Biochemistry of Algae.” Academic Press, New York. Lewis, J. K., Van Dyne, G. M., Albee, L. R., and Whetzal, F. W. 1956. Sourh Dakora, Agr. Exp. Sra., Bull. 459. Lindqvist, I. 1968. Lantbruks-Hoegsk, Ann. 34,377-389. Lochhead, A. G . 1940. Can. J . Res., Sect. C 18,42-53. Lohnis, F. 19 10. “Handbuch der landwirtschaftlich Bakteriologie” Borntraeger, Berlin. Lohnis, F. 1926. SoilSci. 22,253-290.

THE MICROFLORA OF GRASSLAND

43 1

Long, W. G., Sweet, D. V.,andTukey, H. B. 1956.Science 123,1039-1040. Loutit, M., and Loutit, J. S. 1966. New Zealand J . Agr. Res. 9,84-92. Lund, J. W. G. 1945. New Phytol. 44,196-2 19. Lund, J . W. G. 1967. In “Soil Biology” (N. A. Burgesand F. Raw, eds.),pp. 129-147. Academic Press, New York. Lyon, T. L., and Bizzell, J. A. 19 18. Cornell Univ.,Agr. Exp. Sta., Mem. 12, I- 1 15. Lyon, T. L.. and Wilson, J. K. 192 I . New York,Agr. Exp. Sfa.,Geneva, Mem. 40.1-44. Lyon, T. L., Bizzell, J. A., and Wilson, B. D. 1923. J . Amer. SOC.Agron. 15,457-467. Macauley, B. J., and Thrower, L. B. 1966. Trans. Brit. Mycol. SOC.49,509-520. McConnell, W. B. 1950. Can.J. Res., Sect. C 28,600-612. McDougall, B. M. 1968. Trans. 9th Int. Congr. Soil Sci., 1968 Vol. 3, pp. 647-655. Macfadyan, A. 1963. I n “Soil Organisms” (J. Doeksen and J. vander Drift, eds.), pp. 3-16. North-Holland Publ., Amsterdam Mack, E. 1936. Zentralbl. Bakteriol., Parasitenk., Infekfionskr. Hyg., A b t . 2 95,218-261. Mackenthun, K. M. 1965. “Nitrogen and Phosphorus in Water.” U.S. Govt. Printing Office, Washington, D.C. McLuckie, J. 1922. Proc. Linn. SOC.New South Wales 47,3 19-329. Macura, J . 1966. I n “Theoretical and Methodological Basis of Continuous Culture of Microorganisms” ( I . Malek and Z. Fencl, eds.), pp. 12 1-132. Academic Press, New York. Macura, J. 1968. Trans. 9th Int. Congr. SoilSci., 1968 Vol. 3, pp. 619-627. Malcolm, R. L., and McCraken, R. J. 1968. SoilSci. Soc. Amer., Proc. 32,834-838. Martin, J . P., Richards, S. J., and Haider, K. 1967. Soil Sci. SOC.Amer., Proc. 31,657-662. Mathur, S. P., and Paul, E. A. 1967. Can. J . Microbiol. 13,573-580 and 58 1-586. Mayaudon, J . 1971. I n “Use of Isotopes and Radiation in Soil Organic Matter Studies,” F.A.OJ1.A.E.A. Tech. Meeting, 1968. (in press). Mayland, H. F., and Mclntosh, T. H. 1966. Nature 209,420-42 I . Mayland, H. F., Mclntosh, T. H., and Fuller, W. H. 1966. Soil Sci. SOC.Amer., Proc. 30, 56-60. Meiklej0hn.J. 1968. Trans. 9th Inf. Congr. SoilSci., 1968 Vol. 2, pp. 141-150. Meredith, D. S. 1962. Trans. Brit. Mycol. SOC.45,335-347. Metcalfe, G., and Brown, M. F. 1957.J . Gen. Microbiol. 17,567-572. Michniewicz, M. 1951. Ann. Univ. Mariae Curie-Sklodowska, Sect. C 6,19-75. Miller, N. H. J. 1906. J. Agr. Sci. 1,377-399. Minderman, G . 1968. J . Ecol. 56,355-362. Minderman, G., and Daniels, L. 1967. In “Progress in Soil Biology” (0. Graff and J. E. Satchell, eds.), pp. 3-9. North-Holland Publ., Amsterdam. Mishra, R. R. 1965. Trop. Ecol. 6,133-140. Mishra, R. R. 1966. Trop. Ecol. 7,100-1 14. Mishustin, E. N. 1956. Soils Fert. 19,385-392. Mishustin, E. N. 1967. Mikrobiologiya 36,664-685. Mishustin, E. N., Nikitin, D. I., and Vostrov, 1. S. 1968. Trans. 9th Inf. Cangr. Soil Sci., 1968 Vol. 3, pp. 629-637. Molina, J. A. E., and Rovira, A. D. 1964. Can. J . Microbiol. 10,249-257. Monteith, J . L., Szeicz, G., and Yakuku, K. 1964. J . Appl. Ecol. 1,321-337. Moore, A. W. 1966. Soils Ferf. 29,113- 128. Mortensen, J. L., and Himes, F. L. 1964. I n “Chemistry of the Soil” (F. E. Bear, ed.), pp. 206-24 I . Reinhold, New York. Mortland, M. M. 1958. Advan. Agron. 10,325-348.

432

FRANCIS E. CLARK A N D ELDOR A. PAUL

Mrysha, G. N. 1966. Soils Fert. 30,1062. Munro, P. E. 1966. J.Appl. Ecol. 3,227-229 and 23 1-238. Myers, M. G.,and McGarity, J. W. 1968. PlantSoil28,25-37. Myers, R. J., Rice, W. A., and Paul, E. A. 1970. Can. J . SoilSci. 49(in press). Neal, J. L., Jr. 1969. Can. J . Microbiol. 15,633-635. Nicolson, T. H. 1959. Trans. Brit. Mycol. SOC.42,439-448. Nikitin, D. 1. 1960. Izv. Akad. NaukSSSR, Ser. Biol. No. 4,618-625. Novogmdskii, E. D. 1963.Agrobiologiya 5,720-724. Ochilova, M. 1961. Soils Fert. 24,1937: 25,2962. O’Connor, K. F., Robinson, J. B., and Jackman, R. H. 1962. Trans. Int. Soil Conf., 1962 pp. 177-182. Odu, C. T. I., and Vine, 1. L. 197 I . I n “Use of lsotopesand Radiation in Soil Organic Matter Studies,” F.A.0.II.A.E.A. Tech. Meeting, 1968. (in press). Oginsky, E. L., and Umbreit, W. W. 1955. “An Introduction to Bacterial Physiology” Freeman, San Francisco, California. Oleniacz, W. S., and Pisano, M. A. 1968. Appl. Microbiol. 16,90-96. Orpurt, P. A., and Curtis, J. T. 1957. Ecology 38,628-637. Ovchinnikova, M. E. 1965. Sov. Soil Sci. pp. 1564-1573. Paine, F. S. 1927. Mycologia 19,248-267. Parberry, N . H. 194S.Agr. G a z . New South Wales 56,543-544. Parker, D. T. 1962. Soil Sci. Soc. Amer., Proc. 26,559-562. Parkinson, D. 1967. I n “Soil Biology” (N. A. Burges and F. Raw, eds.), pp. 449-478. Academic Press, New York. Parr, J. F., Parkinson, D., and Norman, A. G. 1967. Soil Sci. 103,303-3 10. Paul, E. A. 1970.Advan. Phytochem. (in press). Paul, E. A., and Newton, J. D. 1961. Can.J. Microbiol. 7,7-13. Paul, E. A., Biederbeck, V. O., and Rosha, N. S. 1967. In “Methods of Study in Soil Ecology”IBPSymp., 1 9 6 7 , ~1-10. . Pelczar, M. J., Jr., and Reid, R. D. 1958. “Microbiology.” McGraw-Hill, New York. Petersen, J. B. 193S.Dan. Bot. Ark. 9,l-183. Piercy,A. 1917.Ann. Bor. (London) 31,513-537. Pirt, S. J. 1965. Proc. Roy. Soc., Ser. B 163,224-23 I . Pontovich, V. E. 1938. Mikrobiologiya 7,696-707. Porter, L. K., and Grable, A. R. 1969. Agr0n.J. 61,521-523. Posner, A. M., Theng, B. K. G., and Wake, J. R. H. 1968. Trans. 9rh In?.Congr. Soil Sci., 1968 VOl. 3,pp. 153-162. Prescott, S. C., Winslow, C. E. A., and McCrady, M. H. 1946. “Water Bacteriology” 6th ed. Chapman & Hall, London. Pugh, G. J. F. 1958. Trans. Brit. Mycol. SOC.41,185-195. Rauzi, F., and Hanson, C. L. 1966.5. Range Manage. 19,351-356. Rauzi, F., Fly, C. L., and Dyksterhuis, E. J. 1968. U.S.,Dep. Agr., Tech. Bull. 1390. Ray, R. Y.,and Dwivedi, R. S. 1962. Proc. Nut. Acad. Sci.. India, Sect. B 32,421-428. Reuss, J. O., and Smith, R. L. 1965. Soil Sci. SOC.Amer., Proc. 29,267-270. Rhoades, E. P., Locke, L. F., Taylor, H. M., and McIlvain, E. H. 1964. J . RangeManage. 17,185-190. Rice, E. L. 1964. Ecology 45,824-837. Rice, E. L. 1965. Physiol. Plant. 18,255-268. Rice, W. A., Paul, E. A., and Wetter, L. R. 1967. Can. J . Microbiol. 13,829-836. Richards0n.H.L. 1938.J.Agr.Sci. 2tL73-121.

THE MICROFLORA OF GRASSLAND

43 3

Robinson, J. B. 1963. PlantSoill9,173-183. Robinson, J. B., and MacDonald, G . M. 1964. New ZealandJ. Agr. Res. 7,146- 157. Ross, D. J. 1958. New ZealandJ. Agr. Res. 1,958-967 and 968-973. Ross, D. J. 1960. New Zealand J . Agr. Res. 3,224-236 and 245-249. Ross, D. J. 1965.J . Soil Sci. 16,7345 and 86-94. ROSS, D. J. 1966.3. SoilSci. 17,1-15. Ross, D . J. 1968. Trans. 9th Inr. Congr. SoilSci., 1968 Vol. 3, pp. 299-308. Rotman, B., and Papermaster, B. W. 1966. Proc. Nat. Acad. Sci. U.S . 55,134-141. Rovira, A. D. 1956. Plant Soil 7,178- 194. Rovira, A. D. 1959. PlanrSoil 11,53-64. Rovira, A. D., and Graecen, E. L. 1965.Australian J . Agr. Res. 8,659-673. Rovira, A. D., and McDougall, B. M. 1967.In“Soil Biochemistry”(A. D. MacLarenandG. H. Petersen, eds.), pp. 417-463. Marcel Dekker, New York. Ruinen, J. 1956. Nature 177,220-22 I . Ruinen, J. 1961.PlantSoil 15,81-109. Ruinen, J. 1963.Antonie van Leeuwenhoek, J . Microbiol. Serol. 29,424-438. Ruinen, J. 1965. Plant Soil 22,375-394. Ruinen, J. 1966.Ann. Inst. Pasteur, Paris 111,342-348. Russell, E. J. 1950. “Soil Conditions and Plant Growth,” 8th ed. Longmans, Green, New York. Sadasivan, T. S. 1939.Ann. Appl. Biol. 26,497-508. Samuel, G., and Greaney, F. J. 1937. Trans. Brit. Mycol. SOC.21, I 14-1 17. Sandon, H. 1928. SoilSci. 25,107-121. Sauerbeck, D., and Fuhr, F. 1971. In “Use of Isotopes and Radiation in Soil Organic Matter Studies,” F.A.0.II.A.E.A. Tech. Meeting, 1968. (in press). Saxton, W. T. 1930. South African J . Sci. 27,323-327. Scharpenseel, H. W., and Krausse, R. 1962. 2. Pfanzenernaehr., Dueng., Bodenk. 116, 106-1 15. Scharpenseel, H . W., Tanners, M. A., and Pietig, F. 1968. Trans. 9th Int. Congr. Soil Sci., 1968 Vol. 3, pp. 577-589. Scheffer, F., and Ziechmann, W. 1967.2. PJanzenernaehr.. Dueng., Bodenk. 116,106- 1 15. Schnitzer, M., and DeLong, W. A. 1955. SoilSci.Soc. Amer., Proc. 19,363-368. Schonwalder, H. 1958.Arch. Mikrobiol. 30,162-180. Schuster, J. L. 1964. Ecology 45,63-70. Schweizer, J. 194I . Versl. Verad. Ver Proefst.-person 28,22 1-24 1. Shamoot, S., McDonald, I., and Bartholomew, W. V. 1968. Soil Sci. SOC.Amer., Proc. 32, 817-820. Shantz, H. L.,and Piemeisel, R. L. 1917.J.Agr.Res. 11, 191-246. Shields, L. M., and Durrell, L. W. 1964. Bot. Rev. 30,92-128. Shields, L. M., Mitchell, C., and Frouet, F. 1957.Amer. J . Bot. 44,489-498. Shtina, E. A. 1957. Pochvovedenie 26,12-17. Shtina, E. A. 1963.Agrobiologiya pp. 585-588. Shtina, E. A., Pankratova, E. M., Perminova, G. N., Tretjakova, A. N., and Yang, L. A. 1968. Trans. 9rh Int. Congr. Soil Sci., 1968 Vol. 2, pp. 15 1- 158. Silver, W. S., Centifanto, Y. M., and Nicholas, D. J. D. 1963. Be$ Arch. Microbiol. 28,4568. Simonart, P., and Mayaudon, J. 1966. In “Use of Isotopes in Soil Organic Matter Studies” F.A.O./I.A.E.A.Tech. Meeting 1963,pp. 245-257. Simonart, P., Batistic, L., and Mayaudon, J. 1967. PlantSoil27,153-166.

434

FRANCIS E. CLARK A N D ELDOR A. PAUL

Skujins, J. J. 1967. I n “Soil Biochemistry” (A. D. McLaren and G. H. Peterson, eds.), pp. 371-414. Marcel Dekker, New York. Smit, J., and Wieringa, K. T. 1953. Nature 171,794-795. Sorensen, L. H. 1967. SoilSci. 104,234-241. Sorensen, L. H. 1969. Experienfia 25,20-2 I . Soulides, D. A., and Clark, F. E. 1958. Soil Sci. SOC.Amer., Proc. 22,308-3 1 I . Starkey, R. L. 1958. Bacferiol. Rev. 22,154-172. Steelink, C., and Tollin, G . 1967. In “Soil Biochemistry” (A. D. MacLaren and G. H. Petersen, eds.), pp. 147-169. Marcel Dekker, New York. Stevenson, G. 1959.Ann. Bof. (London) [N.S.] 23,622-635. Stevenson, I. L. 1967. Can.J . Microbiol. 13,205-2 1 1. Stewart, B. A. 1970.J . Envir. Sci. Tech. (in press). Stewart, B. A., Porter, L. K., and Viets, F. G. 1966. Soil Sci. Soc. Amer., Proc. 30,355-358 and 453-456. Stewart, W. D. P. 1966. “Nitrogen Fixation in Plants.” Oxford Univ. PressfAtNone), London and New York. Stewart, W. D. P. 1967.Ann. Bot. (London) [N.S.] 29,229-239. Stewart, W. D. P. 1969. Proc. Roy. SOC.,Ser. B 172,367-388. Stewart, W. D. P., and Pearson, M. C. 1967. Plant Soil 26,348-360. Stockli, A. 1956. M i ff.Schweiz. Landwirt. 8, 125-130. Stone, R. W., Bechold, S. I., McAuliffe, H. D., Murdock, F. R., and Malzahn. R. C. 1943. Pennsylvania,Agr. Exp. Sta., Bull. 444. Stotzky, G. 1965. I n “Methods of Soil Analysis” (C. A. Black, ed.), pp. 1550-1572. Amer. SOC.Agron., Madison, Wisconsin. Stout, J. D. 1958. New Zealand J . Agr. Res. 1,943-957 and 974-984. Stout, J. D. 1960. New Zealand J . Agr. Res. 3,214-223 and 237-244. Stout, J. D. 1961. New Zealand J . Agr. Res. 4, 1-30. Strugger, S. 1948. Can. J. Res., Secf. C 27, 188-192. Szegi, 1. 1967. Acfa Agron. (Budapesf) 16,367-373. Tamm, E., and Krzysch, G. 1965.2. Acker- Pflanzenbau 122,209-224. Taylor, C. B. 1936. Proc. Roy. SOC.,Ser. B 119,269-295. Taylor, C. B. 1938. Soil Sci. 46,307-32 I . Taylor, C. 8. 1942.J . Hyg. 42,23. Tchan, Y. T. 1953.Proc. Linn. SOC.New South Wales 77,265-269. Tchan, Y. T. 1959.Plant Soil 10,220-23 1. Tepper, E. 2. 1963. Mikrobiologiya 32,558-563. Theron, J . J. 1951.J.Agr. Sci. 41,289-296. Theron, J . J., and Haylett, D. G.1953. Emp. J. Exp. Agr. 21,86-98. Thomas, S. B., and McQuillin, J. 1952.J . Appl. Bacteriol. 15,41. Thornton, R. H. 1958. New ZealandJ. Agr. Res. 1,922-938. Thornton, R. H. 1960. New Zealand J . Agr. Res. 3,699-7 1 1. Thornton, R. H. 1965. NewZealandJ.Agr. Res. 8,417-419. Timonin, M. 1. 1935. Can. J. Res. 13,32-46. Topping, L. E. 1937. Zentralbl. Bakferiol, Parasifenk., Infekfionskr. Hyg., Abr. 2 97,289304. Tukey, H. B., Jr., Witter, S. H.,and Tukey, H. B. 1957. Science 126, 120-121. Vandecaveye, S. C., and Katznelson, H. 1938. Soil Sci. 46,57-84. Vasantharajan, G. S.,and Bhatt, J. V. 1968. Plant Soil 28,258-267.

THE MICROFLORA OF GRASSLAND

435

Veldkamp, H. 1967. “Ecology of Soil Bacteria” (T. R. G. Gray and D. Parkinson, eds.), pp. 201-2 19. Liverpool Univ. Press, Liverpool. Vernon, T. R. 1958. New Zealand J . Agr. Res. 1, 985-990. Vincent, J. M. 1965. In “Soil Nitrogen” (W. V. Bartholomew and F. E. Clark, eds.), pp! 384-435. Amer. SOC.Agron., Madison, Wisconsin. Vlassak, K. 1966. Agricoltura 14,49-65. Volk, G . M. 1959. Agron. J. 51,746-749. Volk, G. M. 1961. Agr. Food Chem. 9,280-283. Waid, J. S. 1957. Trans. Brit. Mycol. SOC.40,391-406. Wagner, G . H., and Mutatkar, V. K. 1968. Soil Sci. SOC.Amer. 32,683-686. Waksman, S . A. 1916. Soil Sci. 2, 103-156. Waksman, S. A. 1917. Soil Sci. 3,565-589. Waksman, S . A. 1967. “The Actinomycetes.” Ronald Press, New York. Wang, H. L., and Hesseltine, C. W. 1965. Can. J. Microbiol. 11,727-732. Warcup, J. H. 1951. Trans. Brit. Mycol. SOC.34, 376-399. Warcup, J. H. 1967. In “Soil Biology” (N. A. Burges and F. Raw, eds.), pp. 51-110. Academic Press, New York. Weaver, J. E. 1947. J. Ecol. 28,221-240. Webster, J. 1956. J. Ecol. 44,517-544. Webster, J. 1957. J. Ecol. 45, 1-3. Webster, J., and Dix, N. J. 1960. Trans. Brit. Mycol. SOC.43,85-89. Werkenthin, F. C. 1916. Phytoparhology 6,241-253. Whitman, W. C., and Stevens, 0. A. 1952. North Dakota, Acad. Sci. Proc. 6,73-78. Wiegert, R. G., and Evans, F. C. 1964. Ecology 4549-63. Wiegert, R. G., and Evans, F. C. 1967. In “Secondary Productivity of Terrestrial Ecosystems” (K. Petrusewicz, ed.), pp. 3-15. Krakow, Warsaw. Will, G . M., 1968. Trans. 9th Inr. Congr. Soil Sci., 1968 Vol. 3, pp. 237-246. Williams, R. D. 1964. Ann. Bot. (London) [N.S.] 28,419-429. Winkler, W. 1899. Zentralbl. Bakteriol., Parasitenk., Infektionskr. Hyg., Abt. 2 5, 569579 and 6 17-630. Witkamp, M.,and Olson, J. S. 1963. Oikos 14,138-147. Woldendorp, J. W. 1962. Plant Soil 17,267-270. Wolff. A. 1913. Zentralbl. Bakreriol., Parasitenk., Infekrionskr. Hyg., A b t . 2 39,411-419. Wullstein, L. H., and Gilmour, C. M. 1964. Soil Sci. 97,428-430.

This Page Intentionally Left Blank

AUTHOR INDEX Numbers in italics refer to the pages on which the complete references are listed.

A

Andre, J., 8 1, I I5 Andrew, C. S., 22,23,24,28,29,30,3 1.32, 33, 54, 66, 67, 73 Andrew, M. R., 258, 266,281 Annamalai, P., 320, 326 Antoine, R., 287, 289, 294, 300,326 Apacible, A. R. A., 307, 326 Arai, N., 204, 229, 233 Ariet, M., 256, 260, 281 Arlidge, E. Z., 95, 114 Armitt, J. D., 33, 69 Armstrong. W., 249, 281 Arndt, W., 4, 1 1 , 61, 71, 317,326 Arnold, P. W., 341,370, 373 Arshad, M. A., 412,426 Asher, C. J.. 172, 173, 198, 199 Aspinall, D., 161, 165, 199 Atherton, J. G . , 9, 67 Atkins, J. G . , 3 19, 326 Atwood, S. S., 131, 157 Aubertin, G. M.,244, 245, 246, 248, 252, 283 Austin, L. G., 256, 260, 281 Axley, J. H., 171, 199

Abbott, E. V., 296, 3 15, 324,326, 328 Adams, G. A,, 225,231 Adedipe, N. O., 221, 233 Ahern, G. A., 9 , 6 7 Ahlgren. G . H.. 130, 153 Ahlrichs, J. L., 81, 82, 87, 92, 114, 115, 251,282 Ahmad, N., 54, 66 Ahmada, F., 319,329 Aikawa, J. K., 332, 355,369 Albee, L. R., 386,430 Albrechtsen, R. S., 149, 153 Alburquerque, H., 121. 127, 154 Alers-Alers, S., 307, 32 I , 329 Alexander, C. W., 129, 153 Alexander, M., 38 I , 398,404,426 Alexander, W. P., 323,326 Allcroft, R., 332, 334, 337, 344, 360, 361, 363,365, 366, 367,369,370,371 Allcroft, W. M., 339, 369 Allen, L. A., 377, 379, 426 Allen, R. J., Jr., 46, 69 Allen, R. S . , 142, 156 Allison, F. E., 419, 426 6 Allred, K. R., 140, 153 Ambastha, H. N. S., 288, 326 Andersen, A. J., 192, 198 Babiuk, L. A., 404, 405, 426 Anderson, D. M., 108, 114 Bader, K. L., 136, 137, 138, 153 Anderson, D. W., 322,326 Badoz-Lambling, J., 243, 245, 255, 281 Anderson, G., 95, 114 Bailey, F. M., 13, 66 Anderson, G. C., 332, 335,369 Bailey, G. W., 76, 99, loo, 114, 116 Anderson, G. D., 16, 17, 71 Baker, D. L., 409,429 Anderson, G. R., 42 I , 426 Baker, F. H., 335,373 Anderson, G . W.,139, 157 Baker, J. H., 279, 283 Anderson, J., 121, 153 Baker, L. A. E., 378,426 Anderson, R. S.,322,326 Balch, C. C., 364,372 Anderson, S. R., 124, 136, 137, 138, 148, Baldridge, J. D., 134, 156 153 Balloni, W., 403, 426 Anderssen, K. B., 45, 70 Ballou, G. A., 209, 224,231 Anderssen, R. A., 196, 198 Balls, W., 3 17, 326 431

438

AUTHOR INDEX

Bansal, R. D., 145, 150,153 Barake, N., 78, 115 Barber, S. A., 128,154, 170, 171, 180,198, 199, 201 Barker, J. R., 337,370 Barley, K. P., 161, 162, 166, 167, 171, 172, 175, 176, 178, 181,198, 199 Barnard, C., 8,44,47,66 Barnes, A. C., 292, 308, 321, 323,326 Barnes, B. T., 163, 198 Barnes, R. J., 352, 358,371 Barr, D. J. S., 135, 153 Barrer, R. M., 79, 114 Barrow, N. J., 169, 198 Barskaya, T. A., 216,231 Bartholomew, W. V., 384,433 Barua, D. N., 167, 198 Bashaw, E. C., 58, 59,66,67 Bateman, J. B., 223, 234 Batistic, L., 409, 433 Battell, M. L., 21 1, 212, 231 Baver, L. D., 298, 306, 308, 309,322,326 Bayer, G., 133, 139, 154 Bayliss-Elliot, J. S., 401, 426 Beachell, H.M., 287, 319, 320,326, 327 Beall, L. B., 8, 70 Bear, F. E.,341,370, 372 Beard, W. E., 425,427, 428 Bechet, R., 3 16,329 Bechold, S. I., 379,434 Beck, G. E., 205,233 Beeson, K. C., 340,370 Beeson, W. M., 140, 156 Beetle, A. A., 387, 426 Begg, J. E., 49, 71, 167, 200 Beijerinck, M. W., 376,426 Bell, D. S., 141, 157 Bello, J., 210, 231 Benne, E. J., 142, 155 Bent, F. C., 146, 153 Bernal, J. D., 207, 208,231 Berrill, F. W., 304, 326 Bertram, J., 142, 156 Best, K. F., 416,427 Bhatt, J. V., 417,434 Bhoj, R. L., 306, 3 11,326 Biederbeck, V. O., 409, 414,426, 432 Biemann, K., 110,116

Bienfet, V., 357, 372 Billing, E., 378, 426 Bingham, F. T., 95, 114 Birch, H. F., 49, 53,67, 68 Birkle, D. E., 241, 253,257, 268, 269, 270 27 1,273, 274, 280,281 Bisby, G. D., 400,426 Bissada, K., 87, 114 Bisset, W. J., 44, 72 Bizzell, J. A., 422, 423, 431 Black, J. D. F., 246, 247, 249, 251, 258, 279, 281, 283 Blair, D. N., 226,232 Blakely, E. R.,409,430 Blakemore, F., 346, 370 Blaser, R. A., 161, 198 Blaxter, K. L., 359, 360,370 Bleasdale, J. K. A., 301,326 Blydenstein, J., 57, 67 Bockholt, A. J., 316, 324,326 Bockris, J. O’M., 252, 281 Bogdan, A. V., 5, 6, 16, 17, 44, 45, 46, 47, 48, 49, 67, 68 Bohman, V. R., 339, 358,370,373 Bond,G.,417,419,422,426,427, 430 Bonner, I. A., 18, 70 Bonner, J., 187, 198 Bor, N. L., 50,67 Borzhkovskaya, G. D., 216,231 Botner, P., 410,427 Boughey, A. S., 422,423,427 Bould, C., 340,370 Bouldin, D. R.,179, 198 Bourget, S. J., 270,282 Bowen, G. D., 26,67, 167, 175, 176, 177, 198 Bowyer, J. W., 9.67 Boyer, P. D., 209, 224,231 Boyle, A. J., 50, 67 Bradfield, R., 288,326 Braithwaite, B. M.,9, 67 Brandenburg. N. G., 145, 151,153 Brar, S. S., 423,427 Brassfield, T. S., 2 13,233 Bray, R. A., 22,44,67 Bray, R. H., 159, 198 Bremner,J. M., 111, 112,114,414,427 Brewbaker, J. L., 43.69

AUTHOR INDEX Brierley, W. B., 400, 427 Briggs, R. A., 130, 153 Brindley, G . W., 90, 98, 115 Bristol Roach, B. M., 402,427 Britten, E. J., 35, 67 Broadbent, F. E., 414,424,427 Brodie, H. W., 298, 306, 308, 309,322,326 Bromdep, A,, 40, 72 Brouwer, E., 350,370 Brouwer, R., 161, 174, 185, 198 Brown, A. H. F., 380,427 Brown, C. H., 317,326 Brown, C. S., 125, 130, 153 Brown, D. A., 168,200 Brown, J. C., 171, 199 Brown, M. F., 42 1,431 Brown, P. E., 400, 427 Brown, R., 176, 198, 256, 259, 260, 283 Brown, R. L., 108,114 Brownell, J. R., 340, 371 Bruce, R. C., 21.67 Brummer, K., 79, 114 Brunaugh, H. L., 22 1,232 Bruner, W. E., 166, 201 Brunk, R. E., I3 I , 157 Brustkern, P., 228,232 Bryan, W. W., 7, 14, 16, 19,22, 28,46, 48, 5 5 , 64, 66, 67 Brydon, J. E., 412, 427 Bubar, J. S., 124, 144, 145, 150, 153, 154, 156 Buchanan, H. S., 246, 247, 249, 258, 281 Buckovic, R. G., 124,154 Budd, A. C., 416,427 Bullen, M. R., 147, 155 Bullock, H. C., 176, 199 Buol, S. W., 108, 114 Burau. R. G., 351, 352, 353, 358,370,373 Burger, A. W., 302, 326 Burger, 0. J., 131, 142, 154, 157 Burges, A., 386, 388, 389, 415, 416, 427, 429,430 Burkey, L. A., 377, 378, 379, 390,430 Burns, G. R., 171, 199 Bums, J. C., 352,370 Bums, K. N., 332,334,337,344,360,361, 363, 365, 366,369,370 Burns, R. C., 416,420,429

439

Burri, R., 376, 377,427 Burris, R. H., 416,422,427 Burstrom, H. G., 384,427 Burt, A. W. A., 351, 358,370 Burton, G. W., 5 , 46, 23, 54, 58, 59,67, 72 Butler, A. M., 369,371 Butler, E. J., 334, 336, 344, 345, 346, 370 Butler, G. W., 332, 344, 350,357,362,367, 370,374 Butler, J. A., 410, 430 Butler, .I.E., 387,428 Butterworth, M. H., 51, 67 Buxo, D. A., 3 14,327 Buzzell, R. I., 123, 144, 149, 154 Byrne, G. F., 167,200 Byth, D. E., 39, 67 C

Caddell, J. L., 368, 370 Cairney, 1. M., 335, 370 Calicis, B., 88, 96, 97, 106, 114, 115 Callen, E. O., 120, 135, 153, 154 Calvert, 0. H., 134, 156 Calvet, R., 106, 114 Calzada, B. J., 310, 317, 326 Camargo, A., 57,67 Cameron, D. F., 9, 37, 40, 41.67 Cameron, D. G., 22,68 Cameron, R. E., 402, 422,427, 428 Camp, B. J., 358,370 Camp, T. R., 404,406,427 Campbell, C. A,, 270, 279, 281, 412, 42 I , 427 Campbell, W. F., 302,326 Cano, I. B., 315,329 Care, A. D., 332, 360, 361, 362,370 Carlisle, A., 380, 427 Caro-Costas, R., 56,68 Carrero, J. O., 181, 199 Carslaw, H. S., 177, 179, 183, 195, 198 Carter, R. L., 53, 67 Cartwnght, P. M., 176, 198 Cary, J. W.,227, 233 Casida, L. E., Jr., 406, 427 Cathey, N. N., 221,231 Centifanto, Y. M., 417, 433

440

AUTHOR INDEX

Cessac, G. L., 214,233 Chamblee, D. S., 127, 129, 130, 153, 154 Champion, R. A,, 181, 198 Chandra Mohan, J., 320,326 Chandrasekharan, P., 49, 70 Chang, C.-Y., 225,231 Chang, H., 288, 31 I , 321,326 Chapman, D., 219,231 Chapman, F. M., 161, 165, 199 Chapman, L. S., 3 12,326 Charlot, G., 243, 245, 255, 281 Chase, F. E., 424,427 Chaussidon, J., 80, 81, 87, 106, 114, 116 Chen, C. B., 3 14,326 Cheng, F. S., 87, 114 Chesire, M. V., 414, 427 Chevrette, J. E., 131, 154 Chiang, C., 288, 327 Chichester, F. W. 412,427 Chinlay, T., 3 1 I , 326 Chippindall, L. K. A., 46, 49, 50, 68 Christensen, R. B., 125, 154 Christian, C. S., 21, 45, 68 Christiansen, M. N., 229, 231 Christophersen, J., 2 19, 232 Chu, H. T., 3 IS, 326, 327, 330 Chubnall, A. C., 380,429 Clapp, C. E., 103, 114 Claringbold, P. J., 38, 50, 68 Clark, F. E., 378, 379, 381, 397, 403, 422, 424,425,427, 428, 434 Clark, R. B.. 352,370 Clarke. A. L., 175, 178, 198 Cloos, P., 87, 88, 96, 97, 106, 114, 115 Coaldrake, J. E., 4, 65, 68 Cochrane, V. W., 415,428 Cockroft, B., 171, 198, 267, 282 Coleman, N . T., 105, 115 Coleman, R. G., 35, 68 Collie, T. W., 332, 343, 344, 345, 346, 348, 349, 351, 367,372 Collins, J. L., 287, 294, 300,312, 316, 318, 326 Colom, J., 1 1 I , 114 Coltman, M., 334, 371 Comyn, R. H., 256, 260,281 Connell, J., 367,371 Conrad, J. H.,53, 5 5 , 68, 69 Conway. A., 344,373

Conway, B. E., 252, 281 Cooke, A. R.,35,68 Cooper, C. S., 126, 138, 142, 154 Cooper, J. P., 293,326 Coote, J. N., 36, 70 Corke, C. T., 424,427 Cormack, R. G. H., 165, 198 Corms, W. G., 138,154 Cornelius, C. E., 335, 371 Cornforth, 1. S., 183, 185, 198 Correia, C. A. da Paixiio, 288,326 Court, R. D., 34, 69 Courter, J. W., 296, 326 Covey, W., 240,281 Cowan, 1. R., 189, 198 Cox, C. S., 215,232 Cragg, J. B., 404,430 Cranwell, P. A., 414, 427 Creacen, E. L., 171, 198 Creek, M. J., 5 5 , 57. 71 Crist, J. W., 192, 201 Crookshank, H. R., 336, 338,370, 373 Crosby. F. L., 204,232 Cross, R. F., 34, 35, 71 Crosse. J . E., 378, 379, 428 C ~ U ZM., , 83, 88, 100. 106, I16 Cunningham, 1. J., 339.370 Curtis, J. T., 398, 400, 401, 432 D

Dahlman, R. C., 387, 388, 395, 428 Dalbro. S., 380, 38 I , 428 Daniell, J. W., 204, 232 Daniels, L., 390, 431 Datta, A., 2 16,234 Daubenmire, R. F., 387,428 Davey, L. A., 367,370 Davey, T. B., 29 I , 326 Davies, J. G . , 2, 4, 7, 8, 20, 23, 28, 50, 68, 70 Davies, P., 236, 281 Davies, W., 3,68 Davis, B. L., 226, 232 Davis, C. E., 54,66 Davis, G. K., 332, 356, 370 Davis, M. E., 334,372 Davis, R. E., 336, 337,370

44 1

AUTHOR INDEX

Davis, R. L., 143, 144, 147, 149, 153, 157 Davis, R. R., 131, 140, 141, 154, 156, 157 Dawson,C. D. R., 143, 144, 154 Dawson, J. E., 109, 116 Dawson, P. S. S., 409,430 Decker, A. M., 131, 154 Degens, E. T., 110, 114 De Geus, J. G., 307, 308, 310,326 de Groot, T.. 339,344,345,346,370 Deijs, W. G . , 348, 349,350,357,362,371, 372 Dekking, H. G. G., 104,114 Delahay, P., 244, 282 DeLong, W. A.. 380,433 Delwiche, C. C., 419,428 Demetriadi, M. A., 3 17,327 de Nettancourt, D., 147, 154, 155 Dennis, J., 356, 371 Derrnott. W., 332, 370 de Rothschild, H.,121, 154 Deschuytener. G., 63, 68 de Sornay, A., 287, 294,327 De Vries, A. L., 222,232 De Wit, C. T., 353,371 Dick, T., 322, 327 Dickinson, D. B., 225,232 Dijkrnan, M. J., 18,68 Dijkshoorn, W., 347, 348, 353,371 Di Menna, M. E., 377, 378, 379,421,428 Dirven, J. G. P., 46, 68 Dittmer, H.J., 165, 167, 198 Divakaran, K., 288, 327 Dix, N. J., 389, 435 Dix, R. L., 387,428 Djuricic, M. V., 110, 116 Dobson, A., 361,371 Dodds, E. C., 340,372 Doebbler, G. F., 2 17, 223, 224, 232 Doeksen, J., 390,428 Dollahite, J. W., 358, 370 Domsch, K. H.,405,428 Donaldson, L. E., 34, 69 Doner, H.E., 90, 92, 93, 99, 114 Donovan, L. S.. 144, 154 Dorofaeff, F. D., 347, 372 Dorokhova, N . A., 419,428 Dorsey, N . E., 226, 232 Dougall, H.W., 49, 68 Dougherty, R. W., 125, 154

Dow, B. K., 270,282 Dowdy, R. H., 87, 114 Downes, R. W., 302,327 Doxtader, K. G., 377, 407, 428 Drake, A. D., 24, 69 Drake, C. R.,133, 135, 154, 156 Draper, A. D., 149, 154 Drury, R. E., 225,232 Dubach, P., 409,428 Duell, R. W., 130, 139, 154 Duggeli, M., 376, 377, 378,428 Duncan, W. G., 160,198 Durrell, L. W., 402, 403, 422,428, 433 Du Toit, J. L., 307,327 Dutta, K. N., 167, 198 Dwivedi, R. S., 40 1,428, 432 Dycus, A. M., 222,232 Dye, D. W., 378,428 Dyksterhuis, E. J., 387, 432 E

Eagle, D. J., 332,370 Edgerton, C. W., 296, 298, 327 Edlin, H. L., 289, 300, 327 Edwards,A. P., 111, 112,114 Edwards, D. C., 5 , 46, 48, 68 Edwards, K. J., 293, 326 Edwards, W. M., 186, 198 Edye, L. A., 17, 37, 42, 43, 45, 50, 62, 65, 68 Efferson, J., 287,327 Ehlers, W., 107, 114 Ehlig, C. F., 189, 199 Elich, T. W., 48, 72 Ellern, S. J., 300, 313, 314, 317, 318, 324, 327 Elliot, H. G., 121, 154 Elliott, F. C., 145, 154 Ellis, F. B., 163, 194, 198, 200 Ellis, W., 120, 154 Embleton, T. W., 341, 363, 371 Emmel, M. W., 34,68 England, L. M., 400, 428 Epstein, E., 172, 199 Erdman, L. W., 129, 154 Erickson, A. E., 236, 240, 241, 243, 247. 257, 27 I , 274, 278,282,283

AUTHOR INDEX

442

Eriksson, E., 419,428 Ernst, R.,2 19, 222,232 Erwin, D. C., 30 1, 3 16,330 Eslick R. F., 142, 154 Estermann, E. F., 412, 428 Evans, F. C., 386, 388,435 Evans, H., 166,199 Evans, J., 56,68 Evans, L. F., 226,232 Evans, L. J. C., 300, 301,327 Evans, T. R.,28, 56, 64,67,68 Evans, U. R., 260,282 Evatt, N. S., 287, 290, 300, 301, 319, 320, 326,327 Evenson, J. P., 288, 291, 300, 304, 310, 311, 312, 313, 314, 316, 317, 324,327 Eyles, A. G., 4, 7, 8, 28, 68 F

Fairley, J. L., 222, 232 Falkmer, S., 2 I I , 234 Falshaw, C. P., 414, 427 Farmer, V. C., 77, 82, 84, 85, 86, 87, 88, 89, 91, 92, 97, 102. 114, 116, 117 Farmer, W. J., 82, 87, 92, 107, 114 Farnsworth, R. B., 417,428 Farrell, D. A., 166, 199, 267, 282 Fedorov, M. F., 421, 428 Fehrenbacker, J. B., 186, 198 Felbeck. G. T., 341,371 Fergus, 1. F., 23,69 Ferguson, W. S., 377, 379,426 Fertig, S. N., 132, 133, 139, 154 Fewkes, D. W., 3 14,327 Fiddaman, D. K., 172,199 Field, A. C., 339,373 Figarella, J., 54, 56, 68, 73 Finch, P., 104, 114 Finn, B. J., 270,282 Fitzpatrick, E. G., 162.199 Flaig, W., 410, 414,428 Flanagan, T. R., 132, 139,154 Fleming, A. L., 17 1, 199 Flemons, K. F., 45,68 Flink, E. B., 369,371 Flint, E. A,, 403,428 Floyd, A. J., 414,427

Fly. C. L., 387,432 Folkes, B. F., 253, 265, 271,283 Folkins, L. P., 131, 154 Fontenot, J. P., 367,371 Foot, A. S., 367,371 Forbes, I., 58,67 Ford, G. W., 409, 412,428,429 Ford, R. E., 135, 154 Forrest, J. A., 267,282 Fortman, H. R., 131, 157 Foster, W. N . M., 191, 192, 193, 200 Foy,C. D., 128, 154, 171,199 Fraser, J., 33, 69 Fred, E. B., 406,428 Frederick, S. E., 220,232 French, M. H.,50,68 Frens, A. M., 344,371 Frere, M. H., 171, 183, 199, 200 Freyre, R. H., 35,68 Fribourg, H. A., 127, 154 Fripiat, J. J., 80, 81, 87, 88, 94, 96, 97, 106, 107,114, 115, 116 Frissel, M. J., 95, 115 Frodyma, M. M., 35,67 Frouet, F., 422, 433 Fuhr, F., 410,433 Fuller, W. H., 402, 417, 420, 422, 427, 428, 431 Fulton, H. J., 301, 317,329 G

Gaff, D. F., 2 1 1,232 Galstyan, A. S . , 109, 115, 414, 428 Garns, W., 381,382,429 Ganguly, B. D.. 288, 327 Garber, M. J., 301, 316,330 Garcia D u m , E., 287, 288, 300,327 Gardner, A. L., 121, 127, 154 Gardner, C. A., 121, 154 Gardner, L. I., 369,371 Gardner, W. R., 179, 182, 189, 195, 199, 424,425,429 Garrett, W. G., 80, 116 Gates, C. T., 23, 38, 50, 68, 72, 161, 199 Gausman, H. W., 139, 154 Gautheir, F. M., 131. 154 Gavande, S. A., 270,'282

443

AUTHOR INDEX

Gay, B., 287, 319,327 Geelen, M.J. H., 340,371 Gehenio, P. M.,206, 233 Gentry, H. S., 289,327 Gerdemann, J. W.,134, 155 Gerken, H. J., Jr., 367,371 Gerloff, E. D., 219,232 Germain, R., 2,68 Gershon, D., 146, 148,154 Gershoy, A,, 127, 156 Geurink, J. H., 350, 366,371, 372 Gfeller, F., 224, 234 Ghajar, B. M.,225,232 Gibson, A. H., 26.68 Gibson, T., 397,429 Giddens. J., 423, 427 Giddings, J. C., 163, 199 Gilbert, E. N., 187, 199 Gilbert, G. A., 367,370 Gildenhuys, P. J., 58,68 Gile, P. L., 181, 199 Giles, W.L., 123, 124, 145, 146, 154 Gill, M. S., 321, 327 Gill, W.R., 267, 268, 282 Gilmour, C. M., 425,435 Gist, G. R., 126, 155 Glaeser, R., 87, 115 Glavez, G. E., 315,327 Godbey, E. G., 56, 72 Golley, F. B., 376,429 Gornide, J. A., 53, 5 5 , 68, 69 Gonzales-Tejera, E., 32 I , 329 Gonzalez, V., 43,69 Govindasamy, K. N., 288,327 Gowing, D. P., 318,327 Grable, A. R.,235,282, 420, 422,432 Gradwell, M. W.,253, 270, 27 1, 279, 282 Graecen, E. L., 412, 433 GraR, O., 390,429 Graham. D. C., 378, 379.429 Graham, J. H., 135, 155 Graham, V. W.,332, 343, 344, 345, 346, 348, 349, 351, 367,372 Graham-Bryce, 1. J., 159,200 Grainger, R. B., 335,373 Grant, P. M.,409,412,429 Grant, W.F., 120, 125, 143, 144, 146, 147, 154,155,157

Grant, W.H., 214,233 Grasrnanis, V. 0. G., 176, 199 Graves, D. J., 216,232 Gray, S. G., 17, 18, 41, 42, 43,69, 70 Greacen, E. L., 166, 199, 267,282 Greaney, F. J., 391,433 Green, D. E., 216, 219,232 Green, D. G., 270, 279,281 Green, H. H., 339,369 Greenhill, A. W.,380,429 Greenland, D. J., 76, 78, 93, 94, 96, 97, 98, 111, 115, 116, 409, 412, 425, 428, 42 9 Greenshields, J. E. R., 131, 154 Griffiths, J., 4, 69 Grfiths, T. W.,365,371 Grist, D. H., 288, 327 Grof, B.,45, 46, 48, 56.69 Grossenbacher, K., 110, 115 Grove, H. V., 121, 127,156 Gruber, T., 377,429 Grunes, D. L., 332,339,347,348,349,358, 364,370,371, 372 Guegen, L., 344, 348,372 Guerassimoff, J., 43, 70 Guinn, C., 216, 229, 231,234 Guppy, J. C., 138,155 Gupta, P. S., 288,327 Gurney, E. H., 125, 155 Gyllenberg, H., 381,429 Gyrisco, G. G., 135, 136,156 H

Hacker, J. B., 49, 58,69 Hackett, C., 161, 199 Hagen, C. E., 172,199 Hagihara, B., 415, 429 Hagin, R. D., 132,155 Haider, K., 410, 41 1,431 Hale, R. P., 196, 198 Halick, J. V., 3 19,326 Hall, T. C., 205,233 Hamill, D. E., 43,69 Hamilton, R. I., 33, 34,69 Hammond, M. W.,417,428 Handreck, K. A., 174,199 Hansen, E. H., 409,429

444

AUTHOR INDEX

Hansen, H. W., 124, 155 Hanson, C. L., 386,432 Hanson, D. M., 216, 219,232. Harbaugh, F. G., 356,371 Harding, G. D., 339, 358,370 Hardisson, C., 410,429 Hardy, R. W. F., 416, 420,429 Hare, C. J., 31 1,327 Harland, S. C., 300, 317,327 Harmon, S. A., 225, 232 Harmsen, G. W., 383,424,429 Harney, P. M., 144, 147, 155 Harper, J. L., 186, 199 Harris, J. R., 391, 429 Harrison, C. M., 178, 199 Harrison, J., 377, 379, 426 Hart, R. H., 59,67 Harter, R. D., 81, 115. 251, 282 Hartley, W., 7, 1 I , 69 Harvey, C., 195,199 Harward, M. E., 87, I15 Hasaka, E. Y., 121, I55 Haseman, M., 2 1 I , 233 Hashioka, Y., 288, 327 Hassid, W. Z., 220,232 Hastings, A., 63, 69 Haug, N. F., 63,69 Havoundjian, Z. S., 109, I IS, 4 14,428 Haworth, R. D., 414,427 Haydock, K. P., 38, 43, 48, 50, 51, 5 5 , 56, 68, 69, 70, 72, 73 Hayes, M. H. B., 104,114, 393, 429 Haylett, D. G., 422, 423, 434 Head, G. C., 176,199 Heal, 0. W., 404,430 Heald, W. R.,341,371 Heard. A. J., 409,429 Heber, U. W., 21 I , 216,219,222,223,231, 232 Hebert, L. P., 302, 323,327 Hegarty, M. P., 32, 34, 35, 43, 66, 69 Heinrichs, D. H., 43. 69 Hellreigel, H., 381, 429 Hemingway, R. G., 367, 368,371,372 Hemkes, 0. J., 348, 349, 362,372 Hendricks, S. B., 353, 371 Hendriks, H. J., 340,371 Henke, L. A., 34, 73 Henke, S. L., 34, 35, 71

Henry, T. A., 125,155 Henson, P. R., 127, 133, 139, 149, 150, 155, 156 Henzell, E. F., 23, 34, 45, 46, 54, 5 5 , 57, 68, 69, 70, 73 Herd, R. P., 332, 334, 371 Hernandez, A. R., 58, 72 Hernandez, S. C., 422,429 Hervey, R. J., 378, 379,428 Hesseltine, C. W., 415, 435 Hiatt, A. J., 353, 371 Higashiyama, H., 322,327 Hilborn, M. T., 221,232 Hill, D. H., 63, 71 Hill, D. L., 53, 5 5 , 68, 69 Hillyer, 1. G., 22 I , 232 Himes, F. L., 412,431 Hirst, E. L., 352, 358, 371 Hirst, G. B., 63, 69 Hitchon, K., 409, 429 Hjerpe, C. A , , 332,335,340,359,360,371 Ho, F. W., 298, 306, 307, 321, 329 Hoare, J. P., 244, 249, 266, 282 Hodges, E. M., 46, 69 Hodgkiss, W., 378, 379, 429 Hodgson, G. W.,409, 429 Hoffman, D. J., 103, 114 Hoffmann, G., 406,429 Hofmann, E., 406, 429 Hogg, B. M., 388,429 Hogg, P. G., 64, 69 Holder, J. M.,57, 69 Holsten, R. D., 416, 420, 429 Holt, E. C., 58, 67 Hooker, W. J., 229,232 Hopkins, H., 386, 429 Horvath, D. J., 332, 335, 342, 359, 363, 366,369, 371 Hosaka, E. Y., 14, 69 Hoveland, C. S., 49, 58,69 Howard, G . J., 103, 115 Howell, H. B., 128, 155 Hsieh, C. F., 288,327 Hsieh, S. C., 288,327 Hsu, S. C., 315,326, 327 Hudson, H. J., 388, 389, 429 Hudson, J . P., 305, 327 Hudson, M. A., 206,228,232

445

AUTHOR INDEX

Hughes, C. G., 296, 300, 307, 308, 309, 311,321, 322,323,328 Hughes, H. D., 128, 141, 155, 157 Hughes, J. P., 335, 371 Humbert, R. P., 292, 309, 3 I I , 3 15, 327 Humphnes, E. C., 185,199 Humphreys, L. R., 9, 37, 45, 47,68, 6 9 Hung, S. L., 323, 328 Hung, T. H., 3 14.326 Hunt, D. J., 128, 155 Hunt, J. P., 81, 115 Hurst, H. M., 415, 416,429 Huss, H.,376, 378, 429 Hutcheson, W. L., 163,200 Hutton, E. M.. 5 , 6, 7, 8, 12, 18, 35, 36, 38, 39, 41, 42, 43, 68, 69, 70, 73 1

Idle, D. B., 206, 232 Ilina, T. K., 42 I , 428 Immink, H. J., 350, 371 Ingalls, J. R., 142, 155 Inglis, J. S. S., 334, 371 Inoue. T., 110, 115, 116 Interrante, L. V., 244, 246, 248, 249, 258, 283 Isaacs, R., 32 I . 328 Isely, D., 120, 155 Ishizaki, S. M., 288, 298, 300, 301, 303, 306, 309, 3 13,3 18, 320,329 Iso, E., 288,328 Iswaran, V., 391, 429 Ivanova, T. I., 216, 232 Iwanaga, I. I., 34, 73 Izuno, T., 288, 298, 300, 301, 303, 306, 309, 313, 318, 320,329 J

Jackman, R. H., 424,432 Jackobs, J. A., 130, 139, 156 Jackson, E. K., 416, 420,429 Jackson, R. D., 159, 200 Jackson, R. M., 404, 429 Jackson, W. A,, 181,200 Jacobson, N. L., 142,156 Jacques, W. A., 175, 199 Jaeger, J. C., 177, 179, 183, 195,198

Jagannadha Rao, E., 306, 307, 309,328 Jager, G., 383,429 Jaleel Ahmed, N., 315, 328 James, D. W., 87, 115 James, N., 378, 400,426, 430 Jamieson, N. D., 337,373 Jean, F. C., 192,201 Jelli, A., 81, 115 Jencks, E. M., 332, 335,369 Jenkinson, D. S., 4 12, 4 14, 430 Jenny, H., 110,115, 171,199 Jensen, H. L., 397,400, 404, 421,430 Jha, 1. B., 288,326 Joffe, A., 125,155 Johansson, N., 210,233 John, R. P., 403,430 Johns, W. D., 87, 90, 114, 115 Johnson, 1. J., 123, 124, 145, 156, 157 Johnson, M. O., 287, 300, 304,328 Jones, E. C., 352, 358,371 Jones, F., 258, 266,281 Jones, G. F., 139, 1 5 7 Jones, J. D., 369,371 Jones, L. G., 128, 130, 137, 138, 155, 156 Jones, L. H. P., 174,199 Jones, R. J., 7, 9, 20, 23, 36, 49, 69, 70 Jones, R. K., 32, 70 Jung, G. A., 224,232 Junge, C. E., 419,430 K

Kaack, K., 278, 279,280,282 Kafkafi, U., 169,199 Kai-Chu, Y., 288,328 Kainski, J. M., 133, I55 Kaku, S., 228,232 Kamprath, E. J., 341,372 Kao, S . , 288, 327 Kaper, J. M., 217,232 Karlovsky, J., 347,372 Katznelson, H., 381, 397, 398, 406, 430, 434 Kautsky, J., 172, 199 Kauzmann, W., 206, 207, 208, 209, 210, 232 Kawin, B., 130, 157 K a y , J., 86, I15 Keddie, R. M., 379,430

446

AUTHOR INDEX

Keim, W. F., 143, 144, 146, 147, 148, 149, 153, 155, 156, 157 Kelly, J. M., 386, 387,430 Keltz, A., 219, 225: 226, 232 Kemp, A., 332, 333, 343, 344, 345. 347, 348, 349, 350, 357, 361, 362, 366, 371, 3 72 Kemper, W. D., 159,200 Kendall, M. G., 162, 199 Kendall, W. A., 142, 155 Kendrick, W. B., 388,430 Kennedy, G. S., 358,372 Kennedy, M. M., 26,67 Kennedy, W. K., 140, 142,156, 157 Kenny, T. J., 151,156 Kentzer, T., 22 I , 233 Kerr, H. D., 132, 155 Kerr, J., 365,372 Kessler, B., 222,232 Khrabrova, M.A., 2 16,231 Kiers, H. J., 17, 37, 42, 68 Kiesselbach, J. A., 166, 199 Killinger, G. B., 46, 69, 128, 157 Kim, J. T., 82, 115 Kim, K. S., 217,234 King, H. E., 293,328 King, K. M.,168,200 King, N. J., 292, 296, 300, 307, 308, 309, 31 1, 321, 322,323,328 Kingsbury, C. H., 127,155 Kijne, J. W.. 86, 115 Kitaura, K., 226,227,232 Klein, D. A., 406, 427 Kleist, H., 414, 430 Klingman, D.L., 132,155 Klosterman, E. W., 141, 154, 157 Klostermeyer, E. C., 135, 155 Klute, A., 159, 199 Kluvers, E., 350, 357,372 Knight, B. A. G., 99, 115 Knight, E., Jr., 416, 429 Kodama,H.,95,96, 110, 111,115, 116 Kokkonen, P., 175, 199 Kolthoff, I. M., 243, 244, 249, 282 Kononova, M. M., 76,115, 409,430 Kozawa, A., 244,252, 258,282 Kramer, P. J., 176, 199, 222, 233 Krasavtsev, 0. A., 220,234 Krasil'nikov, N. A., 404,430

Kratzing, C. C., 35, 70 Krausse, R., 414,433 Krehl, W. A., 369,372 Kreitlow, K. W., 133, 134, 155, 156 Kretschmer, A. E., 11, 13, 70 Kretschmer, A. E., Jr., 46, 69 Kreutzer, W. A., 391,430 Knshnaswamy, N., 49, 70 Kristensen, K. J., 238, 239, 240, 241, 247, 253, 254, 256, 257, 259, 261, 263, 268, 269, 270, 272, 273, 274, 278, 279, 280, 282 Kroulik, J. T., 377, 378, 379, 390, 430 Krukovsky. V. N., 140, 142, 155, 157 Krull, E. J ., 2 1 1,232 Krzysch, G., 405,430, 434 Kubota, J., 347, 371 Kucera, C. L., 387, 388, 395,428 Kucera, E., 163,199 Kudrina, E. A., 415,430 Kuile, C. H. H., 109,116 Kuiper, P. J. C., 216, 219, 221, 222, 224 225,232, 233 Kuntz, I. D., Jr., 213, 233 Kunze, G . W., 81, 92, 95, 105, 115, 116 Kuraishi, S., 204, 229, 233 Kurtz, W. G. W., 409,430 Kuster, E., 398,430 Kutschera, L., 165, 166,199 Kutzner, H. J., 398, 410,430 Kyneur, G. W.,17, 70 1

Laby, R. H., 94, 96.97, 98,115 Ladd, J. N., 410,430 Ladna, K. R., 288,328 Lagan, F. B., 288,328 Lailach, G. E., 90, 98. 115 Laitinen, H. A., 244, 255, 261, 265,282 Lid, S. P., 388,430 Lambourne, L. J., 34,69 Lang, A. R. G., 162,199,200 Langer, R. H. M., 301, 302, 303, 3 11,328 Langridge, J., 210, 229,233 Langston, R., 126, 156 Lanuza, E. A., 306,328 Larvor, P., 344, 348,372 Last, F. T., 378,430 Lathwell, D. J., 353, 371

447

AUTHOR INDEX

Latter, P. M., 404, 415, 416,427,429,430 Lauchli, A., 175, 180, 199 Lavy, T. C., 170,199 Law, G. D., 213,233 Law, J. P., Jr., 95, 105, I15 Lawes, J. B., 422,430 Lawrence, D. B., 4 19,430 Lawrence, G. H. M., 286,328 Lawson, N. C., 150, 154 Lawton, K.,130,157 Lazar, V. A., 347,371 Lazenby, A., 302,328 Leben, C., 377,319,430 LeClerg, E. L., 300, 430 Le Croy, W. C., 302, 303,328 Ledoux, R. L., 92, 115 Lee, J. B. S., 293, 294,328 Lee, S. M., 3 15,328 Leech, F. B., 334,372 Leeper, G . W., 180,200 Leeper, R. W., 3 18, 327 Le Febre, V., 214,233 Leffel, E. C., 335,372 Le Grand, F., 302, 303, 328 Leidenfrost, E., 159, 200 Lemon, E. R., 236, 238, 239, 240, 241, 243, 247, 259, 268, 271, 273, 274, 277, 281, 282, 283 Leonard, A., 115 Leopold, A. C., 204, 229,233 Lepage, M., 219,234 Lesperance, A. L., 339, 358,370,373 L'Estrange, J. L., 357,372 Letey, J., 105, 107, 114, 116, 159, 199, 238, 239, 240, 241, 243, 244, 245, 246 247, 248, 252, 253, 251, 267, 268, 269, 270, 271, 273, 274, 278, 279, 280, 281, 281, 282, 283 Levitt, J., 205, 206. 207, 210, 21 I , 212, 2 13, 220,233 Levy, E. B., 121, 155 Lewin, R. A., 402,430 Lewis, C. T., 286, 328 Lewis, D. G., 180, 199 Lewis, J. K., 386, 430 Lindqvist, I., 410, 430 Lingane, J . J., 243, 244, 246, 247, 249, 256, 260, 275, 276,282 Linscott, D. L., 132. 155

Lippincott, E. R., 214,233 Litav, M., 186,199 Little, D. A., 49, 70 Little, I. P., 38, 68 Liu, H. P., 315, 328 Liu, Y. T., 315,326 Livne, A., 224,233 Lloyd, M. K., 367,371 Lochhead, A. G., 381, 391,430 Locke, L. F., 386,432 Loen, E. A., 161,198 Lofgreen, G. P., 142, 155 Loh, C. S., 296, 315,328 Lohnis, F., 377, 414,430 Lomba, F., 357,372 Loneragan, J. F., 172. 173. 178, 198, 199, 201 Long, W. G., 380,431 Loosli, J. K.,140, 142, 155, 157 Lopez, M. E., 3 15,328 Louis, S., 57, 67 Lousse, A., 357,372 Loutit, J. S., 397, 431 Loutit, M., 397,431 Loveday, J., 278,282 Lovelock, J. E., 219, 225, 226, 232, 233 Low, A., 301, 305,328 Lowe, C. C., 131,157 Lowe, L. E., 412,426,431 Lowrey, R. S., 59, 67, 348, 349, 364, 372 Luck, J. M., 209,224,231 Luck, P. E., 20, 70 Ludecke, T. E., 365,372 Ludlow, M. M.,39, 54, 70 Luin, F. G., 209, 224,231 Lulham, A., 23, 73 Lund, F. T., 35,67 Lund, J. W. G., 402,431 Lunt, 0. R., 279,283 Luyet, B. J., 206, 233 Lynch, D. L., 129,155 Lyon, T. L., 383,422,423,431 Lyons, J. M., 229,233 M

McAllan, A. B., 357, 373 McAllister, J. S. V., 365, 372 McArthur, J. M., 33, 70, 142, 155 Macauley, B. J., 388, 431

448

AUTHOR INDEX

McAuliff, C., 105,115 McAuliffe, H. D., 379, 434 McBean, D. S., 270, 279,281 McCaleb, J. E., 46, 69 McCallum, K. J., 412,421,427 McCarty, R. E., 216, 233 McClure, J. W., 195, 199 MacCollom, G. B., 132, 138, 139,154, 155 McConaghy, S., 365,372 McConnell, P., 103, 115 McConnell, W. B., 414, 431 McCown, B. H., 205,233 McCracken. K. J.. 181. 200 McCrady, M. H., 379,432 McCraken, R. J., 380, 381,431 McCreery, R. A., 310,328 McCune, D. L., 121, 127, 156 MacDonald, G. M., 397, 398, 399,432 MacDonald, H.A., 120, 121, 122,123, 124, 125, 126, 128, 130, 137, 138, 139, 140, 144, 155, 156, 157 McDonald, I., 361,371, 384,433 McDonald, P., 346, 372 McDonald, P. W., 142,154 McDougall, B. M., 381, 383,385,431,433 McDougall, W. B., 178, 199 McElgunn, J. D., 178, 199 Macfadyan, A., 375,431 McGarity, J. W., 406,432 McGill, R. F., 359, 360,370 McIlroy, R. J., 63, 71 Mcllvain, E. H., 386,432 Mclntosh, T. H., 417, 420. 422,431 Mclntyre, D. S., 240, 241, 247, 250, 253, 254, 256, 257, 258, 259, 260, 261, 263, 264, 266, 268, 269, 270, 271, 278, 280, 282 Mclntyre, M. P., 4, 70 Mack, E., 378, 431 McKee,G. W., 125, 126, 129, 155 McKee, R., 124, 146, 155 Mackenthun, K. M., 419,431 Mackenzie, D. H., 288, 300, 3 I I , 3 18,328 MacLachlan, E. A., 369,371 McLaren, A. D., 109, 1 1 1, 115, 412,428 McLeester, R. C., 205,233 McLennan, H. A., 144,154 McLeod, M. N., 51, 71 McLuckie, J., 418, 431

McNaught, K. J., 347, 365, 372 McNeal, B. L., 87, 115 McQuillin, J., 377, 379, 434 Macrae, W. D., 334,372 McTaggart, A., 6, 70 Macura, J., 409,431 McVey, D. V., 134, 155 McVey, W. M., 140,156 McWhirter, K. S., 42, 70 McWilliam, J. R., 229, 233 Madsen, N. B., 21 I , 212,231 Maget, H. J. R., 258, 259, 260, 267, 269, 282 Majumder, S. K., 204, 229,233 Makay, K., 88, 96,97, 106, 114. 115 Malcolm, R. L., 380, 381, 431 Malzahn, R. C., 379,434 Mandal, R. C., 288,328 Mangan, J. L., 142, I55 Mangelsdorf, A. J., 292, 315,328 Mann, S. O., 360, 361, 362,370 Markey, S. P., 110, 116 Marriott, F. H.C., 174,199 Marriott, S. J., 45, 47, 70 Marshall, K. C., 109, 115 Marth, P. C., 221,233 Martin, A. E., 23, 55,69, 70 Martin, J. P., 296,328, 410, 41 I , 431 Martin, M. H., 253, 265, 270,282 Martin, R. T., 109, 116 Maruyama, C., 34, 73 Mason, K. R., 335,372 Materassi, R.,403, 426 Matherne, R. J., 323, 327 Mathur, R. B., 136, 155 Mathur, S. P., 415, 416,431 Matsubara, H., 207, 210,233 Matsumoto, H., 34, 35, 67, 71 Maurer, A. R., 22 I , 233 May, L. H., 161, 165, 199 Mayaudon, J., 409,412,414,431,433 Mayland, H. F., 227, 233, 417, 420, 422, 431 Mays, D. S., 142,156 Mazur, P., 206, 213, 217, 218,233 Mead, F. W., 46, 70 Meadows, M. W., 133. 139,154 Means, U. M., 129,154 Mears, K. A., 147, 156

AUTHOR INDEX

Mears, P. T., 9, 70 Meek, R. C., 99, 117 Meggitt, W. F., 88, 100, 102, 115 Mehta, N. D., 409,428 Meiklejohn, J., 421, 422, 423, 427, 431 Melhuish, F. M., 162, 199, 200 Melvin, J. F., 352, 372 Mendelovici, E., 94, 114 Menzies, J. D., 135, 155 Meredith, D. S., 388, 431 Mershon, M. M., 355, 356,372 Meryman, H. T., 2 12,233 Metcalfe, D. S., 124, 137, 138, 153, 156, 157 Metcalfe, G., 42 I , 431 Metson, A. J., 332, 340, 341, 343, 344, 345, 346, 348, 349, 350, 351, 357, 362, 367,370, 372,374 Meyer, H., 334, 338, 339,372 Michniewicz, M., 221, 233, 424, 431 Midgley, A. R., 127, 130, 156 Miles, J. F., 7, 50, 70 Miles, J. W., 196,200 Miles, R. E., 187, 200 Milford, R., 46, 50, 51, 52, 56, 57, 70, 71 Miller, H. P., 1 I , 70 Miller, J. D., 133, 150, 156 Miller, J. G., 335, 372 Miller, M. H., 168, 200 Miller, N. H. J., 422, 431 Miller, T. B., 52, 70 Miltimore, J. E., 33, 70, 142, 155 Minderman, G., 390, 393, 394,431 Minson, D. J., 46, 50, 5 I , 52, 56, 57, 70, 71 Miri, R. K., 144, 145, 154, 156 Misch. M . .I..2 2 5 , 232 Mishra, R. R., 401,431 Mishustin, E. N., 398, 407, 408, 415, 422, 4 31 Mitchell, C., 422, 433 Mitchell, H. L., 221, 234 Mitchell, K. J., 293, 301, 328 Mitra, A. K., 288, 3 2 7 Modlibowska, I., 227, 233 Mohan Rao, N. V., 306, 307, 309,328 Moir, W. W. G., 322,328 Molina, J. A. E., 423, 431 Moller, R. B., 3 14, 328, 329 Molloy, L. F., 350,374

449

Monteith, J. L., 405, 431 Montgomerie, R. F., 340, 372 Montgomery, R. D., 368,372 Moomaw, J. C., 27, 73 Moore. A. W., 21, 23, 71, 419, 421, 422, 431 Moore, W. L., 64, 71 M o m , P. A. P., 162, 199 Morgan, W. C., 279,282 Morita, K., 34, 35, 71 Moms, M. P., 35, 71 Morrison, N. E., 344, 3 6 6 , 3 7 3 Morse, R. A., 137, 156 Mortensen, J. L., 412, 431 Mortland, M. M., 78, 80, 81.82, 83,84,87, 88, 89, 90, 91, 92, 93, 99, 100, 102, 105, 106, 114, 115, 116, 251, 282, 425, 431 Moth G. O., 53, 55, 68, 69, 126, 140, 154, 156 Motta, M. S., 2, 71 Mrysha, G. N., 414,432 Mucke, D., 414,430 Mulder, C. E. G., 312, 328 Muller, L., 244, 283 Mundie, C. M., 414, 427 Mungomery, R. W., 296, 300, 307, 308, 309, 311, 314, 3 2 1 , 3 2 2 , 3 2 3 , 3 2 8 , 3 2 9 Munro, P. E., 422, 423,427, 432 Murdock, F. R., 379,434 Murphy, R. C., 110, 116 Murphy, R. P., 131, 142, 156, 157 Murtagh, G. J., 7, 17, 20, 71, 73 Musgrave, R. B., 142, 155 Mutatkar, V. K., 412,435 Myers, M. G., 406,432 Myers, M. N., 163, 199 Myers, R. J., 419, 420,432 Mylsamy, V., 288, 309, 3 2 9

N Nagai, I., 291, 328 Nahin, P. G., 104, 116 Natanson, N. E., 222,234 Naude, T. J., 312,328 Naveh, Z., 16, 17, 71 Neal, J. L., Jr., 422, 423, 424, 432 Nekrassow, L. N., 244, 283 Nelson, A. I., 212, 213,234

450

AUTHOR INDEX

Nelson, C. J., 143, 157 Nelson, J. C., 121, 156 Neme, N. A., 17,71 Nemethy, G., 207, 208, 209,233 Nestel, B. L., 5 5 , 57, 71 Neunzig, H. H., 135, 136, 156 Newbould, P., 175, 191,200 Newcomb, E. H., 220,232 Newman, E. I., 161, 167, 174, 200, 222, 233 Newman, J., 259,283 Newton, J. D., 421,432 Nezamuddin, S., 288,328 Ng, H., 216,233 Nicholas, D. J. D., 417,433 Nicholson, J. A., 346,370 Nicolson, T. H., 419,432 Nielsen, K. F., 270, 282 Nielson, J. A., 163, 166, 200 Nikitin, D. I., 407, 408, 415, 431, 432 Nilsson-Leissner, G.. 2, 73 Nittler, L. W., 151, 156 Nobbe, F., 200 Noggle, J. C., 353, 371 Noller, C. H., 53, 55,68, 69, 352,370 Nordfeldt, S.,34, 35, 71 Norman, A. G., 409,432 Norman, M. J. T., 4, 11,45,49,60,61,62, 71 Norris, D. O., 19, 23, 24, 25, 26,66, 71 Novogrudskii, E. D., 423,432 Nye, P. H., 173, 174, 177, 179, 191, 192, 193,199, 200 Nylund, R. E., 221, 234 0

Oakes, A. J., 5 , 17, 19, 46, 47, 54, 71 Ochilova, M., 415,432 O’Connor, K. F., 424,432 Oden, S., 243, 245, 249, 250, 251, 253,282 Odu, C. T. I., 420,432 Offutt, M. S . , 127,156 Oginsky, E. L., 406, 432 Ohlrogge, A. J., 160, 198 Okada, T., 163,201 Okorie, I. I., 63, 71 Oleniacz, W.S., 415, 432 Olien, C. R., 206, 218, 220,233 Olness, A. E., 103, 114

Olson, J. S., 392, 393,435 Opstrup, P. A., 386, 387,430 Ormrod, D. P., 22 1,333 Orpurt, P. A., 398,400,401,432 Orsi, S., 121, 156 Ortiz, F. S., 121, 127, 156 Osborn, J. F., 105, 216 Oshumi, F., 3 10,328 Ostazeski, S. A., 127, 133, 134, 135, 156 Osterli, V. P., 128, 130, 137, 156 O’Sullivan, M.,354,372 Ottley, A. M., 120, 156 Oury, B., 319,329 Ovcharenko, F. D., 87, 116 Ovchinnikova, M. E., 414,432 Overstreet, R., 171, 199 Owen, J. B., 357,372 Oxenham, D. J., 55,69 P

Padmanobhan, D., 315,328 Padmos, L., 344,373 Pagan, C., 35, 71 Page, A. L., 95,114 Paine, F. S., 400,432 Palafox, A. L., 35,67 Palaniswamy, K. M., 320,328 Paleg, L. G., 161, 199 Paltridge, T. B., 14, 48, 71 Pan, C. L., 288,328 Panikkar, M. R., 156 Pankratova, E. M., 420,433 Pao, T. P., 323,328 Paolillo, J., Jr., 145, 157 Papermaster, B. W., 407, 433 Paquay, R., 357,372 Parago, J. F., 288,328 Parberry, N. H., 422, 424,432 Parbery, D. B., 288, 300, 31 1, 318, 328, 329 Pardee, W. D., 136, 156 Parfitt, R. L., 87,89,94, 102, 103, 104,116 Parish, D. H., 308,329 Pansi, A. F., 369,373 Park, S . J., 40, 41, 71, 72 Parker, D. T., 393,432 Parkinson, D., 38 1, 409, 432 Parodi, L. R., 49, 71 Parodi, R. A., 49, 71

45 1

AUTHOR INDEX

Parr, J. F., 409,432 Parsons, J. L., 131, 140, 156 Passioura, J. B., 179, 180, 183, 196, 200 Patel, R. M., 315,329 Patil, A. S., 168, 200 Patil, B. D., 13, 71 Patterson, E. K., 178, 200 Paul, E. A., 395, 404, 405, 409, 410, 412, 414, 415, 416, 419, 420, 421, 422, 426 427,431, 432 Pauli, A. W., 166,200 Pauling, L., 207, 234 Pavlychenko, T. K., 167, 200 Payne, W. R., 100, 116 Peacock, H. A,, 148, 149,156 Peake, E.,409, 429 Pearson, M. C., 419,434 Pearson, R. W., 253,283 Peebles, R. H., 3 15, 329 Peele, T. C., 56, 72 Pelczar, M. J., Jr., 4 15, 432 Penefsky, H. S., 216,234 Perminova, G. N., 420,433 Perry, P. W., 99,117 Peters, E. J., 132, 156 Petersen, J. B., 403, 432 Peterson, G. H., 109, 1 1 1 , 115, 116, 412, 428 Peterson, M. L., 128, 130, 137, 156 Pettit, R. E., 134, 156 Pettit, R. M., 210, 233 Peynado, A., 227,234 Philip, J. R., 183, 200 Phillips, J., 3, 71 Phillips, R. E., 168,200 Phillips, R. L., 125, 146, 148, 149, 156 Pichel, W., 212,234 Pick, W., 369, 371 Pielou, E. C., 186, 200 Piemeisel, R. L., 401, 433 Pienkowski, R. L., 136, 156 Piercy, A., 402,432 Pierre, J. J., 130, 139, 156 Pietig, F., 412, 433 Pihl, E., 21 I, 234 Pirt, S. J., 405,432 Pisano, M. A., 415,432 Pisseau, M. A., 3 19, 329 Pissot, P., 307,329 Pitman, M. G., 174, 200

Pittman, U. J., 165, 200 Playne, M. J., 52, 71, 72 Plucknett, D. L., 14, 27, 61, 73, 288, 291, 298, 300, 301, 302, 303, 306, 309, 313, 3 18,320,329 Pomeroy, K.,2 19,234 Poncelet, G., 81, 115 Pontovich, V. E., 414,432 Poostchi, I., 144, 156 Pope, J. D., 100,116 Porter, L. K., 159,200, 393,420,422,432, 434 Posner, A. M.,169,199, 410,432 Potter, E. C., 242, 243, 249, 255, 265,283 Pound, A. W.,35,43,69 Prasad, R. B., 296, 31 I , 329 Pratt, H. K., 229,233 Prescott, S. C., 379,432 Pressey, R., 142, 156 Prince, A. L., 341,370, 372 Prine, G. M., 54, 72 Pritchard, A. J., 38, 39, 40, 42, 49, 50, 5 8 , 59, 72, 73 Pruppacher, H. R., 227,234 Pugh, G. J. F., 388,432 Pullman, M. E., 216,234 Purcell, G . V., 213, 233 Purushothaman, G., 320,328 Purvis, E. R., 341,370 Py, C., 3 19,329 Q

Qualls, M., 126, 154 Quirk, J. P., 78, 94, 96, 97, 98, 115, 116, 169, 178, 180,199,201 Quiros, C. M., 121, 156 Quisenberry, J. H., 34, 73 Quispel. A., 419,430 R

Racker, E., 216, 233, 234 Racz, G . J., 163,200 Radok, J. R. M., 196, 198 Radul, N. M., 87,116 Raheja, P. C., 298,329 Rahman, H., 346,372 Raica, N., 422, 428

452

AUTHOR INDEX

Rainey, R. C., 312, 314, 329 Ralwani, L. L., 288,327 Raman, K. V., 81, 82, 83, 105,116 Raman, V. S., 49, 70 Rama Rao, G., 306, 307, 309,328 Ramstad, S., 352, 358,371 Randles, F. M., 161, 199 Rankin, J. E. F., 365,372 Rapista, E. A., 288,328 Rauzi, F., 386, 387,432 Ray, R. Y., 401,432 Reid, C. S. W., 350,374 Reid, J. T., 140, 142, 156, 157 Reid, R. D., 415,432 Reingold, N., 203, 234 Reith, J. W. S., 342, 347, 365, 372, 373 Renner, G. T., 4, 73 Rennie, D. A., 163, 200, 412, 421,427 Retzer, H. J., 131, 154 Reuss, J. O., 425,432 Rheaume, B., 219,234 Rhoades, E. P., 386,432 Rhodes, A. M., 296,326 Rhykerd,C. L., 125, 126, 156, 352,370 Ricaud, C., 287, 289, 294, 300, 316, 326, 329 Rice, E. L., 400, 422, 423, 424,428, 432 Rice, H. B., 341,372 Rice, R. G., 209, 224,231 Rice, W. A., 419,420,42 I , 422.432 Richards, S. J., 279,282, 410, 41 I, 431 Richardson, H. L., 422,432 Richardson, T., 219, 232 Rickrnan, R. W., 244, 245, 246, 248, 252, 258,259,267,269,271,275,283 Riddiford, A. C., 244,283 Ridgeway, R. L., 136, 156 Rinfret, A. P., 217. 223,232 Rios, E. G., 87, 116 Ripperton, J. C., 17, 72 Ritchey, G. E., 34,68 Ritchie, N. S., 367, 368, 371, 372 Ritson, J. B., 62, 68 Rivera, J. R.,315,329 Roach, S. A., 188,200 Robert-Gero, M., 410,429 Robins, M. F., 29,30,3 I , 32,50.54,66,68 Robinson, A. R., 47, 69 Robinson, D. H., 120, 127, 157

Robinson, J. B., 397, 398, 399, 424, 427, 432 Rochecouste, E., 3 13,329 Rockett, J. A., 256, 259, 260,283 Rodrigues, A., 87, 116 Rogers, 302, 328 Romney, D. H., 54, 72 Ronningen, T. S . , 131, 157 Rook, J. A. F., 332, 337, 338, 339, 356, 364,372, 373 Rose, C. W., 167,200 Rose, L. E., 162, 199 Rosenberg, M. N., 35, 72 Roseveare, G. M., 49, 72 Rosha, N. S., 414,432 ROSS,D. B., 346, 360, 361, 362, 370, 373 Ross, D. J., 5 5 , 73, 397, 406, 421, 423, 432,433 Ross, P.J., 55,69, 70 Rotar, P. P., 14, 27, 36, 40, 41, 61, 71, 72 73 Rothberg, T., 114 Rothlein, R., 258, 259, 260, 267, 269,282 Rotman, B., 407,433 Rovira, A. D., 176, 177, 198. 381, 383, 384,412,423,431,433 Rowe, A. W., 217,223,232,234 Roy, S . K., 320,329 Ruelke, 0. C., 46,69 Ruinen, J., 377, 378, 379, 380, 417, 433 Rurnsey, T. S., 352,370 Russel1.C. P., 163,199 Russell, E. J., 404,422,433 Russell, E. W., 53,72 Russell, J. D., 81, 83, 85, 86, 87, 88,89,91, 92, 96, 97, 100, 102, 106, 114, 116, 117 Russell, J. S . , 22, 72 Russell, R. S., 163, 175, 176, 191,200 Ruth, G., 33,69 S

Sachs, M., 3 13,329 Sadasivan, T. S., 391,433 Sakai, A., 218, 223, 234 Salette, J. E., 5 5 , 72 Salmon, R. C., 340, 341,373 Salmon, S . C., 204,234

AUTHOR INDEX

Salt, R. W., 228, 232, 234 Samuel, G., 391,433 Samuels, G., 307, 32 I , 329 Sanders, D. C., 221,234 Sanderson, J., 176, 200 Sandon, H., 398,433 Santarius, K. A., 2 I I , 216, 222, 223, 23 I , 232 Santhanam, V., 288,328 Santoro, T., 406,427 Saran, A. B., 288, 320,329 Sastry, D. K., 296,329 Satchell, J. E., 390, 429 Sauerbeck, D., 410, 433 Saunders, W. M. H., 332, 343, 344, 345, 346, 348, 349, 35 I , 367,372 Savage, W. H., 340,372 Sawyer, D. T., 244,246,248,249,258,283 Saxton, W. T., 418, 433 Sayre, C. B., 160,200 Scantamburlo, J. L., 49, 71 Scarth, G. W., 206, 234 Scaut, A., 2,68 Schafer, W. P., 321,329 Schank, S. C., 46.69 Scharpenseel, H. W., 412,414,433 Scheffer, F., 4 14,433 Scheraga, H. A., 207, 208, 209,233 Schinckel, P. G., 34, 69 Schleger, A. V., 61, 72 Schmid, A. R., 141, 157 Schmidt, H. L., 414, 428 Schnitzer, M., 95, 96, 110, 111, 115, 116, 380,409,429, 433 Schoenike, R. E., 419,430 Schofield, J. L., 7, :I , 45, 72 Schotield. R . K.. 159. 200 Scholl, J. M., 122, 131, 132, 141, 157 Schonwalder, H., 414, 415,433 Schoth, H. A., 139, 155 Schott, H., 103, 104, 116 Schribaux, E., 121, 157 Schuster, J. L., 388, 433 Schuster, N., 334, 371 Schwartz, W. L., 358,370 Schweizer, J., 380, 433 Scott, D., 361, 371 Scott, D. C., 99, 116 Scott, G. B., 422, 427

45 3

Scott, J. E., 319, 326 Scott, R. F., 267,283 Scotto, K. C., 339, 358, 373 Scrutton, M. C., 216, 234 Sealock, R. W.,216, 232 Seaman, G. V. F., 109,115 Seaney, R. R., 132, 145, 146, 151,155,157 Sears, 0. H., 129, 155 Seawright, A. A., 49, 70 Sedgley, R. H., 162, 167, 198 Seekles, L., 340, 345,371, 373 Sellschop, J. P. F., 204, 234 Semikhatova, 0. A., 216,232 Sen Gupta, P. K., 90,115 Serratosa, J. M., 79, 116 Servais, A., 115 Shafer, N . E., 22 1, 234 Shamoot, S., 384,433 Shanmugasundram, A., 288, 309,329 Shantz, H. L., 401,433 Sharpe, J. P., 46. 55.67 Shaw, N. H., 4, 7, I I , 21, 23, 44, 45, 48, 6 I , 62, 67, 68, 72 Shaw, R. H., 124,156 Shedley, D. G., 300,329 Shelton, D. C., 224,232 Shepherd, H., 414, 427 Sherman, J. K., 217, 234 Sherrod, L. B., 288,329 Shibles, R. M., 126, 157 Shields, L. M., 402, 422, 433 Shih, S. C., 224,232 Shikama, K., 208,209,215,234 Shkol’nik, M. Y.,222, 234 Short, C., 286, 328 Shtina, E. A., 402, 420, 433 Siberg, R. A,, 217,232 Sidhu, B. S., 125, 144, I55 Sieskind, O., 96, 116 Silow, R. A., 145,157 Silva, S., 54, 73 Silver, W. S., 417, 433 Siminovitch, D., 219, 224, 234 Simmonds, N. W.,287, 300, 301, 3 19,329 Simon, E., 225,231 Simonart, P., 409, 412, 433 Simpson, K.,346,372 Sims, A. P., 253, 265, 271,283 Sims, F. H., 336, 338,370, 373

454 Sims, J. R., 95, 114 Singer, R. H., 335,373 Singh, D., 288,329 Singh, 0. N., 13, 71 Singh, S. S., 320,329 Singh, S. V., 13, 71 Single, W. V., 221, 234 Sinha, T. D., 288,328 Sjollema, B., 332, 338, 373 Skaland, N., 130, 131,157 Skerman, P. J., 32, 73 Skidmore, C. L., 3, 68 Skipper, H. D., 106, 116 Skou, J. C., 355,373 Skov, O., 19, 54, 71 Skujins, J. J., 406,434 Slack, S. T., 142, 156 Smillie, L. B., 2 1 I , 2 12, 231 Smit, B., 3 12, 314,329 Smit, J., 318,434 Smith, A., 19, 72 Smith, C. A., 65, 68 Smith, D., 143, 157, 205, 234 Smith, D. H., 425,428 Smith, F. B., 400,430 Smith, J. B., 124, 148, 157 Smith, J. H., 129, 157 Smith, R. E.,140, 156 Smith, R. H., 357, 362,373 Smith, R. L., 425,432 Smyth, P. J., 361,373 Snyder, L. A., 58, 72 Soileau, J. M., 181,200 Sokolowski, M. B., 223, 234 Solomon, D. H., 93, 106, 116 Soper, K., 178,200 Sorensen, L. H., 394, 413,414,434 Soulides, D. A., 422, 425,434 Sowden, F. J., 412,427 Spencer, W. F., 107, 114 Spillner, E., 323, 326 Spiss, L., 145, 157 Sprague, M. A., 130, 153 Srhivasan, V., 320, 326 Stacey, M., 104, 114, 393, 429 Stahmann, M. A., 2 19,232 Standley, J., 393,429 Staniforth, D. W., 132, 157

AUTHOR INDEX

Staples, I. B., 20, 72 Starkey, R. L., 381,434 Stebbins, G. L., 143, 157 Steelink, C., 414, 434 Stein, W. D., 218, 234 Steinberg, M. P., 212, 213,234 Steindle, D. R. L., 292, 3 15,329 Steinlen, H.,288,329 Stevens, 0. A., 417,435 Stevenson, F. J., 110, 116 Stevenson, G., 418,434 Stevenson, 1. L., 406, 415,430, 434 Stewart, B. A., 159,200, 393, 426,434 Stewart, G. A., 60, 62, 71 Stewart, J., 334, 346, 365,370, 373 Stewart, J. McD., 216,229, 231,234 Stewart, W. D. P.,416,417,419,420,421, 422,434 Stickler, F. C., 130, 157, 166,200 Stirk, G. B., 54, 69 Stith, L. S., 288, 329 Stobbs, T. H., 4, 16, 44, 63, 72 Stocker, G. C., 1 I , 72 Stockli, A., 404, 434 Stokes, I. E., 298,329 Stab', L.H., 238,239,240,241,243,244, 245, 246, 241, 248, 252, 253, 251, 260, 261, 268, 210, 271, 213, 274, 218, 279, 280,281, 282,283 Stone, E., 130, 156 Stone, R. W., 379,434 S t o w , J. E., 332, 338, 339, 351, 356, 364, 372,373 Stotzky, G., 109, 116, 406,434 Stout, J. D., 390, 396, 434 Stout, P. R., 351, 352, 353, 358, 370, 373 Stromberg, R. R., 214,233 Strugger, S., 403, 434 Stuckey, 1. H., 131, 157 Sturgess, 0. W., 3 14, 329 Sturtz, J. D., 1 I , 72 Subba Rao, M. S . , 296,3 1 I , 329 Subra, P., 287, 304,329 Sullivan, C. Y.,210, 233 Suman, R. F., 56, 72 Sun, S. W., 288, 330 Surendran, C., 288,327

AUTHOR INDEX

Suresh, S., 288, 309, 329 Suttle, N. F., 339, 373 Swain, F. G., 131, 154 Swan, J. B., 337,373 Sweet, D. V., 380,431 Swincer, G. D., 104, 116 Swoboda, A. R., 81, 92,116 Synhorst, S. H.,142, 156 Szegi, J., 414, 434 Szeicz, G., 405,431 Szokolay, G., 288, 300,329 Szurnkowski, W., 314,329 Szuszkiewicz, T. E., 241, 252, 253, 257, 268, 269, 270, 271, 273, 274, 279, 280, 281,282

45 5

Tesar, M. B., 130, 142, 155, 157 Thangavelu, S., 288, 309, 329 't Hart, M. L., 332, 333, 334, 344, 345, 347, 348,371,372,373 Theng, B. K. G., 78,116, 410,432 Theron, J. J., 422, 423,434 Therrien, H.,224,234 Thomas, D. C., 35 I , 358. 370 Thomas, J. W., 142,155, 332, 359,373 Thomas, R. O., 229,231 Thomas, S. B., 377, 379,434 Thompson, A., 340,373 Thompson, G. D., 321,330 Thompson, H. E., 124, 138,157 Thompson, H.R., 162,200 Thompson, J. F., 347,371 T Thompson, T. D., 98, 115 Thomson, N. J., 296,330 Thorne, P. M.,29, 34,66, 69 Tackette, J. L., 253,283 Thornton, R. H., 401,434 Tahoun, S., 8 1, 84, 87, 89, 116 Thrower, L. B., 388,431 Takahashi, M., 2, 17, 27, 34,35,71, 72, 73 Thurston, H.D., 315,327 Takuchi, K.,409,429 Timonin, M. I., 381, 398, 400, 426, 430, Tamimi, Y. N., 288, 298, 300, 301, 303, 434 306, 309, 313, 318, 320,329 Tinker, P. B., 173,200 Tamm, E., 405,434 Tinsley, J., 414,430 Tanczos, 0. G., 228,234 't Mannetje, L., 22, 24, 37, 39, 40, 45, 71, Tang, K. H.,298, 306, 307, 321,329 72, 73 Tanimoto, T., 298, 306, 308, 309,322,326 Todd, J. R., 332, 342, 343, 344, 363, 365, Tanner, C. B., 245,283 366,371,372,373 Tanners, M. A., 412,433 Tolbert, N . E., 22 1,234 Tappel, A. L., 210, 220,234 Toledo, J., 57,67 Tarasevich, Yu. I., 87, 116 Toledo, R., 212, 213,234 Tatevosian, G. S., 109, 115, 414, 428 Toler, R.W., 316, 324,326 Taylor, C. B., 379, 397, 403,434 Tollin, G., 414,434 Taylor, H. M.,386,432 Toman, F. R., 221,234 Taymen, L. A., 149, 155 Tome, G. A., 123, 145,157 Tomlinson, T. E., 99, 115 Tazaki, T., 204, 229, 233 Tchan, Y. T., 402,434 Topping, L. E., 397,434 Teakle, D. S., 9, 67 Torssell, B. W. R., 167, 200 Teakle, L. J. H., 45, 68, 72 Toth, S. J., 341,370 Teasley, J. I., 100, 116 Tothill, J. C., 44, 73 Teel, M. R., 353,373 Touillaux, R., 87, 114 Templeton, J., 301, 304,312, 314,317,329 Tow, P. G., 38, 73 Tennison-Woods, J. E., 13, 66 Trernillon, B., 243, 245, 255,281 Tepe, W., 159,200 Tretjakova, A. N., 420,433 Tepper, E. Z., 414,434 Trimberger, G. W., 140, 142,156, 157 Terry, M. L.,369,371 Trkula, D., 223, 234

456

AUTHOR INDEX

Troughton, A., 160, 201 Trouse, A. C., Jr., 298, 306, 308, 309, 3 11, 322,326, 330 Trumanov, 1. I., 220, 234 Trumble, H. C., 2, 73 Truong, N. V., 32, 73 Tseng, P. M., 296,328 Tukey, H. B., 380,431, 434 Tukey, H. B., Jr., 380,434 Tulloch-Reid, L. I., 54, 66 Turk, K. L., 140, 142, 156, 157 Turner, H. G., 61, 72 Tzagoloff, A., 216, 219, 232 U

Ueno, M., 163,201 Umbreit, W. W., 406, 432 Underwood, E. J., 332, 336, 356,359, 366, 373 Upadhyay, J., 135, 157 Upchurch, R. P., 99, 117 Urata, U., 40, 72 Ushuima, T., 204, 229,233 Utter, M. F., 216,234 Uytterhoeven, J., 80, 81, 94, 95, 116 V

Vallee, B. L., 332, 373 Vallis, I., 5 5 , 73 Valoras, N., 105, 116, 252, 279, 281, 282, 283 Vandecaveye, S. C., 397, 398,434 van der Boon, J., 228,234 van der Drift, J., 390,428 van der Molen, H., 332, 334,343, 344,373 van Doren, D. M., 241,249,257,274,278, 282,283 Van Dyne, G. M.,386,430 van Es, A. J. H., 348, 349, 362,372 van Gundy, S . D., 260, 271,283 Van Hoof, H. A., 46,68 Van Keuren, R. W., 141,157 Van Olphen, H., 104,116 Van Rheenen, D. L., 340,371 Van Schaik, P. T., 301, 316,330 van Schreven, D. A., 424,429 Varney, K. E., 128, 131, 157 Vasantharajan, G. S . , 417,434

Vavra, J. P., 186, 198 Veiga, F. M., 321, 322,330 Veldkamp, H., 405,435 Venkatachari, A., 296,329 Venkataraman, R., 320,330 Verboom, W. C., 16, 73 Vernon, T. R., 398,435 Vetter, R. L., 122, 141, 157 Vianne, E., 121, 157 Vicente-Chandler, J., 54, 56, 68, 73 Vichurina, G. A., 216,231 Vidyabhushanam, R. V., 288,328 Viets, F. G., 161, 201, 393, 434 Vigil, E. L., 220,232 Vilain, M., 235, 283 Vincent, J. M., 416, 435 Vine, 1. L., 420,432 Vlassak, K., 414,435 Voisin, A., 332, 335, 346, 354,373 Volk, G. M., 426,435 Vostrov, I. S., 407, 408, 415, 431 Vowles, L. E., 360, 361, 362, 370 W

Wacker, W. E. C., 332,354,355, 369,373 Wada, K., 110, 115, 116 Waddington. D. V., 279,283 Wagner, G. H., 412,435 Wagoner, R. E., 128, 155 Waid, J. S., 390, 401, 435 Waite, R. B., 20, 23, 48, 70, 72 Wake, J. R. H., 410,432 Wakefield, R. C., 130, 131, 153, 157 Waksman, S. A., 400, 406,415,428,435 Walker, G. F., 80, 93, 106, 116 Walker, J. M.,180, 201 Walker, J. W., 229, 234 Walker, R. D., 256, 260,281 Wallace, A. T., 128,157 Wallace, L. R., 332, 365, 373 Walshe, M. J., 344, 373 Wang, C. C., 294,330 Wang, H. L., 415,435 Wang, J . H., 216,232 Warcup, J. H., 398,400,402,435 Ward, G. M., 340,373 Warman, H. J., 4, 73 Warmke, H. E., 35, 58, 68, 71, 72, 73 Washko, J. B., 142,156

AUTHOR INDEX

Wassom, C. E., 130, 157 Watson, S. J., 377, 379, 426 Wayman, O., 34, 73 Wayne, N . J., 221,234 Weaver, J. E., 161, 166, 192,201,395,435 Webb, S. J., 214, 215, 223,234 Weber, J. B.. 79, 99. 100, 116, 117 Webster, C. C., 3, 73 Webster, J., 388, 389,429, 435 Wedin, W. F., 122, 141, 157 Weed, S. 9.. 79, 82, 99, 115, 117 Weihing, R. M., 166. 199 Wiersum, L. K., 188,201 Weiss, A., 93, 106, 117 Weiss, E. A., 289, 330 Wene,G. P., 301, 312, 314,330 Weneck, E. J., 223, 234 Wengel, R. W., 278, 279,283 Wergin, W. P., 220, 232 Werkenthin, F. C., 400, 435 Werkhoven, J., 287,330 Wernsman, E. A,, 143, 144, 147,157 West, D. W., 25 I , 279,281,283 Wetter, L. R.,419, 420, 421, 422,432 Whalley, R. D., 45, 68 Wheaton, T. A., 229, 233 Whetzal, F. W., 386, 430 Whitaker, E. S., 178, 201 White, C. H., 125, 155 White, C. L., 4, 73 White, E. J., 380, 427 White, J. L., 76, 83, 88, 92, 99, loo, 106, 114, 115, 116 White, R. E., 48, 73 Whiteman, P. C., 23, 36, 39, 73 Whitman, W. C., 417, 435 Whittet, J. N., 21, 73 Whyte, R. O., 2, 61, 73 Wiebe, H. H., 176,199 Wiegand, C. L., 238, 240, 268, 277, 282, 283 Wiegert, R. G., 386, 388,435 Wieringa, K. T., 378, 434 Wiggans, S. C., 124, 138, 157 Wild, A., 86, I15 Wilfarth, H., 38 I , 429 Wilkinson, H. F., 178, 201 Wilkinson, W. S., 53, 67 Will, F. G., 258, 259, 267, 269, 283 Will, G. M., 390, 435

457

Willers, E. H., 34, 35, 71 Willet, E. L., 34, 73 Witley, C . R., 245,283 Williams, D. E., 172, 201 Williams, I. H., 305, 327 Williams, K. R., 243, 244, 283 Williams, R. D., 385, 435 Williams. R. J., 7, 14, 16, 73 Williams, W. A,, 131, 157 Williamson, R. E., 279, 283 Wilman, D., 357,372 Wilner, J., 224, 234 Wilsie, C. P., 123, 144, 148, 149, 154, 156 Wilson, A. A., 360, 373 Wilson, B. D., 423, 431 Wilson, 9. F., 165, 201, 300, 304, 330 Wilson, G. F., 350,374 Wilson, G. L., 39, 54, 70 Wilson, G. P. M., 7, 17, 20, 48, 71, 73 Wilson, J. K., 383, 431 Wilson, J. R., 23, 72 Wilson, P. N., 3, 73 Winburne, J. N., 286,330 Winch,J. E., 123, 124, 127, 137, 138, 139, 155, 157 Winchester, W. J., 47, 70 Windrum, G. M., 35.68, 70 Winkler. W., 376, 435 Winslow, C. E. A., 379,432 Wiseman, H. G., 377, 378, 379, 390, 430 Withers, F. W., 334,372 Witkamp, M., 392, 393,435 Witter, S. H., 380, 434 Wittwer, L. S., 140, 157 Wohlschlag, D. E., 222,232 Wojciechowska, B., 124, 157 Wolcott, A. R., I l l , 114 Woldendorp, J. W., 422,435 Wolf, A., 377, 378, 379, 435 Wolton, K. M., 364, 365, 374 Wong, C. Y., 288,330 Wood, G. 9.. 256, 260,281 Wood, G. H., 306,330 Wood, G. M., 130,157 Wood, R. A., 306,307, 308,330 Woods, L. E., 63, 73 Woods, R. W., 122, 141, 157 Woods, S. G., 56, 72 Worker, G. F., 288,330 Wullstein, L. H., 425, 435

AUTHOR INDEX

458 Wunsche, U., 221,234 Wutoh, J. G., 38, 42, 73 Y

Yadav, A. S.,388,430 Yakuku, K., 405,431 Yang, K. C., 288,330 Yang, L. A., 420,433 Yang, S. J., 288, 291, 330 Yariv, S., 85, 88, 89, 96, 97, 102, 117 Yates, J. J., 65, 68 Yawalker, K. S., 141, 157 Yoshida, R. K., 18, 33, 73 Yoshida, S., 218,234

Yoshihara, K., 163,201 Young, F. P., 288, 327 Young, R., 227, 231,234 Y o w e , 0. R., 14, 27, 34, 35, 61, 71, 73, 288, 291, 298, 300, 301, 302. 303, 306, 309, 3 13, 3 18, 320,329 Younger, V. B., 279,283 2

Zachev, S., 288,330 Zandstra, I. I., 120, 157 Ziechmann, W., 414,433 Zhmerman, M., 341,372 Zoebisch, 0. C., 35, 72

SUBJECT INDEX c c c , 221 Ceanothus, 417 Cenchrus ciliaris, 5 , 7, 45 Cenchrus setigerus, 45 Centro, 21, 23, 26, 29, 32, 63 Centrosema pubescens, 7, 2 1 Chilling injury, 203-234 Chloris gayana, 5 , 7, 45 2-C hloro-4,6-bis(ethylamino)-s-triazine,1 39 2-CNoroethyl trimethylammoniumchloride, 22 I Chloroplast, 219, 222 Clay-organic complexes, 75-1 17 Cocksfoot, 343 Corn, 341 Cotton, 229, 287, 288, 291, 293-294, 304, 310, 3 ll,316,3l7-3l8 Crested wheatgrass, 335, 342, 352, 353 Cyanogenesis, 125, I34 Cynodon dactylon, 5 , 23, 46, 59 Cyocel, 221

A

Abaca, 287, 3 10 Actinomycetes, 396-398, 4 I7 Agropyron cristaturn, 342 Agropyron desertorurn, 335, 342, 352, 353 Agropyron repens, 388, 389 Alar, 221 Alfalfa, 126, 132, 142, 336 Alnus, 4 17 3-Aminotriazole, 100 Ammonia, 425 Andropogon sorghum, 388 Artemisia lodriciana, 4 17 Avena sativa, 332, 342 B

Bahiagrass, 48 Banana, 287,298, 304 Barley, 353 Barrel medic, 21, 22, 30 Bean, 22 I , 229 Bentgrass, 279 Bermudagrass, 59 Big trefoil, 120 Birdsfoot trefoil, 1 19- 157 Birdwood grass, 45 B-nine, 221 Boehmeria nivea, 287, 288 Brachiaria decumbens. 45, 56 Brachiaria mutica, 7, 44 Brome grass, 353 Bromus inermis, 353 Bromus mollis, 342 Bromus spp., 407 Buffelgrass, 5 , 13, 45, 59. 65 Bulrush millet, 49 2-sec-Butyl-4,6-dinitrophenol, 132 C

Cabbage, 221, 279 Calcium, 364 Calcined magnesite, 364-365, 366 Calopo, 7, 21 Calopogonium mucunoides, 7, 2 1

D

2,4-D, 133 Dactylis glomerata, 335, 342, 343, 377, 388, 389,407 Dalapon, 132, 139 Dallisgrass, 27, 48, 61 Dalrymple vigna, 20, 29 2,4-DB, 132, 133, 139 N-Decenylsuccinic acid, 22 1 Denitrification, 424-426 Desiccation injury, 2 1 1 Desmonium canum, 14, 27 Darmodium gyroides, 14 Desmodium heterophyllum, 14 Desmodium intortum, 9, 14-16, 27, 36, 61, 63 Desmodium sandwicense, 14, 36, 41 Desmodium uncinatum, 9, 16 4-(2,4-Dichlorophenoxy) butyric acid, I32 2,2-Dichloroproponic acid, 132 I ,5-Difluoro-2,4-dinitrobenzene, 22 1 Digitaria decumbens. 5 , 46 Digitaria sp., 7 459

460

SUBJECT INDEX

N,N-Dimethylamino succinamic acid, 22 1 Dimethyl sulfoxide, 225 Dinoseb, 132 Diquat, 99 DMSO, 225 Dolichos axillaris, 20 Dolichos lablab, 7, 9, 20 Dolichos uniflorus, 20 Dolomite, 364

J

Jaraguagrass, 44, 5 I Johnsongrass, 49 K

Kaolinite, 95, 99 Kentucky bluegrass, 13 I , 279 Kikuyugrass, 6, 27, 48, 54 Kraznozem, 17

E

Ecosystem productivity, 375 Eleagnus, 4 17 EPTC, 132 S-Ethyl dipropylthicarbamate, I32 Eucalyptus regnans, 388 F

Festuca arundinacea, 353 Freezing, protection, 220-228 Frost injury, 203-234, 302 G

Gley soils, 28 Glycine, 24, 26, 29, 30, 32, 42 GIycine javanica, 6 Glycine wightii, 6, 9, 16-17, 37-38 Grassland litter, 385-395 Grassland, microflora, 375-435 Grass staggers, 332 Grasses, tropical, 44-50 Greenleaf desFodium, 14, 26, 30, 36, 41 Green panic grass, 13, 54 Guinea grass, 5 , 6, 13, 47, 5 5 H

Hamilgrass, 39 Heteropogon contortus, 44 Hordeum leporinum, 342 Hordeum vulgare, 353 Humus, 385,409-416 Hydrogen bonding, 89-93, 207-208 Hyparrhenia rufa, 44, 63 Hypomagnesium tetany, 332-374 I

Indigofera spicata, 34-36, 43 Ion-exchange, 77-80, 97

1

Latosol, 27, 61 Legumes, 335, 417 tropical, 3, 6, 8-2 1, 36-39 Leucaena, 17-19, 24, 32, 33-34,42-43 Leucaena leucocephala, 17- 19 Lolium perenne, 342, 347 Lotononis bainesii, 7, 19 Lotus corniculatus, 1 20, 12 I , 122, 126, 143 Lotus pedunculatus, 120, 128, 147 Lotus tenuis, 120, 127, 143 Lucerne, 2 1-22, 24, 29.30.43-44, I88 M

Magnesium, 340-342, 354-356 Maize, 192 Medicago sativa, 21, 126, 336 Medicago truncatula, 2 I Melinis minutiflora, 5 , 47 Membrane permeability, 2 16, 2 17-220, 223-226 Microflora, grassland, 357-435 Miles Lotononis, 7, 8, 19-20, 29, 30, 31, 32,64 Mimosine, 33-34 Molassesgrass, 5 , 6. 47. 54 Montmorillonite, 90, 94, 96, 99, 100, 103, 109, 11 I Mouse barley, 342 Murray lathyroides, 13-14, 29, 30, 31, 36, 45 Musa sapientum, 388 Musa textilis, 287 Mycorrhizae, 4 17-4 18 N

Narrowleaf trefoil, 120, 127 Nitrification, 422-424

46 1

SUBJECT INDEX

Nitrogen, 357 Nitrogen fixation, 8, 23-26, 34, 416-422 Nutrient transference, 167- 17 I Nutrient uptake, root system, 159-201 0

Oats, grass tetany, 332, 342 Opuntia fragilis, 4 17 Orchard grass, 332, 342, 343, 377 P

Pangolagrass, 5 , 21, 46-41, 5 1 , 53, 54, 5 5 , 56, 61, 64 Panicum maximum, 6, 7, 39, 47 Panicum maximum var. trichoglume, I3 Paragrass, 7, 44 Paraquat, 99, 3 13 Paspalum commersomii, 48 Paspalum dilatatum, 5 , 27, 48 Paspalum notatum, 5 . 7. 48, 5 8 Paspalum plicatulum, 7 Paspalum scorbiculatum, 288 Pasture, birdsfoot trefoil-grass, 140-142 tropical, 1-73 Patchouli, 287 Pea, 229 Pear, 221 Pennisetum clandestinum. 6, 27. 48 Pennisetum purpureum, 48 Pennisetum typhoides, 49 Phaesolus atropurpureus, 8, 12, 32, 36, 41 Phaseolus lathyroides, 13, 24 Phaseolus vulgaris, 229 Phleum pratense, 335, 342 Phosphorus, 23, 53-54 Phyl!osphere, 376-38 I , 389 Pigeon pea, 192 Pineapple, 287, 293, 294, 304, 309, 316, 3 18-3 I9 Pisum sativum, 166, 229 Podzolic soil, 23, 27, 28 Pogostemon cablin, 287 Potassium, 364 Prickly pear cactus, 417 Protein denaturation, 209-2 I2 Pueraria phaseoloides, 7, 2 I Puero, 7, 21

R

Ramie, 287, 288, 309-3 10 Ratoon cropping, 285-330 Red clover, 126, 132, 342 Rhizosphere, 38 1-385 Rhodesgrass, 5 , 13, 45-46, 54, 5 5 , 57 Rice, 279, 287, 292, 301, 320 Rongai lablab, 7, 20, 33 Root exudates, 382 Root hairs, 178- I8 1 Rye, 332, 342 Ryegrass, 279, 341, 342, 343, 347, 350 5

Saccharurn oficinarum, 388 Sagebrush, 417 Secale cereale, 332, 342 Setaria, 57, 58 Setaria sphacelata, 6, 7, 49, 5 8 , 63 Shepherdia, 4 I7 Signalgrass, 45, 56 Silage, 140, 379 Silverleaf desmodium, 9, 14, 16, 23, 24, 28, 29, 30, 32, 36, 41, 64 Silvex, 133 Simazine, 139 Siratro, 8, 9, 12-13, 24, 29, 30, 32, 36, 3839, 41, 5 1 Soft chess, 342 Soil aeration, platinum micro-electrode measurement, 235-283 Soil microflora, 395-409, 414-416 Solodic soil, 23, 27 Sorghum, 195, 279, 287, 288, 298, 301, 302, 303, 308, 309, 311, 316, 318 Sorghum almum, 49-50,51,54,57,58-59, 65 Sorghum halepense, 49 Speargrass, 44, 52 Stargrass, 5 , 23, 46, 53, 63 Strawberry, 221 Strawberry clover, 30, 3 1 Stylo, 7, 9-12 Stylosanthes guyanensis, 7, 9, 11, 29, 41, 44,63 Stylosanthes humilis, 6, 9, 4 1 Sugarcane, 287, 292, 294, 296-298, 301, 302, 306-309,311, 314, 315-316

462

SUBJECT INDEX

Sulfur, 23 Sweet corn, 221

U

Urochloa sp., 7,50 T V

Tall fescue, 353 Vermiculite, 92,99 Tillering in grasses, 301-303 Vigna luieola, 20 Timothy, 335,342 Tomatoes, 22I W Townsville lucerne, 6 Townsville stylo, 6,7,9-1I , 23.24,29,30, Water, bound, 212-215 31,32,37,40-41,52,61 Wheat, 165,221,279,332,342,382 Transpiration, 174,183 Wheat pasture poisoning, 332 s-Triazines, 99 White clover, 21,22,24,28,29,32,48,64 2-(2,4,5-Trichlorophenoxy)propionic acid,

133 Trifolium pratense, 126,407 Trifolium repens, 2I , 407 Trifolium semipilosum, 9, 22,39 Triiicum aestivum, 165, 332,342 Triiicum vulgare, 388

Y

Yellow nutsedge, 132 2

Zea mays, 192,341