Advances in Agronomy, Volume 26

ADVANCES IN

AGRONOMY VOLUME 26

CONTRIBUTORS TO THIS VOLUME

HERMAN BOUWER

K. 0. RACHIE

R. L. CHANEY

L. M. ROBERTS

M. E. HARWARD

C. W. STUBER

SHERWOOD B. IDSO

B. R. TRENBATH

R. H. MOLL

KOJI WADA

F. J. ZILLINSKY

ADVISORY BOARD

w. L. COLVILLE, CHAIRMAN

(1973)

G. W. KUNZE(1973) D. G . BAKER(1974) D. E. WEIBEL(1974) G. R. DUTT (1975) H. J. GORZ(1975) N. c. BRADY, EX OFFICIO M. STELLY,EX OFFICIO ASA Headquarters

ADVANCES IN

AGRONOMY Prepared under the Auspices of the

AMERICAN SOCIETYOF AGRONOMY VOLUME 26

Edited by N. C. BRADY International Rice Research Institute Manila, Philippines

1974

ACADEMIC PRESS

New York

San Francisco

London

A Subsidiary of Harcourt Brace Jovanovich, Publishers

COPYRIGHT @ 1974, BY ACADEMIC PRESS,INC. ALL RIGHTS RESERVED. NO PART OF THIS PUBLICATION MAY BE REPRODUCED OR TRANSMITTED IN ANY FORM OR BY ANY MEANS, ELECTRONIC OR MECHANICAL, INCLUDING PHOTOCOPY, RECORDING, OR ANY INFORMATION STORAGE AND RETRIEVAL SYSTEM, WITHOUT PERMISSION IN WRITING FROM THE PUBLISHER.

ACADEMIC PRESS, INC. 111 Fifth Avenue, New

York, New York 10003

United Kingdom Edition published by ACADEMIC PRESS, INC. (LONDON) LTD. 24/28 Oval Road, London N W l

LIBRARY OF

CONGRESS CATALOG CARD

NUMBER:5 0-55 98

ISBN 0-12-000726-6 PRINTED IN THE UNITED STATES OF AMERICA

CONTENTS

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

iX

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

xi

CONTRIBUTORS TO

VOLUME 26

GRAIN LEGUMES OF THE LOWLAND TROPICS

K . 0. RACHIEAND L . M . ROBERTS I . Importance and Production

...................................... .................................................... Peanuts ...................................................... Pigeon Peas .................................................. Cowpeas .................................................... Mung Beans .................................................. Secondary Species ............................................. Conclusions .................................................. References ...................................................

I1. Botanical

.

111

IV. V. VI . VII . VIII.

2 7 11 32 44 62 77 91 118

LAND TREATMENT OF WASTEWATER

HERMANBOWER

. . 111. IV. I I1

AND

R . L. CHANEY

Introduction ................................................... Fate of Wastewater Constituents in Soil ............................ Crop Response ................................................. Selection and Design of System . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

133 135 164 167 169

BIOMASS PRODUCTIVITY OF MIXTURES

B. R . TRENBATH I. I1 111 IV

. . . V. VI.

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Comparison of Yields of Mixtures and Monocultures . . . . . . . . . . . . . . . . Theoretical Considerations ....................................... Types of Interaction Causing Nontransgressive Deviations of Mixture Yields from Mid-Monoculture Values ............................. Mechanisms Capable of Causing Transgressive Yielding by Mixtures . . . Conclusions ................................................... References .................................................... V

177 179 183

186 196 205 206

vi

CONTENTS

AMORPHOUS CLAY CONSTITUENTS OF SOILS

KOJI WADA I. I1 111. IV. V VI VII ..

.

. .

AND

M . E. HARWARD

Introduction .................................................. Definition and Scope . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Nature of Materials . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Identification and Quantitative Estimation ......................... Formation and Transformation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Relationship to Soil Properties . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Summary ..................................................... References ....................................................

THE CALIBRATION AND USE

211 212 213 230 233 242 253 254

OF NET RADIOMETERS

SHERWOOD B. IDSO

. . . . .

I I1 111 IV V. VI

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Instruments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Calibration Methods ............................................ Utilizing the Basic Net Radiometer ................................ Modifications for Different Applications ............................ Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References ....................................................

261 262 263 268 269 272 272

QUANTITATIVE GENETICS-EMPIRICAL RESULTS RELEVANT TO PLANT BREEDING

R . H. MOLL AND C . W. STUBER

I . Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Genetic Variability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Inbreeding Depression and Heterosis .............................. Genotype-Environmental Interactions .............................. Response to Selection ........................................... Implications of Quantitative Genetics to Breeding Methodology . . . . . . . . References ....................................................

I1 I11 IV. V. VI .

277 278 284 287 295 305 310

THE DEVELOPMENT OF TRlTlCALE

F. J . ZILLINSKY I . Historical Review .............................................. I1. Breeding and Research in Eastern Europe ..........................

315 318

CONTENTS

. .

I11 IV. V VI.

Breeding and Research in Western Europe . . . . . . . . . . . . . . . . . . . . . . . . . Breeding and Research in North America . . . . . . . . . . . . . . . . . . . . . . . . . . Triticale Improvement at CIMMYT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Recent International Developments ................................ References ....................................................

SUBJECTINDEX ......................................................

vii 322 324 326 338 346 349

This Page Intentionally Left Blank

CONTRIBUTORS TO VOLUME 26 Numbers in parentheses indicate the pages on which the authors' contributions begin.

HERMANBOUWER( 1 3 3 ) , U S . Department of Agriculture, Agricultural Research Service, U S . Water Conservation Laboratory, Phoenix, Arizona R. L. CHANEY(133), US. Department of Agriculture, Agricultural Research Service, Biological Waste Management Laboratory, Beltsville Agricultural Research Center, Beltsville, Maryland M . E. HARWARD (21 1 ), Soil Science Department, Oregon State University, Corvallis, Oregon SHERWOOD B. IDSO(26 1 ), U S . Department of Agriculture, Agricultural Research Service, US. Water Conservation Laboratory, Phoenix, Arizona R. H. MOLL(277), Department of Genetics, North Carolina State University, Raleigh, North Carolina K. 0. RACHIE( 1 ), International Institute of Tropical Agriculture, Ibadan, Nigeria, and The Rockefeller Foundation, New York, New York L. M . ROBERTS ( l ) , The Rockefeller Foundation, New York, New York C. W. STUBER(277), Department of Genetics, North Carolina State University, and US. Department of Agriculture, Agricultural Research Service, Raleigh, North Carolina B. R. TRENBATH ( 177), Waite Agricultural Research Institute, University of Adelaide, Adelaide, South Australia' KOJIWADA(21 1 ), Kyushu University, Fukuoka, Japan F . J. ZILLINSKY ( 315), International Maize and Wheat Improvement Center ( C I M M Y T ) , Mexico City, Mexico

* Present address: Research School of Biological Sciences, Australian National University, Canberra City, Australia. ix

This Page Intentionally Left Blank

PREFACE

Agronomy has again emerged in the eyes of the world as an important profession. The world food production problem has again reared its ugly head and statesmen and laymen alike are looking to crop and animal production scientists for answers to food production problems. The droughts and floods of 1971 and 1972 markedly reduced the supplies of food grains throughout the world. This resulted in unprecedented increases in prices for wheat, corn, and rice and drastically affected the cost of all food products. It also brought to the attention of even the more affluent nations, the grim reality of an ever-present threat of world-wide food shortage. The world once again has been reminded that food production along with population control are mankind's two most serious long term problems. Agronomists are playing a critical role world-wide to help solve these problems. Volume 26 continues the focus of its immediate predecessors in reviewing research concerned with food production. An extensive review of work on edible legumes of the humid tropics illustrates this orientation. Likewise, the paper on the development of the wheat-rye cross, triticale, reviews an important long-range research effort on a new and exciting crop. The review of the biomass productivity of mixtures is significant, not only as it relates to pastures and forages but as it impinges on cropping systems generally. There is increased interest in food crop combinations and sequences which will maximize annual production on limited land resources. The soil as a recipient of municipal and other wastes is given attention in this volume along with articles dealing with more fundamental aspects of soil characteristics, crop improvement, and the measurement of climatic variation. These articles illustrate the variety of research efforts coming from the fertile minds of the world's crop and soil scientists. Their ingenuity will be taxed in the years ahead to provide the knowledge needed if man is to continue to feed himself. N. C . BRADY

xi

This Page Intentionally Left Blank

GRAIN LEGUMES OF THE LOWLAND TROPICS K. 0.Rachie*t and 1. M. Robertst * Infernational

Institute of Tropical Agriculture, Ibadan, Nigeria, ond

t The

Rockefeller

Foundation, New York, New York

I. Importance and Production . . . . . . . . . . . . . , A. The Protein Shortfall .. ... . . . . . . _ . .. B. Worfd Production . . . . . . . . . . . . . . . . . . . , . . . . . . . . . .......... 11. Botanical . . . . . . . , .. . . . . . . . . , . . . .. . . . . . . , . A. Taxonomy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B. Comparative Ecology . . . . . . . . . . . . . . , . . . . . . . . . . . . . . . . . . . . . . , .. 111. Peanuts . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . ............ A. Botanical . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B. Plant Improvement . . . . . . . . . . . .. . . . . . . C. Plant Protection . . . . . . . . . . . . . . . . . . . . . . . . .... . .* . ... D. Growth Process . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . E. Management . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . F. Chemical Composition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . G. Potential . . .. . ... ............................ IV. Pigeon Peas . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A. Importance . . . . . . . . . . . , . . . . . . . .. . . . . . . . .. ., . . .. . . ......... ... B. Plant Improvement . . . . . . . . . . . . . . . . . . . . C. Plant Protection . ... . .. . .. . . . . . .. .. . .. . ...... . . . . . . .. . .. .... D. Physiology and Management . . . . . . , . . . . . . . . . . . . . . . . . . . . . . . . . . . E. Potential . . . . . . . . . . . . . . . . . . . .. . , . . . . . . . . . . . . . . . . . . . . . . . V. Cowpeas . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A. Description and Importance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B. Plant Improvement . . . . . . . . . . . . . , . . . . .. . . . . . . . . . . . .. . . . . . . . . . C. Insect Pests . . . . . . . . . . . . . . . . . . . . . . . . . . D. Diseases and Nematodes ................................ E. Physiology . . . . . . . . . . ........................ ........................ F. Management . . , . . , . . . G . Utilization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ........................ VI. Mung Beans . . . . . . . . . . . ............... A. Importance and Utilization . . . . . . . . . . . . . . . . B. Description and Varieties . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . C. Plant Improvement . . . . . . . . . . . . . , . . . . . . . . . . . . . . . . . . . . . . D. Plant Protection . . . . . . . . . . . . . . . . . , . . . . . . . . . . . . . . . . . . . . . . . . . ..................... E. Physiology . . . . . . . . . . . . ~. . . . . . . . . . . ...... F. Management . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..................... G. Chemical Composition . . . . . . . . . . . . . . H. Potential . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .... VII. Secondary Species . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A. Semiarid Lowland Tropics . . . . . . . . . . . . . . . . . . . . . . . . . . ...... 1

4

8 10 11 12

20 24

29 31 31 32 32 34

38 40 44 44 45 46 52 54

58 60 62 62

68 70 75 76 77 77 78

2

K. 0. RACHIE AND L. M. ROBERTS

B. Subhumid Tropics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . C. Humid Tropics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . D. Very Humid Tropics ........................................ VIII. Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Appendix: Tables . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References ....................................................

I.

80 83 86 91 97 118

Importance and Production

Grain legumes form a major component of lowland tropical cropping systems. Several species are utilized throughout the wet and dry tropics both in monoculture and complex multiple cropping and bush-fallow practices. More than two dozen species are grown to lesser or greater extent depending on the specific uses required of each, but all share similar desirable features. The universal ability to grow vigorously under a wide range of environments and on poor soils without supplemental nitrogen is particularly advantageous in subsistence agriculture in remote areas. The quick growth of some annuals like cowpeas and dry beans, the high consistent productivity of soybeans and peanuts, and the extended fruiting habit of long duration viny species (yam, lima, and velvet beans) and woody perennials (pigeon pea, jack bean, and locust bean) are complementary advantages in complex bush-fallow farming systems. Legumes have several advantages over other food plants in their simplicity of preparation and multiplicity of edible forms, such as tender green shoots and leaves, unripe whole pods, green peas or beans, and dry seeds. Some species, for example, the Mexican yam bean, produce edible tubers in addition to the fruit, and the winged bean is reputed to be utilizable as seedlings, tender green leaves, green pods, dry seeds, and tubers. The excellent nutritional values of most legumes in terms of proteins, calories, vitamins, and minerals are highly complementary in tropical diets comprised of roots and tubers, plantains, cereals, indigenous vegetables, fruits, and minimal animal proteins. Legume seed proteins are also the least exgensive, most easily stored and transported, nonprocessed proteinaceous food concentrate for both rural and urban utilization. A.

THE PROTEINSHORTFALL

Plant sources contribute about 70% of the world’s protein needs, but in many developing countries in the torrid zones this proportion can be even higher, up to 90%. Cereals contribute two-thirds of all plant proteins consumed directly; grain legumes, 18.5% ; and other sources (roots, tubers, nuts, fruits, and vegetables), 13.5%. The production of plant proteins from

GRAIN LEGUMES OF THE LOWLAND TROPICS

3

all sources in 1968 was 153.8 millions of metric tons, or 43.0 kg per capita, but developing countries of the Far East had only 24.1 kg and Africa only 26.0 kg per capita (Tahir, 1970).

1 . Nutrition and Climate Human nutrition often seems to deteriorate proportionately with the decline in elevation and increase in mean annual rainfall. In Nigeria the availability of both protein and caloric energy decreases from the drier north to the subhumid west and humid southeast. In a survey carried out in 1963-1964 by F A 0 (1966), it was found that both energy (2719 cal per day) and proteins (80 g per day) were adequate in the semiarid northern region, but were below recommended nutritional levels in the west (1909 cal and 40 g protein per day), and in the southeast (1774 cal and 33 g of protein per day). Whereas cereals comprised 64% of caloric intake in the north, roots and tubers made up 53 and 68% of the energy sources in the west and east, respectively. The effects of unbalanced nutrition in areas where energy sources may be adequate can be more dramatic. In some humid and subhumid intermediate elevations like Uganda ( 1000-1 200 m) where carbohydrates are more than adequate (3000-4000 cal per day), but proteins are inadequate, there may be as many as five malnourishment deaths per 1000 population-primarily in postweaning children. It might appear that semiarid lowlands and higher elevations with lower population pressures, and where cereals and pulses are more easily cultivated and stored, are better off nutritionally, except that statistics seldom reflect the vulnerability of subhumid and semiarid regions to vagaries of the climate and cyclical famines, which have tended to hold the populations down in the first place. The acute famines in West Africa and Southern Asia in 1972-1973 illustrate this problem. Most vulnerable are those segments of the agricultural society-including nomadic graziersprimarily dependent on domestic animals for their livelihood since they exploit the most arid, and hence, climatically volatile, regions. When a drought continues for more than one season, they begin’ losing their younger, breeding stock, and recovery may require several years. 2 . Constraints

Tropical grain legumes have evolved under high stress conditions or are not genetically capable of responding to favorable growing conditions, and therefore they do not attain reasonable yield levels and good product quality under high temperatures and extreme moisture conditions. In this situation, survival even at low productivity levels is probably more important to both the plant and the peasant cultivator than high yields.

4

K. 0. RACHIE AND L. M. ROBERTS

a. Hazards. Among many hazards limiting productivity in the tropics are pests, diseases, moisture extremes, high temperatures, low insolations, inadequate or unbalanced plant nutrients, and poor soil conditions. These problems may be exacerbated by inefficient plants types with low yielding potential, susceptibility to insects, nematodes, and diseases, soils with extreme pH levels, poor physical structure, and depleted fertility, and poorly distributed rainfall. When it is not possible to relieve these constraints through better management, such as pest control, it may be essential for the plant to have resistance or genetic escape mechanisms like the slowing or cessation of growth processes during dry periods, deep rooting habit, indeterminacy, and photoperiod sensitivity. b. Utilization. Most grain legumes have some specific nutrient deficiencies like the sulfur-bearing amino acids, or contain certain undesirable offflavors, flatus factors, metabolic inhibitors, alkaloids, and other toxic substances. Nevertheless, some otherwise well-adapted, high-yielding and nutritious species are not utilized as a consequent of ignorance or unfamiliarity with their culture and methods of preparation. For example, soybeans with 2-3 times the yielding potential, 60% more protein and 20 times more oil than indigenous legumes have not been accepted in the African tropics in spite of their repeated introduction since the 1920’s. Major deterrents are primarily unfamiliarity with production practices and utilization. However, high world demand for this commodity and urgent need for vegetable oils and animal feedstuffs is providing considerable incentive for increasing tropical soybean production-first, as a cash crop for export and industry, and later for domestic use. B.

WORLDPRODUCTION

Production of tropical food legumes is highly complex owing to the density and distribution of population, climatic/environmental considerations, large numbers of species involved, and inadequacy of available information. Therefore, a cursory analysis has been made on grain legume production and population in tropical regions based on information in Volumes 24 and 25 of the F A 0 Production Yearbook (1971-1972) in order to gain perspective on the problems involved and establish priorities in pulse improvement programs. In this analysis the tropics are defined as countries with the greater part of their territories lying between the Tropics of Cancer and Capricorn. Thus, in the Americas, Mexico is included on the North, but Argentina, Chile, and Uruguay are omitted in the south; in Africa, countries north of the Sahara, and South Africa are omitted; and in southern Asia, India is included, but Pakistan, Bangladesh, Taiwan, and Australia are omitted (Rachie, 1973; Rachie and Silvestre, 1974).

GRAIN LEGUMES OF THE LOWLAND TROPICS

5

1 . Populations in the Tropics

In 1970 approximately 1.36 billion people or 36.6% of the world’s population lived in the tropics. The vast majority, or about 24%, are in Southern Asia, with India making up nearly two-thirds of the 848.5 million people in that region. Tropical Africa and Latin America contribute almost equally to the remainder-260 million (7.2%) and 240 million (6.7% ), respectively.

2 . Production Trends The worldwide production of all grain legumes increased by 49.1 % in area and 103.4% in production between 1948-1952 and 1971. This represents a proportionately greater increase than for cereals and roots and tubers during the same period. It is further observed that 111.6 million metric tons of grain produced on 117.5 million hectares was about 23% above the production for 1961-1965. However, a considerable proportion of this increase (almost 60% ) is attributable to the rapid expansion of soybean cultivation in North America. Further increases are anticipated in 1972 and 1973. Preliminary estimates for 1973 project soybean production at 52.8 million tons on 38.3 million hectares. This would increase total world grain production by 4.12 million tons (3.7% ) to 115.7 million metric tons, allowing for a decline of 412,000 tons of peanuts and dry beans in that year. Producing Regions. Among tropical regions (all elevations) in the early 1970’s, southern Asia contributed 20 million tons of dry grain on 33 million hectares, while tropical Africa and Latin America harvested about 8 million tons each on 12 and 10 million hectares, respectively. Increases in estimated grain legume production in the intermediate/high versus lowland tropics for three separate periods over a 22-year period are presented in Appendix Table I. Production increased by 47.5% in area and 89.7% in tonnage for all elevations between 1948-1952 and 1971. Lowland tropical legumes increased by about two-thirds between 1948-1952 and 1961-1965; and to 190% of 1948-1952 yields by 1971, when production attained 21.6 million metric tons on 33.3 million hectares. Chick-peas (5.7 million tons) and dry beans (5.5 million tons) constituted two-thirds of total pulse production at intermediate and high elevations whereas peanuts (13.0 million tons in shell; or 8.7 million tons kernels) comprised about 40% of all lowland tropical grain legumes in 1971. Pigeon peas were probably the most important lowland pulse, with nearly two million metric tons of estimated production; although the Asian grams collectively were higher (2.5 million tons) in 1971. Proportionately, soybeans increased more rapidly at intermediate to high elevations record-

6

K. 0. RACHIE AND L. M. ROBERTS

ing a 5-fold increased production between 1961-1965 and 1971. At low elevations, cowpeas more than doubled in production between 1948-1 952 and 1961-1965 and by 2.5 times by 1971. The Asian grams increased similarly in the lowland tropics by reaching 2.3 times their 1948-1952 production in 1970. 3. Distribution of Species

More than a dozen species contribute to the production of grain legumes in tropical regions. Of these, dry beans (Phaseolus vuZguris), chick-peas (Cicer arietinum) , some of the soybeans (Glycine max), dry peas (Pisum spp. ) , lentils (Lens esculenta) , and broad beans (Vicia faba) are clearly cool weather and, hence, intermediate-to-high elevation species and are so classified. Similarly, pigeon peas (Cajunus Cajun Millsp. ), cowpeas (Vigna unguiculata Walp.) , peanuts (Arachis hypogaea) , and the Asian grams (mung beans, black gram, rice beans, hyacinth bean, moth, and others included in the “dry beans” category for southern Asia) are usually grown at lower elevations. However, soybeans do occur in both ecologiesat least in southern Asia. Most of the important lowland legumes perform better in the subhumid to semiarid tropics, as evidenced by results and experience in East and West Africa. Among the better known species, pigeon peas seem to occur over a wider range of moisture conditions, while soybeans may have greater tolerance for wet soils than do cowpeas and groundnuts. This implies that grain legumes are not planted as extensively and other protein sources are utilized or available statistics do not accurately reflect the true situation. It is suggested that all three assumptions apply to varying degrees and that per capita intake of proteins is often much lower in humid than in semiarid tropical regions. However, it is also becoming evident that several less familiar species other than those mentioned above are utilized in the humid tropics but are not accounted for in production estimates. Some of these will be described and discussed further in the following sections (see Appendix Table 11). a. Peanut-Producing Regions. Peanuts are more important than all other lowland tropical legumes combined, contributing 13 million tons in shell (about 67% seeds) or about 40% of the total production on the basis of net seed weights. However, peanuts are mainly grown as a cash crop for industrial processing of oil and cake, rather than for direct consumption. Therefore, pigeon peas, cowpeas, and mung beans may contribute more directly.to human diets even in areas where the peanut is a major crop. Several countries, led by India, contributed 54.5% of the world crop and 77.4% of the tropical production in 1971. In southern Asia, India (31.4%), Indonesia ( 2.6%), and Burma (2.8%) contributed 36.8%; in

GRAIN LEGUMES OF THE LOWLAND TROPICS

7

Africa, Nigeria (6.0%), Senegal (5.2%), and Sudan (1.9%) made up 13.1% ; and, in Latin America, Brazil produced 4.6% of the world crop. Between 1961-1965 and 1971, total production in the Americas increased by 36.6%, in Africa by 7.8%, and in Asia (omitting mainland China and the USSR) by 18.0%. b. Pigeon Peas. India produced 1.84 million metric tons of dry grain on 2.65 million hectares in 1971 for 93% of the world crop. Other producers were Uganda (2.0%) and Malawi (1.0%) in Africa, Burma ( 1.4 % ) , and Dominican Republic ( 1.1% ) . Considering unreported and “kitchen garden” plants for home use, these statistics may be underestimated by as much as 10-15%, thereby increasing total world production to as much as 2.25 million metric tons. c. Cowpeas. Africa produced 94.8% of the world crop of 1.14 million metric tons in 1971. Major growing countries were Nigeria (61.2%), Niger (13.1%), Upper Volta (7.4%), and Uganda (5.5%). However, it is quite possible this production was underestimated by 10-15 % considering unreported and “kitchen garden” plantings. This would increase world production by as much as 170,000 metric tons to 1.31 million tons. d. Asian Grams. Mung beans or green gram and close relatives-black gram, yellow gram, rice bean, and moth bean, which have recently been reclassified as Vigna species (Verdcourt, 1970) and possibly horse gram (Dolichos biflorus) , hyacinth, or field bean are presumed reported under “dry beans” in statistical reports. The worldwide tropical production of tGse species is estimated at 2.7 million metric tons on 8 million hectares, of which probably 80% is grown in India where black gram (mash or urad) production is estimated at 0.44 million tons on 1.5 million hectares, green gram (mung) at 0.30 million tons on 1.4 million hectares and horse gram at 0.39 ton on 1.8 million hectares. e. Unspecified Commodities. The category “other and unspecified” species may include both common and less familiar species and are estimated at 80% for lowland tropics or 1.6 million tons from 3.3 million hectares. India is the main producer in this category, with an estimated 70% of the total. II.

Botanical

The food legumes are classified in the Order Leguminosae, and predominantly in the large Family Papilionoideae having 480 genera and 12,000 species, which are widely distributed in both tropical and temperate climates. However, a few economic species do occur in the second and third families of this order, Caesalpiniaceae with 152 genera and 2800 species, and Mimosaceae having 56 genera and 2800 species. The distinguishing

8

K. 0. RACHIE AND L. M. ROBERTS

features of legumes are the following: (1) Leaves are usually alternate and compound, pinnate or trifoliate. (2) Flowers are predominantly hermaphroditic and usually with five sepals and five petals. ( 3 ) Ovary is superior with a single carpel, cavity, and style. (4)Fruit is usually a pod formed by a single carpel and dehisces by both ventral and dorsal sutures into two valves. ( 5 ) Seeds consist of two cotyledons and an embryo containing very little endosperm. Papilionoideae is distinguished from the other two families primarily by the flower petals being imbricate (overlapping) with descending aestivation (order). The upper (adaxial) petal is exterior, usually largest, and forms the standard or vexillum. The two lateral petals are parallel, forming the wings or alae; and the two lowest petals are interior, usually joined by the lower margins to form the keel which encloses the stamens and ovary. There are normally ten stamens, and they may be either monadelphous (all united by filaments) or diadelphous with nine united stamens, the upper or vexillary stamen being free. The anthers have two locules and dehisce lengthwise by slits. The ovary is superior, consisting of one carpel, usually monolocular and sometimes with a false septum; the ovules may be one to many borne on the ventral suture (Purseglove, 1968).

A. TAXONOMY Papilionaceae is divided into twelve tribes, but nearly all of the economic grain legumes occur in VII, Phaseoleae. However, a few also occur in VII, Cicieae, and peanuts belong to IX, Hedysareae. The Phaseoleae may be herbs-erect, procumbent, or climbing; or subshrubs and even small trees. The leaves are pinnately foliate (rarely pentafoliate) and have a terminal leaflet; stipels are present, hairs are never medifixed, stamens are not broadened at the apex, and the ovary is surrounded by a disc. Hedysareae is distinguished from other tribes by having jointed fruits, constricted between the seeds and breaking transversely into one-seeded portions, and stipels are sometimes present (Hutchinson and Dalziel, 1958). Key to the Genera of Tropical Grain Legumes

A simplified key to the warm weather lowland tropical legumes has been prepared and modified after Hutchinson and Dalziel (1958) and Purseglove (1968). Members of the pea family Pisum, Cicer, Vicia, Lens, and Lathyrus) are omitted as being cool season plants and confined mainly to intermediate and higher elevations or as winter crops in the subtropical and temperate regions. In this classification, the old world Asian grams (Phaseolus mungo, P . aureus, P . radiatus, P . acontifolius, P . angularis, and P. calcaretus) have all been transferred to Vigna Savi on the basis

GRAIN LEGUMES OF THE LOWLAND TROPICS

9

of extensive taxonomic studies on foliage morphology, flower structure, pollen grain sculpture, serological tests and electrophoretic analysis of seed extracts as proposed by Verdcourt ( 1970). An adaptation of taxonomic keys to these lowland tropical species is outlined below: A. Fruits ripening underground B. Leaves pinnate with four leaflets; leaflets without stipels; stamens monadelphous, flowers axillary and solitary; jointed fruits constricted between seeds . . . . . . . . ..................... Arachis BB. Leaves trifoliate, not gland dotted; texillary stamens free from near base upward; style bearded; calyx with short broad teeth . . . . . . . . Voandzeia C . Style glabrous; calyx deeply divided into narrow lobes ............................................. Kerstingiella AA. Fruits ripening above ground B. Leaves trifoliate C . Vexillary stamen free from base upward E. Keel of corolla and style coiled through 360" (1-5 turns) ; pollen grains with no obvious sculpture; standard with transverse groove at the top of the claw usually without appendages (but sometimes two) ; fruit . . . . . . . . . . . . . . . . . . . . . Phaseolus or curved but not coiled more than 360"; stipules cordate or appendaged below base; pollen grains strangly reticulated F. Stigma strongly oblique or introrse; roots not tuberous . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Vigna FF. Stigma on inner face of style, subglobose roots tuberous . . . . . . . . . . .Pachyrrhizus D. Style glabrous, has t E. Keel and style bent inward at right angles, beaked F. Stigma surrounded by a ring of hairs . . . .Doliclros DD. Style bearded down one side; stigma without ring of hairs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .Lablab F. Stigma may have a ring of hairs, is laterally oblique, hooded or flattened and broad, more or less spatulate but not appendaged; keel not twisted; stems twining or erect . . . . . . . . . . . . . . Sphenostylis BB. Trifoliate leaves gland-dotted underneath; lanceolate-oblong C . Flowers yellow or orange, borne in subcapitate axillary racemes; vexillary stamen free from near base upward; ovules more than four; fruit obliquely subtorulose; erect, perenniating shrubs

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

Cajanus

BBB. Trifoliate leaves not gland-dotted underneath C . Vexillary stamen free from near base upward E. Bracts and bracteoles small and inconspicuous caducous: F. Keel longer than the standard petal; fruit hispid usually with stinging hairs; flowers in zigzag racemes, short racemes, or somewhat umbellate .................................... Mitcuna FF. Keel shorter than standard petal

10

K. 0. RACHIE AND L. M. ROBERTS

CC. Style glabrous; calyx four lobed, upper lobe entire or shortly twotoothed; nodes of raceme not swollen; standard mainly pubescent; .Glycine very small flowers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . D. Vexillary stamen united in upper part with others, free below E. Fruit square, four winged, 5-6 seeded; leaves 1-3 foliate, herbaceous climber .................... Psophocarpus EE. Fruit not winged; many seeded; trifoliate CCC. Nodes of raceme swollen; apex of fruit not hooked; stamens all fertile D. Calyx lobes unequal in size, upper two rounded and larger than lower three, fruit broad, furrowed along the upper suture . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Canavalia E. Fruit 1-3 seed; nodes of raceme swollen; woody climber . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Dioclea

B. COMPARATIVE ECOLOGY There are two major classes of tropical grain legumes: ( 1 ) Leguminous oilseeds, mainly peanuts and also soybeans, grown primarily in southeast Asia; and (2) pulses-pigeon peas, cowpeas, and mung beans/black gram. A secondary group of crops includes several species with localized use, undetermined potential or unavailable production estimates. Chief among the secondary category are hyacinth bean (Lablab niger), horse gram (Dolichos biflorus), lima beans (Phaseolus lunatus), yam beans (Sphenostylis stenocarpa), rice beans (Vigna umbellata), moth beans (Vigna acontifolia),and velvet beans (Mucuna spp.) In terms of adaptation, these warm weather lowland species could be classified in the following categories (asterisk indicates major species) : I. Semiarid regions-annual

precipitation less than 600 mm 1. Short duration cowpeas (Vigna unguiculata) * 2. Short duration groundnuts (Arachis hypogaea) * 3 . Bambarra groundnuts ( Voandzeia subterranea) 4. Moth bean (Vigna acutifolia) 5 . Horse gram (Dolichos biflorus) 6. Cluster bean (Cyamopsis tetragonolobus) 11. Semiarid to subhumid regions-600-900 mm precipitation 1. Groundnuts-medium and long duration* 2. Cowpeas-medium and long duration* 3. Pigeon peas (Cajanus Cajun)* 4. Mung beans (Vigna radiata var. aureus; var. mungo) * 5. Hyacinth bean (Lablab niger) 6 . Horsegram (Dolichos biflorus) 111. Subhumid to humid regions-900-1 500 mm precipitation 1. Pigeon peas-medium and long duration* 2. Cowpeas-medium and long duration* 3 . Mung beans-medium and long duration* 4. Lima beans (Phaseolus Zunatus) 5 . Haricot beans (Phaseolus vulgaris) 6. Soybeans (Glycine man)

GRAIN LEGUMES OF THE LOWLAND TROPICS

11

IV. Humid and very humid regions-above 1500 mm precipitation 1. Lima beans-dimhing types 2. Yam beans (Sphenostylis stenocarpa) 3. Rice beans (Vigna umbellata; syn. P . calcaretus) 4. Velvet beans (Mucuna pruriens var. utilis and M . sloanei) 5. Pigeon peas-medium and long duration*

A precise definition of an ecology is difficult inasmuch as several factors besides mean annual rainfall are involved including: (1) rainfall pattern (bimodal or monomodal), (2) moisture distribution, (3) temperatures and cloud cover, (4)relative humidity, ( 5 ) soil moisture holding capacity, ( 6 ) soil fertility and physical structure, (7) prevalence of diseases and pests, and ( 8 ) interaction of the species genotype with the total environment. The range of genetic diversity within species is often considerable, sometimes exceeding variability between species. Characteristics like resistance to pests and diseases, quick germination, rapid growth, earliness, tolerance of high temperatures, deep rooting, indeterminancy, day-length sensitivity, yielding potential, and other heritable factors have profound influences on fitness for specific ecological situations. Other aspects must be considered in assessing adaptation. The first is human preference and needs. Often a cultivator will grow a low-yielding, poorly adapted species and cultigen because he prefers its taste, requires the crop for some specific use, or has a cash market for its produce. Moreover, characterization of a particular ecological zone is based on long-term weather records. Therefore, fluctuations in “normal” patterns could result in successful cultivation of otherwise poorly adapted species or cultigens a certain proportion of the time, such as two years out of three seasons out of five. In practice two or more crops are frequently grown in a mixture established after long experience and specific needs, some of which will succeed-although not always the preferred ones. In other situations the grower might wait until the season is underway, or, based on preseason showers, plant more exacting, longer duration species and varieties. In spite of the broad-range genetic diversity and adaptation within species, certain generalities can be assumed regarding botanical characteristics, tolerance of variable stresses, genotype X environment interactions and utilization. These are outlined for 16 genera and 24 species under four distinct ecological zones of the lowland tropics in Appendix Table 111 (Rachie, 1973) . Ill.

Peanuts

There are two important leguminous oilseed crops: peanuts (Arachis hypogeu L.) and soybeans (Glycine max Merr.). Peanuts are of major importance in the lowland tropics, comprising an estimated 60% of all

12

K. 0. RACHIE AND L. M. ROBERTS

tropical grain legumes. In contrast, soybeans have a very minor role in lowland tropics, primarily in southeastern Asia. In Africa, not more than 50,000 tons of soybeans are produced annually in tropical areas as a consequence of lack of an established demand or preference for them as food and some basic problems of management. Nevertheless, the soybean demonstrates exceptional potential for the lowland tropics, and there is increasing demand for industrial protein sources for both human and animal nutrition in developing tropical countries. It is therefore highly likely that increasing emphasis will be placed on adapting and improving this crop for the lowland tropics. A.

BOTANICAL

There are only about 19 species of Arachis indigenous to tropical and subtropical South America from the Amazon through Brazil, Uruguay and Argentina to about 3 5 O south. The cultigen A . hypogaea L. has 2n = 40 chromosomes and is unknown in the wild state; the other species have 2n = 10 chromosomes, are wild and perennial, being used commercially only for forage. All species ripen their fruits underground (Purseglove, 1968). The Portuguese probably introduced the peanut to the west coast of Africa directly from the Caribbean region early in the sixteenth century, while the Spanish brought it from the west coast of Mexico to the Phillippines from whence it spread to Asia, Madagascar, and East Africa (Rachie and Silvestre, 1974).

I . Ecological The highest yields of good quality groundnuts are obtained on well drained, light, sandy-loam soils with a pH above 5.0. Dark soils tend to stain the hulls, and heavy, clayey soils may become too waterlogged to allow optimum growth, or too hard for penetration of pegs (gynophores) and digging to harvest the crop. The most favorable climatic conditions are moderate rainfall during the growing season (annually 1000-3000 mm), plenty of sunshine, and reasonably high temperatures. The heaviest demand for moisture is from the beginning of blooming up to 2 weeks before harvest. However, it should be emphasized that peanuts are not well adapted to the more humid tropics (above 1300 mm) owing to the high incidence of diseases and pests, and other factors. 2 , Description and Classification

The peanut plant is a low-growing annual with a central upright stem readily separated into bunch and runner types. In bunch or erect types the nuts are closely clustered about the base of the plant, whereas the runner types have nuts scattered along their prostrate branches from base to

GRAIN LEGUMES OF THE LOWLAND TROPICS

13

tip. Several investigators, notably Gregory et al. (1951), Bunting (1955, 1958), Krapovickas and Rigoni (1960), Krapovickas (1968), and Gibbons et al. (1972) have contributed to the description and classification of the cultivated forms of Aruchis hypogaea. Distinction between races based on the distribution and ramifications of vegetative and reproductive branches are described as follows: A. Subspecies hypogaea Waldron: This includes the Virginia types. Inflorescences are simple and never borne on the main axis. The first bud of cotyledon axis is always vegetative, and branches have two vegetative and two reproductive buds alternatively and in succession. The main stem can be viny (runner types) or straight. In the latter form the plant presents a bushy appearance owing to the abundance of successive branches. 1. Var. hypogaea: either viny or erect; the main stem is short (less than 40-50 cm) in viny forms; branches are rarely hairy; duration is rather long. 2. Var. hirsura Kohler: erect with long main stem (more than 100 cm); branches are very hairy; it is very late; and susceptible to Cerospora leaf spots. B. Subspecies fastigiata Waldron: This group has sequential branches of the Spanish and Valencia types, but has few branches and the stem is always erect. Inflorescences always occur on the main stem, the first buds of the cotyledon axis are reproductive, and vegetative and reproductive buds succeed each other in an irregular series. 1. Var. fastigiara type Valencia: the branches arising from the main axis do not have branches or branch only at their extreme ends; the inflorescences are simple; and the pods contain 2, 3, or 4 seeds. 2. Var. vulgaris Harz type Spanish: the branches arising from the main stem do have irregular secondary branching; the inflorescences are complex; and the pods are two seeded.

The variety hirsuta is not widely grown, being mainly of botanical interest. Therefore, cultivated sorts can be classified as Virginia, Spanish, and Valencia. Although these have been extensively intercrossed and most advanced cultivars are intermediate, combining characters from all three forms, there appear to be certain associations of characters and categories of utilization, which can be described as follows: A. Virginia: primarily runners or spreading types; indeterminate requiring 120-1 50 days growing season; small leaflets and foliage dark green in color; seeds markedly dormant (1-12 months); two seeds per pod; testa is deep russet brown; resistant to Cercospora leaf spot and some resist rosette disease; seeds reach maximum oil weight before total dry matter during maturation (Schenk, 1971); fat is higher in unsaturated fatty acids than for Spanish and Valencia varieties (Verhoyen, 1960; Gillier and Silvestre, 1969); protein content is lower than sequentially branched forms. These types have higher productivity potential, but require better growing conditions; they produce large fruits extensively used as an edible form. B. Spanish-Valencia: sequentially branched, erect, bunch types; determinate, short season (90-110) days); leaflets larger than Virginia types, and foliage is lighter green; seeds are not dormant; pods have 2-6 seeds; Kernels have a wide range

14

K. 0. RACHIE AND L. M . ROBERTS

of size and testa color; they are highly susceptible to Cercospora leaf spots; and they are higher in protein content and saturated fatty acids than Virginia types. Valencia cultivars can be readily distinguished from Spanish types by having more seeds per pod (usually 3 or 4), and thicker stems with a light reddish or purple tinge.

Virginia varieties produce larger fruits and have higher yielding potential than the Spanish and Valencia types, but are also more exacting in their requirements of moisture, insolation, temperature, and fertility levels. Therefore, production of the large-seeded edible forms is often restricted to the more favorable growing conditions, such as those of Georgia (USA), China, and Senegal (south), ur to the irrigated areas of Egypt, Sudan, and Israel. B.

PLANTIMPROVEMENT

Improvement of the peanut has received considerably more attention than most other grain legumes except soybeans by virtue of its high oil content and industrial potential. Therefore, a substantial amount of information and literature is available on the crop. Since much of this knowledge is already available in various sources, including monographs, this section will deal primarily with some of the more recent literature pertinent to the lowland tropics. I . Breeding Methodology

a. Floral Development and Pollination. The peanut flowers over a 4-6week period beginning about 4-6 weeks after planting. One flower per inflorescence opens on a particular day, others opening successively from one to several days later. The flower bud reaches 6-10 mm in length 24 hours before anthesis, and pollination occurs at sunrise the following morning, coinciding with the time of petal expansion and occurring within the enclosed keel. There are two sterile stamens, four stamens with oblong anthers (dehisce first) and four shorter stamens with smaller globose anthers that elongate and dehisce later. The petals wither 5-6 hours after opening, and the calyx tube is shed, leaving the ovary and base of style forming the fruit. The fruit elongates by means of intercalary meristCm at the base of the sessile ovary forming a peg or carpophore. Cells at the tip of the ovary become lignified and conical in shape to facilitate penetrating the soil via geotropism to a depth of 2-7 cm. The pegs elongate rapidly, reaching a length of 15-16 cm within 7 days after flowering (Ono and Ozaki, 1971). The force exerted by peanut pegs has been measured at about 13 bars, but when grown in compacted soils crop yields are inversely proportiona1 to soil hardness (Underwood ei al., 1971 ) . After reaching maximum depth

GRAIN LEGUMES OF THE LOWLAND TROPICS

15

the ovary swells rapidly and seeas form (Purseglove, 1968). Seeds reach their maximum size 40 days after flowering, but may require 60 days in Spanish types and 80 days or more for full fruit development in Spanish and Virginia types, respectively (Lin et al., 1969). b. Vicinisrn. Peanut flowers are almost totally self-pollinating and frequently cleistogamous. However, some outcrossing does occur and has been recorded in Virginia types at between 0.01 and 0.55% in the United States (Culp et ul., 1968), 0.20% in Senegal (Mauboussin, 1968), 1.67% in Makulu Red Eastern Africa (Gibbons and Tattersfield, 1969), and up to 6.6% in Spanish types grown in Java (Bolhuis, 1951). c. Hand Crossing. Emasculation consists of removing the anthers the evening before dehiscence occurs. Crossing with desired pollen donors is carried out the following morning. Since numbers of F, seeds produced per cross are comparatively few, the F, may be Propagated vegetatively from cuttings. However, cuttings must be taken from lateral branchesparticularly in alternate branched (Virginia) forms-to assure production of inflorescences. d . Breeding Techniques. Early breeding consisted mainly of mass selection within indigenous and introduced germplasm pools. This was followed by pure line selection and recombination of desirable parents. More recently wide crossing, crossing FI’s, recurrent selection, and mutation breeding have been utilized to broaden genetic variability, to more rapidly effect breaking of linkages and increase additive gene action, The use of ionizing radiation has been studied in depth by Gregory (1956), Bilquez et al. (1964), and Patil (1968), while use of diethyl sulfate and other chemical mutagens was investigated by Shchori and Ashri (1970) and by Ashri ( 1972). Preliminary investigations on interspecific crossing within Aruchis species have recently been reported by Gibbons and Bailey (1967) in the 4 area of disease (Cercospora aruchidicolu) resistance; phylogenetic relationships by Raman and Sree Rangasamy (1972) and Raman (1973), and interspecific cross compatibility between A . hypoguea and other Arachis species by Smartt and Gregory (1967). A major but not insurmountable problem in interspecific crossing is that the cultivated A . hypoguea has 2n = 40 chromosomes compared with 2n = 20 chromosomes in the other species. 2. Genetic Considerations Inheritance and gene action have been studied in several important genetic characters. These may be classified into simply inherited characters, cytoplasmic effects and quantitatively inherited characters. In addition there is the problem of linkage and “blocks of genes” which tend to be

16

K. 0. RACHIE AND L. M. ROBERTS

passed on to their progeny and through generations as combinations of characters. This makes it difficult to recombine desired characteristics from different groups, for example, earliness of Spanish and Valencia types with rosette resistance from Virginia types, or to transfer nondormancy characteristics from Spanish to Virginia types (Mauboussin, 1966). Therefore, the backcrossing breeding method has not been successfully used in these and similar situations. a. Character Associations. Several investigators have contributed to information on character interrelationships in recent years. Some of these are summarized below : 1. Pod yield in erect types: positively correlated with pods per plant and number of vegetative nodes per secondary branch and negatively correlated with lateral spread (Sangha and Sandhu, 1970). 2. Pod yield in bunch types: positively correlated with number of primary branches/plant, 100-seed weight and number of pods/plant in that order (Sangha and Sandhu, 1970). Sanjeeviah et al. (1970) observed pod yield to be correlated with number of nodes up to 10 cm above the ground ( I = 0.97% ) and Lin et al. (1969) obtained a correlation with high shelling percent. 3. Pod yield in spreading types: correlated with number of pods per plant, 100-seed weight, number of primary and secondary branches and shelling percent (Sangha and Sandhu, 1970; Raman and Sree Rangasamy, 1970). 4. Shelling percent in interspecific crosses: in various crosses between A . hypogaea, A . glabrata, and A . villosa, positive correlations were observed with pod weight, kernel weight, and percent filled kernels (Ramanathan and Raman, 1968). In another cross between A . hypogaea and A . monticola, shelling percent was highly correlated with number of primary branches (Raman and Sree Rangasamy, 1970). 5 . Other associations with pod yield: Prasad and Srivastava (1968) found pod yield positively correlated with numbers of branches, leaves, nodes, flowering nodes, and pods per plant; and with 100-seed weight. 6 . Interrelationships among yield components : Merchant and Munchi ( 1971) in studies on erect cultivars found positive correlations between leaf length and leaf width, pod length and pod width, pod length and seed length, and between seed length and seed weight. Seed length was negatively correlated with shelling percent. Martin (1969) observed that seed weight was not positively correlated with oil content. b. Simply Inherited Characters. Economically important genetic factors controlled by one or a few genes include the following: (1 ) branching habit; (2) number of seeds per pod; ( 3 ) , cotyledon characters; (4) color of stem, leaf, foliage, and testa; ( 5 ) crimping of leaves; (6) absence of

GRAIN LEGUMES OF THE LOWLAND TROPICS

17

leaf petiole; (7) constriction of the pod; ( 8 ) resistance to rosette; (9) early maturity; (10) oil content. Some simply inherited characters are expressed or modified by geniccytoplasmic effects and have been studied by Ashri (1964, 1968, 1969). He proposed two plasmons designated “V4” and “others” as interacting with two genes Hb, and Hb, to produce the runner (trailing) or bunch (erect) growth habits. In “V4” plasmon Hb, and Hb, produce runners while either recessive condition produces as bunch habit, whereas in “others” plasmon the dominant alleles are additive and possibly complement ary. Hb Hb ,Hb,Hb?, Hb,H b ,Hb,h b:, and Hb,hb ,H b,Hb, produce runners and all other combinations give bunch plants. Ashri further proposed genetic control of growth habit operated through production of antigibberellins similar to abscissic acid. c. Polygenic Inheritance. Several economically important characters are complex and controlled by several genes acting quantitatively. These include seed yield, number of pods per plant, weight of pods, and size of leaves. The environment also interacts directly and profoundly with these characters. Therefore, the most rapid progress in breeding should be realized through population improvement and recurrent. selection under favorable growing conditions of moisture, fertility, temperatures, and soil conditions. d . Heritabilities and Coefficients of Variation. Heritability in some important botanical characters has been studied by several investigators. Asoka-Raj (1969) found high heritability estimates for days to flowering, number of leaves per main stem, pods per plant and 100-pod weight out of 14 characters studied. The coefficients of variation were high for days to flowering, pods per plant, and haulm weight per plant. Majumdar et al. (1969) likewise observed high genetic coefficients of variation for number of leaves, number of nodes, number of peg-bearing nodes, number of pod-bearing nodes, and length of pod. Heritability estimates in the broad sense ranged from 49.6% for pod yield to 98.6% for days to maturity. Numbers of branches, leaves, and nodes had particularly high genetic advance potential; whereas numbers of pods and pod yield had low heritability values. Oil content and shelling percent were both observed to be highly heritable characters; oil content was controlled by two pairs of genes; shelling percent was governed by a single pair of genes without dominance; and 100seed weight, which is closely correlated with shelling percent but not with oil content, was controlled by five pairs of genes, four of which had isodirectional effects in genetic studies carried out by Martin (1969). In heritability analysis of maturity as measured by light transmittance through the seed oil, Gupton and Emery (1970) obtained higher rates of genetic gain

18

K. 0. RACHIE AND L. M. ROBERTS

in later maturing groups and suggested that most rapid gains could be made by selecting for rapid maturation in later pegging segregants. e. Combining Ability. Recent studies on heterosis and combining ability were carried out by Parker et af. (1970) and Wynne et ul. (1970) in diallel crosses of six lines representing the three major groups Virginia, Spanish, and Valencia. Virginia X Valencia crosses had the highest combining ability. Estimates of general combining ability were significant for 8 out of 17 characters studied, and estimates of specific combining ability showed significance for 16 characters. Specific combining ability was greatest for yield and most fruit characters. Specific combining ability effects were greatest for leaves on mainstem (at 15 days) ,days to first flower, and petiole length (at 33 days). The greatest portion of genetic variance was associated with general combining ability. Heterosis for vegetative characters was greatest in Virginia X Valencia crosses, and for seed yields in Valencia X Spanish crosses. 3. Breeding Objectives There are several important breeding objectives being actively pursued in the important peanut growing regions of the world, depending on their relative importance in that locality. Overriding all other considerations is yielding potential and stability of productivity over a reasonably broad range of ecological conditions. The most important specific breeding objectives of present plant improvement programs are briefly summarized below: 1. Yield potential- a highly complex character with very high genotype X environment interactions. Yield and stability of yield must remain at the highest level of priority for all breeders. 2. Oil content- Virginia types have up to 47% or more oil contents, but higher IeveIs of unsaturated fatty acids than Spanish-Valencia types, and crosses between Virginia and Spanish types have recorded oil contents up to 5 8 % in Upper Volta (IRHO, 1972). Oil content is a highly heritable characteristic (Martin, 1969) , but also strongly influenced by environment (Holley and Hammons, 1968). In India, higher oil content was recorded in TMV-1 (Virginia) and TMV-2 (Spanish) cultivars grown on upland conditions than under irrigation (Gopalswamy and Veerannah, 1968). Genotype, environment, and stage of maturity directly influence the quantity and fatty acid composition of peanut oil (Worthington and Hammons, 1972; Young et al., 1972). In the latter investigations, concentrations of stearic and oleic acid were higher, and the linoleic, arachidic, and bebenic fatty acids were lower in mature than in immature nuts, 3. Protein content and quality-comparatively little attention has been given to improving protein content and quality in the peanut. Crude protein

GRAIN LEGUMES OF THE LOWLAND TROPICS

19

content commonly ranges between 24 and 35% and is negatively correlated with oil content (Holley and Hammons, 1968). Perhaps the most serious problem in peanut protein is its poor quality and deficiencies not only in the sulfur-bearing amino acids, but in tryptophan and lysine as well (Harvey, 1970). 4. Edible peanut-the confectionary type of nut is mainly found in large seeded (100 seeds = 65 g ) Virginia types. These should have constricted pods to eliminate the flattened ends of the seeds, but the constriction should not be too tight to avoid rupturing the locules. They should also have attractive flavor and aroma. Some excellent confectionary varieties include the following: (a) United States types (Virginia) : NC 2, NC 4, NC 17, GA 119-20, FLORIGIANT, and FLORUNNER. The latter has a particularly desirable flavor and aroma. (b) African/Indian types (Valencia) : Valencia 247 from Madagascar, A 1241 B from Central Africa, and WHITE ACHOLI from India. 5 . Earliness-reducing the vegetative cycle in Virginia types by crossing with Spanish-Valencia strains has been an important objective in several breeding programs (Goldin, 1970; IRAT, 1969). Several short-duration types (105 days) have been developed in this way in Israel and Senegal; but combinations of characters have proved somewhat of a block in these efforts. 6. Seed dormancy-dormancy from Virginia types could be very useful in Spanish-Valencia types in tropical regions with bimodal rainfall patterns for growing during the first season, or where dry weather does not otherwise coincide with maturation. Combining seed dormancy with short duration has been of limited success in Senegal, Sudan, and India (Gillier and Silvestre, 1969; Ramachandran et al., 1967), but dormancy has not yet been satisfactorily introduced into Spanish-Valencia types for the more humid regions, nor for the intermediate elevations of the tropics. 7. Drought resistance-while drought resistance is a very valuable attribute in the peanut, there is some difficulty in effectively screening and evaluating this character. Relationships between drought resistance, short internodes, and root volumes were demonstrated in India (IARI, 1970). High pressure osmotic germination tests, relative transpiration rate, high temperature, suction pressure, number of stomata and rapidity of stomata regulation have been used with varying success to measure drouth tolerance. Gautreau ( 1969) demonstrated that rapidity of stomata regulation is more important than numbers of stomata in drought tolerance. Two drought resistance cultivars developed in Senegal utilizing high osmotic pressure germination tests are 55-437 and 59-127. 8. Resistance to rosette disease-major sources of resistance have been found in later maturing Virginia types in Africa, Two rosette-resistant

+

20

K. 0. RACHIE AND L. M. ROBERTS

strains such as 48-37,28-206 RR, and 1040, have been developed at Bambey and Niangoloko experiment stations and are well adapted throughout West Africa (Daniel and De Berchoux, 1965). Resistance is physiological, recessive, and governed by two pairs of homologous genes in the presence of a modifier (Dhery and Gillier, 1971; Mauboussin, 1970; IRHO, 1972). Several lines extracted from Mwitunde stock have not proved as satisfactory in resistance as the West African sources (Klesser, 1967). 9. Resistance to Cercospora-Virginia types are usually more tolerant of Cercospora arachidicola than Spanish-Valencia strains. There also appears to be an association between resistance to rosette and Cercospora. West African strains combining moderate resistance to both diseases with regional adaptation include 55-460,48-37, and 52-14 (Fowler, 1970). The resistance mechanism may result from hypersensitivity or small stomata size. Resistance factors have also been found in wild species of both Rhizomatosae and Extranervessae groups including A . repens, A . glabrata, A . hagenbeckii and A. villosa (Abdou, 1966; Gibbons and Bailey, 1967). 10. Resistance to rust-resistance to Puccinia arachidis is physiological in nature and has been found in several lines including PI 314817, a selection out of PI 298115, NC 13, and others; and in A . glabrata (Bromfield and Cevario, 1970; Bromfield and Bailey, 1972; Cook, 1972). 11. Resistance to Aspergillus flaws-production of aflatoxin should be controlled by resistance to this fungus. Moderate resistance has been found in Kaboka and Mwitunde (Rao and Tulpule, 1967; Kulkarni, 1967); and a higher level of resistance was found in two Valencia strains from Argentina, PI 337394 and PI 337409 (Mixon and Rogers, 1973). Resistance may result from inability of the pathogen to penetrate the seed coat. 12. Resistance to other diseases: (a) Verticillium dahliae-lines derived from Mwitunde and Schwartz 21 (Frank and Krikun, 1969); (b) Sclerotium bataticola-Punjab 1 and TMV 3 (Mathur et al., 1967); (c) Sclerotium rolfsii and Macrophomina phaseoli-variety 28-204 (Garren, 1964; Bouhot, 1967) ; (d) Pseudomonas solanacearum-Schwartz 21 strains, CES 101, PI 341884, PI 341886 (Bolhuis, 1955; UPCA, 1968; Simbwa-Bunnya, 1972); (e) Aspergillus niger-U4-47-7 (EC 21 115) from the Sudan (Aulakh and Sandhu, 1970). C.

PLANTPROTECTION

The peanut is attacked by several pests and diseases, although damage by foliage- and pod-feeding insects is rather less than for many other tropical grain legumes. However, some insects and diseases are interrelated through spread of virus or by predisposing the plant to invasion of fungi and bacteria.

GRAIN LEGUMES OF THE LOWLAND TROPICS

21

1. Insect Pests Insects can attack the peanut throughout its growing cycle and in storage. Several factors influence the severity of attack and susceptibility of the crop. The more common of these pests are included in the accompanying tabulation (Stanton, 1966) : Growth stage

Pests

Region

Myriapoda: Peridontopygae spinosissima Silv. Africa All Coleoptera: several species USA Stems and leaves Caterpillars: Anticursia gemmatalis Hubn. India Amsacta albiatriga Walk. India Stomopteryx nerteria Meyr. Prodenia, Leucania, Laphygma, and Amsacta species Africa Beetles: Pontomorus leucoloma USA India Sphenoptera perotelti G . Africa Capsids: IIaltieus minutus Reut. Africa Thrips: Hercothrips femoralis Rem. Africa Aphids: Aphis craccivora Koch. Africa Aphis laburni Kalt. Africa Flowers Mylabrid: Decapotoma afinis Bibb. Africa Pods/stems (in soil) Termites: Eutermes parvulus Sjostedt USA Rootworm: Diabrotica undecimpunctata howardi Pods and seeds (atorage) Bruchid: Coredonfuscus sp. Africa Africa Corya cephalonica Staint. Seedings

Aphids like Aphis craccivora are particularly important in their dual role of sucking the plant sap and spreading the rosette virus, whereas the Myriapoda and termites predispose the plant to mildew and other fungi. a. Controls. Seed treatment helps prevent attack of young seedlings by various pests. Soil insecticides like heptachlor, dieldrin, Thimet, Furadan, and Dyfonate are effective against termites and a broad range of soil-inhabiting grubs (Smith, 1971;Sharma and Shinde, 1970). Several systematic insecticides, such as dimethoate, malathion, and other organophosphates control aphids; and contact insecticides, like gamma BHC, thiodan, and Fenitrothion, control foliage feeders when these build up excessively. However, foliage insecticides are seldom used, nor are they always necessary in many tropical regions. One of the most effective controls for Aphis craccivora is a dense stand of the crop since the insect is attracted to spots of bare soil, where colonization occurs. If wide row spacings are utilized, seeds should be planted thickly within the row to increase ground coverage.

22

K. 0, RACHIE AND L. M. ROBERTS

b. Host Plant Resistance. This form of control has recently been explored mainly for foliage-feeding caterpillars. Leuck and Skinner (1971 ) found Southeastern Runner 56-15 and 40 other lines out of 1700 tested to be partially resistant to fall armyworm (Spodoptera frugiperda) in the field. In other studies Leuck and Harvey (1968) observed variation in resistance to the lesser cornstalk borer (Elasmopalpus lignosellus) and that seedling survival was highest in PI 259777. However, Smith and Porter (1971) found only low levels of resistance to the southern corn rootworm (Diubrotica sp.) in field and greenhouse studies.

2 . Diseases There are five major diseases affecting peanuts in tropical regions. In addition there are several secondary pathogens and nematodes that may become important in certain regions or under special circumstances. In some cases the incidence of the disease is dependent on attack by insects or other pests or in physically spreading the disease as in rosette virus. A brief description of these diseases is given below: 1. Cercospora leaf spots-The causal organisms are C. arachidicola Hori C . personata Ellis and Everh., and, to a lesser extent, C. canescens (Fowler, 1970, 1971). These organisms are dark spots surrounded by a yellow ring and are most predominant on the lower and older leaves. Sometimes they can cause complete defoliation. Yield losses may range from 15 to 60% depending on conditions and locality. 2 . Sclerotiurn rolfsii Sacc.-this produces a wilt that causes death of branches or the entire plant. Reddish fruiting bodies may be formed on the stem at the ground line. Although widespread it may be most serious in the United States and Australia. 3 . Aspergillus flavus-this organism attacks the stored seeds with moisture contents of 15-25% and has also been found on living plants. It produces aflatoxin some forms of which are highly toxic causing fatalities in turkeys and possibly also in humans. It has also been found to cause carcinoma of the liver in experimental animals. 4. Puccinia arachidis-this pathogen does not yet occur in Africa but is endemic in southeastern Asia, India, and the United States (Van Arsedel and Harrison, 1972). Build up of the disease in the United States occurred when a change was made in fungicides used to control Cercospora. 5. Rosette virus-this is the most serious disease of the peanut in Africa and is spread by aphids ( A . craccivora and A . laburni). The whole plant is severely stunted, and the younger leaves are chlorotic and mottled, with successive leaves becoming smaller, curled, distorted, and yellow.

GRAIN LEGUMES OF THE LOWLAND TROPICS

23

6. Secondary diseases and nematodes: a. Stem and root rots-these are caused by several secondary pathogens, such as Rhizoctonia bataticola, Asperigillus niger, Rhizopus nigricans, Macrophomina, and may cause breaking of gynophores and withering of the plant. A . niger and R . nigricans are mainly responsible for crown rots. Insects or other injury and drought following germination can predispose the plant to invasion by these pathogens. b. Miscellaneous virus diseases-several virus diseases are recorded including mottle, foliar spotting, bunchy top, chlorosis and stunt in the United States (Kuhn, 1965; Miller and Troutman, 1966), in India (Sharma, 1966), and in Australia and Senegal (Bouhot, 1968). c. Nematodes-Pratylenchus brachyurus attacks the pods and peg tissues of peanuts in the southeastern United States. VIRGINIA BUNCH 67 and GEORGIA 186-26 were not as susceptible as other varieties tested in Georgia (Minton et al., 1970). Controls. Diseases are most practically controlled through host plant resistance. However, growing adapted cultivars “in season,” sanitation, and control of predisposing pests all contribute to reducing the incidence of diseases. Secondary stem and root rots are often reduced by fungicidal seed dressings or avoided by planting on well drained soils in less humid areas or seasons. Fungicide-insecticide seed dressings have been demonstrated to increase yields by up to 4 0 4 0 % when germination and seedling growing conditions are unfavorable (IRAT, 197 1; Lewin and Natarajan, 1971). Cerospora leaf spots can be reduced by removing debris from previous crops and by crop rotation (Fowler, 1971; Mazzani and Allievi, 1971) and by spraying with fungicides like Dithane M 45 at the rate of 2.2 kg a.i. (active ingredient)/ha in 400 liters of water whenever regular rainfall has been recorded for 8-10 days (McDonald, 1970a; Corbett and Brown, 1966). The most dramatic recent development in control of Cercosporu is foliage application of benomyl (Benlate). Applied three times at the rate of 8 oz of 50 WP per acre, it increased yields from about 20% to 10 times that of unprotected plots in Virginia in 1968 and 1969 (Porter, 1970). Miller et al. (1970) likewise obtained excellent control in Florida trials in 1968-1969. The most practical control of rosette disease is probably cultural, although resistant varieties (like IRAT 48-37 and 1040) are becoming available. Vigorous, thick stands without gaps and possibly systemic insecticides are probably the best means of preventing the aphid vectors from building up (Kousalya et al., 1971). Development of Aspergillus flavus in storage can be greatly reduced t y the following means: ( 1 ) selecting well-adapted varieties and growing to

24

K. 0. RACHIE AND L. M. ROBERTS

obtain a good-quality, uniform harvest, ( 2 ) avoiding damage to the ripening nuts in the ground by tillage practices and at harvest, (3) cleaning and drying the nuts as rapidly as possible after harvest. It is particularly important to harvest as soon as ripening occurs and to carefully hand or machine sort the pods and seeds, removing blemished and injured ones. A pneumatic sorter has been developed in Senegal to mechanically remove injured pods (Gillier, 1970). D.

GROWTHPROCESS

Physiological processes in the peanut have been rather intensively studied throughout the growing regions for this crop. Greatest emphasis has been given to the areas of: (1) temperature and light effects, (2) water relationships, (3) mineral nutrition, and (4) hormonal and enzymatic effects. However, it has been difficult to derive definition conclusions in many aspects for various reasons. Field experiments are usually conducted with local varieties of undetermined genetic origins or poorly defined characteristics; a large number of uncontrollable variables interact with genotypes and treatments; and when growth is not retarded by flowering (as in the peanut) the plant often has a compensatory mechanism in response to ecological hazards. 1 . Light and Temperature Responses A . Plant Structure. Investigations at Yaounde, Cameroons, with 5 16 mm of rain in the 100-day vegetative cycle on the cultivars 55-437 and Minkong planted at 50 x 10 cm (200 th/ha), 25 X 20 cm (200 th/ha), and 40 X 40 cm (62.5 th/ha) showed numbers of leaves developing on the main stem to be positively correlated with mean temperatures; maximum leaf area index was 4.0; net assimilation rate reached a maximum of 0.85-0.90 mg DM/cm2 per day; the DM content of the leaf blade was 3.0-4.5 mg/cm2 and was not affected by genotype or growth stage, all parameters being higher than in the drier savannah (Forestier, 1969). b. Daylength and Light Quality. Initiation of the flowering process is largely unaffected by photoperiodism (Prbvot, 1949). However, light quality and intensity does affect floral development. For this reason, elite strains can frequently be grown over a wide range of latitude and seasons. Reduced light and shading tend to inhibit growth and fruit formation particularly during early stages of development (On0 and Ozaki, 1971). c. Direct Temperature Effects.Peanut seeds will germinate between 15O and 45°C but the optimum is 32-34°C. Optimal growing temperatures are 24-33°C (Catherinet, 1956; Bolhuis and De Groot, 1959). However, ex-

GRAIN LEGUMES OF THE LOWLAND TROPICS

25

treme differences between day and night temperatures of more than 20°C tend to limit flowering, and night temperatures below 10°C delay maturation (Shear and Miller, 1955). 2. Assimilating Processes a. CO, Pathways. Experiments tracing the pathways of CO, using 14C have shown that during the vegetative stages, developing leaves assimilated most of thelT, while fully expanded lcaves exported most of their C to developing apices, young expanding leaves, and roots. Immediately following peg formation the developing pods become the main sinks, and at this stage assimilates were mainly translocated from the leaves of a branch to the pods of the same branch (Khan and Akosu, 1971) . b. Protein Synthesis. Accumulation of dry matter in developing seeds begins about 4 weeks and continues for more than 12 weeks. Moisture content declines from the twelfth week, but some protein accumulation continues for two more weeks. Synthesis of proteins is associated with DNA and RNA production during the early part of this period. DNA content reaches its maximum by the eighth and tenth week after pegging in embryonic axes and cotyledons, respectively, and decreases thereafter. Cotyledonary RNA increases for 8 weeks after pegging, decreases from 8-1 1 weeks corresponding to increased RNase activity, and then increases until maturity, whereas embryonic RNA increases throughout the maturation period (Aldana, 1969; Aldana et al., 1972). c. Oil Accumulation. The oil content increases rapidly from the fourth to twelfth weeks after pegging, going from about 15 mg per kernel to 360 mg per kernel from the beginning to the end of this period. Toward maturation the carotenoid concentraton decreases markedly, this is attributable to rapid increase in oil (Pattee et al., 1969). 3. Water Relationships

Water requirements for peanuts grown on drylands have been estimated as 500-600 mm per season. During the first months of growth, daily requirements increase from 1.5 to 4 mm per day; they reach 5-7 mm during the peak of growth, decrease to 4 mm by the last month of the vegetative cycle, and finally drop to 2 mm daily during ripening (Ilyana, 1959; Ochs and Wormer, 1959; Mantez and Goldin, 1964). a. Relative Humidity. Experiments carried out in Texas by Lee et al. (1972) demonstrated the beneficial effects of high relative humidity at flowering. Plants transferred from 50% RH to 95% R H 50 days after planting had much better flowering, peg growth, formed more ethylene, and had more gibberellins than plants transferred from 95% R H to 50% RH. Low relative humidity (5 % ) was found less satisfactory for retaining

26

K. 0. RACHIE AND L. M. ROBERTS

seed viability in storage than higher relative humidity. Seed remained viable for three years in storage at 30% RH and 4°C in studies carried out by Gavrielit-Gelmond ( 1970). b. Drought Resistance. The peanut is highly resistant to drought and is able to extract soil moisture under quite extreme conditions. In the sandy soils of Senegal, yields are reduced when soil moisture reaches 70% of its retention capacity, and the permanent wilting point is estimated at 40% of that level (Dancette, 1970). Nevertheless, moisture deficiency has a direct effect on yields, patticulary when it occurs during flowering. 4. Enzyme Activity and Hormones Seed Dormancy. Dormancy is primarily related to genotype in the peanut. Lin and Chen (1970) found wide differences in seed dormancy in 56 cultivars studied and divided them into four classes: ( 1 ) nondormant types, (2) dormant for 2-4 weeks, (3) dormant for 5-8 weeks, and (4) dormant more than 9 weeks. Virginia types tended to have greater dormancy than Spanish cultivars. Ascorbic acid was found associated with the germination process in studies carried out by Screeramulu and Rao (1970, 1971). They observed a rapid increase in ascorbic acid from 35 mg in the entire seed to 284 mg in the 5-day-old seedlings of the freshly harvested nondormant TMV 2. In contrast, fresh seed of the dormant variety TMV 3 did not germinate and showed very slight ascorbic acid activity. Sreeramulu and Rao ( 1971) further observed growth promoting substances to increase from 20-30 days after the pegs touched the ground in both acidic and neutral fractions of both dormant and nondormant seeds. From 30 days onward growth promotors decreased while growth inhibitors increased. In the dormant type (TMV-3) growth inhibitors were higher than the promotors in the acid fraction of the seeds in contrast to the condition in nondormant seeds. Dormancy in both apical and dorsal types of seeds can be broken by and ethylene gas at 8 ppm and by 2-chloroethylphosphonic acid at 5X M,increasing germination to 100% in both types of seeds after 48 hours. Gibberellic acid at 5 x stimulated ethylene production in apical seeds, increasing germination to 40% above the control (Ketring and Morgan, 1970).

5. Mineral Nutrition Mineral requirements of the peanut plant have been extensively studied in all major growing regions. Foliar diagnosis has made it possible to determine precisely the requirements of the plant (Pr6vot and Ollagnier, 1961; Martin, 1965). These can be summarized for each 1000 kg of kernels in shell of yield (see tabulation).

GRAIN LEGUMES OF THE LOWLAND TROPICS

Foliar/stems Element

(ks)

N

10-12 1.5-2.0 10-12 8-12 8-10

P20b

K2O CaO MgO

27

Pods/seeds (kg) 30-35 6 6-10 1-62

2

a. Nitrogen and Rhizobial Symbiosis. Peanuts utilize considerable nitrogen, which is almost totally provided by the rhizobial system. Therefore, nitrogen fertilizer applications beyond about 19 kg of nitrogen per hectare have not been economic. However, instances of severe nitrogen deficiency have occurred in acidified sandy soils when normal rhizobial populations disappeared (Blondel, 1969). Inoculation can be important in certain regions where peanuts have not been grown previously for some time. However, there is apparently no need to provide rhizobial cultures in most of tropical Africa or Madagascar. Some strains of peanuts, like Asiriya Mwitunde in India, nodulate better than other strains (Narsaiah et al., 1969). Attempts are also underway to improve the symbiont through mutation breeding utilizing ultraviolet radiation and N-methyl-N-nitro-N-nitrosoguanidine(Raina and Modi, 1969). However, improved mineral nutrition, particularly by supplying adequate amounts of calcium, phosphorus, molybdenum, magnesium, and potassium has shown positive responses under field conditions (Nair et al., 1970, 1971). b. Phosphorus. This is probably the most important mineral nutrient required by the peanut in much of the tropics. Good response is observed even to applications as low as 11 kg of P,O, per hectare as presently recommended in northern Nigeria. ”However, in soils markedly deficient in phosphorus, comparatively heavy applications of up to 150 kg of P,O, per hectare every few ( 3 ) years have been economic and have a “saturation” effect on the soil complex (Carriere de Belgarric and Bour, 1963; Goldsworthy and Heathcote, 1964). c. Potash. Peanuts make a heavy drain on soil potash, but seldom respond to this element except in leached ferralitic and intensively cropped soils. Potash deficiency is evident when a high proportion of the pods produce only one seed. d . Calcium. This element is very important in the formation and development of the seeds and in nodulation. It can be absorbed both by the roots and the gynophores. A deficiency of calcium decreases the shelling percent

28

K. 0. RACHIE AND L. M. ROBERTS

and results in a high proportion of “pops” or empty pods. Salinity can inhibit uptake of calcium, as observed in studies carried out by Kamana and Rao (1971). Calcium can be added in the form of lime- gypsum-, or calcium-bearing fertilizers, like single superphosphate. Sometimes gypsum is dusted on the foliage at flowering to increase the uptake of both calcium and sulfur. e. Sulfur. This element contributes to nodulation and helps prolong flowering. The normal requirement for peanuts is 12-15 kg of sulfur per hectare, and it can be absorbed both by the roots and the foliage. There is usually an adequate quantity of sulfur in sulfur-bearing fertilizers like ammonium sulfate and single superphosphate when used at reasonable levels in the crop rotation to satisfy the requirements of peanuts (Brzozowska and Hanower, 1964; Hanower, 1969; Bromfield, 1973). Sulfur significantly increased the yield of unshelled nuts and the oil, sulfur, and methionine contents of mature seeds in experiments carried out in India by Singh et al., (1970). f. Iron. In highly calcareous soils iron chlorosis may occur; it has been observed in Negev, Israel, with a soil pH 7.6-8.3 containing up to 21.4% CaCO,. Yields of shelled nuts increased by up to 250% through soil or foliage applications of chelated iron (Fe EDDHA). The most effective treatment was one or two foliar sprays or 10 kg of iron chelate per hectare 3-6 weeks after planting (Hartzook et al., 1971, 1972). g. Boron. Boron deficiencies cause “hollow heart,” blackening of the embryo, and, occasionally, cracking of the stems that may appear to be a secondary effect of drought. The deficiency is correctable by applying 5-10 kg of borax per hectare (Harris and Brolmann, 1966; Gillier, 1969). Excessive boron also causes toxicity by inhibiting the uptake of iron (Gopal, 1970, 1971). h. Molybdenum. This element acts directly on the rhizobial process, increasing the number and weight of nodules formed, and, therefore, influences nitrogen availability. Deficiencies can be corrected by seed treatment or application of as little as 28 g of molybdenum per hectare (Gillier, 1966; Pillai and Sen, 1970). i. Manganese Toxicity. Manganese toxicity may occur in acid soils, which change manganese into an exchangeable form, and excess amounts are taken up by the plant. It aggravates the unbalanced soil condition in the absence of calcium. It was first observed in Zaire (Prkvot et al., 1955) and sometimes occurs in autoclaved soil (Boyd, 1971). j . Fertilizer Use in the Tropics. Comparatively little fertilizer is applied directly to the peanut crop itself in the tropics. Rather, the plant has to rely on residual fertility applied to previous or associated crops. In areas, where cash crops like cotton, tobacco, vegetables, or heavily manured

GRAIN LEGUMES OF THE LOWLAND TROPICS

29

cereals precede the peanut crop, soil nutrients may be more than ample at present levels of production. However, higher yield levels will be necessary in the future for peanuts to compete better with other high-yielding crops. The present nutrition level for a l-ton crop will hardly be adequate for yields of 4-5 tons/ha.

E. MANAGEMENT The peanut has made an enormous impact as a cash crop in peasant farming systems. As such it is frequently interplanted with other crops or is included late in rotations with cotton, tobacco, maize, or other cereals. Although possessing the dual advantage of being both a cash and subsistence crop, an estimated 7 5 9 0 % of the produce is milled for oil and cake, or exported unprocessed. Optimizing husbandry practices will be discussed briefly in the section to follow. I . Tillage and Planting Peanuts prefer loose, friable sandy soils and can be planted either on the flat or on ridges. Ridges offer better drainage and facilitate lifting, but impose limitations on the populations that can be grown and may increase difficulties in pegging (particularly narrow ridges). Rows are frequently mechanically cultivated during early growth. Populations and Spatial Arrangements. Growth habit tends to govern the planting rates. Higher populations and uniform stands generally produce higher yields, in addition to reducing the incidence of rosette virus and suppressing late flowering. Thus, maturation is more uniform and results in better quality. Populations of 250,000 plants per hectare or individual plant spacings of 15 cm X 30 cm and seeding rates of 65-90 kg of shelled nuts per hectare are recommended for hand placement of seeds, or 10 cm X 60 cm for mechanical planting and cultivation for bunch varieties. Spreading or runner types are usually planted at lower populations of about 55,000 plants per hectare at spacings of 30 cm X 60 cm and a seeding rate of 45 kg/ha. In hand planting it is preferable to drop two seeds per hill to assure better stands and permit wider row spacings (Purseglove, 1968). Alternatively, double rows 15 cm apart with 92 cm between pairs of rows has been advantageous in Australia (Wood, 1970). Seeds for planting should be stored in the shell; just before planting they should be carefully shelled to avoid injury, and treated with a fungicide/insecticide combination. Mercurials or thiram are the preferred fungicides in combination with aldrin or dieldrin. Early planting is essential for high yields and the young crop must be carefully weeded, as seedling growth is slow. Weeds can greatly re-

30

K. 0. RACHIE AND L. M. ROBERTS

duce yields by competing for moisture, nutrients, and light and also cause difficulty in harvesting. Mechanical cultivation and hoeing must be terminated after about 8 weeks to avoid damaging the developing gynophores (pegs). 2 . Weed Control Weeds compete for nutrients, moisture, and light and can easily become the major deterrent to increased productivity. In the Sudan, Ishag (1971) found weeds decreased the number of branches per plant and pods per plant, and yields were increased nearly 5-fold by handweeding 30 and 60 days after planting. Chemical Control. Several preemergence herbicides including trifluralin at 1.1 kg a.i./ha and benefin at 2.2 kg a.i./ha gave excellent control of grasses-mainly Brachiaria and Digitaria spp. in Australia (Wood, 1970). Prometryne at 1.7-2.2 kg a.i./ha, linuron at 1.7 kg a.i./ha, and nitrofen at about 2.2 kg a.i./ha are promising under a wide range of conditions in India (Sindagi et al., 1972; Patro et al., 1970), Taiwan (Wang et af., 1971), Australia (Wood, 1970), and Ghana (Takyi, 1971). 3 . Harvesting

Most of the tropical peanut crop is harvested by hand. This consists of pulling and inverting the plants to facilitate drying of the pods in the sun before stripping off the nuts. Harvesting of the optimum time is essential to maximize yields, minimize losses from shedding, and eliminate sprouting in bunch types (Purseglove, 1968; Young et al., 1971). Considerable losses occur if the soil becomes hard before lifting. In most tropical regions peanuts are hand harvested because mechanization is too expensive. However, several shelling machines have been designed for hand or power operation. Yields. Peanut yields are reasonably high considering the conditions under which they are grown. Worldwide yields are 9.5 Q of unshelled nuts per hectare and up to 23 Q/ha for the United States; but much higher yieldsbeen obtained under favorable conditions. exceeding 4000 kg/ha-have The shelling percent runs to about 80% for bunch types compared with 60-75 % for runner (spreading) cultivars.

4 . Combined Technology Adoption of a combined improved technology of “package of practices” can increase yields remarkably. Bray (1970) found the mean yields of unshelled nuts in the Sine-Saloum area of Senegal increased from 937 kg/ha in 1960-1963 to 1 1 18 kg/ha in 1964-1967 as a consequence of using an improved variety, treating the seeds, sowing with a drill, obtaining better stands, and applying fertilizers. Application of fertilizers (phosphorus,

GRAIN LEGUMES OF THE LOWLAND TROPICS

31

potassium, calcium) alone in Sambwa, Tanzania increased yields of unshelled nuts by 20% from 1503 to 1835 kg/ha (Anderson, 1970). At Katherine in northern Australia, Wood (1970) increased yields of unshelled nuts by 270 kg/ha merely by increasing the seeding rate from 55 to 100 kg/ha. In Madras, India, Mohan (1970) obtained higher yields (3221 kg of pods per hectare) by irrigating whenever moisture was down to 60% of field capacity (1.05 atm tension) ; and, in another trial, yields were increased from 3086 to 3291 kg of pods per hectare by mulching with Glyricidiu leaves. However, maximized economical production is nearly always achieved by adopting the complete improved technology.

F. CHEMICAL COMPOSITION Peanut seeds contain mainly nondrying oil and protein. Virginia types are usually 3 8 4 7 % oil, and Spanish types are 47-50% oil. Carbohydrates are about 11.5%, ash is 2.3%, and water is 6.0%. Decorticated cake contains about 10% water, 47% protein, 6-7% fat, 23% carbohydrate, 6.5% fiber, and 6.0% ash. The oil normally contains 53% oleic and 25% linoleic acid, but genotype and environment strongly influence oil quality. The principal proteins are arachin and conarachin. The peanut is also rich in vitamins B and E. a. Protein Quality. Peanut protein is frequently low in sulfur amino acids and in lysine, but considerable variation has been observed in both methionine (3.7-8.7 mg per gram of seed) and lysine, suggesting that these components might be increased through genetic manipulation (Chopra and Bhatia, 1970; Heinis, 1972). b. Fatty Acids. Fatty acid composition in 101 local and introduced peanuts were investigated by Worthington and Hammons (1971 ). They found linoleic acid to range from 14 to 40%, lower in large-seeded Virginia than in Spanish types and positively correlated with levels of palmitic, behenic, and lignoceric acids. c. Other Constituents. Thiamine (B,) occurs in raw peanuts, but very little is strongly bound to the proteins, according to Dougherty and Cobb (1 970). Some peanut cultivars are comparatively low in intestinal gasforming sugars in studies carried out by Hymowitz et al. ( 1972). The varieties ARGENTINA, EARLY RUNNER, SPANCROSS, TIFSPAN, and VIRGINIA BUNCH 67 had stachyose contents less than 0.1 g/lOO g seed and are believed to be non gas-formers.

G . POTENTIAL The peanut has become a major contributor to the economy of tropical peasant agriculture. It enjoys the dual advantages of being both a cash

32

K. 0. RACHIE AND L. M. ROBERTS

and subsistence crop permitting the small holder considerable flexibility in buffering adversities in weather and general growing conditions. That is, he can hold in reserve the quantity required for food, while having a good stable market for the surplus. Peanuts are comparatively easy to grow and normally do not require sophisticated technology nor massive inputs of fertilizers or plant protection. Moreover, they are well adapted to both hand and mechanized cultivation. They are reasonably free of pests and have few serious diseases. Underground fruiting avoids devastating attacks by pod feeding insects that plague many other tropical grain legumes. Fortunately, there appear to be reasonably effective solutions to some of the major constraints to increased productivity levels in the tropics. These include discoveries in host plant resistance to the most serious diseases; recent developments in systemic fungicides and insecticides; and better understanding of the nutritional and water requirements of the crop. Perhaps the major constraint to expanded use and productivity would be limitations in genetic diversity and moving off the present yield plateau. Nevertheless, the immediate future for peanuts should be very bright. IV.

Pigeon Peas

Origin and Spread. Pigeon pea, red gram, arhar, tur, Congo bean, or gandul (Cajanus cujun Millsp.) is probably a native of Africa as a wild species occurs in the Sub-Saharan region. Seeds have been found in Egyptian tombs of the XIIth Dynasty, and it was cultivated there before 2000 BC, when trade relations had been established with both Africa to the south and Arab countries to the east. Pigeon peas were cultivated in Madagascar from a very early period and must have reached India in prehistoric times. Pigeon peas rzached the New World shortly after Columbus, perhaps in the 16th cen ury, but were not spread into the Pacific until much later and are recoLded to have been introduced into Guam in 1772. However, this crop is now widely spread throughout the lowland tropics and is the most important pulse (non oilseed grain legume) in these regions.

IMPORTANCE Production. About two million tons of seeds were produced on three million hectares in 1970-1 971. Southern Asia-primarily India-produces 95% of the world crop; but statistics for the African tropics may be low by a factor of 2 to 4 times since pigeon peas are extensively grown in compound or kitchen gardens throughout the humid to semiarid low and intermediate elevations. Field plantings are reported mainly from Malawi A.

GRAIN LEGUMES OF THE LOWLAND TROPICS

33

and Uganda, but also occur in most other countries throughout central Africa. In the tropical Americas, the pigeon pea is an important vegetable in the form of cooked green peas, but it is also consumed as a dry pulse. Major producing countries are the Dominican Republic, Puerto Rico, and Venezuela, but other Caribbean and Latin American countries grow substantial quantities of this crop (FAO, 1972; Rachie, 1970). 1. Morphology

The pigeon pea plant is a woody, short-lived perennial shrub which can be grown either as an annual or a perennial (some bushes survive more than 10-12 years). It has a deep taproot with longer laterals in the spreading than in erect types. The narrowly lanceolate and finely pubescent trifoliate leaves are borne in a spiral arrangement with a M phyllotaxis on the main axis and on branches. Some cultivars tend to produce long undivided primary branches on a shorter main axis with leaves carried the entire length and which fruit along the terminal one third to half the length. In other strains there is profuse secondary and tertiary branching, but some genotypes branch very little, producing leaves and fruits directly on the main axis, being similar in structure to the modern soybean. There are both erect (acutely angled branches-less than 30") and spreading types (60" branching). Inflorescences are smaller than cowpeas and predominantly terminal (basipetalous) or mainly axillary (acropetalous) and racemes can be 4-12 cm long. The flowers are about 2.5 cm in length, yellow or with the dorsal side of the standard red, purple, deep orange, or yellow with red or purple veins. The pods are somewhat flattened and 2-8 seeded (commonly 4) and 4-10 cm X 0.6-1.5 cm in length and width, Unripe pods may be solid green (recessive), purple or maroon, or green blotched with purple or maroon. Seeds vary in size, are usually globular in shape, may be white, grayish, red, brown, dark purplish or speckled in color, have a small white hilum and commonly weigh 7-15 g/100 seeds. Seeds germinate hypogeally, but dormancy may occur in some species (Purseglove, 1968). Related Wild Species. The most closely related wild forms are probably Atylosia species, especially the erectoid forms. Morphological, taxonomic, and cytological evidence, as well as homology of genetic characters and a high degree of fertility of the intergenetic hybrids indicates a close affinity between Cajanu,s and A tylosia-particularly the erectoids, A . sericea and A . scarubaeoides W. and A. (Deodikar and Thakar, 1956; Kumar et al., 1958; Roy and De, 1965). They suggest that structural changes in the chromosomes may have played a major role in the differentiation of the two species. In crosses between C. Cajun and A . lineata these structural changes

34

K. 0. RACHIE AND L. M . ROBERTS

appeared to be responsible for partial seed abortion, low percentage of germinating pollen and chromosomal abnormalities (quadrivalents, bridges, and fragments at meiosis) in the intergenetic hybrid. However, in view of the comparative ease in crossing and similarity in chromosome number ( 2 n = 22) Roy and De (1965) suggest incorporating Atylosia into Cujanus with appropriate cytotaxonomic revision of the latter. 2. Adaptation Pigeon peas are widely adaptable to climate and soil conditions, but perform better when annual precipitation exceeds 500 mm, the soils are not markedly deficient in lime, and waterlogging does not occur. They have a very deep tap root and are highly drought and heat resistant (Gooding, 1962). However, the plant also grows in subhumid ecologies where ripening can occur during the dry season. Thus, it may span the range of lowland tropical conditions better than most other legumes. Most pigeon peas are highly photoperiod sensitive, although some lines have been shown to be highly insensitive. The range of maturities is from 90 to 250 days depending on genotype, time of planting and other factors when grown at low elevations in the tropics. Since early growth is very slow, pigeon peas are frequently mixed with other crops-mainly short-term, hot weather cereals or other grain legumes-and continue to grow and fruit after the shorter duration companion crop has ripened and been harvested. Competition with weeds (and companion crops) is very poor during the first 4-6 weeks, but is excellent once a canopy has been established and a leaf litter builds up which not only suppresses weeds but also reduces erosion. High yields can be obtained when there are good rains during the first 2 months, but no further precipitation occurs during the remaining 2-4 months before harvest.

+

B. PLANTIMPROVEMENT Pigeon peas are comparatively highly outcrossed from 3 to 40%, averaging about 20%. Wilsie and Takahashi (1934) observed 14.0 to 15.9% outcrossing between adjacent rows of contrasting varieties, Deshmukh and Rekhi (1962) found 25.0% natural crossing in central India; and Abrams (1967) found 5.8% natural crossing between rows 8 feet apart in Puerto Rico. Selfing may be accomplished by bagging terminal racemes with thin cloth bags or mesh (plastic or metal) “sleeves” which permit light and air penetration, but exclude larger insectivorous pollinators; or by sealing the petals with melted candle wax (Kelkar and Pandya, 1934). Higher levels of “controlled” natural outcrossing can be attained by using genetic male sterility or sprays of male gametocides. Kaul and Singh (1967) obtained

GRAIN LEGUMES OF THE LOWLAND TROPICS

35

100% pollen sterility with minimum yield reduction by a foliar spray of a 1 % solution of FW 450 applied prior to floral bud initiation. Thrips (Taeniothrips spp.) are believed to be responsible for both selfing and some outcrossing as well (Prasad and Narasimhamurthy, 1963; Sen and Sur, 1964). Hand Pollination. Hand manipulation can be difficult under certain conditions and a high proportion of flowers shed before setting fruit. Blooming occurs over several weeks and flowers normally open between 1 1 AM and 3 PM and remain open for about 6 hours. It is important to emasculate before 3 AM on the day before the flowers open as pollen is shed later that morning (the day before opening). Rain reduces fertilization. There appear to be few if any barriers to wide crossing within and between species in Cajanus.

1 . Plant Types Early botanists recognized two basic plant types: var. flavus DC-the tur varieties of peninsular India which are shorter and earlier, have predominantly yellow flowers, green pods, light-colored seeds, and are usually 3-seeded; and the second basic type, var. bicolor DC-the arhar varieties of Northern India, are large, bushy, late-maturing perennials with red or purple or darkly veined flowers and hairy maroon or purple unripe pods with 4-5 dark colored or speckled seeds when ripe. However, most permutations of these characters have been observed in the world collection and in breeding lines, as there appear to be few if any barriers to recombination save the physical ones of time and space. Perhaps a more useful classification of the cultivated species was developed by Akinola and Whiteman (1972) in studies on 95 accessions of Cajanus. They used 3 1 attributes including plant and leaf morphology, growth, flowering patterns, disease tolerance and components of seed yields to form 15 classes and three major groups according to a hierarchial program (MULTCLAS) and Euclidean system. The group categorization included: Group A-inflorescence basipetalous, comparatively early maturity and pod ripening was both extensive and intensive Group B-these types were early in maturity and pod ripening was extensive Group C-maturity was very late and pod ripening was intensive Fifty widely diverse plant types were studied by Sharma et al. (1971) and classified into five categories based on studies of growth habit and yields: ( 1 ) tall and compact, ( 2 ) tall open, ( 3 ) medium tall compact, (4) medium tall open, ( 5 ) dwarf bushy. Early and intermediate maturing types tended to be in groups (4) and ( 5 ) , whereas late cultivars occurred

36

K. 0. RACHIE AND L. M. ROBERTS

mainly in group ( 1 ) . Seed yields were found positively correlated with spread of the plant, number of secondary branches, effective pod-bearing length of branch and pod number. 2. Breeding Methods The most widely used method of improving and genetically manipulating Cajanus is by simple and mass selection in collections and pools of genetic stocks. However, considerable controlled outcrossing and hand-manipulated pollinations with or without selfing or isolation are done. Other techniques include using mutagenic agents like ionizing radiation and chemicals such as EMS, HNO,, HCl, and chloral hydrate, polyploidizing agents (colchicine) , and wide crossing. Objectives in breeding include characters like (1 ) high-yielding potential; ( 2 ) earliness; (3) perenniality of determinancy; (4) resistance to drought; ( 5 ) clustering of fruits (for convenience in harvest) ; ( 6 ) resistance to diseases like wilt (Fusurium udum) , leaf spots, and stem rots, viruses and nematodes; ( 7 ) resistance to insects, such as leaf feeders and pod borers; (8) better quality seeds and improved cooking quality. a. Mutation Breeding. The effect of ionizing radiation has been studied in Puerto Rico by Abrams and VClez Fortufio ( 1961, 1962). They obtained a wide range of genetic variation in the R, of seeds treated with gamma rays or neutrons in the early-flowering KAKI and late-maturing SARAGATEADO cultivars. Most lines tended to be taller and included both earlier and later and higher yielding types than their parents. The effects of mutagenic chemicals on pigeon peas in India were reported by Deshmukh and Phirke (1962) and Phirke (1966). Seeds of EB3 and El338 treated with HNOB,HCl, and chloral hydrate proved to be diploids.Their progenies showed variation in plant height, pod size, yield, and grain weight, but crossed successfully with untreated normals. A new flattened pod character was discovered and proved to be a point mutation. b. Polyploidy. Both natural occurring and induced polyploids-tetraploids ( n = 22) and hexaploids ( n = 33)-have been studied by investigators in India (Pathak and Yadava, 1951; Bhattacharjee, 1956; Joshi, 1966; Dafe, 1966; Shrivastava et al., 1972). They observed total sterility in the hexaploid forms and varying levels of sterility in the tetraploids. The tetraploids were usually later in maturity and shorter in height, had longer and thicker leaves, more branches, and thicker stems, and were more erect. Flowers, stoma, pollen grains, pod and seed sizes were generally larger, and leaves and seeds contained more nitrogen than their parents. Varying levels of multivalence were observed in cytological examination. Since considerable variation in fertility occurred in both natural and induced auto-

GRAIN LEGUMES OF THE LOWLAND TROPICS

37

tetraploids it should be possible to increase fertility through appropriate breeding methods like recurrent selection. Wide Crossing. “Intergeneric” crossing between Cajanus and A tylosia has been reported from India (Deodikar and Thakar, 1956; Kumar et al., 1958; Roy and De, 1965; Sikdar and De, 1967). A . lineata, A . sericea, and A . scurufcxeoides have been utilized to introduce and enhance combining ability, perennial growth habit, tolerance of drought and resistance to pests and diseases. In particular, A . lineata and A . sericea have shown resistance to the pod borer (Exelastic atumosa) and wilt (Fusurium udum). Since these crosses have been comparatively easily made, it is proposed to combine Atylosia with Cajanus. 3. Genetics

There is comparatively limited information on heterosis, inheritance, associations of morphological characters, linkage groups, nature of gene action, and cytology in Cajanus. Heterotic effects were reported in ten F, hybrids grown at Bijapur, India (Solomon et a!., 1957). They obtained grain yield increases up to 24.5% over the mean of the parents, but the best-yielding hybrid was less productive than the best parental type in this study. a. Character Associations and Heritability. Studies of five hybrids between four pigeon pea cultivars demonstrated that seed yields were highly positively correlated with number of pods per plant, to a lesser extent with plant height and 100-seed weight, and negatively with days to flowering. However, there was very little variation in seeds per pod compared with seed weight, plant height, and flowering date. Pods per plant had a low heritability: 45.3% in the F, and 52.1% in the F,; but flowering date, plant height, and seed weights were highly heritable (Muiioz and Abrams, 1971). In other studies carried out in India, Sharma et al. (1971) found seed yields to be positively correlated with plant spread, number of secondary branches, effective pod bearing length of the branch and pod number. Beohar and Nigam (1972) confirmed these results and also found a positive correlation between number of branches and number of pods per plant; but number of pods per plant were negatively correlated with pod length. b. Genotype X Environment Interactions. The nature and magnitude of variance components for yield, date of flowering, plant height, and seed weight were studied in 20 cultivars grown in southern and northwestern Puerto Rico over a three-year period by Abrams et al. ( 1969). They found the variety and variety X year components to be significant for all characters with the latter being greater than the variety component. The variety X locality x year interaction was also highly significant for all characters except yield, but was smaller in magnitude than the varietal components. How-

38

K. 0. RACHIE AND L. M. ROBERTS

ever, the variety x locality component was negative and nonsignificant except for date of flowering. It is suggested that the two locations may have been quite similar ecologically (since Puerto Rico is a comparatively small island) insofar as interacting with the genotypes in this study. c. Inheritance. Several botanical characters have been studied for their inheritance, and linkages with other factors have been determined. Some of these are listed in summary form in Appendices Tables IV and V. C.

PLANTPROTECTION

I . Insect Pests Catepillars of Heliothis armigera Hubn., plume moth Exelastis atomosa, the pod fly Agromyza obtusa M . , and pulse beetle Bruchus sp. are serious pests of pigeon peas in India; whereas Heliothis sp., Maruca testulalis, and Laspeyresia pychora occur in Africa; the pod borer Elasmopalpus rubedinellus (Zell.), Ancylostomia stercorea (Zell.), and Heliothis virescens ( F . ) are serious pests in the West Indies; and in Peru, Korytkowski and Torres (1966) reported fourteen pests attacking this crop. In India, yield losses from pod-feeding insects (pod fly, plume moth, and pulse beetle) were studied in eight cultivars by Rawat and Jakhmola (1967). They observed greatest seed yield reduction in the cultivar Jabalpur Local (34.9% ) and lowest in Type 148 (5.5%). However, Bindra and Jakhmola (1967) observed no varietal differences among eleven cultivars in tolerance to the same three pests. Nymphs and adults of hemipterous pests (coreids), like Clavigralla gibbosa Spin. in India, suck the cell sap from green pods and seeds, causing shriveling and reduced germination (Choudhary, 1969). Chemical Controls. Insecticidal sprays of such materials as malathion, BHC, thiodan, dimethoate, Gardona, and Furadan (seed or seed furrow treatment) are among the most effective insecticides used in India and Africa (Thevasagayam and Canagasingham, 1960; IITA, 1973). Under intensive management, chemical insecticides are probably the only practical controls and are the most important management input for pigeon pea pests in the tropics at the present time. However, possibilities for biological and cultural controls deserve much greater attention than is presently accorded. The most ideal control would be host plant resistance or tolerance as the major form of augmenting other controls when the problem is highly intractable such as pod feeding pests. 2. Diseases a. Fungi. Pigeon peas are often less susceptible to diseases than many other grain legumes in the lowland tropics. In India, the most serious fungus disease is a wilt caused by Fusarium udum Butl. However, there

GRAIN LEGUMES OF THE LOWLAND TROPICS

39

appears to be good host plant resistance to different isolates of the fungus in C . l l , C36, and NPWR.15 (Subramanian, 1963a; Ramanujam, 1972). In these and other investigations Subramanian ( 1963b) inoculated plants of both the wilt-resistant NP.15 and the susceptible NP.24 and found greater tolerance associated with a smaller reduction in contents of chlorophyll, ascorbic acid, iron: manganese ratio and total carbohydrates and less rapid decrease in transpiration. Inoculated NP.24 showed a marked increase in free reducing sugars. Other important fungus diseases of Cajanus in various parts of Africa and Madagascar include root and stem rots caused by Macrophonzina phaseoli and Phaeolus manihotis. Collar and stem rots caused by Physalospora cajanae and rust caused by Uromyces sp. are important disease in the Caribbean and South America. Leaf spots by Cercosporu sp. and Colletotrichum cajanae and downy mildew caused by Leveillula taurica occur in some areas with largely undetermined consequences. b. Viruses. Very few viruses have been reported on pigeon peas except for the sterility virus in India and witches broom apparently caused by a microplasma in the Caribbean. Sterility “disease” is characterized by: (1) reduction in leaf size, ( 2 ) bushy growth habit, ( 3 ) light yellowishgreen foliage, and ( 4 ) suppression of flowers and fruits (Alam, 1933). More recent investigations on this or another form of virus-induced sterility indicate that it may be transmitted by nematodes since it was not transmissible by sucking insects nor with dodder, and the incidence of infected plants was greatly reduced in DD or Nemagon fumigated soil and appeared to be related to populations of Rotylenchus reniformis and Tylenchorhyachus sp. in the rhizosphere (Narayanaswamy and Ramakrishnan, 1966). Search for varietal resistance to sterility mosaic within a limited range of genetic diversity at Coimbatore, India did not produce resistance (Kandaswamy and Ramakrishnan, 1960).

3. Disease and Pest Control The options for controlling or reducing losses by diseases are fewer than for insect pests. Application of fungal sprays to the foliage is seldom as economical nor practical as for chemical insecticides. The latter are sometimes the most important and essential management input in the growing of tropical legumes. The most practical approaches to disease control are through (1) resistant varieties, (2) cultural practices, and ( 3 ) seed treatments. While resistant varieties provide the most practical control mechanisms, proper management in terms of crop rotations, sanitation, planting dates, and control of disease vectors (insects and nematodes) can greatly reduce the incidence of disease and its consequences even in susceptible species and cultivars.

40

K. 0. RACHIE AND L. M . ROBERTS

Seedling diseases are often reduced by fungicidal seed dressings-a highly economic and convenient practice. To these can be added the newer systemic fungicides (Benlate, Demosan, and others), which greatly improve the effectiveness and extend the period of protection. Similarly, seed dressing contact insecticides like dieldrin or aIdrin and the systemics like Furadan, phorate and Lanlate show considerable potential. Furadan seed treatments ( 1 % active ingredient) at Ibadan have given up to 6 weeks of protection from root and foliage insects in addition to its possible nematocidal properties (IITA, 1973 ) .

D.

PHYSIOLOGY AND

MANAGEMENT

There are comparatively few definitive physiological investigations in Cajanus, although agronomic studies have been reported from all the major growing regions. While the range in genetic diversity is very great and the influence of environmental factors is considerable in this species, some generalizations can be made.

I . Growth Characteristics Seedling growth in Cajanus is very slow for the first 30 to 50 days. The seedling plant is very slender and fragile, tending to grow upward more rapidly rather than spreading out. This characteristic may be desirable in predominantly mixed cultivation, but can be a handicap in monoculture. However, root development is both extensive and deep, enabling the plant to tap moisture and nutrients at greater depths than the more herbaceous tropical legumes. In comparisons of root development in soybeans, cowpeas, and pigeon peas at Ibadan, only the latter could penetrate the compacted gravel layer underlying these soils at depths ranging from a few centimeters to 1 meter. The erect, self-supporting Cajanus plant, with comparatively small, lanceolate leaves, should confer high levels of photosynthetic efficiency in comparison with other legumes. In fact, some preliminary findings suggests leaf area indexes of 7.0 and higher to be optimal compared with LAI’s of 3.0 and 4.0 for other large-leaved tropical pulses. a. Drought Tolerance. Once established, pigeon peas grow exceptionally well on residual moisture. In southern Nigeria on upland soils the crop has been observed to grow vigorously and fruit profusely 75 days after cessation of rains (plantings established with 50-60 days of rains). The only other short-term field crop with similar capabilities in this area is cassava. Moreover, perennial types tend to be deciduous following fruiting in the dry season and will bear new leaves and resume growth with the return of rains even after several months of dry weather.

GRAIN LEGUMES OF THE LOWLAND TROPICS

41

b. Light and Temperature Response. Very little is known about light and temperature response in Cajanus. Most cultivars are probably responsive to variations and interactions of both factors. However, daylengthinsensitive genotypes are available and breeding efforts at IITA ( 7 N latitude) have produced several strains with “normal,” predictable maturities and heights when planted in all seasons (IITA, 1973). Other investigators have demonstrated genotype x environment interactions in Caianus (Derieux, 1971). In Trinidad, Spence and Williams (1972) utilized short days through December plantings and high populations (165,000 plants per hectare) to reduce mature plant growth to about 1 meter and increase determinancy , thereby facilitating mechanical harvesting. 2. Mineral Nutrition

It is often difficult to demonstrate response of pigeon peas to fertilizers or rhizobial inoculation in most tropical soils having a reasonable pH and good drainage (Pietri et al., 1971). However, moderate applications of phosphate and potash could be expected to produce economic returns on soils deficient in those elements. Increases in dry matter and absorption of mineral nutrients occurs continuously in the plant, reaching peak assimilation/accumulation rates between flowering and seed set. At all stages of growth, calcium and magnesium are greater in the leaves than other organs, and seeds are richer in nitrogen, phosphorus, and potassium than other tissues. The nutrients exported in a crop producing 1630 pounds of dry matter per acre were: N = 29 Ib; P = 9 lb; K = 10 Ib; Ca = 12 lb; and Mg = 5 Ib (Mehta and Khatri, 1962). Most studies indicate phosphorus to be the first limiting element under tropical conditions and recommend applying 20-80 kg of P,O, per hectare (Khan and Mathur, 1962; Bhatawadekar et al., 1966). In India yields were incre?sed by 13.5% in Madras by applications of 5 tons of compost plus 22.5 kg of P,O, per hectare (Veeraswamy et al., 1972a), and at Delhi from 1.29 to 2.76 tons of dry seed per hectare by applications of up to 100 kg of P,O, per hectare (Chowdhury and Bhatia, 1971a). Moreover, several elements appear to be essential for satisfactory nodulation and production of rhizobial nitrogen. Nichols ( 1965) demonstrated that deficiencies of calcium, phosphorus, and magnesium had a direct and greater effect on reducing plant growth and nodulation than nitrogen, potassium, or iron deficiencies. Singh and Archana ( 1964) investigated requirements for molybdenum observing the beneficial effect of molybdenum at 0, 1.25 and 2.5 ppm in pot culture on elongation of roots and shoots, dry matter accumulation, and free amino acid levels. The highest molybdenum treatments maximized accumulations of arginine, glycine, serine, asparagine, histidine, and lysine

42

K. 0. RACHIE AND L. M. ROBERTS

but decreased p-alanine in the shoots by day 41 after germination. Sulfur applied at 100 ppm alone or in combination with phosphorus was observed to increase the number, dry weight, and nitrogen content of Cajanus root nodules, and increased the methionine content and dry weight of plants in pot trials by Oke ( 1969 ) . Symbiotic Nitrogen Fixation. Pigeon peas utilize the same rhizobial complex as the cowpea group, but considerable variability in the bacterial strain and in the host X symbiont relationship does occur (Ramaswami and Nair, 1965). Nitrogen fixation and transfer of nitrogen to other parts of the plant can be quite efficient according to experiments carried out in Nigeria by Oke (1967). He found fixation to reach a maximum of 14.5 mg per day per plant in Cajanus compared with 10.3 and 4.6 mg per day per plant for Centrosemu and Stylosanthes. Younger plants were more effective than older ones in the fixation process. The beneficial effects of phosphorus, potassium, calcium, magnesium, molybdenum, and sulfur on nodulation as well as plant growth have been previously mentioned. However, nitrogen applications above 20 lb/acre tended to decrease yields of Cujunus in mixed cropping with millet in India-perhaps, in part, by increasing competition of the millet (Bhatawadekar et ul., 1966).

3. Enzyme Activity and Hormones The occurrence and activity of urease in Cajanus were demonstrated by Malhotra and Rani (1969). They recorded a specific activity of 1500 +- units per milligram of protein, and observed it to be inhibited at high substrate concentrations in Tris.acetic acid buffers and by alkali metal and nitrate ions. They concluded the activity of urease on urea and its derivatives to be complex and that the substrate binds to the enzyme through hydrogen bonding involving urea protons. Acid phosphatase activity was observed only in the nucleus and nucleoli of radicle cells of germinating seedlings by Kathju and Tewari (1968). They considered this finding unique since the autonomous cytoplasm inclusions-lysosomes and mitochondria-assumed to be the only centers of activity had none of the enzyme. Seed treatments with up to 0.5% solutions of B-nine (Ndimethylaminosuccinamic acid) were observed to inhibit shoot growth of Cajanus proportionately to the levels applied by as much as 70% in length and weight at the highest treatment levels in experiments conducted by Mishra and Mohanty (1,966). 4 . Management

Pigeon peas are used both in short- and long-term cropping systems for food, forage, cover, or multiple purposes. In many regions of India

GRAIN LEGUMES OF THE LOWLAND TROPICS

43

and Africa they may be used as hedges in family kitchen gardens and to support climbing vegetable and pulse plants (Grubben, 1970). In field plantings pigeon peas are often intermixed with other crops like maize, sorghum, or millet and are left to mature on residual moisture after the cereal is harvested. Therefore, row spacings vary widely depending on the companion crop and cultivar used. In mixed cropping pigeon peas may be planted after every 2-4 rows of the main crop; but in pure stands, spacings normally vary from 30 to 90 cm in the row and 90 to 300 cm between rows. As a forage crop, defoliating and slashing back to 90 cm stubble, heights at 3-5-month intervals produced highest vegetative dry matter yields. Population Experiments. Long-season types usually respond best to low populations of 7000-10,000 plants per hectare. Mukherjee (1960) obtained highest dry grain yields of 3530 kg/ha at spacings of 2 ft X 2 ft (about 27,200 plants/ha) and highest seed yield per plant at spacings of 4 ft X 4 ft (6800 plants per hectare). Hammerton ( 1971 ) investigated the effects of planting date and plant populations ranging from 4300 to 47,900 plants per hectare (0.21-2.32 m?/plant) on green pod and seed yields and components of yield in two dwarf cultivars grown in Trinidad. There was no effect of planting date on yield norits components, but a marked effect of plant populations. Increase in area per plant decrefiled pod yield per hectare, but increased pod yield per plant. The highest populations produced highest green pod yields of 8000 kg/ha, increased plant height at flowering and harvest but had no effect on the yield components (seeds per pod, mean pod and seed weights, and seed:pod ratio). Exceptionally high plant populations (165,000 plants per hectare) of determinate dwarf cultivars planted under short days in December in Trinidad were shown to give satisfactory seed yields (2.5 tons/ha), reduce plant height at harvest to about 1 meter and greatly facilitate mechanical harvest. Yields were comparable to those obtained from longer-duration cultivars planted at 6600 plants per hectare (Spence and Williams, 1972).

5 . Utilization and Composition The green pea makes an excellent vegetable constituting 45% of the weight of the whole pod. In this form it is about two-thirds water (normally harvested when alcohol-soluble solids reach 25% ), 20% carbohydrates, 7.0% protein, 3.5% fiber, 1.5% fat, and 1.3% ash. Dry, ripe seeds contain about 10% water, 23% protein, 56% carbohydrate, 8.1% fiber, and 3.8% ash. The protein is of reasonably good quality, but, like most grain legumes, is somewhat deficient in sulfur amino acids and tryptophan. However, the seeds are comparatively low in metabolic inhibitors and flatus sugars; and the testa is free of lipoxidase, which can cause off-flavors in soybeans and

44

K. 0. RACHIE AND L. M. ROBERTS

other legumes. Normally the seeds are split and testas are removed in the preparation of Indian dhal, but since the cotyledons contain most of the important nutrients-primarily proteins, phosphorus, copper, and ironremoval of the seed coat and testa during milling does not substantially lower the food value of this pulse ( S . Singh et al., 1968; Kurien and Parpia, 1968). However, storage pests, birds, and rodents do cause substantial losses prior to milling (Khare el al., 1966). Nutritional Values. A trypsin inhibitor in Cajanus has been isolated and characterized by Tawde ( 1961) . It was found quite active over p H range of 2.5-10.1 and was fairly heat stable. The deficiency of Cajanus seeds and meal in sulfur amino acids and tryptophan has been well established. However, it has also been shown to have a high biological value-comparable with black gram-when fed to mice at a 20% protein level (Daniel and De Berchoux, 1965; Ahsan et al., 1968). Braham et al. (1965) showed that 20-minute autoclaved meal at 121"C supplemented . with 0.1% tryptophan and 0.3% methionine was comparable to casein in rat diets fed atlO% protein level.

E.

POTENTIAL

The pigeon pea has exceptional potential for use over a wide range of tropical conditions from subhumid to semiarid regions. It is particularly valuable in mixed cropping and in bush-fallow systems of agriculture where a perennial crop of three to four years is desirable. It can be used both as field and garden crops for producing green seeds as a vegetable, dry seeds as a pulse, green leaves for cooking and for forage or as a cover crop. Green pod yields of 1000-8000 kg/ha (conversion ratio of green pods to dry peas is 3.3 :1) have been recorded; and good dry peas yields of 500-1000 kg/ha are realizable. However, favorable growing conditions can result in high yields of 1600-2500 kg/ha; while an exceptional yield of 5000 kg/ha of dry seeds was reported from India (RPIP, 1967), and Akinola and Whiteman (1972) obtained highest dry seed yields per year (7600 kg/ha) based on two major harvests from a single planting of cultivar UQ 50. V.

Cowpeas

The cowpea (Vigna unguiculata Walp.), also known as southern pea, blackeye pea, beans (West Africa), lubia, niebe, coup&,or frij6le appears to have originated in West Africa, very likely in Nigeria, where a profusion of wild and weedy species abound in both savannah and forested zones.

GRAIN LEGUMES OF THE LOWLAND TROPICS

45

It exists as herbaceous, erect, semiupright, prostrate spreading, and twiningclimbing forms. The cowpea was gathered or cultivated in prehistoric times in tropical Africa and must have reached Egypt, Arabia, and India very early, since it is known from Sanskritic times. The early Greeks and Romans knew of it, and it was introduced by the Spaniards into the West Indies in the 16th century, reaching the United States about 1700. A.

DESCRIPTION AND IMPORTANCE

The cowpea and its closely related weedy and wild relative have 2n = 22 and 2n = 24 chromosomes, but 22 is the more common condition. Outcrossing is low, depending on season and activities of pollen vectors. In subhumid parts of West Africa, the first rains are often characterized by a high level of bee activity resulting in outcrossing to the extent of 10% or more, whereas in the second rains insect activity is reduced and outcrossing may be less than 1%. Seed germination is epigeal, quick (48-72 hours), and usually very high. Adaptation and Production The cowpea is a predominantly hot-weather crop well adapted to the semiarid and forest-margin tropics. It is frequently mixed with other crops like maize, sorghum, millet, and cassava, but it is sometimes grown as a pure crop. Cowpeas are grown on a wide range of soil types from sands to heavy, expandable clays. Most cultivars do not tolerate waterlogging as well as soybeans. However, some forms like yard-long bean both tolerate and require higher rainfall than other cowpeas. Some varieties are daylength insensitive, while others require short days to mature within a reasonable time. Maturities may range from less than 60 days up to 7 or 8 months depending on genotype, and environment. Cowpeas tend to grow and spread very quickly thereby forming a quick cover to prevent soil erosion. Producing A reus. Cowpeas are grown extensively throughout the lowland tropics of Africa in a broad belt along the southern fringe of the Sahara, and in eastern Africa from Ethiopia to South Africa. They are mainly confined to the hot semiarid to subhumid areas with significant production in Nigeria, Niger, Upper Volta, Uganda, and Senegal. Nigeria alone produces about 61 % of the world crop or about 760 thousand tons annually. They are also extensively grown in India, southeastern Asia, Australia, the Caribbean, lowlands and coastal areas of South and Central America, and in the southern United States (primarily in the southeast, with some production in California). Frequently, the production of cowpeas is included under the general category “dry beans.”

46

K. 0. RACHIE AND L. M. ROBERTS

B. PLANTIMPROVEMENT Yields of cowpeas in West Africa are very low as a consequence of several constraints including: 1. The climate: Insufficient, poorly distributed or excessive moisture; low insolations; and extremes of temperature. 2. The soil: Poor physical structure, low water holding capacity, deficiency of organic matter; extremes of pH; low or unbalanced fertility; and unfavorable microbiological conditions. 3. Plant protection: Large numbers of insect pests at all stages of growth; and a complex of diseases including fungi, bacteria, viruses, and nematodes. 4. Weeds: Uncontrolled weed growth competition for moisture, nutrients, light, and space. 5. Cultural practices: Land preparation, planting methods, planting dates, populations, spatial arrangements, and fertilizer applications (mainly phosphorus amd potassium). 6. Genetics : Low productivity efficiency; limited range of adaptation; agronomic deficiencies (shattering and lodging) ;susceptibility to pests and diseases; and poor acceptability and nutrient values. I . Breeding Objectives

A strategy for breeding cowpeas has been described by Ebong ( 1970a,b). He stressed the importance of assembling and maintaining collections of geneticaIly diverse materials and breeding for ( 1) high yield,

(2) acceptable quality, (3) day neutrality, (4) erect growth habit, (5) long peduncles (above foliage), and (6) resistance to diseases (anthracnose, seedling blights, stem rot, viruses, and leaf spots). In francophone Africa, research on cowpeas was started in 1953 when variety trials, fertilizer experiments, and populations were studied and germplasm was collected (734 accessions; working collection is 222 entries). Most intensive efforts on “niebe” were made during 1962-1966 ( S h e and N’Diaye, 1970). Bambey, Senegal has been the center for hybridization and breeding with emphasis on erect, determinate plant types, while observations and selection are being carried out in Niger, Upper Volta, North Cameroons, and Dahomey (Silvestre, 1970b). In India, emphasis has been given to both pulse and forage types. Forage varieties like FOS-10, K. 397, and FOS-1 were found high yielding in green matter and useful as a dairy feed (Ralwani et d.,1970). Other pulse breeding programs from Punjab to Madras have concentrated on the dryseeded pulse types with drought resistance and maturity within 100 days (H. B. Singh et al., 1968; Veeraswamy et al., 1972b). Cowpeas have also

GRAIN LEGUMES OF THE LOWLAND TROPICS

47

received attention in the Philippines and Indonesia, where both vegetable ( V . sesquipedalis) and pulse types are grown, in the Caribbean and Central America (Aguirre and Palencia, 1968), and even in Russia (Medvedev, 1949). 2 . Improvement Methodology Breeding of cowpeas has largely followed conventional lines. Assembling, introducing, and testing germplasm constitute the all-important first link in this effort. Recombination of desirable characters through pedigree, backcross, and multiple crossing has been the major breeding technique employed. More recently, bulk-pedigree, single-seed descent, mutation breeding, and various population improvement systems have been considered. Wide crossing and mutation breeding have not been attempted seriously considering the vast array of genetic diversity already available. 3. Assembling and Evaluating Germplasm

Partial germplasm collections have been assembled by breeding programs in several countries (India, the United States, Nigeria, and others). These have recently been coordinated by the International Institute of Tropical Agriculture located at Ibadan, Nigeria. About 6800 different accessions had been assembled or collected by mid-1974. These will be grown out in total and partially in uniform nurseries to evaluate and catalog their important botanical characters, Records will be maintained on electronic cards for computer analysis and information retrieval. Genetic Variability. A collection of 1072 lines was evaluated in Jhansi, India, by Kohli et al. (1 971 ), and Mehra et a f . (1969). Variability in several characters was studied and appeared to be related to geographical source. Greatest range in number of days to flowering (42.7-77.6 days) was observed in Indian cultigens; African lines had the greatest range in dry matter per plant (14.2-74.6 g). African sources had the highest green forage weight, plant heights, and main branch lengths. Far-Eastern accessions had greater stem girths, high primary branch numbers (almost equal to African cultigens), but lowest dry matter content (11.1 g per plant). Metroglyph analysis on the analysis of characters produced groupings of 11 distinct plant types. 4 . Mutation Breeding

Increasing variability through mutation breeding has been explored to a limited extent. Uprety (1968) obtained a significant stimulation in plant growth and increase in protein nitrogen and proteinase activity in plants grown from seeds irradiated with 4 kr of gamma rays. Irradiation also accelerated flower initiation by 8 days. Irradiation greater than 4 kr ad-

48

K. 0. RACHIE AND L. M. ROBERTS

versely affected these characters, but 2 kr of gamma irradiation had no effect. Ojomo and Chheda (1971) found significant reduction in survival at irradiation levels of 20 kr of X-rays and 12.2 X 10" thermal neutrons/cm2. However, chromosomal disorders, leaf spots, and gross foliage distortion tended to disappear as growth advanced. Cekalin and Zelenskaja ( 1970) induced sterility by mutagenic treatments. Chemical mutagens-dimethyl sulfate (DMS) , ethyl methane sulfonate (EMS) , and N-nitrosomethylurea (NMU)-were studied by Sharma (1969). He observed reduction in fertility in the M, plants and obtained a late, giant-type mutant with large leaves, peduncles, and fruits. It also had thick, occasionally fasciated stems, large seeds with uniform black or mottled testas, trailing growth habit, and was very late. These mutants constituted 11.3% of all mutants recorded, maintained high fertility, and were high yielding. From these observations the author suggested the presence of a mutator gene with a wide range of activity causing other genes to mutate with differential but specific activity. Yield of mutants were about equal for DMS and EMS; but NMU was about twice as effective (18.2% mutations).

5. Developing Elite Strains Hand Crossing. Improuement in tropical cowpea varieties has been accomplished mainly through selection and recombination followed by simple or mass selection in segregating generations. Genetic recombination in cowpeas is possibly the easiest among the self-pollinated legumes. The flowers are large, the keel is not twisted, emasculation is quick, and seed setting can be very high (up to 50% or more) when conditions are favorable. Reasonably high humidity and moderate temperatures appear to favor setting in hand-manipulated flowers; but there is a strong genetic component as well since some parents are much easier to cross than others. Naphthalacetic acid in talc dusted into emasculated flowers reduced blossom drop and resulted in 30% setting in hand crosses (Barker, 1970). It is also highly desirable to do the crossing in the greenhouse or where pollinating insects are excluded; and to avoid high winds, rain, moisture stress, and high temperatures which cause heavy flower and bud drop (IITA, 1973). However, wide crossing with wild or other cultivated species of Vigna has been largely unsuccessful as either pollen germination fails or union of gametes does not occur. Sometimes when apparent fertilization takes place the embryos collapse soon afterward. 6 . Genetic Aspects

Genetic investigations have been comparatively limited in Yigm species. However, the work that has been done on inheritance, heritability, charac-

GRAIN LEGUMES OF THE LOWLAND TROPICS

49

ter correlations, nature of gene action, and combining ability has been very useful in support of plant improvement activities. Perhaps the most limited area of research is in evolutionary relationships, wide crossing and cytology, although Mukherjee (1968) has made a pachytene analysis and described the eleven chromosome pairs. a. Znheritance of Characters. The mode of inheritance of several simply inherited characters in cowpeas is briefly described in Appendix 6; and 14 linkage groups are listed in Appendix Table VII. b . Variability and Correlations. The major components of dry seed yield are pods per plant, seeds per pod, and 100-seed weight. Genetic variance studies have demonstrated a very wide range in variability of these characters, particularly for pods per plant. Variance estimates were also high for secondary characters like branches per plant, bunches (clusters) of pods per plant, days to maturity, peduncle length, pod length, and weight of nodules per plant; but they were lower for days to flowering and number of seeds per pod (Doku, 1970; Singh and Mehndiratta, 1969; Trehan et al., 1970). Heritability estimates were high for 100-seed weight, but medium for most of the other important primary and secondary yield components, except for days to flowering, which was low. Sbne (1968) found 100-seed weight to have a broad sense heritability of 0.80, and to appear to be controlled by six pairs of genes acting additively in the cross N58-25 X N5840 with 100-seed weights of 8-9 g and 19-20 g per 100 seeds, respectively. Correlation and regression analyses in varying numbers of cowpea cultivars have been carried out by Singh and Mehndiratta (1969, 1970), Doku (1970), Trehan et al. (1970), and Janoria and Ali (1970). Genotypic correlations were generally higher than phenotypic correlations. These results can be summarized as follows: 1 . Seed yield: highly and positively correlated with pods per plant, pod clusters per plant, seeds per pod, 100-seed weight, number of inflorescences per plant, days to maturity, and peduncle length. 2. Weight of 100 seeds: positively correlated with pod length; but negatively correlated with numbers of inflorescences per plant, pods per plant and seeds per pod. 3. Number of pods per pIant: highly correlated with pod clusters per plant; and to a lesser extent with seeds per pod. 4. Days to flowering: correlated with days to maturity. Partial regression analysis carried out by Janoria and Ali (1970) showed that pods per plant, seeds per pod, and 100-seed weight accounted for 83% of the variation in yield, whereas pods per plant and 100-seed weight accounted for 64% of the variation in this character. Singh and Mehndiratta (1970) found these three components of yield together accounted

50

K. 0. RACHIE AND L. M. ROBERTS

for 68% of the yield variation in path coefficient analysis; and that selection based on discriminant function involving the three components was 33% more efficient than selecting directly for yield. c. Combining Ability. Studies on combining ability in a diallel cross of four cowpea cultivars were carried out by Kheradnam and Niknejad (1971) in Iran. They found general and specific combining ability effects were significant for yield per plant, pod clusters per plant, number of seeds per 25 pods, seed weight and flowering date, but not significant for branches per plant. The ratio of general to specific combining ability was close to one for yield and pod clusters per plant, but general combining ability was more important for seed weight, days to flowering, and seeds per 25 pods.

C. INSECTPESTS Insects attacking cowpeas in all stages of growth and in storage are probably the major limiting factor in cowpea production in the low humid tropics. Effective control of insect pests in these circumstances often returns 10-30 times the productivity of unprotected crops (IITA, 1973). In Africa and Asia about 15 major and more than 100 minor species attack the cowpea crop. Among these the most serious control problems involve the foIlowing: Stage of growth 1. Early seedling growth, foliage, flowers, and pods 2. Green foliage (chewing,

rasping)

3. Flowers and floral buds 4. Floral buds and pods

5. Stored seeds

Insect species

Empoasca fascialds Taeniothrips sjostedti Sericothrips occipetalis Ootheca mutabilis Zonocerus spp. Spodoptera spp. Thrips (same as above) Hemiptera spp. Coreid spp.

Maruca testulelis Laspeyresia ptychora Melanagromgza vignalis Heliothis spp. Bruchideae Laspeyresia Others

Other species in addition to those mentioned above and important in the drier regions of West Africa have been noted by Delassus (1970): ( 1) Melangromyza pktaseoli-the bean fly attacks the young developing

GRAIN LEGUMES OF THE LOWLAND TROPICS

51

shoot. ( 2 ) Sphenoptera sp. (Buprestide)-also attacks the young stems; ( 3 ) Hemiptera/coreid species puncturing the floral buds and developing pods include: Anoplocnemis, Acanthornia, and Tassidedes; (4) pod and seed borers-including Piezotrachelus varium, Dendorix sp., and Lampides SP. In the Americas the cowpea curculia (Chalsodermus aeneus Boh.) is a very serious pest in the southeastern United States; and the bean leaf beetle (Cerotoma ruficornis Oliv.) is a vector for cowpea mosaic in the Caribbean. Otherwise, Phaseolus bean pests can also be significant problems on cowpeas.

1. Losses from Insects Direct losses in grain yields resulting from uncontrolled insect attack have been estimated in experiments carried out in Ibadan by Dr. W. K. Whitney (IITA, 1973). Individually, these losses may be as high as in the following cases.

Insect

Yield reduction (%)

Thrips Maruca-flower damage Maruca-pod damage Hemiptera-seed damage 5 . Laspeyresia-seed damage

1. 2. 3. 4.

50 20 LO

35 50

Collectively these estimates exceed loo%, and productivity is virtually nil without some control in certain seasons in the humid tropics. 2. Chemical Insecticides

The basic approach to insect problems of cowpeas has been through chemical insecticides. The best of these from several points of view are endosulfan (Thiodan 50 WP) at 0.15% a.i. in water or 0.9 kg a.i./ha; lindane 50% WP (Gammalin) at 0.14% a.i. concentration or 0.6 kg a.i./ha (if emulsifiable compound is used, concentration should not exceed 0.5 % concentration) ; azinophosmethyl 25 % WP (Gusathion M ) at 0.14% a.i. concentration or 0.6 kg a.i./ha; dimethoate 30% EC (Rogor 40) at 0.03% a.i. concentration or 0.2 kg a.i./ha; and Gardona 75% WP at 0.07 to 0.14% a.i. concentration or 0.3 to 0.6 kg a.i./ha. Combinations of Thiodan and Rogor 40 applied six to eight times during the growth cycle provides a high level of control, but often as few as two or three sprays are highly profitable under commercial practice. Gardona alone or

52

K. 0. RACHIE AND L. M. ROBERTS

mixed with Thiodan may be more effective against pod-boring species during the postflowering period (IITA, 1973). Some newer insecticides show considerable promise, including methomyl 90% (Lannate), Orthene 75 SP, Dupont 1410 20% EC and Carbofuran 75% WP (Furadan). Carbofuran and methomyl applied to the seed as a pelleting treatment at 0.5 to 2.0 g per 100 g of seeds or in the seed furrow at 1 kg a.i./ha have been shown to protect the plants for up to 6-7 weeks after planting. Some applications are comparatively inexpensive and, when combined with 2 or 3 postflowering foliar applications (Gardona or Gardona Thiodan), should provide a high level of economical plant protection throughout the growing period of the crop. Finally, the harvested, threshed, and properly dried seeds can be safely stored in tightly closed plastic bags of 0.3 mm thickness, holding 40-50 kg of grain, and to which 18 g of carbon tetrachloride have been added (Caswell, 1968).

+

3. Cultural and Biological Controls

The possibilities for cultural and biological controls and host plant resistance must not be overlooked and should be investigated intensively in the future. There already appears to be a genetically controlled mechanism for low level tolerance or resistance to thrips (IITA, 1973); but the possibility of resistance to pod borers has not yet been established. Lorz (1970) suggested that ZIPPER CREAM with thick pod walls might have resistance to pod-puncturing insects. Todd and Canerday (1969) observed Fla 453-01 to be least damaged by the cowpea curculio, and Ala 963-8 and Va 59-119 impaired larval deveIopment of the pest. Chandola et at. (1969) found T.2 to be resistant to Bruchus sp. Generally then, the pests of cowpeas in the tropics are indeed formidable, and problem solving must proceed on all fronts-both in the direction of rapidly developing chemical protectants as well as the longer-term cultural, biological, and host plantresistant aspects. D.

DISEASES AND NEMATODES

There are several major disease problems of cowpeas in the lowland tropics, although overall reduction in yield may be less than insect predations, at least in West Africa (IITA, 1973). The major problems in southern Nigeria include the following: I. Fungal and bacterial diseases 1 . Seedling blights and wilts a. Rhizoctonia solani b. Pythium aphanidermatum c. Secondary-Colletotrichum theobromae

sp.,

Fusariuin

sp.,

and

Botryodiplodia

GRAIN LEGUMES OF THE LOWLAND TROPICS

53

2. Stem blight-Colletotrichurn lindemutliianurn 3. Leaf spots a. Cercospora cruenta b. Cercospora canescens c. Bacterial pustule (Xanthornonas vignicola) 11. Virus diseases 1. Cowpea green mottle virus (green blister) 2. Cowpea yellow mosaic virus (or yellow flecks) 111. Nematodes 1. Root nematode (Meloidogyne incognita) 2. Root lesion nematode (Pratylenchus sp.) 3. Spiral nematode (Helicotylenchus pseudorobructus) 4. String nematode (Belonlaimus gracilis) IV. Phanerogram parasites Striga gesnerioides-a parasitic weed on the roots of cowpeas in tropical Africa

Delassus (1970) also mentions several species occurring in the drier francophone regions of Africa: 1. Rust or blight-Urornyces appendiculatus 2. General wilting or withering-Neoscosrnopora vasinfecta 3. Affecting stems, foliage and pods: Cercospora sp., Hetmintliosporiurn sp., Leptosphearda sp., Choenephora sp., and Rhizoctonia bataticola

1 . Losses in Production Separately and collectively, diseases can result in reduced grain yields and loss of foliage. In southern Nigeria losses in stand and grain productivity have been estimated as high as those tabulated below. Disease 1. Seedling blights (fungus) 2. Anthracnose

3. Leaf spots (Cercospora) 4. Bacterial pustule 5 . Viruses-yellow mosaic 5 . Nematodes-root knot

Reduction 75% 50% 30% 10% 50% 95%

Stand Yield Yield Yield Yield Yield

Seedling blights do not reduce yields directly, and if the ldss in stand occurs early, adjacent plants tend to compensate for the dying plant in terms of increased and extended branching and fruiting (IITA, 1973). In studies on Cercosporu leaf spots at Ibadan, direct yield losses from C. cunescens and C . cruenta were observed to be 18 and 42%, respectively (IITA, 1973). A single benomyl (Benlate) spray applied 5 weeks after planting controlled these diseases in a determinate variety, but reduced yields by 20% in an indeterminate variety.

54

K. 0. RACHIE AND L. M. ROBERTS

2. Fungicides The most practical and promising approach to disease problems generally is through host plant resistance and by chemical seed dressings. A new systemic fungicide, chloroneb (Demosan 65 W) alone or in combination with thiram at the rate of 2 g per kilogram of seeds has given excellent results in experiments carried out at Ibadan (IITA, 1973). Potassium azide at 2000 ppm used for soaking seeds for 5-10 minutes protected seedlings for up to 21 days from Rhizoctonia sp., Sclerotium rolfsii and Pythium sp. in experiments carried out by Gay (1970) in Georgia, USA. Other fungicides like Dithane M-45, benomyl (Benlate) , carboxin (Vitavax) or the copper-based compounds like Perenox and Kocide can be used as foliage sprays to control several diseases but are seldom economic on a commercial scale. Host Plant Resistance. Preliminary evaluation of comprehensive germplasm collections indicates that several sources of tolerance or resistance to all major diseases and viruses in southern Nigeria are available (IITA, 1973). Earlier investigations have identified many sources for resistance to various diseases, but more recent reports describe the following specific host plant resistance: 1. Rust (Urornyces phaseoli var. VignUf?)+UEEN ANNE is immune to all races; previously PmKEYE PURPLE HULL, TEXAS CREAM, and CREAM 40 were resistant (Heath, 1971; Gay, 1971). 2. Viruses (bean yellow mosaic, cucumber mosaic, mottle viruses)-ALABAMA 3-6-5, ALABAMA 91-7, and PRINCESS ANNE were resistant (Harrison and Gudauskas, 1968); cowpea mosaic virus in India-MS 9081, EC 2085, EC 4216, and EC 4203 were resistant (Govindaswamy et al., 1970; Khatri and Chenulu, 1971).

The nature of rust immunity in QUEEN ANNE results from cell necrosis and formation of calloselike sheaths enclosing the haustorium; in other varieties, resistance was due to hypersensitivity and death of the invaded host cells in observations made by Heath (1971).

E. PHYSIOLOGY Information on physiological processes in the broader sense includes both developmental aspects and the influence of external factors interacting with those processes. Among these external factors are daylength, light quality, temperature, elevation, humidity, water relationships, pests and diseases, mineral nutrition, soil physical and chemical characteristics, harvesting procedures, and other factors.

GRAIN LEGUMES OF THE LOWLAND TROPICS

55

1. Reproductive Ontogeny Flowers are borne in racemose inflorescences at the distal ends of peduncles arising from leaf axils. They are arranged alternately in acropetal succession with up to 8-12 pairs per inflorescence, although usually only the first two pairs develop. The flower is large, has a straight keel, and is normally white or blue. There are ten diadelphous stamens, the vexillar one being free; the ovary is sessile and multiloculate, and the style is bearded along the inner side, ending in an oblique stigma. A high rate of abortion occurs in the cowpea plant which normally produces 100-500 flower buds of which 70-88% are shed before anthesis. Of the remaining 12-30% up to half abort prematurely, so that only 6-16% of the total flower buds produce mature fruits (Ojehomon, 1968a,b). Flower loss is compensated for by setting more buds on secondary and tertiary branches, but this mechanism is limited. Ojehomon (1970) demonstrated a 43% reduction in seed yield per plant by removing all flowers for 12 days after anthesis. However, individual flowers represent very little loss in dry weight (0.01 g) and therefore do not constitute a very large sink (Summerfield el at., 1973). 2 . Light and Photoperiod Short-day cultivars are commonly grown in the higher tropical latitudes to assure maturation toward the end of the rainy season with more efficient utilization of moisture and production of better seed quality through regulation of flowering toward the end of the rainy season. Photoinsensitive varieties are often grown in low tropical latitudes and in long-day temperate regions. However, short-day plant types will often produce excessive vegetation to the detriment of grain yield when planted earlier than optimal for that cultigen (Summerfield et al. 1973). The optimal photoperiod for induction of flowering in cowpeas is 8-14 hours from studies carried out by Wienk (1963) on 14 cultivars and 15 photoperiods varying from 6 to 24 hours. Moreover, many cultivars respond to light quality by becoming etiolated with twining leaders under reduced light. This effect results in the climbing habit when grown in association with other crops and weeds. Under artificial lighting, the use of “daylight” fluorescent tubes with a 5 % tungsten supplement was found to normalize growth in cabinet experiments in England (Summerfield and Huxley, 1972). Photosynthesis. Studies on absorption of radiation at the 2-leaf and 4-leaf stages in Guadeloupe (French West Indies) showed that the greatest proportion of radiation is adsorbed in the morning when conditions are favorable for photosynthesis, but at midday adsorption was less, thereby

56

K. 0. RACHIE AND L. M. ROBERTS

preventing a buildup of high temperatures in the leaves (Varlet-Grancher and Bonhomme, 1972). 3. Temperatures

Radiant energy directly affects both air and soil temperatures, which in turn can limit the growth process in various ways. In some recent experiments carried out at the University of Reading, Summerfield et al. (1973) have amply demonstrated the profound effects of night temperatures (19" and 24°C) on both vegetative and reproductive development in terms of growth, days to first flower, and seed yields in 30 cultivars studied. Moreover, air temperatures also influence rhizobium activity and nodulation as shown in experiments conducted by Dart and Mercer (1965). Maximum dry matter production occurred at 27°C day and 22°C night temperature in the cowpea rhizobial system in the range of 21" to 36°C day and 16" to 3 1 "C night temperatures studied. They concluded that air temperature is of considerably greater importance than either light intensity or nitrogenous fertilizers in the symbiotic system. The conclusions reached by these investigators are that initiation of the reproductive process occurs as a balance between vegetative growth and concentration of the flowering stimulus, and that t,emperatures could act to either increase vegetative growth or reduce the floral stimulus. 4 . Water Requirements

The cowpea can be highly drought resistant, being grown in subhumid to semiarid conditions. However, it may also be reasonably tolerant of high soil moisture from preliminary studies on hydromorphic soils at Ibadan (IITA, 1973). Most of the cowpea crop is grown under rainfed conditions, but it may also be grown with surface or sprinkler irrigation in different parts of the world; or on residual moisture on soils with high water-holding capacity (e.g., after rice). Moisture stress resulting in temporary wilting can reduce productivity considerably during the period from emergence to first flower, but may not significantly affect yields thereafter in determinate varieties (Huxley and Summefield, 1973). However, Doku (1970) found nodulation to be reduced when moisture was limiting particularly when combined with long days ( 16 hours vs 1 1 hours, 48 minutes).

5. Mineral Nutrition The requirements of cowpeas for nitrogen, phosphorus, potassium, calcium, magnesium, and sulfur have been partially established under certain conditions and for specific genotypes. However, there is very litle information on minor nutrient requirements, except for assumptions based on re-

GRAIN LEGUMES OF THE LOWLAND TROPICS

57

sults with other grain legumes. Thus, molybdenum, manganese, copper, zinc, and boron may be required for effective nodulation and increased productivity. It has been estimated that each ton of cowpea seeds exports the following minerals: N = 40 kg, P,05 = 17 kg, K,O = 48 kg, CaO - 16 kg, MgO = 15 kg, and S = 4 kg. Information available on major nutrients other than nitrogen is summarized as follows: 1. Phosphorus-field applications of 100-200 kg/ha of superphosphate are commonly recommended (Sellschop, 1962). Absorption of phosphorus occurs primarily at the end of the growing period and is mainly transported to the seed (Jacquinot, 1967). Foliar application may be more efficient than soil application (Mohta and De, 1971). 2. Potassium-this element is transported mainly to the stem in early growth and later to the seeds (Jacquinot, 1967). Field response to potassium by cowpeas has been low in Africa although K,O applications of 40 kg/ha have increased nodulation in eastern Nigeria (Tewari, 1965). In the southeastern United States, Worley et al. (1971) recommended that the level of elemental potassium not exceed 42 kg/ha. 3. Calcium-this element stimulates nodulation and may affect the release of molybdenum in acid soils. Most calcium is taken up during the first 40 days of growth, but may accumulate in the leaves in replacement of potash during later growth (Jacquinot, 1967). On acid soils cowpeas may benefit from application of 1.6-2.8 tons of lime per hectare (Sellschop, 1962). 4. Magnesium-uptake is maximal during the last third of the growth stage and foliar concentrations are slightly higher than in other plant organs (Jacquinot, 1967). Growth responses to magnesium have not been studied. 6. Nitrogen Fixation

Among the major elements, very little response to nitrogenous fertilizers is realized when seeds are properly inoculated or the appropriate rhizobial cultures occur in the soil. Therefore, it is usually more efficient to improve conditions tending to maximize the rhizobial process: 1. Inoculation with efficient strains of rhizobium if nodulated cowpeas or related species have not recently been grown on the land. Efficient strains may double yields in comparison with some indigenous rhizobium (IARI, 1971). 2. Improving soil moisture and mulching increased nodule production, but excessive cultivation decreased nodulation in Malaya (Masefield, 1957). 3. Temperatures of 24°C were optimum for primary root nodulation; but 33°C was found optimum for secondary root nodulation by Dart and

58

K. 0. RACHIE AND L. M. ROBERTS

Mercer (1965). Number of plants nodulating and nodulation per se were decreased in a linear manner as temperatures increased from 3 1“C to 42°C (Philpotts, 1967). 4. Photoperiod affects nodulation and the latter is reduced by photoperiods longer than 16 hours although moisture may not be limiting (Doku, 1970). 5 . Applications of phosphorus increases nodulation (Tewari, 1965), but nodulation is reduced by high soil nitrogen during early growth (Ezedinma, 1964). 6. Selfing tended to increase nodulation when two lines were crossed factors” and the possibility together, suggesting the accumulation of of increasing nitrogen-fixing efficiency through breeding (Doku, 1970). Fixation of nitrogen by cowpeas was estimated by Nutman (1971) at 73-240 kg/ha per annum. Nitrogenous substances are usually concentrated in the leaves during vegetative growth and then transported to the seeds during the grain filling period. It is further estimated that about 40 kg of nitrogen are exported from each ton of cowpea seeds harvested from a hectare of land (Jacquinot, 1967). Generally then, it can be assumed that the symbiotic nitrogen fixing process is adequate for cowpea production at current productivity levels if conditions for nodulation are favorable.

“+

F. MANAGEMENT Productivity in cowpeas grown in Africa is very low being only 100-300 kg/ha of dry seeds. This is attributable to its being cultivated in subsistence agriculture as a secondary crop in association with cereals like sorghum, millet, or maize. In this situation cowpeas are frequently planted broadcast at 22-33 kg/ha after the cereal is about 50 cm tall. After germination the seedlings are often thinned out (and used as a pot herb) depending on the cultivator’s judgment on availability of moisture which may be in excess of the cereal’s needs that season, and to maximize dry seed yields of the “beans.” In this system, cereals constitute the major energy source for the cultivator and his dependants and are given first priority in the season’s activities. If additional lands are cultivated beyond the yearly subsistence needs they are usually planted to a cash crop like cotton or groundnuts. 1. Associated Cropping

Mixed cropping has been demonstrated to be more highly productive than sole cropping of millet and cowpeas in Niger. The mix grown on 1 hectare plots produced yields of 682 kg and 1525 kg of peas and millet, respectively, compared with 1072 kg of peas and 905 kg of millet from pure stands of one half hectare each. However, profits from sole cropping

GRAIN LEGUMES OF THE LOWLAND TROPICS

59

were greater as a consequence of the higher market value of cowpeas (Nabos, 1970). Of course, cowpeas can also be quite successful and profitable grown as a sole crop when pests and diseases are controlled. Yields ranging from one to three tons per hectare are obtained in small plots, and at these levels would more than compensate for both the costs of protection and the returns from the associated crop at current market valuations. Nevertheless, adoption of monoculture systems may depend on developing both high-yielding cereals and legumes together with “packages” of reliable production practices to enable nonmechanized agricultural units to attain higher levels of productivity on limited cultivated areas. 2. Date of Planting Day neutral cowpeas can be planted any time of the year in lower tropical latitudes when moisture and fertility are adequate and satisfactory pest control is achieved. At Ibadan, good growth of photoperiod-insensitive cultigens occurs at any time of the year, including the dry season, late November to March, if irrigation is available (IITA, 1973). However, it is highly desirable for maturation to occur during bright, sunny weather to reduce pod and seed damage from both insects and diseases (McDonald, 1970b). Since most cultivars commence flowering optimally from 35 to 70 days after germination, date of planting should be so timed that protracted rainy periods are over by the time the crop begins flowering. Thus, at Ibadan, late May and late August are better for day neutral types, whereas at higher latitudes in monomodal or monsoon type climates, a late June or early July planting may be preferable. Daylength-sensitive strains should not be planted in the first season in bimodal rainfall regions.

3. Populations and Spacings In mechanized agriculture, cowpeas are usually planted in rows 75-100 cm apart, 7-10 cm within the row, and at a seed rate of 17-28 kg/ha. For forage, cowpeas may be broadcast at seed rates of up to 100 kg/ha and mixed with sudangrass, sorghum, or maize (Sellschop, 1962). In African mixed cropping systems cowpea seeds are frequently broadcast at a seed rate of 22-33 kg/ha. In francophone Africa, hill plantings (2-3 seeds per drop) are recommended at spacings of 50 x 50 cm or 50 X 60 cm for early cultivars, and wider for late or spreading varieties (Silvestre, 1970b). 4 . Fertilization Fertilizer experiments in West Africa have shown low but significant responses to the three major nutrients nitrogen, phosphorus, and potassium (Nabos, 1970). The most common recommendation is for phosphorus at 20-60 kg of P,O, per hectare, but potash may be included at the rate

60

K. 0. RACHIE AND L. M. ROBERTS

of 30-60 kg of K,O per hectare if known to be deficient; and perhaps a light application of nitrogen-I 5-30 kg of nitrogen per hectare-can be profitable. Worley et al. (1971) made definite recommendations on plant nutrients for southern Georgia, USA: nitrogen, 27-55 kg/ha; phosphorus, 12-24 kg/ha; and potassium, not more than 46 kg/ha. They further observed contents of copper, zinc, aluminum, and titanium in the foliage to be negatively correlated ( r = -0.36) with seed yields. In India, foliar sprays of 60 kg P,O, per hectare were even more effective than the same application banded beside the row (Mohta and De, 1971 ).

5 . Weed Control Weed competition becomes the major constraint when other factors are not limiting. Mechanical cultivation or hoeing may be the most practical means of control under most tropical conditions, but several weedicides have been tried with varying success in different regions. Trifluralin at 0.56-1.12 kg/ha applied presowing and immediately harrowed or rotovated in has given good control in the United States (Ogle, 1967). Chloramben (Amiben) at 3 kg/ha has generally given good results; but hand weeding, for the first month after planting of short duration, determinate cultivars, has given as good yields as clean weed control plots in southern Nigeria (IITA, 1973). 6. Cover Cropping The effectiveness of cowpeas in rapidly covering the soil surface and preventing loss of topsoil has been amply demonstrated in runoff experiments in southern Nigeria. Cowpeas proved superior to maize and other cereals for this purpose (IITA, 1973) . 7 . Harvesting the Leaves The young shoots, leaves, and even roots of cowpeas are used as pot herbs in most parts of Africa. If the tender green leaves are plucked before the reproduction phase begins, the plant continues to produce new leaves. Mehta (1971) demonstrated that it was possible to remove all tender leaves up to a maximum of three times at weekly intervals during the vegetative stage of growth without reducing the final seed yield. G . UTILIZATION The primary use of cowpeas is for the dry pulse, but the green pods, green seeds, seedlings, and tender young leaves are often used as pot herbs. Canning and freezing shelled green peas has become an important industry in parts of the United States in recent years, having exceeded 40 million pounds by 1971. The vegetation also makes excellent hay, and the surplus culled and broken seeds can be used as a protein concentrate for domestic

GRAIN LEGUMES OF THE LOWLAND TROPICS

61

animals. Cowpeas cook more easily and quickly than Phaseolus beans and are therefore favored when fuel is scarce. Cowpea hay is high in nutrients, and its fiber is more easily digested than lucerne fiber. Moreover, it is excellent for grazing by milk-producing animals. Cowpeas are the preferred pulse crop in many regions, particularly in tropical Africa. This is fortunate since they do provide an important source of proteins, caloric energy, and other nutrients with a minimum amount of cooking or preparation. Moreover, the levels of toxic substances and antimetabolites like the trypsin inhibitor, hemagglutinins and flatus factors are minimal in the cowpea (Liener, 1969). 1 . Food Preparations In Africa cowpeas are consumed in three basic forms of which there may be many variations. Most frequently they are cooked together with vegetables, spices, and other ingredients to make a thick soup or gruel which is eaten in association with the basic staple such as preparations of cassava, yams, plantain, or cereals. The second preparation would be as deep-fried cakes (akara balls) prepared from a dough of decorticated cowpea flour to which onions and seasonings are added. The third preparation is steamed bean cakes (moin-moin in Nigeria) prepared from decorticated cowpea flour to which chopped onions and seasonings have been added. In preparing the flour, the testas are removed first by soaking in water and rubbing. Rough or wrinkled testas are preferred, as they soak quickly and are easily removed. 2 . Nutritive Qualities Although low in toxic substances, cowpeas have been shown to contain trypsin and chymotrypsin inhibitors (Ventura and Filho, 1967) and may have a cyanogen as high as 2 mg per 100 ml of extract (Montgomery, 1964). Therefore, cooking is needed to inactivate these undesirable principles. In terms of proximate principles, the dry pulse contains the following constituents. Constituent

Percent

Water Protein Carbohydrate Fat Fiber

11 .o 23.4 56.8 1.9 3.9 9.6

Ash

Contents of calcium (90 mg/100 g), iron (6-7 mg/100 g), nicotine acid (2.0 mg/100 g) and thiamine (0.9 mg/100 g ) are high and contribute substantially to these requirements in tropical diets (Platt, 1962).

62

K. 0. RACHIE AND L. M. ROBERTS

VI.

Mung Beans

Green, golden, and black gram are now considered to belong to the same species, Vigna radiata (L.) Wilczek formerly placed in Phaseolus as P . aureus Roxb. (mung bean, green and golden gram) and P . mungo (L.) Hepper, (black gram, urad, mash, woolly pyrol). Verdcourt (1970) has recommended retaining the subspecies designation to maintain the separate identities of the two races as V . radiata var. aureus for mung bean/green gram and V . rudiata var. mungo for black gram. The putative wild ancestors are believed to be Vigna (Phaseolus) radiata var. sublobata (Roxb.) Verdc. and/or Vigna (Phaseolus) trinervius (Wight and Am.) Verdc., which occur wild in India. The two subspecies can usually be distinguished from each other by the pod characteristics. The variety aureus has spreading or reflexed pods with short hairs, globose seeds, and flat seed hilums, whereas in the variety mungo the pods are erect or suberect with long hairs, the seeds are larger, oblong, and smooth, and the hilum is concave. For purposes of this paper both subspecies will be referred to under the term “mung beans” unless otherwise specified as green or black gram. A.

IMPORTANCE AND UTILIZATION

Mung beans are important crops in southeastern Asia, and particularly in India, where about 0.30 million tons of green gram and 0.44 million tons of black gram are grown annually on 1.4 and 1.5 million hectares, respectively. They are particularly esteemed for their excellent quality, high digestibility and freedom from the flatulence effect associated with other pulses. They are frequently fed to children, convalescents, and geriatrics or used when “breaking” a long fasting period, owing to their ease of digestibility. Black gram is particularly highly prized in the vegetarian diets of high caste Hindus. The haulms are used for fodder, and the bean husks and small broken pieces are useful as a feed concentrate. The crop is also grown for hay, green manure, and cover crop. Green gram makes better hay than black gram, as the stems and leaves are less hairy. 1 . Food Preparations The dried pulse can be split or eaten whole after cooking and made into a soup of dhal (porridge) to be eaten with a cereal. The beans can also be used in various deep-fried and spiced dishes like noodles, balls, or snacks, or baked in bread and biscuits after parching removing the testa and grinding into flour. They are also widely relished as bean sprouts, which are prepared by soaking the dry beans overnight, draining, placing

GRAIN LEGUMES OF THE LOWLAND TROPICS

63

them in containers in a dark place and sprinkling with water every few hours. In about a week the sprouts are ready to eat. One kilogram of dry beans makes 6-8 kg of sprouts. The green pods and seeds can be cooked as vegetables (Purseglove, 1968). 2 . Ecology Mung beans are grown in the southeastern Asia at low to intermediate elevations, on rainfed lands, and frequently following rice. They perform best on good loamy soils with a well distributed rainfall of 750-900 mm per year, but are reasonably resistant to drought and somewhat susceptible to waterlogging. However, black gram is well adapted to clayey soils and is frequently grown on black cotton soils in India. Mung beans are also grown to a limited extent in Eastern Africa, primarily to cater to the Asian demand, and show considerable agronomic potential in West Africa as well (IITA, 1973). a. Alkalinity and Salinity. Both black and green gram grow well under both alkaline and saline conditions. One cultigen of black gram (T.9) was quite tolerant of both salinity (up to 4.3 x 10 13 EC) and alkalinity (90% exchangeable calcium) in Uttar Pradesh, India (Mehrotra and Gangwar, 1964). In other experiments Sharma et al. (1971) obtained good germination of cultivars Nos. 19, 21, and 3 in solutions of NaHCO, and NazC03 from 1-6 mmhos/cm at 25°C. Other cultivars, Nos. 36 and 92, RS 4, and RS 5 , were considered suitable for saline soils when CaC1, was not the predominant salt. b. In Cropping Sysrems. Mung beans are frequently grown as short-term crops following the main crop, such as rice, utilizing end-of-season precipitation and residual moisture, or are intercropped with other species like cereals, sugarcane, or cotton. In Taiwan, mung beans have been highly profitable intersown with spring-planted sugarcane (Tse and Hsueh, 1965). They can also be undersown with a cereal, like maize, in north India, and when planted within 2 weeks of maize produced above 200 kg of dry seed per hectare (Pathak er al., 1968). Mung beans having small seeds and quick early growth make excellent cover and green manure crops in Australia, producing up to 9 tons of green material for plowing down (Chapman and Garioch, 1966; Gonzales, 1962). B.

DESCRIPTION AND

VARIETIES

Morphology

The mung bean is an erect or suberect, deep-rooted, much branched, rather hairy, annual herb, 0.3-1.5 m tall. In some respects it resembles

64

K. 0. RACHIE AND L. M. ROBERTS

cowpeas, but tends to be more erect and is less twining. The leaves are alternate, trifoliate, and dark or medium green; the leaflets are ovate and are sized about 1.5-12 X 2-10 cm; and the petioles are long. The inflorescence is an axillary raceme, and the peduncle is 2-13 cm long. The standard is yellowish and 1.1-1.7 cm in diameter. The keel is spirally coiled with a hornlike appendage. Germination is epigeal. Other distinctive characters useful in distinguishing these two subspecies from other Vigna and from each other are listed in the accompanying tabulation.

Subspecies Character Plant height Stipule shape Inflorescence type Florets per peduncle Calyx bracts Pod attitude Mature pod size Mature pod color Pod hairiness Seeds per pod Seed shape Weight of 100 seeds Seed color Hilum Testa

aureus Up t o 1.5 m Ovate Axillary raceme 10-20 Same length as calyx Spreading, reflexed 0.4-0.6 X 4-10 cm Gray or brownish Moderate, short hairs 10-15 Seeds Globular 5-5 g; up t o 8 g Green, yellow, blackish Round, white, flat Has fine, wavy ridges

mungo Up t o 0.80 m Falcate Raceme may be branched 5-6 Longer than calyx Erect or suberects 0.6 X 4-7 cm Buff to dark brown Profuse longer hairs 6-10 Seeds Oblong, square About 49 g Black, occasionally green White, concave Smooth, without ridges

0 Pods of var. mungo are shorter, thicker, hairier and have a characteristic short, hooked beak.

a. Embryology. Embryology and seed structure follow that of other Vigna species and were described by Misra and Sahu (1970). They observed the ovule to be campylotropous, bitegmic, and crassinucleate and that the embryo with cotyledons occupies the entire seeds. The nuclear endosperm disappears upon maturation, and the testa is formed from an outer palisadelike epidermal layer and some cells from the outer integument. b. Varietal Types. There are two major types in var. aureus depending on seed color: (1) yellow or golden gram has yellow seeds, is generally low in seed production, has a tendency to shatter and is used mainly for forage or as a cover crop; ( 2 ) green gram has dark or bright green seeds, is more prolific, ripens more uniformly, has less ten-

GRAIN LEGUMES OF THE LOWLAND TROPICS

65

dency to shatter and is more commonly planted as a pulse. The bright green types are preferred for sprouting. C. PLANTIMPROVEMENT The major volume of information and improvement work on mung beans has been done in India in recent years. There have also been limited reports from southeastern Asia and tropical Americas. The following sections will treat recent progress in plant improvement in green and black gram during the later 1960's and early 1970's. A bibliography of 344 references comprising the world literature on mungbean has been compiled by Poehlman and Yu-Jean (1972) of the University of Missouri. 1. Pollination Mung beans including black gram are highly self-pollinated and have about 42 % cleistogamy. Flowering usually commences 6-8 weeks after planting. Pollen is shed on the evening before the flowers open. By the afternoon of opening, the petals fade and drop off. Rain is detrimental to good pod setting. Therefore emasculation must be done on the morning of the day before flower opening. Sometimes pods form but have no seeds. Hand emasculation is somewhat complicated by twisting of the keel, which usually does not exceed 360". The chromosome complement is 2n = 22. 2. Germplasm Evaluation

More than 2000 mung bean accessions had been assembled by mid-1973 at the Asian Vegetable Research and Development Center (AVRDC) located in Taiwan. The AVRDC will become a global center for assembling, maintaining, and evaluating germplasm for this species (McKenzie, 1973). However, smaller collections have been studied at other locations, chiefly in India, where 878 accessions were assembled and evaluated by the Regional Pulse Improvement Project (RPIP, 1967, 1968, 1969); in Azerbaijan, USSR (Rimikhanov, 1968) ;and at the University of Missouri, where 249 strains were assessed for several botanical characters in both replicated and unreplicated trials (Yohe et al., 1972; Yohe and Poehlman, 1972; Watt et aE., 1973). Variation in botanical characters and components of yield have been evaluated in studies on 16 green grams and 12 black gram cultivars representing a broad range of genetic diversity in the two subspecies at Hissar, India (Chowdhury et al., 1968; Chowdhury ct al., 1969). In the Missouri studies the range of variability in 12 plant characters was evaluated over a three-year period from 1970 to 1972 at Colombia,

K. 0. RACHIE AND L. M. ROBERTS

66

Missouri. Variability in all characters was considerable, as demonstrated in the accompanying tabulation. 1971 (203 cvs)

1972 (70 cvs)

Character

Range

Mean

Range

Mean

Seed yield (kg/hs) Plant type' Leaf sizeb Days t o flower Days t o ripe" Plant height (cm) Branch length (cm) Pods per plant Seeds per pod 1000 seed weight (g) Virus sccred Mildew scoree Protein/ Lysineg

12-2548 1-6 2.8-7.5 40-119 56-120 24-82 13-76 4-255 5-14 24-75 1-100 1-5 22.1-31 . 2 5.90-8.45

924 3.2

10.7-1966 1-5 5.0-7,9 43-84 59-112 42-94 34-85 20-192 7-14 22-86 1-80 1.7-5.0

1273 2.7 7.2

~~

~

~~

~

5.7

59 880 58 51 115 10.7 45 34.1 3.6 26.5 7.27

58

79 67 61 106 11.5 58 18 3.2

~~

1 = prostrate; 5 = erect. 1 = 6.1 ema; 4 = 26.0 cm2; 8 = 145.4 cm2. c Days t o first ripe pod. d Proportion of plot showing virus symptoms. 6 1 = resistant; 5 = susceptible. f Kjeldahl analysis of seeds. Expressed as percent of protein. a

3 . Breeding Methodology

Conventional methods have been used almost exclusively in breeding mung beans. These include pedigree selection, mass selection, backcrossing, multiple line crossing, mutation breeding, and interspecific crossing. Radiation with X-rays was reported by Van Emden (1960, 1962) and resulted in developing an early maturing, highly branched, profusely fruiting line named Jumbo Mung 1000 R/252. Ultraviolet and infrared radiation was applied to root tips of both green and black grams by Prasad (1967). He found that infrared radiation had a greater inhibition on germination than ultraviolet treatment. However, 48 % of the UV-radiated plants developed chimera1 branches with quadri- and quinquefoliate compound leaves. Utilization of the male gametocide 2,2-dichloropropionic acid produced male sterility in cultivar T.51 and was attributable to the excessive stickiness of chromosomes resulting in bridges, laggards, and consequently to formation of micronuclei and polysporads (Kaul, 1970).

67

GRAIN LEGUMES OF THE LOWLAND TROPICS

Crosses between green and black grams have been highly successful (ICAR, 1952). However, other “interspecific” or wider crosses have not been widely recorded as feasible since Strand (1943) reported crosses between P . vulgaris and P . mungo.

4 . Genetic Investigations There is only limited information on genetic aspects like inheritance, heritabilities, genetic variation, heterosis, and gene action in the mung bean. Most of the genetic studies on this species have been carried out in India and have focused on some aspects important in breeding this crop. a. Simply Inherited Characters. Some simply inherited characters have been studied including growth habit, daylength sensitivity, shattering, leaf shape, and plant coloration. The more recent findings are tabulated below.

Character 1 . Pod veining (ventral

suture) a. Ripe pod color 3. Stem color 4. Growth habit 6 . Twining habit 6 . Photosensitivity

7. Shattering

a. Leaf shape (var. mungo) 9. Black-spotted testa (Bsp)

10. Shiny testa

Inheritance

Reference

Purple-red dominant to absence of veins Black dominant to light green Purple dominant to green Semispreading dominant to erect Nontwining dominant to twining Insensitivity dominant to photosensitivity (one gene pair) Pod shattering dominant to nonshattering Hastate leaves dominant to ovate (two gene pairs) Spottedness dominant t o no spots Nonshiny testa dominant t o shiny; shiny expressed only in presence of Bsp

Pathak and Singh (1963) Pathak and Singh (1963) Pathak and Singh (1963) Pathak and Singh (1963) Pathak and Singh (1963) Verma (1971)

Verma (1969) Singh and Singh (1971) Singh and Singh (1970) Singh and Singh (1970)

b. Variability. Considerable genetic and environmental variability was observed by Joshi (1969) and Singh and Malhotra (1970a). In the latter study, a collection of 75 indigenous and exotic strains were evaluated for eight quantitative characters contributing to seed yield. Wide genotypic and phenotypic variability was observed for all characters, but genotypic correlation coefficients were greater than phenotypic or environmental coeffi-

68

K. 0. RACHIE AND L. M. ROBERTS

cients. Singh and Malhotra (1970a) concluded that selection for 100-seed weight, which had the highest variability and very high genetic advance, would be the most effective character. However, they also observed genetic advance to be high for pod number, bunch (cluster) number, and seed yield, but these characters had low heritability estimates. In other investigations, Empig et al. (1970) found heritability estimates low for most yield components, except for number of days to flowering and maturity. Tomar el al. (1972) obtained higher heritability estimates for pod length or seed number per pod in the rainy season than in the dry season, but heritability estimates for branch number were higher in the dry season. c. Associations of Characters. Several simple and multiple character correlations have been reported recently by Singh and Malhotra (1970b) and Tomar et al. ( 1972). These can be summarized briefly as follows: 1. Seed yield-positively correlated with pods per plant, pod clusters per plant, pod length, seeds per pod, and seed size. 2. Seed size-negatively associated with number of seeds per pod and pods per plant. 3. Branch numbers-positively correlated with plant height and pod number. Path coefficient analysis showed that pods per plant, seeds per pod, and seed size had greatest direct influence on seed yield, assuming constancy of other yield components. d. Heterosis. Considerable heterotic effects that sometimes persisted into the F,’s were observed by Singh and Jain (1970) in the F,’s of various mung bean combinations in a 7-line diallel series. Heterosis was observed in seed yield, pod length, and branch number, and the best combinations were Hyb. 45 X 305 and T.51 X D. 45-6. e. A utotetruploidy. Colchicine-induced autotetraploids were studied by Kumar (1945). He observed that the tetraploids exceeded their diploid progenitors in length and width of the flower petal and in the pod and seed diameters, but other components of yield were reduced, resulting in lowered yields. He concluded that colchicine-induced tetraploidy would have limited value in improving the mung beans.

D.

PLANTPROTECTION

The mung bean is susceptible to many of the same diseases and pests that attack other legumes in southeastern Asia. However, in the United States and’in Africa it is somewhat less susceptible to the problems confronting Phaseolus beans, soybeans, or cowpeas. Heavy incidence of virus has occurred in certain genotypes and seasons in West Africa, but the hairy

GRAIN LEGUMES OF THE LOWLAND TROPICS

69

pods and smaller seeds may confer somewhat lowered attractiveness to cowpea pod-boring pests in Africa. 1 . Insect Pests

Some of the same insect pests of beans and cowpeas also attack the mung bean in the tropics-although perhaps to a lesser extent. In India and southern Asia the hairy caterpillar (Diacrisia obliqua), bean fly (Melangromyza phaseoli COQ), pulse beetle (Caffosobruchuschinensis Linn. ) and other species attack these crops (Jakhmola and Singh, 1971; Sepswadi and Meksongsee, 1971). In West Africa cowpea pests attack these crops but at a much lower incidence than for V . unguiculata (IITA, 1973). Use of contact and systemic foliage and seed dressing chemicals as specified for other pulses is highly effective. 2 . Diseases Mung beans are susceptible to nematodes; a root rot caused by Scferotium rolfsii Sacc; downy mildew (Erisiphe polygoni DC.) ;rust [Uronmyces uppendiculatus (Pers.) Unger]; leaf spots caused by Cercospora spp. and Macrophomina phaseoli; halo bright (Pseudomonas phaseolicola) ; and several viruses (yellows, mosaic, crinkle, stunt, and flower abortion). Control. Host plant resistance is the most practical control measure for diseases attacking the well developed plant. Sources of resistance to some of these diseases can be summarized as follows: 1. Macrophomina leaf spot: BR-68 and T-29 (Kumar et al., 1969). 2. Viruses (various): in the United States M101, M238, M330, M118, M221, M174, M235 were resistant (Yohe et al., 1972; Watt et al., 1973); in India T.65 and T.67 appeared to be resistant (Srivastava et a/., 1969). 3. Downy mildew: In the United States M221, M243, M210, M319, M330, M358, M366, M81, M183, M90, M238, M4, M195, and M409 were resistant (Yohe e t a / . , 1972; Watt et a/., 1973). 4. Sclerotirim leaf spot: four strains resistant, 1 1 strains moderately susceptible out of 21 tested (Mishra et al., 1971). 5 . Halo blight: Peruvian sources showed resistance in Ohio (Schmitthenner et a!., 1971). 6. Cyst nematode: resistance to Heterodera glycines observed in Jumbo, but OKLAHOMA 12 and KILOGA were susceptible in studies carried out by Epps and Chambers (1959).

It is particularly encouraging that reasonably good resistance to the major disease problems has been found with relative ease and in comparatively limited collections of germplasm. It is also interesting that germplasm evaluation in Missouri identified strains like M238 and M330 with good resistance to both downy mildew and viruses (Watt et al., 1973),

70

K. 0. RACHIE AND L. M. ROBERTS

E. PHYSIOLOGY There have been reasonably intensive investigations into the metabolic processes and enzymatic systems in the mung bean, but relatively limited information is available on effects of daylength, quality of light, temperatures, mineral nutrition, and water on plant growth and development. Recent findings on physiological processes are discussed briefly in the sections to follow, with major emphasis on information of greatest use to plant improvers.

1 . Light and Pholoperiod The existence of both day-neutral and short day types in mung beans have been recorded. Rimikhanov (1967) grew both mung beans and cowpeas under 9-hour and natural (14.5-15.0 hour) daylengths in Azerbaijan, USSR. He observed a shortening of the growth by 11-22 days, decrease in size of plant and reduced number of seeds per plant under shortday conditions. In other experiments on dates of sowing in India, Sen and Chedda (1960) proposed a black gram breeding system utilizing shortening daylengths for purposes of synchronizing flowering and hybridization, and utilizing long days to screen for day neutrality. 2 . Metabolic Processes in Seedlings

Metabolic processes and enzymatic systems during germination and in etiolated mung bean seedlings have been studied intensively in reference to their utilization for bean sprouts. Quantitative determinations of cell wall constituents of growing seedlings showed marked changes during early growth in experiments conducted by Franz ( 1972). a. Promoting Germination and Rooting. Optimum germinating temperatures for mung beans were 24-32"C, but black gram germinated almost as well at 16"C, although more slowly, in experiments carried out by Knapp (1966). Several investigators have studied the effects of growthpromoting substances in stimulating rooting in mung bean cuttings. Hormones like 2-thiouracil, 5-bromodeoxyuridine, indoleacetic acid (IAA), abscissic acid, and ethephon applied to hypocotyl cuttings markedly increased root formation, whereas uracil, thymidine, IAA, GA,, and kinetin reduced the number of length of roots formed (Chin et al., 1969; Jackson and Harvey, 1970; Krishnamoorthy, 1970; Anzai et al., 1971). Soaking seeds in 0-fluorophenoxy-a-methylacetic acid or N-benzyl-0-fluorophenoxyacetamide at 50 ppm for 6 hours drastically reduced radicle growth and development during the first 72 hours of growth and also decreased the activities of p-glycerophosphatase and phytophosphatases (Tewari,

GRAIN LEGUMES OF THE LOWLAND TROPICS

71

1968; Tewari and Kathju, 1969). Paul et aE. (1970) observed that exogenous GA, promoted a-amylase and ribonuclease, whereas chloramphenical, actinomycin D, and 5-fluorouracil were inhibitory to net enzyme synthesis during germination. b. Cell Wall Constituents. Cell walls of seedlings 2-4 days old were composed of 30% a-cellulose, 50% hemicellulose, and 20% pectin. After 4 weeks the constituents had changed to 60% a-cellulose, 30% hemicellulose, and 10% pectin. c. Oxygen Uptake. Oxygen uptake during germination of 30°C was investigated by Morohashi and Shimokoriyama ( 1972b). They observed that O2uptake increased rapidly for 4-5 hours, remained constant for 1-2 hours, and then increased again. Citric and malic acid were the major organic acid constituents, aspartic acid being converted into malic acid during the first 9 hours. Ethanol fermentation occurred early, but CO, fixation was low in the early stages. d . Light Eflects in Germinating Seedlings. Etiolated mung bean seedlings contain several carotenes and xanthophylls, but illuminated seedlings increased more rapidly in xanthophylls than in carotenes. Among the latter, lutein increased more rapidly than p-carotene in studies carried out by Valadon and Mummery (1969). Dodge et al. (1971) demonstrated rapid formation of chloroplasts in 7-day-old etiolated seedlings when supplied with diuron and sucrose together with light at 500-2000 lux. Jaffe (1970) and Yunghans and Jaffe (1972) found acetylcholine to be present in all organs of light- and dark-grown seedlings of mung bean, the highest concentrations occurring in buds and secondary roots. Exposure of roots to red light in presence of acetylcholine caused a rapid utilization of ATP pools, whereas far red light appeared to inhibit this utilization. Tanada (1972) in other red light experiments concluded that phytochrome in bean seedling tissues acts in conjunction with growth regulators like IAA and ABA to produce rapid changes in root tip surface changes-ABA inducing positive and IAA negative potentials. Racusen and Miller (1972) confirmed these observations and further noted effects of light qualities. They observed no effects of green light below 880 mV nor of red light below 220 mV on electrical potentials of root tips. In other radiation experiments, Murray and Newcombe (1970) noted the inhibitory effects of low level X-ray (100 R and 1000 R ) treatments on seedling growth and development. Ting and Ho (1971) subjected germinating seedlings to magnetic fields of 5000 gauss at 30°C and observed an accelerated breakdown and translocation of phosphorus-containing materials and secretion of nitrogen-containing compounds by the embryo. e. Enzymatic Processes in Seedlings. The literature on this subject is voluminous and secondary in the improvement and production of mung

72

K. 0. RACHIE AND L. M. ROBERTS

beans. The reader is therefore referred to the following recent papers pertaining to various subject fields in this area: 1. DNA and RNA synthesis: Ong and Jackson (1970); Parekh et al. (1969); Kobayashi and Yamaki ( 1972). 2. Starches and sugars: sucrose and fructose-Grimes (1969), Delmer and Albersheim (1 970), Heuser and Hess ( 1972); glucose metabolism-Franz (1972), Clark and Villemez (1972), Ikeda (1968); mannose-Vessal and Hassid (1972), Heller and Villemez (1972), and Brar and Elbein (1971, 1972). 3. Phospholipid synthesis: Katayama and Funahashi ( 1969). 4. Phosphorus metabolism: Ong and Jackson (1972a), Mandal and Biswas (19701, Mandal et al. (1972), Delmer and Albersheim (1970), Majumdar er al. (1972). 5. Organic acids: malic, citric, and aspartic acids-Morohashi and Shimokoriyama (1972a); alicylic acid-Minamikawa et al. (1968). 6. Amino acids: alanine-Kasai et al. (1971); methionine-Sakai and Imaseki ( 1972), Truelsen ( 1972) ; aspartate transcarbamoylase-Ong and Jackson (1972b) ; aromatic amino acids-Gilchrist et al. ( 1972). 7. Other compounds: chalcones-flavanones isomerases-Hahlbrock et al. ( 1970) ; carotenoids-Valadon and Mummery ( 1969) ; isozymes in hypocotyl cuttingsChandra et al. (1971).

3. Soil Moisture and Temperature Effects of soil moisture on forniation of hard seeds in mung bean (cultivar BLACK MAPTE No. 1 ) were studied by Ishii (1968, 1969). He found that when soil moisture was reduced from 50 to 20% during the floweringripening process that more than 90% hard seeds were formed, particularly at the lower nodes. 4 . Mineral Nutrition Limited studies on mineral nutrition in plant tissues of mung beans have been reported. Most of these investigations were carried out in field trials, but some experiments have included tissue analysis and tracing of physiological processes. Recent information on specific elements is summarized under the respective sections to follow. a. Phosphorus. The most economic level of phosphorus application in commercial production of mung beans and black gram is 15-40 kg P,O per hectare with overall yield increments of about 25% (Singh and Virk, 1965; Sekhon et al., 1966; Behl et al., 1969; Sreenivas et al., 1968; Rajagopalan et al., 1970; Chowdhury and Bhatia, 1971b; Mandloi and Tiwari, 1971; Prasad et al., 1968). However, on some lateritic soils with high phosphate fixation capacity, good response is obtained up to 100 kg of P,06 per hectare. Simple superphosphate has generally been more effective than more concentrated forms, probably as a response to sulfur contained in superphosphate (Deshpande and Bathkal, 1965). Banding the fertilizer has been superior to broadcast application (Chowdhury and

GRAIN LEGUMES O F THE LOWLAND TROPICS

73

Bhatia, 1971b) ; and foliar applications of 11 kg of P,O, per hectare have been more effective in increasing yields than 22 kg of P,O, per hectare applied to the soil (Deshpande and Bathkal, 1965). Compost and farmyard manure at up to 5.6 tons/ha further increased yields when added to P,O, applications (Sreenivas et al. 1968; Rajagopalan et al., 1970). The response to phosphorus in mung bean, as in other legumes, is quite complex. It is sometimes considered the pivotal element without which other nutrients are ineffective or even detrimental. Phosphorus deficiency in the presence of adequate amounts of other nutrients produces stunted plants with small, dark green leaves with high contents of total and soluble nitrogen but low in protein associated with an accumulation of arginine in experiments carried out by Pandey (1968). Phosphorus also had a beneficial effect on both rhizobial activity and plant growth in studies carried out by Iswaran et al. (1969). They observed better growth and greater uptake of phosphorus in well nodulated pot grown plants supplied with P,O, at the rate of 80 kg/ha. Shanker and Kushwaha (1971 ) obtained increased uptake of nitrogen, potassium, calcium and magnesium in addition to phosphorus from applications of 44.8 kg P,Os/per hectare to black grams in experiments in north India. Contents of nitrogen and phosphorus were highest 30-50 days after sowing, whereas potassium and calcium were highest at 30 days and magnesium increased up to 70 days of age. b. Sulfur. Addition of sulfur to mung bean in sand culture was investigated by Arora and Luthra (1971a,b). They observed increasing sulfur in the nutrient solution up to 90 ppm, with or without additional nitrogen, increased the contents of total, soluble, and proteinaceous nitrogen; but decreased the amide, amino, ammonium, and nitrate N in plant tissues. Moreover, the sulfur contents of leaves increased with increasing sulfur application, reaching a maximum at 50 days after planting. The sulfur contents of leaves were significantly correlated with contents of methionine, cystine, and cysteine in the mature seeds up to a maximum of 90 ppm in the nutrient solution. c. Other Nutrients. There are comparatively few recent reports on nutrients other than phosphorus and sulfur in mung beans. However, application of micronutrients investigated by Abutalybov and Samedova ( 1966) showed that cobalt, molybdenum, or manganese increased the total amounts of amino acids in mung bean leaves at the 7-leaf stage; whereas, zinc and copper applications decreased the amino acid contents. Sodium humate up to 0.1% of soil with or without nitrogen markedly increased the length and growth of shoots and roots, resulting in increased dry matter production in uninoculated pot-grown mung beans. The effects were most pronounced when nitrogen and humate were applied together (Khandelwal

K. 0. RACHIE AND L. M. ROBERTS

74

and Gaur, 1970). In other experiments in India, Iswaran et al. (1972) observed foliage sprays of 1% sucrose solutions applied at weekly intervals significantly increased seed yields of inoculated, pot-grown mung beans. 4. Nitrogen Fixation

Inoculation with appropriate rhizobium has been found beneficial where plants of the same host range have not been grown recently. Structure and development of P. mungo root nodules including the enveloping membrane of the bacterium were studied and described by Narayana and Gothwal (1964) and Prasad and De (1971). In experiments carried out in USSR by Dorosinskii and Lazareva (1967) on several species to determine host range, the following results were obtained: Rhizobium source Peauuts Cowpeas Lupines Blackgrain Soybeans

Species with effective activity

Survival only

Cowpeas-black gram Black gram

Lupines Peanuts-lupines Peanut-cowpeas Peanuts Peanut-gram-cowpeas

Cowpeas-lupines Lupines

In central India, Singh and Choubey (1971 ) demonstrated a wide range of rhizobial acceptability in mung bean using local strains A, B, or C, and inoculation was roughly equivalent in seed yields (1.23-1.36 tons/ha) to between 20 and 40 kg of nitrogen per hectare applications in inoculated plots. a. Nutrients on Rhizobial Development. In Australia, Brockwell (1971 ) found mung beans inoculated with rhizobial strain CB 756 had significantly lower nodulation when fertilized with 75 kg of nitrogen per hectare. However, addition of nitrogen did increase seed yields by 27% or 641 kg at the first harvest in the absence of inoculation. Iswaran et al. (1969) demonstrated the beneficial effect of 80 kg of P,05 per hectare on inoculated mung beans in terms of increased dry matter content and phosphorus uptake by the plant. The nitrogen content of P . mungo nodules was increased significantly from 3.97 to 5.01% by removing flower buds as they developed and fertilizing with phosphorus and potassium in experiments carried out by Kulimbetov (1968). b. Soil Conditions. Rhizobial sensitivity to extremes of soil pH was studied by Yadav and Vyas (1971). They observed that P . mungo and P . acontifolius rhizobia were sensitive to chlorides and sulfates of sodium and potassium, but P . azireus rhizobia were unaffected by up to 3% saline solutions. Magnesium salts in less than 1% solution were stimulatory, but

GRAIN LEGUMES OF THE LOWLAND TROPICS

75

NaHCO, at concentrations of 0.4-0.6% was critical for all rhizobia tested. Rhizobia survived at p H 10, but were inhibited at pH 3.5 or lower. c. Insecticides Residues. DDT in the soil at levels up to 40 ppm in pot trials had no adverse effects on Rhizobium in mung bean roots, nor on the amounts of leghemoglobin in root nodules. However, at 100 ppm DDT nodulation was prevented and seed/plant yields were reduced to about a tenth of the control except where 1 ppm of farmyard manure was added to the soil (Gaur and Pareek, 1969; Pareek and Gaur, 1970).

5 . Growth Regulators Growth regulators sprayed on black gram ( P . mungo) at early and full flowering stages were investigated in India by Mehrotra et al. (1968). They obtained increases in seed yields of 35% from NOA (p-naphthoxyacetic acid) applied at 50 ppm p-CPA (para-chlorophenoxyacetic acid) at 5 ppm; 36% increase from NAA (naphthaleneacetic acid) sprayed at 25 ppm; 39% increase from p-CPA applied at 5 ppm; and 56% increase from NOA sprayed at 50 ppm. Yield increases were ascribed to an increase in number of pods and seeds per plant, but had very little effect on seed sue. 6. Seed Storage

+

Investigations on storability of grain legumes in sealed and open containers for up to 18 years were carried out in the USSR by Gvozdeva and Zhukova (1971). They found that P . aureus survived storage better than P. vulgaris var. Triumf, which retained its viability intact for the full period in hermetically sealed containers at 10% moisture. Open containers were much less satisfactory for preserving viability than hermetically sealed ones.

F. MANAGEMENT Mung beans require good soil tilth. In India, black grams are frequently grown after rice or mixed with rice and other crops both in summer or winter (in the south). They may be broadcast or planted in rows at the rate of 11-17 kg/ha. In rows 25-90 cm apart the seed rate can be 5-9 kg/ha for green gram and 11-13 kg/ha for black gram. The crop normally matures in 80-120 days but has a tendency to shatter; therefore, the first harvest should be picked after about 2 months in early strains. Yields of 300-500 kg/ha are common, with occasional exceptional yields of up to 1100 kg/ha obtained under favorable growing conditions. However, experimental plot yields up to 2700 kg dry seed per hectare have been obtained (Watt et al., 1973). Yields of dry hay in golden gram may range from 2.2 to 6.0 tons/ha, but is usually less for green gram.

76

K. 0. RACHIE AND L. M. ROBERTS

1. Populations and Spacings Field triaIs of mung beans planted in hills 25 cm apart with 75 cm between rows and varying numbers of plants per hill were investigated by Natribhop et al. ( 1972). They observed increased seed yields of approximately 55 kg/ha for each additional plant per hill up to 5 plants per hill (53,500 plants per hectare) with maximun seed yields of 617 kg/ha. In northern India, Sharma (1969) found 60-cm rows superior to 30-cm or 45-cm rows in June plantings, but 30-cm rows were superior for July 15 plantings of black gram. However, seeding rates of 15-25 kg/ha produced similar yields (470-5 10 kg/ha). 2 . Chemical Weedicides The effects of MCPE on mung beans and other species were investigated by Gupta and Mani (1964). They found berseem, chick-peas, and cowpeas highly susceptible to “normal” applications. However, at the same dosages green gram, black gram, and hyacinth beans (Lablab niger) , although temporarily stunted, recovered later without loss in final yields. Pre- and postemergence herbicides trials on several legume species carried out by Gentner and Danielson (1965) showed the best preemergence herbicides overall to be trifluralin, diphenamid, pebulate, siduron, CDAA, CPD, CP 3 1393, RP-2929, chlorpropham, and dinoseb.

G. CHEMICALCOMPOSITION Dried seeds of mung beans contain about 9.7% water, 23.5% protein, 1.1% fat, 57-58% carbohydrates, 3.3-3.8% fiber, and 4.0-4.8% ash (Purseglove, 1968). Both carbohydrate and protein fractions are highly digestible and used in feeding infants, geriatrics, and convalescents and in “breaking” religious fasts in India (Pant and Tulsiani, 1968). Protein Quality. Seed proteins have been analyzed by several investigators. Many experiments were carried out without specifying genotype, environment, and management practices. Generally, the mung bean is similar to other grain legumes-especially other Vignas in protein quality. Gonzalez et al. (1964), studied 12 amino acids in mung bean and found the highest to be isoleucine and the lowest cystine. Sayanova (1970) identified ten salt-soluble protein fractions, and Kasai et al. (1971) found high concentrations of gamma peptides but not in the seedlings. They found no significant relationship between the amino acid composition of the protein and the free acid composition of the protein and the free amino acid composition of seeds or seedlings. It appears that protein quality in terms of sulfur amino acids can be influenced to a degree by sulfur fertilization (Arora and Luthra, 1971a,b). However, Totawat and Saxena ( 1971) observed that saline irrigation water

GRAIN LEGUMES OF THE LOWLAND TROPICS

77

could reduce contents of lysine, arginine/histidine, aspartic acid/glutamine, threonine, alanine, proline, and cystine in mung bean.

H. POTENTIAL The mung bean-both green and black gram-has definitely not received the attention deserved in proportion to its potential in the lowland tropics. Present yield levels are among the lowest of the major grain legumes. T o a degree, this results from poor response to improvements in cultural practices. Therefore, plant improvement efforts should be directed primarily toward increasing the yield potential and physiological efficiency in growth processes. Other important objectives would be insensitivity to daylength and other environmental factors, and stabilization of productivity over a wide range of growing conditions. Associated with these basic characters must be resistance to the major diseases and pests of the region while retaining consumer acceptability and improving the inherent nutritional qualities of the plant. The mung bean has many desirable attributes, chief among which are a high consumer preference for pulse and vegetable forms in Asia, and a worldwide demand for bean sprouts used in Chinese dishes. The United States alone consumes about 25 million pounds of mung beans annuallyprimarily for sprouting. It therefore has export potential as a cash crop and improved productivity levels would markedly increase the production and use of this highly promising crop. VII.

Secondary Species

There are at least 20 grain legume species and subspecies representing

14 genera of largely undetermined importance in the lowland tropics in addition to peanuts, pigeon peas, cowpeas, and mungbeans. Of these, about eight species, including bambara groundnuts, moth bean, cluster beans (guar), hyacinth bean, dry beans, soybeans, lima beans, and African yam beans, are known to be the predominating species in certain areas. Many of the secondary species, however, have exceptional productivity potential, unique characteristics of adaptation, known resistance to pests and diseases, and significant nutritive qualities. Production statistics for these legumes, if available at all, are usually extrapolations from rough estimates in limited areas and are often included under the general heading “Other and Unspecified Species” in official reports. Secondary grain legumes are described and categorized according to ecological zones in which they are presumed to be best adapted and most extensively utilized. The lowland tropics are defined as areas lower than about 600-800 m elevation lying between the Tropics of Cancer and Capricorn. They include these moisture zones: (1) semiarid tropics: less than

78

K. 0. RACHIE AND L. M . ROBERTS

600 mm annual precipitation; (2) subhumid: 600-1000 mm; ( 3 ) humid: 1000-1500 mm; (4) very humid: above 1500 mm annual precipitation. It should be emphasized that these are only approximations and that there is considerable overlapping of conditions and species adaptation within and between moisture zones and elevations depending on soil conditions, proximity to coastal areas or deserts, moisture distribution, relative humidities, ambient temperatures, cloud cover, indigenous vegetation, prevalence of pests and diseases, genotypes available, and local demands. A.

SEMIARID LOWLAND TROPICS

Sustained cultivation of all rainfed crops is hazardous in the semiarid tropics, especially below the 500 mm rainfall belt; it is almost nonexistent in areas with less than 300 mm precipitation, except for low areas, or under irrigation. Nevertheless, distribution and reliability of rainfa11 and other conditions are very important considerations in this as in other tropical cropping zones. Generally the higher the latitude, the more concentrated is the rainfall pattern and therefore the more favorable the conditions for cultivated crops. The more important secondary hot-weather grain legumes grown in the semiarid Iowland tropics include moth and cluster (guar) beans in southern Asia; bambara groundnuts, Kersting’s groundnut, and locust beans in Africa; and tepary beans in both the Old and New Worlds. These species (described in the following sections) are usually secondary to short-season groundnuts, cowpeas, and even pigeon peas. 1. Moth Bean Mat or moth bean, Vigna acontifolia Marechal (formerly Phaseolus acontifolius Jacq.) is an important pulse of the semiarid regions adjoining tropical deserts. It has 2n = 22 chromosomes, is highly self-fertilized, has epigeal germination, and is native to India, Pakistan, and Burma, where it occurs both wild and cultivated. It is a short, compact plant and grows best under uniform high temperatures and well distributed rainfall of up to 750 mm per annum, but resists dry periods by remaining dormant. It is a slender, trailing, and hairy herb with deeply divided leaflets which easily distinguish it from other grams. It is usually grown mixed with cereals like pearl millet or sorghum and is used as a pulse and for hay. Seed yields average around 300-400 kg/ha, but up to 1500 kg of seeds per hectare, and dry hay yields of 6-8 tons per hectare have been recorded. 2. Cluster Beans (Guar)

Guar, Cyamopsis tetragonolobus (L.) Taub (syn. C. psoralides D O ) , is frequently grown in southern Asia for its tender green pods cooked as

GRAIN LEGUMES OF THE LOWLAND TROPICS

79

vegetables, as well as for its dry seeds, fodder, and as a cover crop in the drier tropics. The flour from dry seeds is mucilaginous (mannogalacton) and has exceptional viscosities, possessing 5-8 times the thickening power of corn starch. The dry seeds contain 33.3% protein and 40% carbohydrates. Description. The genus Cyamposis includes three species with 2n = 14 chromosomes and is indigenous to Africa and Asia. Guar probably originated in India. It is very hardy, drought resistant, and grows very well on alluvial and sandy loams in hot weather. It is a robust, bushy annual, 1-3 m tall, has stiff branches with white hairs, and bears small flowers in dense axillary racemes or clusters. Pods are linear in stiff, erect clusters. It is frequently broadcast at 10-20 kg of seed per hectare and commences bearing after 12-14 weeks. Yields of 10,000 kg green fodder and 600-800 kg of dry seed are obtained per hectare under dryland, and double this quantity under irrigation in India. 3 . Tepary Bean

This species (Phaseolus acutifolius Gray var. latifolius Freem. ) originated in the New World, probably in the southwestern United States and northwestern Mexico, where it occurs wild. It is particularly suited to hot, arid, and low humidity conditions, and will usually produce a crop when other beans fail, maturing out as quickly as two months. Under dryland conditions yields of 500-700 kg of dry seeds per hectare can be obtained, whereas with irrigation 800-1500 kg/ha may be realized. It has 2n = 22 chromosomes, germinates epigeally, and the seeds absorb water very easily. In most soils the testa wrinkles within 5 minutes; in warm water it wrinkles in 3 minutes. It is believed to have been introduced into the Old World fairly recently. It is grown to a limited extent in the drier regions of southern Asia and in both West and Eastern Africa as far south as Lesotho and Botswana. 4 . Bambara Groundnut This crop [Voandzeia subterranea (L.) Thou.] is also known as the Congo goober, earth pea, kaffir pea, jug0 bean, Madagascar or stone groundnut, guerte/gertere (Arabic), and voandzu. Voandzeia is most extensively cultivated in Africa, where as much as a third of a million tons may be grown on an estimated 400 thousand or more hectares. Major producers are Nigeria (estimated 100,000 tons), Niger (30,000 tons), and Ghana (20,000 tons); but it is grown widely in Eastern Africa as well. This crop is most extensively grown on very poor sandy soils which are marginal for other pulses and groundnuts. It is grown primarily for its seeds, which can be eaten semiripe or as a pulse after soaking and cooking, or parched and ground into flour. The ripe seeds contain about 20% pro-

80

K. 0. RACHIE AND L. M. ROBERTS

tein, 6 7 % fat, and 50-60% carbohydrate. It may be grown intermixed with cereals like pearl millet or sorghum or in pure stands and matures in about 4 months. It has few diseases and pests and normal yields are 500 kg of dry seeds per hectare, although occasional production levels exceeding 2500-3000 kg of shelled nuts per hectare have been reported from Malawi and Rhodesia (Stanton, 1966). a. Adaptation and Description. Bambara groundnuts are indigenous to Africa with a broad range of variation occurring in West Africa or East Africa and Madagascar. It reached Brazil and Surinam in the 17th century and was later taken to the Philippines and Indonesia. It is an annual herb with short, creeping, highly branched stems, rooting at the nodes, and with very short internodes giving the plant a bunchy appearance. The chromosome complement is 2n = 22. It is outwardly similar to ordinary peanuts, except that it is trifoliate, bearing elongated leaflets on long, hairless petioles carried at a wide angle to the stem. The style is short, bent, hairy along the surface, and the stigma is small and laterally positioned. Flowers are cleistogamous and after fertilization the peduncle, with a swollen tip bearing a brush of hairs behind, bends into the soil pulling the developing pods along. The plants are usually earthed up to facilitate this process. Fruits are rounded, wrinkled when mature, about 2 cm in diameter, and usually 1- or 2-seeded. Seeds are often patterned, sometimes with an eye, range in color from white to red or brown, are round, smooth, hard, and may be up to 1.5 cm broad and weigh 50-75 gm/ 100 seeds. b. Recent Investigations. Several recent studies on culture and growth habits of Voandzeia have been carried out in Ghana by Doku (1968, 1969) and Doku and Karikari (1970a,b) and in Rhodesia by Johnson (1968). 5. Kersting’s Groundnut This species (Kerstingiella geocarpa Harms) is superficially similar to Voandzeia and is important in Dahomey, Upper Volta, and Sudan. It is also a herb with prostrate rooting branches that fruits below ground. It can be distinguished from Voandzeia subterranea by its deeply divided calyx with narrow lobes and glabrous style. The axillary flowers are subsessile, have two-seeded fruits about 2 m long which are buried by carpophores similar to Arachis hypogaea, and it is therefore different from Voandzeia. Nutritionally it has good amounts of the essential amino acids, but like other legumes it is somewhat deficient in S-bearing amino acids.

TROPICS B. SUBHUMID The largest group of tropical grain legumes-seven species including the four major crops-occurs in this agro-climatic zone (600-1000 mm annual precipitation). In addition to peanuts, pigeon peas, cowpeas, and mungbeans, this group also includes horse gram, hyacinth bean, and locust

GRAIN LEGUMES OF THE LOWLAND TROPICS

81

bean. However, genotypic representations from all seven species occur in the semiarid zone and some extend into the humid zone as well. Moreover, Phaseolus beans, soybeans, rice beans, and jack beans are also cultivated in the subhumid zone, particularly in areas with a better distribution of rainfall. Since most tropical regions have distinct wet and dry seasons, late plantings-perhaps following a main crop such as rice or maize-make it possible to successfully grow even semiarid crops in humid or very humid regions. However, most, if not all, grain legumes produce higher seed yields of better quality when final maturation coincides with bright, sunny weather. The three secondary species to be discussed briefly in the following sections include two herbs, and a noncultivated tree whose fruits and seeds are gathered. The locust bean is not used directly as a pulse, but rather as a flavoring in sweet preparations and native stews. However, it does make an important contribution to nutritional requirements of millions of inhabitants of tropical savannahs, particularly in West Africa. It also represents the stabilizing potential for perennial and tree crops in providing proteinaceous and other essential nutrients in human diets.

I. Horse Gram The horse or Madras gram, Dolichos uniflorus Lam. or D. bifiorus Auct., is indigenous to the Old World-about 100 species have been reported. It is considered the poor man’s pulse crop in southern India, a country producing 390,000 tons on 1.8 million hectares. It is consumed like other pulses, but in Burma it may also be fermented to make “soy sauce.” It is grown as a hardy, drought-resistant dryland crop in areas of moderate rainfall (less than 900 mm) or planted after the rains have ceased, since the developing buds and pods are susceptible to rotting in humid weather. It matures in 4-6 months, producing only 150-300 kg/ha of dried seeds plus animal forage. Description. The horse gram is a low-growing, slender, suberect annual herb with slightly twining, downy stems and branches 30-50 cm high. Flowers are borne in axillary racemes clustered at nodelike thickenings of the peduncle. The flowers are about the same size as cowpeas, and the keel is uncoiled, but bent inward at right angles, rather than curved as in Vigna. The pod is large, somewhat hairy, flat, usually curved and beaked, and the seeds are smaller than those of the hyacinth bean. The chromosome complemeqt is 2n = 24. 2. Hyacinth Bean

The species Lablab niger Medik. is also known as the bonavist, dolichos, lablab, seim lubia, India butter, and Egyptian kidney bean (syn. Dolichos

82

K. 0. RACHIE AND L. M. ROBERTS

lablab L.; Lablab vulgaris Savi). The young pods and tender green beans are used as a vegetable, and the dry seeds as a pulse. It is also grown for forage and as a cover crop in rotation with sorghum and cotton in the Sudan Gezira. It is a hardy, drought-resistant, dryland crop for low rainfall areas (600-1000 mm) in the tropics-Egypt, Sudan, India, and southern Asia. It is an herbaceous perennial herb usually grown as an annual and is often twining, but bush forms do occur. The pod is similar to that of a lima bean, but normally contains 3-5 seeds. The seeds are similar in size to those of a medium lima bean, but plumper and have a characteristic arid or projection from the hilum extending one-thiTd the circumference of the seed making it easy to identify. Two botanical varieties are recognized: (1 ) var. ZabZab (var. typicus Prain.)-short lived, twining, perennial herb; pods are longer, more tapering with seeds parallel to the suture; grown mainly for green pods; (2) var. lignosus (L.) Prain-longer lived, semierect, bushy perennial also called Australian pea. Pods are shorter, more truncated and the long axis of the seeds is at right angles to the suture. The plants have a strong unpleasant smell and are used mainly as a dry pulse and fodder. These variety identifications are confused, however, and frequent hybridization occurs giving rise to numerous intermediate types.The chromosome complement is 2n = 22 and 24. Germination is epigeal requiring 5 days, self-pollination is predominant, but some outcrossing may occur. Lablab niger is apparently closely related to Dolichos uniflorus, as hybrids between the two species have been obtained. Adaptation.. The hyacinth bean can tolerate poor soils if well drained and is usually photoperiod sensitive-some cultivars require 6-7 weeks to flower, depending on planting date. The garden type hyacinth bean is often planted in heavily manured and irrigated pits at the rate of 6-10 seeds per hole in midsummer and harvested for green pods from December to March, after which they may be kept for a second year. Field crops are often intersown with cereals Iike ragi (Eleusine coracana) and receive little attention. Yields of dry seeds average 400 kg/ha in mixtures, and up to 1300 kg/ha in pure stands.

3. Locust Bean Fernleaf, nitta tree, nere, or nele are some of the common names for Parkia spp., particularly P . filicoides Welw. (syn. P . clappertonia) and P. biglobosa Benth. (also P . oliveri), which become large trees but are seldom planted. They occur throughout the African savannah zones. The seeds are not used as a dietary staple, but are cooked and fermented to make a condiment widely used as a food flavoring. The fruit pulp from immature pods is sweet; it is a popular flavoring in desserts and drinks and is particu-

GRAIN LEGUMES OF THE LOWLAND TROPICS

83

larly nourishing by virtue of its high content of sulfur amino acids-up to 2.9 g of methionine and 3.8 g of cystine per 16 g of nitrogen, dry weight (Busson et al., 1958). The locust bean also contributes to soil fertility through its ability to extract plant nutrients from the deeper soil layers and by leaf shedding. “Pure” strands of this species are estimated to contribute yields of 350-500 kg of dry seeds per hectare planted like an orchard (Stanton, 1966).

C. HUMIDTROPICS The humid tropics classification of 1000-1500 mm annual rainfall is arbitrary, and many leguminous species, including the four major ones, span the range from less than 600 to more than 1000 mm precipitaton. Perhaps the most widely adapted species is the pigeon pea, which occurs in all four zones. However, species and genotypes adapted to high rainfall areas tend to be sensitive to environment, so that flowering and maturation occur at the end of the rains or well into the dry season. Five species and four genera, including soybeans, Phaseolus (dry) rice, jack and sword beans are considered to occur in this group. Most of these species do not require 1000 mm of rainfall to produce satisfactorily but perform better when moisture is well distributed. They are therefore classified for the humid zone, where better distribution of moisture occurs. Jack beans are not only drought resistant, but also tolerate waterlogging and salinity better than many other grain legumes, and hence are classified for this zone. 1. Common Bean The common, dry, dwarf, kidney, french, navy, snap, runner, salad or string bean (Phaseolus vulgaris L.) is the most widely grown and best known of all Phaseolus species. Although extensively cultivated at intermediate and higher elevations, it is grown to a very limited extent in the lowland tropics (usually in the vegetable form) being highly susceptible to pests, diseases, high temperatures and even short periods of water stress. Although certain black-seeded cultigens are better adapted to the lowlands, their acceptability is less than that of the lighter colored seed types. Nevertheless, considerable plantings are made both for vegetables and as a pulse, with variable results. Commercial yields of 500 to 1200 kg of dry seeds per hectare are reasonable for short-term production of 56-90 days. 2 . Soybeans The soybean, Glycine max Merr., is predominantly a crop of temperate regions and intermediate elevations in the tropics and is probably the most

84

K. 0. RACHIE AND L. M. ROBERTS

advanced and best developed of all legumes. It has been extensively grown for a long time as a basic food crop of the low elevations in southeastern Asia (Indonesia, Philippines, Malaysia). More recently, investigations in India, the West Indies, and both East and West Africa have demonstrated that soybeans can be very successfully grown in the lowland tropics under favorable conditions. At present there is no other species that can so consistently produce on a hectare per day basis both high yields of good quality protein and oil. The major deterrent to increasing production of this species in many tropical regions is lack of markets and understanding of its cultivation and utilization. However, there is a rising demand for both industrial proteins and vegetable oils, together with an increasing consumption of processed foods in the rapidly growing urban areas of the tropics. In fact, it is usually easier to transplant industrial processing into new areas and developing regions than to change traditional agricultural practices or social habits. a. Adaptation and Problems. Soybeans grow best at maximum temperatures between 27°C and 32"C, have a wide range of adaptation of soil types, but thrive best on sandy or clayey loams in areas with hot, damp weather. They are somewhat less drought resistant than cowpeas, but tolerate waterlogging better. They are mostly short-day plants requiring 14-16 hours of darkness to flower, but a wide range of maturities and determinancies exist. Maturities may range from 75 to 200f days depending on the genotype and environment. They respond well to fertilization and require a special strain of Rhizobium japonicum for proper inoculationparticularly when grown in new areas. In tropical Africa diseases and insect pests that affect the soybean are fewer than for cowpeas, but in southeastern Asia (Indonesia) the reverse may apply. Bacterial pustule (Xanthomonus phaseoli Bows. var. miensis Hedges), viruses (soybean mosaic and yellow bean mosaic), root knot nematode (Meloidogyne incognita), and cyst nematode (Heteroderu sp. ) may be the most serious problems. In southern Nigeria there has been considerable difficulty in obtaining good stands during hot, dry weather as a result of high soil temperatures (above 35°C) causing the seed to rot. Resistance to shattering and lodging in commercial varieties are also essential attributes in the tropics. b. Utilization. The rapid increase in soybean production appears highIy likely in view of its export potential and currently exceptionally high world market prices. This situation will tend to familiarize the crop in tropical areas outside of the regions of traditional use. It is suggested, however, that rather substantial inputs in terms of education and extension will be required to induce people unfamiliar with the crop to use the soybean directly for food owing to the rather sophisticated methods of preparation

GRAIN LEGUMES OF THE LOWLAND TROPICS

85

required to make it palatable. First, whole beans require heating or boiling intact for 15-20 minutes to inactivate the enzyme lipoxidase in the testa (which produces an off-flavor, beany, or painty taste), hemagglutinins, trypsin inhibitor, and other toxic substances. In southeastern Asia they are frequently eaten as green beans (vegetable), split, sprouted, processed into soy milk (cooking and pressing), fermented into sauce utilizing Aspergillus oryzae, or made into curds and cheese. In Indonesia the boiled beans are fermented with Aspergillus sp. to make a cheeselike preparation called tempe. c. Recent Investigations. Perhaps the most successful campaign to introduce soybeans and find solutions to production and utilization problems has been in India with assistance from a USAID-sponsored contract with the University of Illinois. In Africa, French-sponsored research organizations have centered their activities mainly in Madagascar with testing and management experiments in the Cameroons and Centralafrique (Silvestre, 1970a; Marquette, 1970). In anglophone Africa genetic recombination has been employed in breeding programs at Nachingwea, Tanzania (Aukland, 1966, 1967), Makerere University (Radley, 1971), and in Nigeria (Van Rheenen, 1972; Ebong, 1970a; IITA, 1973). Major objectives in soybean improvement for the lowland tropics have been the following: ( 1 ) wide adaptation and stability of yields-particularly insensitivity to daylength and temperature fluctuation; (2) earlier maturation to better accommodate short bimodal rainfall patterns and permit multiple cropping; ( 3 ) improvement of quality for direct food use; (4) resistance to shattering and lodging; ( 5 ) germinability under high soil temperatures; (6) resistance to insect pests like thrips, foliage feeders, and pod bores; (7) resistance to diseases, particularly nematodes, viruses, and bacterial pustule. Broadly based and multifaceted improvement programs are required to realize these objectives. However, considerable progress has already been made, particularly in India for the higher tropical latitudes. In Africa, yield testing and other experiments have identified several widely adapted, highyielding cultivars, including: BOSSIER, HARDEE, CLARK 63, CHUNG HSING 1, IMPROVED PELICAN, and CES 486 in West Africa; DAVIS, DARE, CLARK, KENT, NATHO, and MORISSONNEAU in Madagascar; some of these, BUKALASA 4 and recombinations from crosses between HERNON 237 and LIGHT SPECKLED from Tanzania (Silvestre, 1970b; Marquette, 1970; Leakey, 1970; IITA, 1973). Among varieties and strains showing photoperiod and temperature insensitivities very useful in breeding programs are: GRANT, CLARK 63 and SRF 300 from the USA; FISKEBY V from Sweden; TOCKACHINAGAHA from Japan; and HSHI HSHI from Taiwan (Summerfield and Huxley, 1973; Radley, 1971).

86

K. 0. RACHIE AND L. M. ROBERTS

3. Rice Bean This species, Wgna umbellata (Thumb.) Ohwi and Okashi, was formerly classified as Phaseolus cakaretus Roxb. It occurs wild from the Himalayas and central China to Asia and is cultivated as a dried pulse in India, Burma, Malaysia, the Philippines, and other parts of southeastern Asia, but it is grown to a very limited extent in Africa. The beans are frequently cooked with or instead of rice. The green pods and tender leaves are used as vegetables and the whole plant may be used for fodder or as a cover crop. It is a vigorously climbing or suberect annual which produces long, slender, glabrous, shattering pods, with various colored, medium-small (8 mm-long) seeds. It is frequently grown in rotation with rice, producing a crop in as little as 60 days. Yields are usually low at 200-300 kg per hectare, but it performs well under humid conditions with fewer pests and diseases than most other legumes. It also tolerates high temperatures and is moderately drought resistant.

4 . Jack and Sword Bean Jack beans (Canavalia ensiformis L. and C . plagiosperma) and sword beans (C. gladiata Jacq.) are used as a green vegetable (immature pods), dry pulse, and the vegetative portion may be grown for forage, green manure, or cover crop. However, dry C. gladiata seeds may contain toxic substances. All three species are very hardy, deep-rooted, and exceptionally drought resistant, and tolerate waterlogging, shade, and saline soils better than many other grain legumes. The extremely large tough pods and hard seeds (2-3 cm long) germinate epigeally and very quickly (48-72 hr) ; and the developing plant is exceptionally free of insects and diseases under some tropical conditions. The large dry seeds are hard to cook and can be somewhat toxic requiring boiling in salt water for several hours with a change of water. The jack bean is also an important source of urease and of the lectin (cell-ag glutinating agents) concavalin A, which occurs at levels of 2.5-3.0% by weight in dry seeds and is being used in medical research (Sharon and Lis, 1972). This particular lectin has a special affinity for agglutinating cells transformed by DNA tumor viruses or carcinogens. Dry seed yields of jack beans can be considerable even on poor soils where up to 2000 to 2500 kg/ha are reported. Production is scattered generally throughout West Africa and in Zaire and Angola.

D. VERYHUMIDTROPICS There are comparatively few well-adapted grain legumes for the very humid tropics, that is for the humid Guinean and Rain Forest Zones where

GRAIN LEGUMES OF THE LOWLAND TROPICS

87

annual rainfall exceeds about 1500 mm and humidities are constantly in the high 90"'s F as a result of heavy surrounding vegetation or location effect. However, some of the species allocated to the humid tropics are or could possibly be extended into the more humid regions. These might include in approximate descending order of adaptation (1 ) jack beans, ( 2 ) rice beans, ( 3 ) pigeon peas, and (4) soybean. Of these, jack beans and soybeans tolerate waterlogged and partially saline conditions better than most grain legumes, whereas rice beans and pigeon peas require good drainage but withstand high humidities during grdwth. However, all species produce better yields and seed quality when rainfall is light during flowering and they mature out in dry weather. There are about five species confined mainly to the humid tropics. Three of these-African yam bean, Mexican yam bean, and winged bean-all produce edible tubers similar to the sweet potato, which may be eaten fresh or cooked. Sometimes the inflorescences are nipped off to increase tuber size and productivity. The five humid, lowland tropical species are further described separately. I . Lima Beans

The species Phaseolus lunatus L. is also known as butter bean, sieva bean, Madagascar bean, and Burma bean. There are several forms which are grown for both dried and green shelled beans and are considered to originate in Central America-probably in Guatemala, where wild endemic forms occur and from whence the large white types were spread southward to Peru by the Incas, the small-seeded forms northward through Mexico to the southern United States and eastward to the West Indies, and from there to Brazil (tropical, perennial types) according to Mackie ( 1943). The latter are short-day plants with high contents of hydrocyanic acid. The large-seeded types ( P . lunatus f. macrocarpus) have been found in Peruvian excavations dating to 6000 to 5000 BC. Early Spanish explorers carried the Caribbean type limas with them across the Pacific to the Philippines and southern Asia, but African limas trace their origin back to Brazil, although large types grown in Madagascar came originally from Peru. It is one of the major pulse crops of the humid rain forests of Africa, being most extensively grown in Madagascar where as much as 30,000 hectares are planted annually. Burma is a major producer in Asia. a. Adaptation and Problems. There are two basic plant types-bush types 30-90 cm tall and twining, climbing herbs 2-4 cm tall-both weakly perennial in growth habit. They tolerate wetter weather during active growth than P . vulgaris but require moderately diry weather to produce good quality dry seeds. High temperatures tend to inhibit fruit setting, but the small or sieva group are more resistant to hot, arid conditions. Early bush types mature within 100 days but large-seeded Peruvian types may require

88

K. 0. RACHIE AND L. M. ROBERTS

7-9 months to reach maturity. Under humid tropical conditions (in West Africa), lima beans are much less affected by diseases and insect pests than most other grain legumes, although in some areas pod borers and leafhoppers attack them. Major diseases reported are downy mildew (Phytophthora phaseoli Thaxt. ) , pod blight (Diaporthe phaseolorum Sace. ), anthracnose (Colletotrichum lindemuthianum Bri. and Cav.), fusarium root rot, rust (Uromyces phaseoli Arth.) and viruses. b. Improved Strains. Varieties adapted to the tropics include the bush types, FORDHOOK 242 with large thick seeds, and BURPEE BUSH. Recommended pole limas for green beans include KING OF THE GARDEN and FLORIDA SPECKLED BUTTER. For production of dry seeds in California, VENTURA having large flat beans, and WILBAR and WESTERN with Small thin seeds, are the most important types, and both are semiclimbers. Dry seed yields exceeding 2000-2500 kg/ha have been observed in southern Nigeria (IITA, 1973). c. Botanical. The chromosome complement of P. lunatus is 2n = 22; germination is epigeal, and up to 20% outcrossing occurs. Hand crossing is sometimes difficult owing to diminutive flowers and a high degree of abortion. However, genetic male sterility has been utilized to facilitate the genetic recombination procedure. The seeds vary in size from 45 to 200 g per 100 seeds; range from flat to rounded “potato” types; and may be white, ivory, red, purple, brown, or black in solid colors or mottled. 2. Winged Bean This plant (Psophocarpus tetragonolobus L. ) is also known as goa bean, four-angled bean, Manila bean, princess pea, and asparagus pea (not to be confused with Lotus tetragonolobus L. also called asparagus or winged pea). The winged bean probably originated in tropical Asia and is fairly extensively cultivated by the Melanesians in New Guinea. It is grown primarily for its immature pods cooked like French beans, but the partially and fully ripe seeds are eaten after parching in Java and New Guinea. The tubers are smaller than Pachyrrhizus spp. and are eaten raw or cooked, particularly in Burma; and the young leaves, shoots, and flowers may also be eaten as a vegetable. It may also be grown for fodder, as a cover crop or for green manure because of its exceptionally good nodulation. a. Adaptation. Winged beans perform best in hot humid cIimates and are remarkably free of pests and diseases in southern Nigeria (IITA, 1973). Loamy soils are best and the crop does not tolerate waterlogging. However, it does require ample, well-distributed moisture-perhaps in excess of 1500 mm annual rainfall or irrigation to perform well. It has exceptional ability to nodulate-one plant studied by Masefield (1961) produced 585 g of fresh nodules, compared with only 1.5 g of fresh nodules

GRAIN LEGUMES OF THE LOWLAND TROPICS

89

for P. vulgaris plants. However, experiments with winged beans in southern Nigeria have sometimes produced comparatively slow growth accompanied by markedly chlorotic, light-green foliage, suggesting poor nodulation or unavailability of the most effective rhizobial strains. b. Description and Nutritive Values. The winged bean is a twining, glabrous perennial herb usually grown as an annual. The pods are large varying 5rom 16 to 36 cm, have four jagged-edged wings and enclose 8-17 medium to large, globular seeds (up to 300 g/100 seeds). The ripe seeds are rich in both protein (up to 37.3%) and oil (15.0-18.1%) from investigations carried out by Pospisil et al. (1971). The tubers may also contain up to 24% crude protein on a dry weight basis (Burkhill, 1967). The oil is high in unsaturated fats with only 28.8% saturated fatty acids, and 126 mg per 100 mg of tocopherol (dimethyltocopherol-both alpha and beta forms). The protein is also high in essential amino acids: cystic acid = 2.6% of protein; lysine = 8 % of protein; histidine = 2.7% of protein; threonine = 4.5% of protein; and methionine = 1.2% of protein (Pospisil et al., 1971 ) . Investigations in Southern Nigeria (IITA, 1974) demonstrated a wide range in productivity of seeds and tubers in the winged bean cultigens. Dry seed yields from small plots planted in mid-May at Ibadan ranged from 948 to 2010 kg/ha. One of these cultigens, TPI 6 produced dry seed yields of 1653 kg/ha and fresh tuber yields of 1288 kg/ha. Tubers from the four cultigens observed were analyzed biochemically and contained: 41.4% dry matter, 18.7% crude protein, 50.4% starch, 1.1% ether extract, 2.0% ash, and 19.0% crude fiber. Although not all the crude protein is assimilable, the tuber constitutes another form of nutrient which is producible in areas or seasons unfavorable to seed development, such as high incidence of pests and diseases and where humidities remain high throughout the ripening period. Moreover tubers are partially storable in the ground until required for consumption. 3 . African Yarn Beans This pulse (Sphenostylis stenocarpa Harms.) is a slow growing, herbaceous climber frequently grown in association with the common yam (Dioscorea spp.) and beans, in humid and forested regions of the African lowland tropics. It is grown both for its seeds and tubers which resemble the sweet potato tuber in size and shape (7-13 cm long), but tastes more like an Irish potato. a. Description. The plant may be procumbent, twining, or erect, and the leaflets are lanceolate or ovate. The flowers are multiple and borne on axillary peduncles. The calyx is broadly cupular, and shortly, undulately lobed. The stigma is flat and broad like a spatula. The fruits are linear,

90

K. 0. RACHIE AND L. M. ROBERTS

up to 30 cm long, glabrous and become tough and hard when ripe. The seeds are ellipsoidal, smooth, shiny, and hard when ripe (Hutchinson and Dalziel, 1958). b. Utilization. The tough pods and hard seeds appear less susceptible to pest attacks either in the field or storage than beans or cowpeas, but the seeds require several hours of soaking before cooking. However, the seeds may have an excellent protein with exceptionally high levels of methionine (up to 1.92 g per 16 g of nitrogen) and cystine (1.44 g per 16 g of nitrogen). They are also high in glutamic acid and highly palatable. Ten tuber-bearing yam beans averaging fresh tuber yields above 2 tons/ha were grown and analyzed biochemically at Ibadan, Nigeria in 1973 (IITA, 1974). These ten lines averaged 25.7% dry matter, 15.7% crude protein, 58.1 % starch (of which 27.1% was amylose), 0.56 ether extract, 4.8% crude fiber and 3.6% ash. Nicol (1959a,b) studied the utilization of yam bean in Nigeria and found the crop grown in such disparate regions as Okuta (West of Ilorin near the Dahomey border); at Bida and Mbanegi near Obuda in the Eastern Region primarily for the tubers; whereas at Esike near Enugu (East Central State) it was grown mainly for the seeds. It is also cultivated in Central African Republic, Cameroons, Zaire, Ethiopia, and various other parts of East and Central Africa. c. Potential. The African yam bean is nearly always grown interplanted with other crops, usually with staked yams. Attempts at sole cropping have not been very successful owing to spread of diseases (wilt) in pure stands. There are no records of seed yields nor productivity levels in this crop; although preliminary observation trials at Ibadan, Nigeria in 1973 suggest that tuber-producing ability is at least partially dependent on genotype. Significant tuber production was realized in 28 out of 64 yam bean collections; the six highest yielders averaged fresh tuber yields of 2717 kg/ha, the highest, TSs 20, producing 421 1 kg/ha in small plots, although nearly all plots were heavily infected by virus (IITA, 1974). 4 . Mexican Yam Bean Two of the six species of Pachyrrhizus are cultivated for their edible tubers-P. erosus spring. and P . tuberosus. P. erosus was spread from its origin in Mexico to the Philippines in the 17the century, and P. tuberosus appears to have originated in the Amazon headwaters region. Adaptation and Use. Yam beans (jicama) are indigenous to Central America and occur wild in Mexico. They are best adapted for growing in the hot, humid tropics on well-tilled, loose sandy soils. The high moisture tubers may be cooked or eaten raw with salt and spice, and the young pods of P. erosus may be eaten like French beans; but the roots, mature seeds and leaves contain a toxic substance-rotenone. The young pods

GRAIN LEGUMES OF THE LOWLAND TROPICS

91

of P . tuberosus are not eaten, as they are covered with irritating hairs, but this species produces larger tubers. P . erosw is widely grown in home and market gardens in southeastern Asia, particularly in Singapore, southern China, Thailand, and even in Hawaii (Purseglove, 1968). Exceptionally high yields of fresh tubers-22 tons/ha in addition to 2.65 tons/ha yields of dry seeds (nonedib1e)-were obtained from replicated experiments conducted in southern Nigeria in 1973 (IITA, 1974). When analyzed these tubers were found to contain 12.1% dry matter and 9.9% crude protein on a dry weight basis.

5. Velvet Beans There are two species of Mucuna grown as pulse in various parts of tropical Africa. Both are vigorous, herbaceous climbers with long duration, often requiring 8-12 months to fruit. The fruits of velvet beans ( M . pruriens var. utilis) are medium in size, whereas the fruits of horse eye bean ( M . sloanei Adars., syn. M . urens) are large. However, the pods of both species are distinctive in appearance and frequently covered with profuse, stiff, golden, urticating hairs giving them a velvety appearance.This feature together with extreme hardness of testa in horse eye bean render them unpleasant to harvest and thresh, but may impart a measure of protection against insects and animal pests. The seeds are cooked or roasted, husked, ground into flour and used in soups and stews. In Indonesia, the seeds may be fermented and used like tempe from soybeans. Apparently the flour has good thickening properties as less than 10-12 large seeds of horse eye beans are reported to make a “gallon of thick soup.” They are a secondary species in such diverse regions as southeastern Asia-primarily Indonesia-and in tropical regions of Africa, such as southern Nigeria, Dahomey, Senegal, Upper Volta, Sudan, and Mozambique. A very similar woody climber, Dioclea reflexa Hook., also has extremely hard seed testas like those of M . sloanei. It has been reported to occur from Sierra Leone and Guinea eastward to the Cameroons, However, there is comparatively little evidence of its extensive use for food, and it may be only an occasional “gathered” crop used sporadically or in times of scarcity. VIII.

Conclusions

Twenty-four grain legume species contribute significant amounts of food, animal feeds, and industrial products in the lowland tropics. However, only four species-peanuts, pigeon peas, cowpeas and mung beans (both green and black grams)-made up 87% of both the cultivated area and total production on a worldwide basis. Peanuts alone comprise 52% of this production, although an estimated 80-90% of peanut production in the tropics went into industry or was exported.

92

K. 0. RACHIE AND L. M. ROBERTS

Among the lowland tropical legumes the improvement of peanuts in the tropics has received major emphasis owing to their industrial and export potential, and it now appears feasible to draw on extensive knowledge and genetic matelrials from temperate soybean improvement programs-at least in the beginning. Comparatively little in-depth plant improvement has been done on cowpeas, pigeon peas, and mung beans; and virtually nothing has been done on secondary species like bambara groundnuts and yam beans which are so important in the African tropics; on winged beans, velvet beans and jack beans in the more humid areas; nor on moth or tepary beans for the semiarid regions. Fortunately, there is a modest background on lima beans improvement in temperate regions, and hyacinth bean and horse gram have received a little interest in India. 1 . Comparative Features of Tropical Species Some botanical and adaptive characters of tropical lowland grain legumes can be summarized in tabular form to facilitate direct comparisons of their potential for specific situations (Appendix Table 111). Although species are grouped according to their presumed ecological use patterns, considerable overlap occurs in adaptation and in microclimates of particular locales within regions. a. Plant Types. Most grain legumes are twining, semiprostrate or climbing in their unimproved state. However, several species do have erect, self-supporting structures, including: locust bean-a tree; pigeon peas-semiwoody shrubs; and jack beans. There are three small bunchy herbs with underground fruiting: peanuts, bambara, and Kersting’s groundnuts. Several others have erect or semierect bushy forms with implications of photosynthetic efficiency, high yielding potential, convenience of growing, and adaptability for mechanization. These include: soybeans, mung beans, cowpeas, horse gram, Phaseolus beans, jack beans, lima beans, moth beans and tepaq beans. Twelve species are classified as annuals and thirteen are more or less perennial, although both forms frequently occur in the same species. Lodging occurs frequently in the erect and semierect Vigna and Phaseolus species and in soybeans, but is much less of a problem in pigeon peas and jack beans. Shattering is also a major problem in grain legumes, particularly in the more primitive and wild species, and where frequent wetting and drying out occurs. Shattering is most serious in soybeans and also in the tepary bean, rice bean, and wild/weedy Vigna species. Less susceptible are pigeon peas (except the long-podded PHILIPPINE cultivars) , cultivated cowpeas and mung beans. b. Soil and Climate Preferences. Most species require well-drained sandy to sandy loam soils. However, a few, like jack bcans, soybeans, black grams, and rice beans, tolerate heavy soils well and even waterlogging to

GRAIN LEGUMES OF THE LOWLAND TROPICS

93

some extent. Black gram, mung beans, and rice beans are frequently grown after rice. Jack/sword beans also have some tolerance of salinity, and certain species of Canavalia are used in coastal regions, where they are exposed to salt sprays and possibly brackish water for binding sand and sand dunes. The semiarid species are usually more sensitive to waterlogging, but also perform better under low humidities and bright sunshine. If grown under more humid conditions they are often subject to heavy disease and insect attack. Some of the long duration climbers like yam, velvet, horse-eye, and Dioclea beans may be tolerant of shade and very “moist” soils, as they are usually grown in the humid forested zones with high amounts of cloud cover. c. Susceptibility to Pests and Diseases. A subjective and somewhat arbitrary assessment of susceptibility to pest and diseases suggests that several species, including Voandzeia, Kerstingiella, Cyamopsis, Parkia, Canavalia, Vigna umbellata, Phaseolus lunatus, Psophocarpus, Sphenostylis, Pachyrrhizus, Mucuna, and Dioclea have comparatively fewer pest and disease problems when grown in their regions of adaptation. However, peanuts, cowpeas, and Phaseolus beans are frequently highly susceptible to various diseases and pests even in their optimal ecologies. In these situations plant protection is often the major management input. For example, in the subhumid regions of Nigeria, an effective insect control program can increase yields of cowpeas on the order of 5-10 times or more. d. Yielding Ability. It is interesting to consider authentic yield records as one indication of the potential of different species, since “average” yields may indicate that the crop was grown outside of its adaptive ecology, was intermixed with other species, used for other purposes (leaves, green pods, or tubers), or grown as a catch crop. IHowever, there is good evidence to confirm that soybeans on both an average and maximum record basis have the highest yielding potential at present, both in terms of quantity and quality (high protein and oil content). However, it is very interesting (hat some hitherto relatively unimproved species like pigeon peas (3000-5000 kg/ha of dried beans) and jack beans (4600 kg/ha dry seeds) can produce high yields under certain circumstances. Even short-term crops like bambara groundnuts (2600-3000 kg/ha), cowpeas (2800 kg/ha), mung beans (2600 kg/ha), Phaseolus beans (2500 kg/ha) have high yielding potentials for the periods they occupy lands. At Ibadan, where it is comparatively “easy” to obtain 2800 kg/ha of soybeans or about 32 kg of dry seed per ha/day, short duration cowpeas (65-70 days) will normally produce 1500 kg or up to 25 kg dry seeds per ha/day (IITA, 1973). 2 . Utilization Aspects

The total benefits from growing grain legumes are frequently ignored both in commercial cropping systems and in the decision-making process

94

K. 0. RACHIE AND L. M . ROBERTS

when allocating resources for crop research. In order to emphasize the positive aspects of growing tropical legumes, the broad array of uses for these crops as well as their beneficial effects both on the soil and crops to follow are briefly summarized in the sections below. a. Utilization for Food. The many uses of grain legumes and the broad array of dishes that can be prepared from dried pulses or tender green pods and seeds are reasonably familiar. However, it is not so widely recognized that extensive use is made of young seedlings and tender green leaves as pot herbs and as such contribute excellent quality protein (up to 35% crude protein) to the diet. Moreover, the use of tubers from winged and yam beans is hardly known, nor are there any definite reports on productivity levels and nutritive values of leguminous tubers, although results at Ibadan (IITA, 1974) suggest that winged bean and African yam bean tubers can range from 17-20% and 13-18% crude protein, respectively, on a dry weight basis. Many grain legumes have special features of utilization and of nutritional value. Some also contain toxic properties, metabolic inhibitors, off-flavor enzymes, or flatulence factors. Fortunately, the most important of these factors are dissipated in the cooking process, except the flatulent sugarsstachyose and raffinose. b. Forage and Cover Cropping. The important role of grain legumes on succeeding crops or their use for forage is frequently ignored. Some of the best forage and grazing comes from several tropical species like the procumbent or semierect cultivars of cowpeas, tepary beans, horse gram, mung beans, rice beans, velvet beans, guar, soybeans, and others. It is not unusual to obtain more than 20-40 tons of high protein green matter or 5-10 tons of dry hay within a comparatively short time (60-80 days) from thickly-planted grain legumes. Similarly, these and other species are valuable for their fertility-restorative abilities, and as cover crops to protect erodable soils. Spreading types of cowpeas have shown excellent cover in erosive situations, and the winged bean has such excellent nodulating and nitrogen-fixing abilities that sugarcane is reported to give up to 50 % higher yields following Psophocarpus than in ordinary rotations in Burma. In southern Georgia (USA) seed cotton yields were increased from 918 Ib/acre following cotton to 1578 lb/acre following velvet beans plowed down as green manure (Martin and Leonard, 1967). 3. Opportunities for Improvement

It is both surprising and encouraging that some “neglected” pulses like pigeon peas, lima beans, bambara groundnuts, mung beans, and jack beans occasionally demonstrate remarkable productivity potential and often without serious problems or substantial’management inputs. This sug-

GRAIN LEGUMES OF THE LOWLAND TROPICS

95

gests that comparatively modest investments on improvement could pay off handsomely and quickly. Nevertheless, it would probably be unrealistic to activate major improvement efforts on more than four or five of these secondary species at the present time at least at the international level. However, it may become expedient over the longer term as demand increases, more marginal lands are brought under cultivation and resources become available to mount programs on some of the more promising of these “secondary” species. Furthermore, localized interests may well decide that emphasis on some presently obscure species lies within the realm of their national goals. a. Collecting Germplasm. Since secondary species are quickly lost with the expansion of more sophisticated farming systems, there is an urgent, immediate need to thoroughly and systematically collect and maintain indigenous germplasm. It should also be noted that the large-seeded legumes may be more closely related than heretofore suspected (the majority have 2n = 22 chromosomes). Therefore, even obscure species could have important breeding potential when techniques are developed to readily combine diverse genetic stocks. b. Selection of Species. The choice of species worthy of in-depth improvement must be made on a rational basis considering their present importance, preference €or food, intrinsic problems (pests and diseases), inherent productivity potential, nutritional qualities, genetic diversity available and ease of genetic manipulation. Using these guidelines the first order of priority should be focused on the four species discussed in detail in this section: ( 1 ) peanuts, (2) pigeon peas, ( 3 ) cowpeas, and ( 4 ) mung beans. These four crops must receive broadly based support at both international and national levels. The second order of priority for species with regional potential at the outset is considered to be ( 1) lima bean for its exceptional range of adaptation, freedom from diseases and pests, high yield potential and excellent nutritional quality; ( 2 ) soybean for its industrial and market demands, exceptional yielding potential, broad range of adaptability, and reasonable freedom from pests and diseases; ( 3 ) hyacinth bean for adaptation to subhumid and semiarid conditions, broad range of uses, and productivity on poor soils; and (4) bambara groundnut for its widespread importance in tropical Africa, excellent adaptation to areas with low and unstable moisture conditions, freedom from pests and diseases and excellent grain quality. A third order of priority is assigned to species with more localized adaptation and use but with considerable potential for expanded production. It is assumed that these species would be of greater immediate concern to national interests. These include: (1) African yam beans-for adaptation to the humid tropics, exceptionally good quality protein (high in essen-

96

K. 0. RACHIE AND L. M. ROBERTS

tial amino acids) and excellent taste qualities; ( 2 ) winged bean-exceptionally broad array of uses, high quality of product (leaves, green pods, dry seeds, and tubers), and freedom from pests and diseases; ( 3 ) velvet bean-broad range of uses, vigorous growth and freedom from pests and diseases; (4) jack beans-exceptional adaptive qualities, high yield potential, and medicinal purposes; (.5) horse gram-productivity under very poor soil conditions and high yields of both seed and forage; (6 ) moth bean-exceptional tolerance of extreme conditions of drought and high temperatures; (7) cluster bean-adaptation to high temperatures and low moisture, exceptional production of seeds, pods, and forage and possible commercial production of mucilage,

4 . Strategy for Diversification The range of environmental conditions and crop hazards in the lowland tropics is nearly infinite. It is reasonable to assume that many of the major agronomic problems of specific crops can be solved through combinations of broadly based research in the areas of plant improvement, plant protection, growth processes and management. However, some constraints on individual species may not be amenable to solution at prevailing levels of technology. It may be much more expedient to offer other species for those situations. Moreover, complex multiple and mixed cropping systems require several productive and trouble-free crops. Since optional secondary crops inevitably have deficiencies of their own requiring attention, there is a question of allocating research efforts. Diverting resources from established major objectives could result in net loss in overall productivity, since human needs and tropical environments with attendant hazards are dynamic, requiring constant vigilance. While it is unrealistic to expect all 24 species to receive full research support, then are certain mitigating factors favoring an increased interest in secondary legumes. The rapidly rising food prices and current world shortage of energy and fertilizers provides incentives to explore more efficient sources of nutrition. Moreover, the fact that the global network of international agricultural research institutes is committed to including the four major lowland species and Phaseolus beans in their improvement activities should relieve some of the pressure on national programs. It is hoped that individual countries and regional organizations could divert a portion of their research resources to some of those secondary tropical legumes with localized interest. Since these developments will take some time, it is essential to assemble, collect, and maintain the germplasm of these species before they are irretrievably lost in the accelerated ecological changes occurring with rapidly increasing population pressures.

97

GRAIN LEGUMES OF THE LOWLAND TROPICS

Appendix: Tables

TABLE I Trends in the Estimated Production of Grain Legumes for Intermediate-High and Lowland Tropical Regions for the Periods 1948-1952, 1961-1965, and 1971 1948-1952

Area Crops and species

Prod.

(M ha) (M tons)

Intermediate-high elevations 1. Dry beans (Fhaseolus vulgaris) 4. Dry peas (Pisum spp.) 3. Broad beans (Vicia

1961-1965

Area

Prod.

(M ha) (M tons)

1971

Area

Prod.

(M ha) (M tans)

4.24

2; 31

6.55

3.74

8.38

5.53

1.53 0.30

1.05 0.23

1.76 0.40

1.s2 0.29

1.50 0.47

1.18 0.32

8.25

4.27

9.79

5.88

8.50

5.69

0.70 0.08

0.31 0.07

0.96 0.44

0.45 0.49

0.95 1.96

0.48 2.60

1.14 16.24 -

0.40 __ 8.65

0.44 __ 12. 62 45.9

0.80 -

-

1.01 __ 20.91 28.70

22.56 37.2

0.38 16.03 101.0

8.98

6.40

14.48

11.52

14.87

13.01

4.26

1.08

7.43

2.12

7.80

2.54

2.41

1.40

2.69

1.76

4.91

2.61

1.31

0.44

2.42

0.96

2.99

1.08

0.42 4.05 21.42

0.29 1.60 __ 11.a1

0.66 4.03 __ 31.60 47.00

0.45 1.78 18158 65.8

0.70 4.01 __ 33.28 55.4

0.51 1.91 21.64 93.0

Total (both elevations) 37.66 Increase over 19481952 (%)

19.86

52.51 39.40

31 .20 57.10

45.56 47.5

37.67 89.7

faW 4. Chick-peas (Cicer

arietinum) 5. Lentils (Lens esculenta) 6. Soybeans (Glycine max)b 7. Other (20%) Subtotal Increase over 19481952 (%o)

Lotoland tropics 1. Peanuts (Aruckis hypogaea)c 2. Asian grams (dry beans) 3. Pigeon peas (Cajanus cajan) 4. Cowpeas (Vigna unguiculata) 5. Soybeans* 6 . Other (80%) Subtotal Increase over 1948-

~

-

-

195%(%I

4

M ha

-

= million.hectares; M tons = million tons ;Prod, = Production. African and American tropics. Production expressed in shell. “Dry bean” production in Asia. Soybeans in tropical Asia.

TABLE I1 Regional Production of Loudand Tropical Grain Legumes in I971 in Millions of Hectares and Metric Tons Southern Asia Crop

Actual

Pcrcenta

8.19 7.10 2.72 1.87 0.03 0.02 7.80 4.52 3.30 1.70 22.04 13.E l

55.1 54.6 94.0 71.7 1.0 1.9

Tropical Africa Actual

Tropical Americas

Percent“

Actual

Percent”

38.9 37.2 5.15 27.2 99.0 98.15

0.90

6.1

1.07 0.04 0.03

11.1 1.4 1.15

World total Actual

Percenth

14.87 13-01 2.91 2.61 2.99 1.08

44.7 60.1 8.7

0 p

12.1 9.0

8

7.80

93.4 11.6 14.2 11.2 59.99 57.49

P 1. Peanuts? 2.

3. 4. 5.

Aread Prdd Pigeon peas: Area Prod Cowpeas: Area Prod Asian grams? Area Prod Unspeci6ed:f Area Prod All lowland pulses: Area Prd

5.78 4.84 0.15 0.71 2.96 1.06

100

100 70.1 70.2 66.2 61.0

1.34 0.68 10.23 7.29

28.4

28.1 30.7 33.7

-

-

0.07 0.04 1.01 1.14

1.5 1.7 3.0 5.3

2.52 4.71 2.4% 33.28 21.64

5.0

Percent of total production of that species grown in the lowland tropics. Percent of all grain legumes (all species) grown in the lowland tropics. Since peanuts are one-third shell, net kernel production is estimated a t 8.9 million metric tons, or 52.8% of all lowland grain legumes. Area in millions of hectares and production (Prod) in millions of metric tons given in columns headed “Actual.” e “Dry bean” production in Southern Asia is assumed t o be predominantly Asian g r a m (mungbeans, black gram, rice bean, and others). f Includes soybeans, presumed t o be a lowland crop in tropical Asia. 9 Proportion of all tropical legumes including intermediatehigh elevations.

*

z r

z

zid

1

v1

TABLE I11 Some Important Adaptive Features and Botanical Characteristics of Selected Lowland Tropical Grain Legumes" Part A. Semiarid Regions (less than 500-600 mm annual rainfall) Dry seed pruductivity levels

Region and name (presumed origin) 1.

Bambara groundnut (Africa)

e. Kersting's groundnut (Africa)

3. Moth bean (India/

Burma)

Scientific name

Cbromosomes (an = )

DuraPerention nialityb (daya)

Voandzeia subterranea

42

Kerstingiella geocarpa

ee

p.

Yigna oconfifolia

24

-4

A

Plant tgpe/size

Small, bunchy herb; prostrate, rooting 150 branches; underground fruiting 90 Small, bunchy herb; to prostrate, rooting la0 branches; uuderground fruiting 65 Slender, trailing hairy to herb 10-90 cm tall 90

to

Soil and climate preference/ tolerance

Pest/ diseases Masisuscep Average mum tibility. (kg/ha) (kg/ba)

Dry, poor soils; high temperatures

VL

750

2600

Unripe seeds eaten fresh; and ripe seeds used as a pulse

Dry, poor sandy soils; high temperatures and sunshine

VL

500

-

Unripe and mature seeds used as a pulse

Dry, light sandy soils

M

300

1600

to 400

90

4. Tepary bean (Mexico)

5. Cluster bean (India)

Plaseolus acutifolius var. latifolius Cyamopsis tefragonolobua

ee

h

60

to 14

A

90 90

to

1eo

Suberect herb, bushy or recumbent, %5em high Robust busby herb

Purpose and utilization

Dry soils; does not tolerate waterlogging Alluvial/sandy soils; high temperatures

M

400 to

700

VL

400

to 600

Green pods as vegetable; ripe seeds whole or split as pulse. Forage, hay, and manure 1500 Dry seeds for pulse; forage: 5-10 tons of dry hay 1600 Green beans as vegetable; leaves and stems for forage; dry seeds for mucilage (Continued)

Table 111 is from Rachie, 1973. A = annual. C VL = very low: M = medium.

TABLE I11 (Continued) Table 111, Part B. Semiarid t o Subhumid Regions (600-1000 mm annual rainfall) Dry seed productivity levels

Region and name (presumed origin)

Scientific name

ChromoDurasomes Peren- tion (4n = ) nialityn (days)

1. Peanut (Brazil)

Araehis hupogaea

40

A

9. Pigeon peas

Cajanua cajan

4%

P

(E. Africa)

(44)

Plant type/aize

Soil and climate preference/ tolerance

Friable aandy loams 100 Low bunchy herb; to underground fruiting 150 Well-drained sandy/ 100 Semiwoody shrub 1.5 clayey loams to to 5 m tall

P&/ diseaaes Marisuaeep Average mum tibilitya (kg/ha) &&ha) ME

600 to 800

L

400

to 500

900

9. Cowpeas (Nigeria)

4 . Mung beans/black

gram (India/Burma) 5. Horse gram (So.Asia)

6 . Hyacinth bean

Vigna unguiculala

Vigna radiala and var. munpo Dolichoa uniporua

Lablab niger

(So. Asia)

a4 44

44

A or SP

A

44 (44)

80

to 140

(44)

94

Twining, climbing, or procumbent herb; or 200 erect bush: 40-140 cm tall 65

to

A

SP

Well-tilled l o a m to Erect-suberect, hairy clays blaek cotton herb; 50-130 cm tall soils

140 Low, slender, semito erect herb 180 75

to

Herbaceous twining and bush forms

900

7. African locust bean

P a r k a spp.

-

P

(Africa) a

A = annual; P = perennial; SP = short-term perennial. VL very low; L = low M = medium; MH = medium high.

-

WeU-drained sandy loam; bigb tcmperatures

Tree: 10-30 m

Tolerates very poor soils

EI

300

to 400

M

400

to 500

M

200

to 300

Well-drained; tolerates poor soils and low fertility

M

Wide range. alluvial soils

VL

400

to 500

950 to 500

Purpose and utilization

9000 Industrial: oil, seed cake: dry see& for

cooking, condiments 9000 Dry seeda for pulse; unripe seeds as vegeto table; forage crop 5000 and cover 2800 Dry seeds as pulse; tender green aeedlings, leaves, pods and seeds as vegetables: forage and green manure crop 4700 Dry seeds as puhe, split or sprouted; green pods as vegetable; forage 800 Dry seeds as pulse and animal feed; dry forto age and green 1400 manure 1500 Young pods and green beans as vegetables; dry seeds for puhe and feed for livestock; forage - Dry seeds fermented as flavoring: fruit nulD also cooked

Table 111, Part C. Subhumid Regions (1000-1500 mrn annual rainfall)

Pest/ Region and name (presumed origin) 1. Pbaseolus beans

(C. America)

Scientific name Phaseoh8 oulgaria

ChromoDuraPeren- tion somes (4n =) nialityn (days)

44

A

60 to

Plant type/size Dwarf bush to twiniug/climbing

Soil and climate preference/ tolerance Light sands and peat, to clayey soils

diseases Maxisuscep- Average mum tibilityb (kg/ha) (kg/ba) VH

Glycine moz

40

A

China)

80 to

700

Erect bush; also twining 9W180 cm

Tolerates some waterlogging

M

Vigna umbellata

49

SP

60

Erect-suberect/ twining 150-800 ern

Light to heavy soils (after rice)

L

400 to

to

Tolerates some waterlogging

VL

800

300

90 4. Jack/sword beans

(C. America and Africa)

-

Canaaolia spp. C. en~forformis C. gladiata

49

P

(44)

annual; P = perennial; SP = short-term perennial. b VL = very low; L = low; M = medium; V H = very high.

.A

180 Bushy, erect 1-4 m; to large climber 300

600

to 1000

900

3. Rice beans (S.E. Asia)

500

to

100

4. Soybeans (S.E. Asia/

Dry seed productivity levels

to 1000

Q

gcr Purpose and utilization

Dry seeds as a pulse; green pods and beans as vegetable; also for forage 5000 Industrial-protein and to oil; green seeds as 6000 vegetable; dry seeds as a pulse; forage from leaves stems 1400 Dry seeds as pulse: green seeds and pods asa vegetable,fodder 4600 Green pods as veget4500

able; ripe seeds as pulse; medicinalurease and lectiu; vegetation for forage and cover (Continued)

5

r m

$

E

% +I

0"

* $

+I

gcr

c1

C A

c.L

0

TABLE 111 (Continued)

h,

Table 111, Part D. Humid and Very Humid Regions (above 1500 mm annual rainfall) Dry seed productivity levels

Region and name (presumed origin) 1. Lima beans

( C . America)

4. Winged bean

(tropical Asia)

Scientific name Phaaeolus lunatus

Psophoearpua tetragonolobue

Chromommes

(en = ) 42

DuraPeren- tion nialityo (days)

P

100

to 270

-

P

180

to

Plant type/size Twining climbers; or bush types

Soil and climate preference/ tolerance Humid; well drained; aerated soils

Pest/ Mandiseases suscep Average mum tibilityb (kg/ha) (kg/ha)

VL

to 600

Twining, glabrous herb, 2-4 m long

Humid climate: loamy soils

VL

Sphenostylis stcnocarpa

-

4. American yam bean

Pachynhizus eroaus

28

P

Mucuna pruriens var. utilia

-

P

Mucuna eloanei

-

(Mexico and C. America) 5. Velvet bean (Africa)

6 . Horse-eye bean

P

Twining, climbing or procumbent herb, 300 3-6 m Herbaceous climber, 2-5 m to

to

P = perennial.

* VL = very low; L = low.

Eerbaceous climber, 3-8 m

300

P

440

to 360 a

500

150

240

400

to

270

3. African yam bean (West Africa)

500

Herbaceous climber, 3-10 m

Humid, well drained loams

L

Humid: well tilled, sandy loams

VL

Humid, poor, sandy loams; high temperatures

VL

Humid, well drained soils; high temperatures

VL

300

to 500 -

700

to 1000

-

Purpose and utilization

2800 Dried beans as pulse; green beans, young pods and leaves as vegetable (seeds may have HCN) 4500 Fresh peen pods, leaves as vegetable; tubers; dry seeds as pulse; also green manure and forage 1400 Dry seeds as a pulse; tubers fresh or cooked Tubers: raw or cooked; green pods: vegetable Seeds used as pulse; crop also grown for green manure, cover, and forage Ripe seeds are used as a pulse in thickening soups

9

z

tr

r

F % l

0

103

GRAIN LEGUMES OF THE LOWLAND TROPICS TABLE IV Inheritance of Some Important Genetic Characters in Pigeon Peas (Cajanus cajan Millsp.) Character

Symbol

Plant architecture Cotyledon shape

Normal trifoliate leaf

TT

Pointed leaf apex (lanceolate)

Mc

11 12

Stature Short

Growth habit Creeping

Erect

Reference

Determined either by pleio- Deksmukh and tropic action of a leaf Rekhi (1961) shape gene or a gene closely linked to it, pointed leaf apex and lanceolate cotyledon being dominant over rounded apex and ovate cotyledon

Foliate condition Unifoliate (pointed leaflet) Oval-oblong trifoliate Trifoliate with pointed leaves

Leaf mutants Obcordate leaflets

Mode of inheritancepigeon peas

I n crosses of these types, the Deshmukh and trifoliate condition is Rekhi (1960) monogenic and dominant over unifoliate; pointed apex is dominant over the rounded apices and also monogenic. The two gene pairs segregated independently Dominant to unifoliate; monogenetically inherited Dominant to round apex; Rekhi (1966) monogenetically inherited (obovate)

A spontaneous mutant having obcordate leaflets with mucronate apices and filiform flower keel depends on pleiotropic duplicate factors 11 and 12

Deshpande and Jeswani (1956)

Shaw (1986) Dominant to tall stature of type 80-monofactorial segregation observed in Fx of both pairs of characters Segregation data of FI showed 13 creeping: 3 erect, suggesting two factors, one of which has inhibiting action Erect branching, dominant t o spreading habit; monogenetically inherited I

Shinde et al. (1971)

Rekhi (1966)

(Continued)

104

K. 0. RACHIE AND L. M. ROBERTS TABLE IV (Continued)

Character

Symbol

Prostrate

Fasciation

Plant color Stem color Purplish stem Green stem Growth/development Time of flowering Lateness

Steriles Steriles Sepaloid mutant

Weak mutant

Reference

In one cross, erect habit was Shaw (1936) only partially dominant t o spreading habit True breeding mutant; may Deshpande and Jeswani (1959) be useful as cover crop and soil conservation Chaudhari and Patil (1953)

Spreading

Dwarf, bushy plant Late, brittle stalks

Mode of inheritancepigeon peas

d

A single recessive gene desig- Sen et al. (1966) nated d appears t o be involved; pollen fertility in the mutant was only 70% Bhatnagar et al. Mutant had weak, curved stems (purple), with (1967) branches fused t o the main stem a t place of emergence; 11% pollen sterility though many seeds produced; fasciation was recessive t o normal

Incomplete dominance over green pigmented stem

Ganguli and Srivastava (1967)

Completely dominant in one Ganguli and cross over earliness; inSrivastava (1967) completely dominant over earliness in another cross

Simple leaves replaced normal trifoliate ones and were associated with a sepaloid condition of the flowers Simple leaves on lower part of plant and none on upper part with rudimentary floral organs in addition t o dwarf habit and thin, straggling branches

Jeswani and Deshpande (1969)

105

GRAIN LEGUMES OF THE LOWLAND TROPICS

TABLE IV (Continued) Character Cleistogamous mutant

Symbol

Inferior stigma

Mode of inheritancepigeon peas Reference Possessed thick, puckered trifoliate leaves; overall abnormal condition is monogenically recessive t o normal, segregates independently of obcordate/ lanceolate leaflet gene. A t Niphad (India) a sterile Patil and Sheikh (1957) plant found to have stigmas positioned below anthers instead of above them

Injoreseence Flowering conditions Nonflowering Flowering

Monogenically recessive t o Joshi and flowering condition; does Ramanujam not appear t o be linked t o (1963) pleiotropic locus controlling trifoliate versus simple leaf and normal versus sepaloid flower

Inflorescence Crowded Open

Crowded inflorescence of type 5 dominant t o open inflorescence of type 80 on a 3 : 1 ratio

Pistil Multicarpellate condition

Flower color Basic color Absence of venation Interacts with 21 locus

Shaw (1936)

Monogenetically recessive t o Joshi and normal unicarpellate conRamanujam (1963) dition-the allele appears also t o control development of supernumerary petals, the development of stamens into petals or, carpel-like structures and exposed ovules; mutant plants are female sterile, with 80% stainable pollen Y

r p

Loci 2: and p found t o be linked with a recombination frequency of 29.7%; and genotypes ppVV, ppVv showed incomplete penetration of the V allele, resulting in 13-527% percent of deep-veined individuals recorded under light-veined class

Jain and Joshi (1964)

(Continued)

106

K. 0. RACHIE AND L. M. ROBERTS

TABLE IV (Continued) Character Flower petal color Yellow-entire Yellow with light red veins

Yellow with dark red veins Purple streaked Blood red (solid White 5owers (mutant)

Symbol

Ap cevs, ap eevs up ce V s

A p Ce va

Mode of inheritancepigeon peas

Reference

Recessive t o all other condi- Menezes (1956) tions Shinde et al. Data showed 3 yellow with deep red veins: 1 yellow (1971) with light red veinsindependent assortment Dave (1954)

Ap C E Va Dominant t o plain yellow: monogenetically inherited Ap C E VY Simple dominant to all yellow and yellow with purple Interaction of two duplicate Patil and D’Cruz genes W1 and W2 and (1962) spontaneous mutation of the inhibitory gene I , conditioning yellow

Wing color Orange Yellow

Dominant over yellow

Ganguli and Srivastava (1967)

Pods and seeds Unripe pod color Green withlblack diffused Green with/black streaks All green All purple

Dark green

Lrd Ld Id LD

Data showed 3 green with black diffused: 1 green with black streaks-independent assortment Recessive t o all others Dominant t o all others. “D” controls color distribution; incomplete dominance over “d” Dominant t o light green

Shinde et ul. (1971)

Menezes (1956)

Sen et al. (1968) Menezes (19.53)

107

GRAIN LEGUMES OF THE LOWLAND TROPICS

TABLE IV (Continued)

Character Seed coat color Purplish black Chocolate Spotted White

Dark purple with blotches Brown

Seeds/pod Four-seeded Three-seeded

Disease resistance Resistance t o wilt

Symbol

PR PR Pr

P

Mode of inheritancepigeon peas

Reference

Color is expressed as an inDave (1934) teraction of two loci: Menexes dominance is simple; (1956) black is dominant t o chocolate, spotted, and white Incomplete dominance over Ganguli and chocolate and light brown Srivastava (1967) Brown is partially or incompletely dominant over white seed coat (monogenic)

Rekhi (1966)

Four-seeded pods dominant t o three-seeded pods; monogenetically inherited

Rekhi (1966)

Shaw (1996) Inherited independent of flower color, erect or spreading habit of growth, short or tall stature of plant, crowded or open inflorescence and brown or gray markings of the seeds Inheritance of wilt susceptibility suggests its control by 2 or 3 factors not linked with any of the morphological characters studied

108

K. 0. RACHIE AND L. M. ROBERTS TABLE V Linkage Groups of Some Important Genetic Characters in Pigeon Peas (Cajanus cajan Millsp.) Linkage groups in pigeon peas

1. Erect-Black pod-Lanceolate leaflet shape, Z--B1p-L~t. The 3 characters have shown recombination values of 40.8% between factors Z and Llr;35.7% between I and Bl, and E.9% between BZp and Zl, 2. There is a complete linkage between orange yellow flowers and purplish black seeds; and between yellow flowers with back of standard having purple veins, the base diffused purple, and purple-green pods. There is a close linkage between yellow flowers with the backs of their standards purple and maroon blotched pods a. Orange yellow flowers-purplish black seeds b. Yellow flowers, back of standard with purple veins, base diffused with purple and purple-green pods c. Yellow flowers, backs of standards purple, close linkage with maroon blotched pods 3. Linkage of venation with pigmentation: v = absence of venation, p = interacts with 8 , y = basic color (flower color and venation) Loci p and v were found t o be linked with a recombination frequency of %9.7%, and genotypes ppVV, ppVv showed incomplete penetration of the V allele, resulting in a range of 13-'27% of the deep veined individuals recorded under the light-veined class

Reference Pati1 (1965)

Dave (1934)

Jainand Joshi (1964)

109

GRAIN LEGUMES OF THE LOWLAND TROPICS TABLE VI Inheritance of Some Important Genetic Characters in Cowpeas (Vigna unguiculata Walp.)

Character

Symbol

Mode of inheritancecowpeas

Reference

Plant architecture Stem Swelling Normal

Leaf shape Narrow leaf

Hastate leaves Rhomboid leaves

Growth habit Vining Tallness

Vininess

sw

Swelling a t base of stem due Roy and Richharia to an increased amount of (1948) parenchyma in the phloem in ssp. v. sinensis ‘Tanganyika’ was monogenically dominant (Sw)over normal stems

Nlb

Determined by incompletely Saunders (1960) dominant gene like ancestral forms Dominant over rhomboid Jindla and Singh leaves: LS1 is essential, (1970) while any two of LS2, LSI, or LSd produce hastate leaves

LSI, LS2 LSa, LS4

V T

VI, Vz

Crossed between Vigna Kovarskii (1939) sinensis and V . catjang (close to wild spp.) showed dominance for “wild” characters: vining, earliness, dark-green leaves, dark mottled seeds, resistance to mosaic, and generally vigorous growth Governed by duplicate genes Kolhe (1970)

Plant color Foliage color Pale green (light)

Lga, Lgb

Normal

rr

Basic plant color

R

Inherited independently as a single recessive gene Two complementary genes govern foliage color Plants are green with white flowers and cream seeds; sap-soluble pigment produced only in presence of basic gene for color ( R , 7 )

Saunders (1960) Kolhe (1970) Sen and Bhowal (1961) ; Saunders (1960)

(Continued)

110

K. 0. RACHIE AND L. M. ROBERTS TABLE VI (Continued)

Character Stem pigmentation Base of primary branches Petiole base

Growth/development Time of flowering Earliness Lateness

Symbol

Pbr

Pb

Efi, Efz

Photoperiod response Short day Day neutral

Steriles and lethals Male sterility

Mode of inheritancecowpeas

Purple base of primary branches simple dominant to green Purple petiole base-simple dominant to green

Reference

Sen and Bhowal (1961)

Ojomo (1971) Early flowering was dominant t o late flowering; number of days to flowering appeared to be controlled by the action of duplicate dominant epistasis between two major genes (Efl and Efi) in the presence of some minor modifying genes Roy and Richharia In another experiment, the (1948) F1was intermediate between the two parents (49 and 94 days) with a tendency toward earliness. The FZdata suggested that time of flowering may be determined by two complementary factors Sine (1967) Short-day response simple dominant to day neutrality

Ms ms

Sen and Bhowal A spontaneous male sterile (1962) mutant arose in Vigna a'nesis ssp catjang 'Poona'; pollen meiosis did not proceed beyond early diacinesis. Sterility is controlled by the recessive condition of a single pair of genes (msms) Sterility discovered at IITA IITA (1974) is also controlled by a recessive pair of genes (ms2 ms2)

GRAIN LEGUMES OF THE LOWLAND TROPICS

111

TABLE VI (Continued)

Character

Symbol

Lethal genes

Injlorescence Type Compound inflorescence Simple inflorescence

C

Compound inflorescence of ssp. V . catjang “Poona” was monogenically recessive ( c ) to simple inflorescence

Sen and Bhowal (1961)

These factors are pleiotropic with those governing seed testa colors and patterns Dark is dominant t o pale or tinged

Jindla and Singh (1970)

WHO who WHO WhO WHO Who H or D

Tinged Dark Violet

~

Dark flower color is epistatic to pale and tinged Sen and Bhowal Violet flowers with self(1961) colored seeds in presence of the gene R Tinged flowers with holsteineyed seeds in presence of the gene R Violet flowers dominant over very light violet, D, enhances the intensity of color in the presence of L

h or L

Tinged, light blue or violet

~~

Seedlings from the cross Saunders (1952) PORTUQUESE WHITE X LIGHT RED began t o show reduced growth and wilting when about 2 weeks old; all succumbed within 6 weeks. Two complementary genes, L1 and Lz, carried by PORTUQUESE WHITE and LIGHT RED, respectively, were responsible. Two other lethals, occurring at a much earlier stage, were discovered in the F1 of the cross LIQHT RED X NI and are designated La and Ld

w, H , 0

Pale

~

Reference

LI, LZ La, Lc

C

Flower color

Mode of inheritancecowpeas

~

(Continued)

K. 0. RACHIE AND L. M. ROBERTS

112

TABLE VI (Continued)

Character

Symbol

Mode of inheritancecowpeas

Reference

~~~~

Flower color Standard petal Calyx color

Pod characters Unripe pod color Purple

Pf Ystp

BGY

Pu P p or Pc

Cerise Straw

PC

Drab

rr

Red-tipped straw

Pb

Green pods and purple-tipped pods

Pt

Green pods with a purple ventral suture Purple pods with both sutures green Green pods with both sutures purple

P'

P3

PC

Expressed in presence of R Expressed in presence of R Expressed in presence of R

Kolhe (1970)

Pp produces purple pods

El-Murabaa and Mustafa (1970)

only in the presence of R ; purple pod is epistatic t o both cerise and strawcolored pod

Straw-colored phenotype is also produced by p in the presence of either R or rr Drab pod color of Sudani is controlled by a single gene and is dominant t o straw pod color of varieties Asmidi and Fitreiat (rr) Red-tipped straw pods formed by Pb only in the presence of R; in combination with rr the pods are straw-colored Sen and Bhowal Allele pt may represent 2 alleles, one for a more (1961) purple tip and black seeds, and the other for a less purple tip and nonblack seeds. The dominance relationship between pg and pv a t the unripe stage was reversed at the half-ripe stage Respectively, plants homoSaunders (1960) zygous for p s and p u bore green pods and green pods with faintly purple sutures and tips. Pod color genes appeared pleiotropic with colors of stem, petiole calyx, and standard petal

GRAIN LEGUMES OF THE LOWLAND TROPICS

113

TABLE VI (Continued)

Character

Brown Speckled Yellowish green Green Dark green

Number of chloroplasts

Symbol

Y 9

G GD

g

G Ge

Ripe (dry) pod color Brownish-straw Amber-straw Pod condition Inflated Constricted

Mode of inheritancecowpeas

Reference

Dominant t o green and other Saunders (1960) lighter colors I n order of increasing domi- Sen and Bhowal nance effects the grade of (1961) chlorophyll production in calyx, leaves, and dorsal surface of standard The number of chloroplasts Sen and Bhowal per cell and their intensity (1961) of color and size progressively increased in unripe pods representing the series gL,GL, Ge (see unripe pod color) Both monogenically dominant over straw (rr) colored ripe pods

Sen and Bhowal

Inflated in cv. Sudani controlled by single gene dominant to constricted pods of Azmirli and

El-Murabaa and Mustafa (1970)

(1961)

Fitreiat Pod size Length (long, short) Size (large, small)

Pod surface

Partial dominance observed Jindla and Singh (1970) for pod length; appeared t o be under multiple gene control I n other studies the F,’s be- Roy and Richharia tween short and long pods (1948) were intermediate, with a tendency toward short pods W p a , W p b Two complementary genes Kolhe (1970) responsible for expression

Seed characters Seeds per pod

Heterosis for seeds per pod was exhibited in Fl’s of 13 and 16 seeds per pod parent cross. The FI produced a n average of 18.0 -t 0.52 seeds per pod

Roy and Richharia (1948)

(Continued)

K. 0. RACHIE AND L. M. ROBERTS

114

TABLE VI (Continued)

Character Grain deposition

Symbol

Dgda Dgdb

Seed shape Cylindrical Kidney-shaped Cordate

Seed length

Lg

Testa thickness

Th

Co

Mode of inheritancecowpeas Two complementary genes are responsible

Reference Kolhe (1970)

Two multiple factors deter- El-Murabaa and mine the difference beMustafa (1970) tween the cylindrical seeds of BUDANI and the kidney/ cordate shaped seeds of AZMIRLI and FITRELAT Long seeds (grain) dominant Kolhe (1970) t o short seeds Two major genes with possible cumulative interaction appear to govern testa thickness in Vigna unguiculata The presence of additional Ojomo (1972) minor genes is postulated to account for the variabilfty observed in all phenotypic classes

Testa color patterns WV,H,O Self or solid Watson

WHO Who

Holstein

WHO

Small eye

who

Hilum ring

WHO Who

These factors appear to be pleiotropic with those governing flower colors Dominant H is self-colored The symbol w is also used for Watson-eyed, being recessive to TI (selfcolored) Another symbol for holsteineye ( H H ) is hh WW Another symbol for smalleyed (hh) is hh ww These four genotypes produce a hilum ring

Saunders (1960)

WHO

Large eyed Testa color Gray White

who Wwhh

Big-eyed (or Hh) Black (2)is dominant and epistatic to brown (B), which in turn is epistatic to red ( R ) .Black is dominant to red

Saunders (1959)

GRAIN LEGUMES OF THE LOWLAND TROPICS

115

TABLE VI (Continued) Character Green

Symbol

Buff Red Brown Black Speckled Mottled Blotched (testa pattern)

Seed eye color Black eye Brown eye Smoke eye

B

z

S

U

Mode of inheritancecowpeas Reference Solid colored testa is brought Roy and Richharia (1948) about by the complementary action of two domiSaunders (1959) nant genes, H and W; and the patterns “variegated” or “holstein” ( H I Z ) , large eye (Hh), small eye (hh) find expression in the absence of W, which governs a color pattern with an indistinctly limited eye; intensity genes may also play a part Mackie (1934) Epistatic t o brown and smoke eyes The eyed character is a sim- Mackie (1939) ple recessive so that it can be easily retained in the true-breeding condition

Retktances Fusarum wilt (F. oxysporum,f .

trachciphilum)

Resistance to race 1 is domi- Mackie (1934) nant; sources of resistance: IRON, VICTOR, and BRABHAM Earliness may permit escape; Hawthorne (1943) IRON CLAY also resistant. Resistance t o race 1 but sus- Hare (1959) ceptible or tolerant to races 2 and 3: EXTRA EARLYBLACKEYE,PURPLE HULL BUNCH,

and

MISSIS-

SIPPI CROWDER

Resistance to all three races: IRON, MISSISSIPPI 755, and MISSISSIPPI 5 7.1

MAES (1959)

Resistant is dominant; sources are: IRON, VICTOR,

Mackie (1934, 1939)

Charcoal rot

(Rhizoctonia 6ataticola)

BRABHAM

Stem rot

Field resistance: MALABAR, QIANT, and CRISTANDO (when inoculum level is low); immunity: BLACKEYE 6 , HAVANA, and SANTIAGO. POONA is susceptible.

PURSS(1958)

(Continued)

116

K. 0. RACHIE AND L. M. ROBERTS

TABLE VI (Continued) Character Bacterial pustule (canker) (Xanthomonas vignieola)

Powdery mildew

Symbol

Mode of inheritancecowpeas

Reference

Resistance is dominant to Lefebvre and susceptibility (3: 1) Sherwin (1950) Sources: 451 out of 578 IITA (1974) tested a t IITA in 1973 Resistance is recessive to Fennel1 (1948) susceptibility and dependent on multiple factors. Sources: sesquipedalis; also CHINITO and No. 0199

v.

Leafspot diseases (Cercospora cruenta)

Southern bean mosaic virus

Resistance is dominant: 454 IITA (1974) out of 578 lines tested a t IITA were free of disease Out of 37 lines tested in Verma and Pate1 India, 2%were resistant (1969) Resistance is dominant and Kuhn and Brantley (1963); Brantley single gene. Resistance in 9 and Kuhn (1970) varieties; also found in crosses of IRON X QROIT; and in CLAY and CLAY X TOPBET

Cowpea yellow mosaic

Tolerance attributed to three additive factors:

Bliss and Robertson (1971)

ALABUNCH

Resistance also found in a selection from DIXIELEE and appeared to be dominant (single gene) Resistance was observed to be recessive-single gene pair in PI 297562 Immunity identified in: TVu 39, 45, 99, 103, 106 and

Reeder et ul. (1972)

IITA (1974)

1190

Cowpea mottle virus

Cucumber mosaic virus

Tolerance was dominant to susceptibility with either one or two genes involved Resistance due to a single dominant gene in “Black SR” Other tolerance: OAKBANCITO, AZUL QRANDE,

CHINITO,

and No. 0199

Bliss and Robertson (1971) Sinclair and Walker (1955); DeZeeuw and Crum (1965) Fennel1 (1948)

117

GRAIN LEGUMES OF THE LOWLAND TROPICS

TABLE VI (Continued)

Character

Symbol

Mode of inheritancecowpeas

Reference

Resistance is dominant and DeZeeuw and linked to CMV susceptiCrum (1963); bility; a n epistatic recesDeZeeuw and BaIlard (1958, sive inhibitor may also be 1959) present. Resistant: “Black RS” and mutant of “Black”

Tobacco ringspot virus

Root knot nematodes (Meloidogyne incognita)

Resistance is dominantsingle gene pair: IRON, VICTOR, BRABHAM Other resistance: PURPLE HULL CROWDER, PINK-EYE; also, M 855, M 455, M 755, and M 855. These were also resistant to M . jauanica and M . arenatia MISSISSIPPI755 and 57-1 reported to be resistant Early maturity helps plants escape nematodes Widecrossing with resistant wild species not successful

Mackie (1934,

Resistance to root diseases and nematodes in IRON, VICTOR,and BRABHAM attributed to high quantity of suberin in root cortex. Dark leaf color was also associated with disease resistance Multiple resistance of MISSISSIPPI 755 and 57-1 obtained from PURPLE HULL CROWDER includes resistance to Fusatium wilt, nematodes, and viruses

Mackie (1954,

Resistance is a monogenic dominant

Kolhe (1970)

1939)

Ivanoff (1963)

MAES (1959,1961) Hawthorne (1943) Saunders and Laubscher (1945)

General resistance

Beetle (Ceratoma trifurcata)

Rb

1939)

MAES (1959,1961, 1963)

118

K. 0. RACHIE AND L. M. ROBERTS

TABLE VII Linkage Groups of Some Important Genetic Characters in Cowpeas ( V g n a unguiculata Walp.) Linkage groups in cowpeas

Reference

1. Flower color (pf)-standard petal color (yatp): crossover Kolhe, (1970) value = 20.4% 2. Flower color (pf)-grain color ( B T ~ ) : crossover value = 20.3% 3. Grain deposition (Dgda)-pod surface (Wpa): crossover value = 5.8% 4. Speckled pod color-speckled seed coat Saunders (1960) 5. Pale green plant color-seed color pattern (W) 6. Spindly growth habit-purple plant color ( P ) 7. Speckled seed coat @)-brown pod (Y) 8 . Hilum ring pattern-purple pods 9. Basic color gene @)-mottled seed (0) 10. Black seed coat @)-cerise pod ( p C ) 11. Gray seed color-late maturity Roy and Richharia (1948) 1%.Pod length-fibrousness 13. Resistance to root knot nematode-poor seed set (cream Ivanoff (1962) and crowder types): weak linkage, but can be broken 14. Resistance to tobacco ringspot virus (TRSV) is linked to cqcumber mosaic virus (CMV) susceptibility

REFERENCES Abdou, Y. A. M. 1966. “The source and nature of resistance in Arachis sp to Mycosyhaerella arachildicola and M . berkeleyii and factors influencing sporulation of these fungi.” Thesis, North Carolina State University, Raleigh. Abrarns, R. 1967. I . Agr. Univ. P . R. 51, 1-21. Abrarns, R.,and Velez-Fortuiio, I. 1961.I . Agr. Univ. P.R. 45, 197-204. Abrams, R.,and Velez-Fortuiio, J. 1962.I. Agr. Univ. P . R . 46, 34-42. Abrams, R., Velez-Fortuiio, J., and Garcia, J. L. 1969. J . Agr. Univ. P . R. 53,

61-66.

Abutalybov, M. G , , and Sarnedova, A, D. 1966. Izv. Akad. Nauk Azcrb. SSR, Ser. Biol. Nauk 2, 16-22. Aguirre, A., and Palencia O., J. A. 1968. Agronomia (Guatcmala) 2, 27-42. Ahsan, R., Fatina, R., Zain, B. K., and Zain-ul-Abedin, M. (1968). Pak. J , Biochem.

1(1/2), 16-20.

Akinola, J. O., and Whiternan, P. C. 1972. Aust. 1. Agr. Res. 23, 995-1005. Alarn, M. 1933. 20th Annu. Meet. Indian Sci. Congr. Patna Sect. Agr., Abstract No. 43, pp. 15-16. Aldana, A. B. 1969.Diss. Abstr. Int. B 30, 2047B. Aldana, A. B.,Fites, R. C., and Pattee, H. E. 1972. Plant Cell Physiol. 13, 515-521. Anderson, G.D. 1970. Exp. Agr. 6, 213-222.

GRAIN LEGUMES OF THE LOWLAND TROPICS

119

Anzai, T., Shibaoka, H., and Simokoriyama, M. 1971. Plant Cell Physiol. 12, 695-700. Arora, S. K., and Luthra, Y. P. 1971a. Plant Soil 34, 91-96. Arora. S. K., and Luthra, Y. P. 1971b. Plant Soil 34, 283-291. Ashri, A. 1964. Genetics 50, 363-372. Ashri, A. 1968. Genetics 60, 807-810. Ashri, A. 1969. Res. Rep. Hebrew Univ. Jerusalem, Sci., Agr. 1, 1 . Ashri, A. 1972. In “Proceedings of the Latin American Study Group on Induced Mutations and Plant Improvement,” pp. 253-264. Asoka-Raj, P. C. 1969. Sci. Cult. 35, 408-409. Aukland, A. K. 1966. J . Agr. Sci. 67, 109-119. Aukland, A. K. 1967. 1. Agr. Sci. 69, 455-464. Aulakh, K. S., and Sandhu, R. S. 1970. Plant Dis. Rep. 54, 337-338. Barker, L. N. 1970. “A Review of Cowpea Breeding at the University of Ife ( 1966-1970) Paper presented to Ford Foundation/IRAT/IITA Seminar IV. Grain Legume Research in West Africa, University of Ibadan, Nigeria. Behl, N. K., Sawhney, J. S., and Moolani, M. K. 1969. J. lies. Punjab Agr. Univ. 6, 608-613. Beohar, A. B. L., and Nigam, P. K. 1972. JNKVV Res. 1.6, 37. Bhatawadekar, P. U., Chiney, S. S., and Deshmukh, K. M. 1966. Indian 1. Agron. 11, 243-246. Bhatnager, P. S., Sengupta, P. K., Gangwar, L. C., Saxena, J. K., and Kumar, V. 1967. Sci. Cult. 33, 120-121. Bhattacharjee, S. K. 1956. Caryologia 9, 149-159. Bilquez, A. F., Magne, C., and Martin, J . P. 1964. “Bilan de six ann6es de recherches sur l’emploi des rayonnements ionisants pour l’amtlioration des plantes au Senegal,” pp. 586-601. Meeting, F A 0 and IAEA, Rome. Bindra, 0. S., and Jakhmola, S. S. 1967. Indian J . Agr. Sci. 37, 177-186. Bliss, F. A., and Robertson, D. G. 1971. Crop Sci. 11, 258-262. Blondel, D. 1969. Agron. Trop. (Paris) 22, 864-887. Bolhuis, G. G. 1951. Landbouwk. Tijdschr. 63, 447-455. Bolhuis, G. G. 1955. Oleagineux 10, 157-160. Bolhuis, G. G., and De Groot, W. 1959. Neth. 1. Agr. Sci. 7, 317-326. Bouhot, D. 1967. Agron. Trop. (Paris) 22, 864-887. Bouhot, D. 1968. Agron. Trop. (Paris) 23, 1126-1127. Boyd, H. W. 1971. Plant Soil 35, 133-144. Braham, E. J., Maddaleno Vela, R., Bressani, R., and Jarquin, R. 1965. Arch. Vcnez. Nutr. 15, 19-32. Brantley, B. B., and Kuhn, C. W. 1970. J . Amer. SOC. Hart Sci. 95, 155-158. Brar, S. S., and Elbein, A. D. 1971. Phytochernistry 10, 2099-2102. Brar, S. S., and Elbein, A. D. 1972. Phytochernistry 11, 651-656. Bray, M. 1970. Agron. Trop. (Paris) 25, 192-204. Brockwell, J. 1971. Plant Soil 35, 265-272. Bromfield, A. R. 1973. Exp. Agr. 9, 55-58. Bromfield, K. R., and Bailey, W. K. 1972. Phytopathology 62, 748. Bromfield, K. R., and Cevario, S. J. 1970. Plant Dis. Rep. 54, 381-383. Brzozowska, J., and Hanower, P. 1964. Oleagineux 19,663-672. Bunting, A. H. 1955. Emp. J . Exp. Agr. 23, 158-170. Bunting, A. H. 1958. Emp. J . Exp. Agr. 26, 254-258.

.”

120

K. 0. RACHIE AND L. M. ROBERTS

Burkhill, I. H. 1967. “Dictionary of the Economic Products of the Malay Peninsula,” 2nd ed. Ministry of Agriculture and Cooperatives, Kuala Lumpur. Busson, F., Toury, I., and Bergeret, B. 1958. “Composition en acides amines des graines des principales legumineuses tropicales utiliskes dans l’alirnentation,” Doc. No. 5. Laboratoire de Recherches des Bernice de Sante de la France d’Outre Mer, Marseilles et Organisme de Recherche sur I’Alimentation et la Nutrition Africaines, Dakar. Carriere de Belgarric, R., and Bour, F. 1963. Agron. Trop. (Paris) 18, 863-875. Caswell, G. H. 1968. Proc. Agr. SOC.Nigeria 5, 4-6. Catherinet, M. 1956. France. Dir. Agr. Elevage Forets, Off. Rech. Sci. Tech. OutreMer Bull. Agron. 16, 9398. Cekalin, N. M., and Zelenskaja, L. A. 1970. Rep. All Union Lenin Acad. Agr. Sci. 9, 13-14. Chandola, R. P., Trehan, K. B., and Bagrecha, L. R. 1969. Curr. Sci. 38, 370-371. Chandra, G. R., Gregory, L. E., and Worley, J. F. 1971. Plant Cell Physiol. 12, 3 17-3 25. Chapman, L. S., and Garioch, D., 1966. Cane Grow. Quart. Bull. 30, 57-59. Chin, T. Y., Meyer, M. M., and Beevers, L. 1969. Planta 88, 192-196. Chopra, A. K., and Bhatia, I. S. 1970. 1. Res. Punjab Agr. Univ. 7 , 69-74. Chaudhari, B. B., and Patil, J. A. 1953. Curr. Sci. 22, 153. Choudhary, M. J. 1969. Labdev, Part B 7, 200-201. Chowdhury, J. B., Kumar, R., Bhat, P. N., and Kakar, S. N. 1968. J . Res. Punjab Agr. Univ. 5 , 4 6 3 4 7 0 . Chowdhury, R. K., Chowdhury, J. B., Bhat, P. N., and Kakar, S. N. 1969. J . Res. Punjab Agr. Univ. 6, 867-874. Chowdhury, S. L., and Bhatia, P. C. 1971a. Indian Farming 20, 27-30. Chowdhury, S. L., and Bhatia, P. C. 1971b. Indian Farming 21, 18-19. Clark, A. F., and Villemez, C. L. 1972. Plant Physiol. 50, 371-374. Cook, M. 1972. Plant Dis. Rep. 56, 382-386. Corbett, D. C. M., and Brown, P. 1966. Rhodesia, Zambia, Malawi J . Agr. Res. 4, 13-21. Culp, T. W., Bailey, W. K., and Hammons, R. 0. 1968. Crop Sci. 8, 109-1 1 1 . Dafe, G . A. 1966. Ann. Agr. Res. Abstr. PG-Res. Work, Nagpur Coll. Agr. Mag., Spec. Res., p. 107. Dancette, C. 1970. Agron. Trop. (Paris) 25, 225-240. Daniel, C., and De Berchoux, C. 1965. Oleagineux 20, 373-376 and 4 4 1 4 4 6 . Dart, R. J., and Mercer, F. V. 1965. Aust. J . Agr. Res. 16, 321-345. Dave, B. B. 1934. Indian I . Agr. Sci. 4, 674-691. Delassus, M. 1970. “Crop Protection: Major Problems Observed and Studies on Edible Legumes in Various French Speaking Countries.” Paper presented to Ford Foundation/IRAT/IITA Seminar IV. Grain Legume Research in West Africa, University of Ibadan, Nigeria. Delmer, D. P., and Albersheim, P. 1970. Plant Physiol. 45, 782-786. Deodikar, G. B., and Thakar, C. V. 1956. Proc. Indian Acad. Sci., Sect. B 43, 37-45. Derieux, M. 1971. Ann. Amelior. Plant. 21, 373-407. Deshmukh, N. Y.,and Phirke, T. S. 1962. Nagpur COIL. Agr. Mag. 36,46. Deshmukh, N. Y., and Rekhi, S. S. 1960. Curr. Sci. 29, 237-239.

GRAIN LEGUMES OF THE LOWLAND TROPICS

121

Deshrnukh, N. Y., and Rekhi, S. S. 1961. Indian J . Agr. Sci. 31, Suppl., 24-26. Deshmukh, N. Y., and Rekhi, S. S. 1962. Proc. Bihar Acad Agr. Sci. 8-9, 135-139. Deshpande, A. M., and Bathkal, B. G. 1965. Indian J . Agron. 10, 271-278. Deshpande. R. B., and Jeswani, L. M. 1952. Indian J. Genet. Plant Breed. 12, 50-51. Deshpande, R. B., and Jeswani, L. M., 1956. Curr. Sci. 25,201-202. DeZeeuw, D. J., and Ballard, J. C. 1958. Plant Dis. Rep. 42, 898. DeZeeuw, D. J., and BaIlard, J. C. 1959. Phytopathology 49, 332-334. DeZeeuw, D. J., and Crum, R. A. 1963. Phytopathology 53,337-340. Dhery, M., and Gillier, P. 1971. Oleagineux 26, 243-251. Dodge, A. D., Alexander, D. J., and Blackwood, G. C. 1971. Physiol. Plant 25, 71-74. Doku, E. V. 1968. Exp. Agr. 4 , 4 1 4 8 . Doku, E. V. 1969. Ghana J . Agr. Sci. 2, 91-95. Doku, E. V. 1970. Ghana 1. Agr. Sci. 3, 145-149. Doku, E. V., and Karikari, S. K. 1970a. Ghana J . Agr. Sci. 3, 17-26. Doku, E. V., and Karikari, S. K. 1970b. Ghana J . Sci. 11, 47-56. Dorosinskii, L. M., and Lazareva, N. M. 1967. “Biologicheskiy Azot i Ego Rol‘v Zernledelii” (Symbiotically Fixed Nitrogen and its Role in Agriculture). Institute of Microbiology, Academy of Sciences, U.S.S.R., Moscow. Dougherty, R. H., and Cobb, W. Y. 1970. J. Agr. Food Chem. 18, 921-925. Ebong, U. S. 1970a. “On the Nigerian Grain Legume Gene Bank.” Paper presented to Ford Foundation/IRAT/IITA Seminar IV. Grain Legume Research in West Africa, University of Ibadan, Nigeria. Ebong, U. S. 1970b. “Grain Legume Improvement in West Africa.” Paper presented to Ford Foundation/IRAT/IITA Seminar IV. Grain Legume Research in %st Africa, University of Ibadan, Nigeria. El-Murabaa, A., and Mustafa, I. 1970. Alexandria 1. Agr. Res. 16, 39-51. Ernpig, L. T., Lantican, R. M., and Escuro, P. B. 1970. Crop Sci. 10, 240-241. Epps, J. M., and Chambers, A. Y . 1959. Plant Dis. Rep., 43, 981-982. Ezedinma, F. 0. C. 1964. Trop. Agr. (London) 41, 243-249. Fennel], J. L. 1948. J . Hered. 39, 275-279. FAO, 1966. “Agricultural Development in Nigeria, 1965-1980,” pp. 392-400. Food and Agriculture Organization of the United Nations, Rome. FAO. 1972. “Production Yearbook,” Vol. 25, pp. 175-176. Food and Agriculture Organization of the United Nations, Rome. Forestier, J. 1969. Cah. ORSTOM (Off. Rech. Sci. Tech. Outre-Mer) Ser. Biol. 9, 33-63. Fowler, A. M. 1970. Samaru Agr. Newsl. 12, 66-69. Fowler, A. M. 1971. Samaru Agr. Newsl. 13, 11-13. Frank, Z. R., and Krikun, J. 1969. Plant Dis. Rep. 53, 744-746. Franz, G., 1972. Planta 102, 334-347. Ganguli, D. K., and Srivastava, D. P. 1967. Ranchi 1. Agr. Res. 2, 23-25. Garren, K. H. 1964. Phytopathology 54, 277-281. Gaur, A. C., and Pareek, R. P., 1969. Experientia 25,777. Gautreau, J. 1969. Oteagineux 24, 339-342. Gavrielit-Gelmond, H. 1970. Proc. Int. Seed Test. Ass. 36, 159-171. Gay, J. D. 1970. Plant Dis. Rep. 54, 604-605.

122

K. 0. RACHIE AND L. M. ROBERTS

Gay, J. D. 1971. Plant Dis. Rep. 55,384-386. Gentner, W. A., and Danielson, L. I. 1965. Proc. Amer. SOC. Hort. Sci. 87, 359-362. Gibbons, R. W., and Bailey, B. E. 1967. Rhodesia, Zambia, Malawi J . Agr. Res. 5, 57-59. Gibbons, R. W., and Tattersfield, J. R. 1969. Rhodesian J . Agr. Res. 7 , 71-85. Gibbons, R. W., Bunting, A. H., and Smartt, J. 1972. Euphytica 21, 78-85. Gilchrist, D. G., Woodin, T. S., Johnson, M. L., and Kosuge, T. 1972. Plant Physiol. 49, 52-57. Gillier, P. 1966. C . R . Acad. Agr. Fr. 52,446-449. Gillier, P. 1969. Oleagineux 24, 79-81. Gillier, P. 1970. “L’arachide de bouche au Senegal.” Paper presented to Ford Foundation/IRAT/IITA Seminar IV. Grain Legume Research in West Africa, University of Ibadan, Nigeria. Gillier, P., and Silvestre, P. 1969. “L’arachide.” G. P. Maisonneuve et Lgrose, Paris. Goldin, E. 1970. World Crops 22, 220-224. Goldsworthy, P. R., and Heathcote, R. G. 1964. Emp. J . Exp. Agr. 31, 351-366. Gonzales, T. T., 1962. Philipp. J . Agr. 27, 131-143. Gonzalez, 0. N., Banzon, E. A., Liggayu, R. G., and Qunitio, P. H. 1964. Phillipp. J . Sci. 93, 47-56. Gooding, H. J., 1962. Field Crop Abstr. 15, 1-5. Gopal, N. H. 1970. Indian J . Plant Physiol. 13, 90-98. Gopal, N. H. 1971. fndian J . Exp. Biol. 9, 524-526. Gopalswamy, A., and Veerannah, L. 1968. Madras Agr. J . 55, 209-211. Govindaswamy, C. V., Mariappan, V., Murugesan, S. S.,Padmanabhan, C., Thangamani, G., and Janaki, I. P. 1970. Madras Agr. J . 57, 405-414. Gregory, W. C. 1956. Brookhaven Symp. Biol. 9, 177-190. Gregory, W. C., Smith, B. W., and Yarbrough, J. A. 1951. In “The Peanut, the Unpredictable Legume, a Symposium,” pp. 28-88. National Fertilizer Association, Washington, D.C. Grimes, W. J. 1969. Diss. Abstr. Int. B 30, 2579. Grubben, G. J. H. 1970. “Observations sur les tuteurs vivants les haies et les plantes grimpantes dans les clotures au Sud-Dahomey.” Commun. Found. NEDERF, Amsterdam. Gupta, 0. P., and Mani, V. S. 1964. Indian J . Agron. 9, 98-99. Gupton, C. L., and Emery, D. A. 1970. Crop Sci. 10, 127-129. Gvozdeva, Z. V., and Zhukova, N. V. 1971. Tr. Prikl. Bot., Genet. Selek. 45, 161-168. Hahlbrock, K., Wong, E., Schill, L., and Grisebach, H. 1970. Phytochemistry 9, 949-958. Hammerton, J. L. 1971. Trop Agr. (London) 48, 341-350. Hanower, P. 1969. “Repercussion de la deficience en soufre sur certains aspects du metabolisme de l’azote chez I’arachide,” Thesis, Natur. Sci. ORSTROM, Paris. Hare, W. W. 1959. Phytopathology 49, 318. Harris, H. C., and Brolmann, J. B. 1966. Agron. J . 58, 575-582. Harrison, A. N., and Gudauskas, R. T. 1968. Plant Dis. Rep. 52, 34-36. Hartzook, A., Fichman, M., and Karstadt, D. 1971. Oleagineux 26, 391-395. Hartzook, A., Karstadt, D., and Feldman, S. 1972. SABRA0 Newsl. 4, 91-94.

GRAIN LEGUMES OF THE LOWLAND TROPICS

123

Harvey, D. 1970. “Tables of the Amino Acids in Food and Feeding Stuffs,” 2nd ed., pp. 65-75. Commonwealth Bureau of Animal Nutrition, Aberdeen. Hawthorne, P. L. 1943. Proc. Amer. SOC.Hort. Sci. 42, 562-564. Heath, M. C. 1971. Phytopathology 61, 383-388. Heinis, I. L. 1972. Oleagineux 27, 147-152. Heller, I. S., and Villemez, C. L. 1972. Biocliern. J . 128, 243-252. Heuser, C. W., and Hess, C. E. 1972. J . Amer. SOC.H o r f . Sci. 97, 392-396. Holley, K. T., and Hammons, R. 0. 1968. “Strain and Seasonal Effects on Peanut Characteristics,” Res. Bull. No. 32. University of Georgia, Athens. Hutchinson, I., and Dalziel, I. M. 1958. “Flora of West Tropical Africa,” Vol. I, Part 2, pp. 505-587. Whitefriars Press, Ltd., London and Tonbridge. Huxley, P. A., and Summerfield, R. J. 1973. Ann. AppL Bot. (in press). Hymowitz, T., CoIlins, J., Panczner, J., and Walker, W. M. 1972. Crop Sci. 12, 7 10-7 1 1 . IARI. 1970. “A New Technology for Dry Land Farming.” Indian Agricultural Research Institute, New Delhi. IARI. 197 1. “New Vistas in Pulse Production,” Bull. Indian Agricultural Research Institute, New Delhi. ICAR. 1952. “Annual Report of the Indian Council of Agricultural Research for 1950-1951.” ICAR, New Delhi. IITA. 1973. “Grain Legume Improvement Program of the International Institute of Tropical Agriculture,” Published reports of the IITA, 1971 and 1972. Ibadan, Nigeria. IITA. 1974. “Grain Legume Improvement Program of the International Institute of Tropical Agriculture.” Published report of the IITA, 1973. Ikeda, C. J. 1968. Diss. Abstr. B 28, 2749. Ilyana, A. I. 1959. Oleagineux 14, 89-92. IRAT. 1969. Cah. Agr. Prat. Pays Chauds 3, 127-129. IRAT. 1971. “Institut de Recherches Agronomiques Tropicales ed des Cultures Vivrikres,” Rapports Annuels. IRAT, Paris. IRHO. 1972. “Institut de Recherches pour les Huiles et Oleagineux,” Rapports annuels. IRHO, Paris. Ishag, H. M. 1971. J . Agr. Sci. 77, 237-242. Ishii, Y. 1968. Jap. J . Trop. Agr. 12, 4-8. Ishii, Y. 1969. Jap. J . Trop. Agr. 12, 158-163. Iswaran, V., Kavimandan, S . K., and Kamath, M. B. 1969. Curr. Sci. 38, 251-252. Iswaran, V., Sen, A., and Singh, N. P. 1972. Plant Soil 36 (31, 717-718. Ivanoff, S. S. 1962. Miss.Farm Res. 25, 7-8. Ivanoff, S . S., 1963. Miss. Farm Res. 26, 8. Jackson, M. B,, and Harvey, P. M. 1970. Can. J . Bot. 48, 943-946. Jacquinot, L. 1967. Agron. Trop. (Paris) 22, 575-640. Jaffe, M. J. 1970. Plant Physiol. 46,768-771. Jain, S . K., and Joshi, B. C. 1964. Genetics 49, 611-615. Jakhmola, S. S., and Singh, L. 1971. JNKVV Res. J . 5, 104-106. Janoria, M. P., and Ali, M. A. 1970. JNKVV Res. J . 4, 15-19. Jeswani, L. M., and Deshpande, R. B. 1962. Indian J . Genet. Plant Breed. 22, 23 6-240. Jindla, L. N., and Singh, K. B. 1970. Indian J . Hered. 2 , 4 5 4 9 . Johnson, D. T. 1968. Rhodesia Agr. J . 65, 1 - 4 .

124

K. 0. RACHIE AND L. M. ROBERTS

Joshi, B. C., and Ramanujam, S. 1963. Indian J . Genet. Plant Breed 23, 64-66. Joshi, K. S. 1966. Nagpur Coil. Agr. Mag., Spec. Res. p. 11 1. Joshi, S. N. 1969. Zndian J . Agric. Sci. 39, 1010-1012. Kandaswamy, T. K., and Ramakrishnan, K. 1960. Madras Agr. J . 47, 440-441. Kasai, T., Sakamura, S., and Sakamoto, R. 1971. Agr. Biol. Chem. 35, 1603-1606. Katayama, M., and Funahashi, S. 1969. J . Biochem. [Tokyo) 66, 479-485. Kathju, S., and Tewari, M. N. 1968. Curr. Sci. 37, 83-84. Kaul, C. L. 1970. Genetica 41, 316-320. Kaul, C. L., and Singh, S. P. 1967. Indian J . Agr. Sci. 37, 69-76. Kelkar, S. G., and Pandya, P. S. 1934. Poona Agr. Coil. Mag. 26, 108-111. Ketring, D. L., and Morgan, P. W. 1970. Plant Physiof. 45, 268-272. Khan, A. A., and Akosu, F. 1. 1971. Physiol. Plant. 24, 471-475. Khan, A. R., and Mathur, B. P. 1962. Indian J . Agr. Sci. 32, 35-38. Khandelwal, K. C., and Gaur, A. C. 1970. Sci. Cult. 36, 110-111. Khare, R. N., Krishnamurthy, K., and Pingale, S. V. 1966. Bull. Grain Techno[. (India) 4, 125-132. Khatri, H. L., and Chenulu, V. V. 1971. Abstr. Pap., Znt. Symp. Plant Pathol., 2nd, 1971, pp. 64-65. Kheradnam, M., and Niknejad, M. 1971. Z. Pflanzenzuecht. 66, 312-316. Klesser, P. J. 1967. S. A fr . J . Agr. Sci. 9, 731-733. Knapp, R. 1966. Angew. Bot. 39,230-241. Kobayashi, K., and Yamaki, T. 1972. Plant Cell Physiol. 13, 49-65. Kohli, K. S., Singh, C. B., Singh, A., Mehra, K. L., and Magoon, M. L. 1971. Genet. Agr. 25, 231-242. Kolhe, A. K. 1970. Poona Agr. Coil. Mag. 59, 126-127. Korytkowski, C., and Torres, M. 1966. Rev. Peruana Entomol. 9, 3-9. Kousalya, G., Ayyavoo, R., Krishnnamurthy, C. S., Kandaswamy, T. K., and Bhaskaran, S. 1971. Madras Agr. J . 58, 495-505. Kovarskii, A. E. 1939. Selek. Semenovod. (Moscow) 10-11, 9-13. Krapovickas, A. 1968. Actas Mem. Congr. Int. Amer., 37th Vol. 2, pp. 517-534. Krapovickas, A., and Rigoni, V. A. 1960. Rev. Invest. Agr. 14, 198-228. Krishnamoorthy, H. N. 1970. Plant Cell Physiol. 11, 979-982. Kuhn, C. W. 1965. Phytopathology 55, 880-884. Kuhn, C. W., and Brantley, B. B. 1963. Plant Dis. Rep. 47, 1094-1096. Kulimbetov, T. D. 1968. Dokl. Akad. Nauk Uzb. SSR 3, 50-51. Kulkarni, L. G. 1967. Indian Farming 17, 4-7. Kumar, L. S. S. 1945. Proc. Indian Acad. Sci., Sect. B 21, 266-268. Kumar, L. S. S., Thombre, M. V. S., and D’Cruz, R. 1958. Proc. Indian Acad. Sci., Sect. B 47, 252-262. Kumar, S. M., Khare, M. N., and Shrivastavia, S. K. 1969. Mysore J . Agr. Sci. 3, 472-474. Kurien, P. P., and Parpia, H. A. B. 1968. J. Food Sci. Technol. 5, 203-207. Leakey, C. L. A. 1970. “Brief Summary of Work on Soyabeans in East Africa.” Paper presented to Ford Foundation/IRAT/IITA Seminar IV. Grain Legume Research in West Africa, University of Ibadan, Nigeria. Lee, T. A., Jr., Ketring, D. L., and Powell, L. D. 1972. Plant Physiol. 49, 190-193. Lefebvre, C. L., and Sherwin, H. S. 1950. Phytopathology 40, 17-18. Leuck, D. B., and Harvey, J. E. 1968. J . Econ. Entomol. 61, 583-585.

GRAIN LEGUMES OF THE LOWLAND TROPICS

125

Leuck, D. B., and Skinner, J. L. 1971. J. Econ. Entomol. 64, 148-150. Lewin, H. D., and Natarajan, S. 1971. Madras Agr. J . 58, 395-404. Liener, I. E. 1969. “Toxic Constituents of Plant Foodstuffs,” Chapters 1-3. Academic Press, New York. Lin, H., and Chen, C. C. 1970. Taiwan Agr. Quart. 6, 56-65. Lin, H., Chen, C. C., and Lin, C.Y . 1969. Taiwan Agr. Quart. 5, 45-64. Lorz, A. P. 1970. Flu., Agr. Exp. Sta., S-210. McDonald, D., 1970a. “Field Trials for Control of Cercospora Leaf-spot Disease on Groundnut.” Paper presented to Ford Foundation/IRAT/IITA Seminar IV. Grain Legume Research in West Africa, University of Ibadan, Nigeria. McDonald, D. 1970b. “Survey of Cowpea Market Samples for Seed Borne Fungi.” Paper presented to Ford Foundation/IRAT/IITA Seminar IV. Grain Legume Research in West Africa, University of Ibadan, Nigeria. McKenzie, D. R. 1973. Personal communication concerning the mung bean collection assembled at Asian Vegetable Research and Development Center (Taiwan). Ibadan, Nigeria (IITA). Mackie, W. W. 1934. Phytopathology 24, 1135. Mackie, W. W. 1939. Phytopathofogy 29, 826. Mackie, W. W. 1943. Hilgardia 15, 1-29. MAES. 1959. Mississippi Agric. Exp. Sta. Annu. Rep. 72. MAES. 1961. Mississippi Agric. Exp. Sfa. Annu. Rep. 74. MAES. 1963. Mississippi Agric. Exp. Sta. Annu. Rep. 75. Majumdar, A. N. L., Mandal, N. C., and Biswas, B. B. 1972. Phytochemistry 11, 503-508. Majumdar, P. K., Prakash, R., and Haque, F. 1969. Indian J . Genet. Plant Breed. 29, 291-296. Malhotra, 0. P., and Rani, I. 1969. Indian J . Biochem. 6, 15-20. Mandal, N. C., and Biswas, B. B. 1970. Plant Physiol. 45, 4-7. Mandal, N. C., Burman, S., and Biswas, B. B. 1972. Phyfochemistry 11, 495-502. Mandloi, K. K., and Tiwari, K. P. 1971. JNKVV Res. J . 5 , 54-55. Mantez, A., and Goldin, E. 1964. Isr. J . Agr. Res. 14, 103-210. Marquette, J. 1970. “Report on Grain Legumes in Madagascar.” Paper presented to Ford Foundation/IRAT/IITA Seminar IV. Grain Legume Research in West Africa, University of Ibadan, Nigeria. Martin, G. 1965. Oleagirieux 20, 241-242. Martin, J. H., and Leonard, W. H. 1967. “Principles of Field Crop Production,” 2nd ed., pp. 643-704. Macmillan, New York. Martin, J. P. 1969. Cah. ORSTOM (Off. Rech. Sci. Tech. Outre-Mer) 7, 3-35. Masefield, G. B. 1957. Emp. J. Exp. Agr. 25, 139-150. Masefield, G. B. 1961. Trop. Agr. (London) 38, 225-229. Mathur, S. B., Singh, A., and Joshi, L. M. 1967. Plant Dis. Rep 51, 649-651. Mauboussin, J. C. 1966. “Amelioration de I’arachide. Etudes en cours sur la dormance.” Reunion, Samaru. CRA, Bambey. Mauboussin, J. C. 1968. “Etudes sur le taux d’allogamie chez I’arachide” (non publii), Annual Report. IRAT, Senegal. Mauboussin, J. C. 1970. “Groundnut Selection at the IRAT/Senegal.” Paper presented to Ford Foundation/IRAT/IITA Seminar IV. Grain Legume Research in West Africa, University of Ibadan, Nigeria.

126

K. 0. RACHIE AND L. M. ROBERTS

Mazzani, B., and Allievi, J. 1971. Agron. Trop (Maracay, Venez.) 21, 329-332. Medvedev, P. F. 1949. Selek Sernenovod. (Moscow) 16, 41-43. Mehra, K. L., Singh, C. B., and Kohli, K. S. 1969. Indian J . Hered. 1, 53-58. Mehrotra, C. L., and Gangwar, B. R. 1964. 1. Indian SOC.Soil Sci. 12, 75-83. Mehrotra, 0. N., Saxena, H. K., Roy, A. N., and Nath, S. 1968. Exp. Agr. 4, 3 39-344. Mehta, B. V., and Khatri, P. D. 1962. J . Maharaja Sayajirao Univ. Baroda 11, 109-122. Mehta, P. N. 1971. Tech. Comm. Int. SOC.Hort. Sci. 21, 167-171. Menezes, 0. B. de. 1953. Rev. Agr. Sao Paolo 38, 281-284. Merchant, N. M., and Munshi, Z. A. 1971. We$t Pak. J . Agr. Res. 9, 34-41. Miller, C. R., Dickson, D. W., and Lipscomb, R. W. 1970. Soil. Crop Sci. SOC. Flu., Proc. 29, 273-277. Miller, L. T., and Troutman, J. L. 1966. Plant Dis. Rep. 50, 139-143. Minamikawa, T., Oyama, I., and Yoshida, S. 1968. Shokubutsugaku Zasshi 81, 13 5-140. Minton, N. A., Hammons, R. O., and Parham, S. A. 1970. Phytopathology 60, 472-474. Mishra, D., and Mohanty, S. K. 1966. Curr. Sci. 35, 340-341. Mishra, R. P., Khare, M. N., and Chand, J. 1971. Indian Phytopathol. 24, 609-611. Misra, R. C., and Sahu, R. C. 1970. Indian J . Agr. Sci. 40, 216-226. Mixon, A. C., and Rogers, K. M. 1973. Oleagineux 28,85-86. Mohan, J. C. 1970. Indian Farmizg 19, 9-11. Mohta, N. K., and De, R. 1971. Indian Farming 20,27-28. Montgomery, R. D. 1964. West Indian Med. J . 13, 1-1 1. Morohashi, Y., and Shimokoriyama, M. 1972a. J . Exp. Bot. 23, 45-53. Morohashi, Y., and Shimokoriyama, M. 1972b. J . Exp. Bot. 23, 54-61. Mukherjee, D., 1960. Indian J . Agr. Sci. 30, 177-184. Mukherjee, P. 1968. Sci. Cult. 34, 252-253, Muiioz, A. A., and Abrams, R. 1971. 1. Agr. Univ. P . R . 55, 23-43. Murray, S. A., and Newcombe, C. L. 1970. Radiat. Bot. 10, 563-567. Nabos, J. 1970. “Grain Legumes in Niger-Present State of the Research.” Paper presented to Ford Foundation/IRAT/IITA Seminar IV. Grain Legume Research in West Africa, University of Ibadan, Nigeria. Nair, K. S., Ramaswamy, P. P., and Perumal, R. 1970. Madras Agr. J . 57, 307-310. Nair, K. S., Ramawamy, P. P., and Perumal, R. 1971. Madras Agr. J . 58, 5-8. Narayana, H. S., and Gothwal, B. D. 1964. Proc. Indian Acad. Sci., Sect. B 59, 350-359. Narayanaswarny, P., and Ramakrishnan, K., 1966. Proc. Indian Acad. Sci., Sect. B 63, 288-296. Narsaiah, D. B., Kulkarni, L. G., and Rao, K. A. 1969. Sci. Cult. 39, 699-700. Natribhop, S., Lochaiyukoi, P., and Beech, D. F. 1972. “Thai-Australian Chao Phya Research Project. 1st Report to the Ministry of Agriculture of the Kingdom of Thailand,” Part B, pp. 73-75. NCSC (North Carolina State College). 1968. Seed World 103, 22. NFDAR. 1967. “Preliminary Green House Studies of Some of the Effect of Day-

GRAIN LEGUMES OF THE LOWLAND TROPICS

127

length on the Morphology and Development of Three Varieties of Cowpea (Vigna sp.),” No. 84. Mem. Dep. Agr. Res., Ibadan. Nichols, R., 1965. Ptant Soil 22, 112-116. Nicol, B. M. 1959a. Brit. J . Nutr. 13, 293-306. Nicol, B. M. 1959b. Brit. I . Nutr. 13, 307-320. Nutman, P. S. 1971. Sci. Progr. (London) 59, 55-74. Ochs, R., and Wormer, T. H. M. 1959. Oleagineux 14, 281-291. Ogle, W. L. 1967. Proc. Amer. Soc. Hort. Sci. 90, 290-295. Ojehomon, 0. 0. 1968a. J . West A f r . Sci. Ass. 13, 227-234. Ojehomon, 0. 0. 1968b. In “The Physiological Basis of Varietal Yield Differences in Cowpea (Vigna unguiculata (L) Walp.),” p. 44, Agr. SOC. Nigeria, 4th Conf., Zaria. Ojehomon, 0. 0. 1970. J . Agr. Sci. 74, 375-381. Ojomo, 0. A. 1971. Trop. Agr. (London) 48, 279-282. Ojomo, 0. A. 1972. J. Hered. 63, 147-149. Ojomo, 0. A., and Chhedda, H. R. 1971. Radiat. Bot. 11, 375-381. Oke, 0. L. 1967. Exp. Agr. 3, 315-321. Oke, 0. L. 1969.Exp. Agr. 5 , 111-116. Ong, B. L., and Jackson, J. F. 1970. “Enzymes of DNA Synthesis in the Developing Radical of Mung Bean, Phaseolus aureus,” Waite Agr. Res. Inst. Biennial Rep., 1968-1969. University of Adelaide, Australia. Ong, B. L., and Jackson, J. F. 1972a. Biochem. J . 129, 571-581. Ong, B. L., and Jackson, J. F. 197213. Biochem. J. 129, 583-593. Ono, Y., and Ozaki, K. 1971. Proc. Crop Sci. SOC.Jap. 40, 480-485. Pandey, R. M. 1968. Flora (Jena), Abt. A 159, 299-306. Pant, R., and Tulsiani, D. R. P. 1968. Curr. Sci. 37, 74-75. Parekh, L. J., Anthony, A., and Ramakrishnan, C. V. 1969. Phytochemistry 8, 1167-1172. Pareek, R. P., and Gaur, A. C. 1970. Plant Soil 33, 297-304. Parker, R. C., Wynne, J. C., and Emery, D. A. 1970. Crop Sci. 10,429-432. Pathak, G. N., and Singh, B. 1963. Indian J. Genet. Plunt Breed. 23, 215-218. Pathak, G. N., and Yadava, R. S. 1951. Curr. Sci. 20, 304. Pathak, R. D., Ghosh, T. K., and Srivastava, V. C. 1968. Ranchi Univ. J . Agr. Res. 3, 4-6. Patil, J. A. 1965. Indian J . Genet. Plant Breed, 25, 234-238. Patil, J. A., and D’Cruz, R. 1962. Curr. Sci. 31, 119-120. Patil, J. A., and Sheikh, K. N. 1957. Curr. Sci. 26, 253-254. Patil, S. H. 1968. Can. J . Genet. Cytol. 10, 545-560. Patro, G. K., Mishra, A., and Tosh, G. C. 1970. Indian J . Agr. Sci. 40, 626-629. Pattee, H. E., Purcell, A. E., and Johns, E. B. 1969. J . Amer. Oil Chem. SOC. 46, 629-631. Paul, A. K., Mukherji, S., and Sircar, S. M. 1970. Oesierr. Bot. Z. 118, 311-320. Philpotts, H. 1967. Aust. J . Exp. Agr. Anim. Hiisb. 7, 371-376. Phirke, T. S. 1966. Nagpur Coll. Agr. Mag., Spec. Res. p. 101. Pietri, R., Abrams, R., and Julia, F. J. 1971. J. Agr. Univ.P . R . 55, 474-477. Pillai, R. N., and Sen, A. 1970. Indian J . Agr. Sci. 40, 1017-1018. Platt, B. S. 1962. Med. Res. Counc. (Gt. Brit.), Spec. Rep. Ser. 20.

128

K. 0. RACHIE AND L. M. ROBERTS

Poehlman, J. M., and Yu-Jean, F. M. M. 1972. Mo., Agr. Exp. Eta., Res. Bull. 991.

Porter, D. M. 1970. Plant Dis. Rep. 54, 955-958. Pospisil, F., Karikari, S . K., and Boamah-Mensah, B. 1971. World Crops 23, 260-264. Prasad, D. N., and De, D. N. 1971. Microbios 4, 13-20. Prasad, M. K. O., and Narasimhamurthy, B. L. 1963. Andhra Agr. J . 10, 161-167. Prasad, R., Bhendia, M. L., and Bains, S. S. 1968. Indian J . Agron. 13, 305-319. Prasad, S. 1967. Sci. Cult. 33, 66-67. Prasad, S., and Srivastava, D. P. 1968. Ranchi J. Agr. Res. 3, 12-16. PrCvot, P. 1949. “Croissance d6veloppement et nutrition minerale de l’arachide,” Oleagineux, Coloniaux, Ser. Sci. No. 4. SETCO, Paris. PrBvot, P., and Ollagnier, M. 1961. Znt. Hort. Congr., Proc. 15th 1958. pp. 2 17-228. Prevot, P., Ollagnier, M., Aubert, G., and Brugiere, J. M. 1955. Oleagineux 10, 23 9-243. Purseglove, J. W. 1968. “Tropical Crops: Dicotyledons 1,” pp. 199-332. Longmans, Green, New York. Purss, G. S. 1958. Queensl. J . Agr. Sci. 15, 1-14. Rachie, K. 0. 1970. “Progress in Cajanus Investigations at Makerere University (Kampala, Uganda) .” Paper presented to Ford Foundation/IRAT/IITA Seminar IV. Grain Legume Research in West Africa, University of Ibadan, Nigeria. Rachie, K. 0. 1973. “Relative Agronomic Merits of Various Food Legumes for the Lowland Tropics,” pp. 123-139. Potentials of Field Beans and Other Food Legumes in Latin America. Series Seminar No. 2E. pp. 388. CIAT, Cali, Colombia. Rachie, K. O., and Silvestre, P. 1974. “Production and Improvement of Grain Legumes for the Lowland Tropics with Emphasis on West Africa,” Sect. IV, Part 1. Tropical Products Institute, London. Racusen, R., and Miller, K. 1972. Plant Physiol. 49, 654-655. Radley, R. W. 1971. “Soybean Production.” Paper presented to the FAOIUNICEF sponsored Seminar on the Production, Utilization and Marketing of Soybeans, Kampala, Uganda. Raina, J. L., and Modi, V. V. 1969. J. Gen. Microbiol. 57, 125-130. Rajagopalan, C. K., Venugopal, K., and Sheriff, N. M. 1970. Madras Agr. J . 57, 271-273. Ralwani, L. L., Kurar, C. K., and Bagga, R. K. 1970. Zndian J. Agron. 15, 166-168. Ramachandran, M., Loganathan, N. S., Sridharan, C. S., Chandrasekharan, N. R., and Krishnaswami, R. 1967. Indian J . Agr. Sci. 37, 429-436. Raman, V. S. 1973. Oleagineux 28, 137-140. Raman, V. S., and Sree Rangasamy, S. R. 1970. Madras Agr. I . 57, 571-577. Raman, V. S., and Sree Rangasamy, S. R. 1972. Madras Agr. J . 59,262-271. Ramana, K. V. R., and Rao, G. R. 1971. 2.Pflanzenphysiol. 64, 91-92. Ramanathan, T., and Raman, V. S. 1968.1. Indian Bot. SOC.47, 113-116. Ramanujam, S., 1972. Indian Farming 21, 12-14. Ramaswani, P. P., and Nair, K. S. 1965. Madras Agr. J . 52, 239-240. Rao, K. S., and Tulpule, P. G. 1967. Nature (London) 214, 738-739. Rawat, R. R., and Jakhmola, S. S. 1967. Madras Agr. 54, 601-602.

GRAIN LEGUMES OF THE LOWLAND TROPICS

129

Reeder, B. D., Norton, J. D., and Chamblis, 0. L. 1972. I. Amer. Soc. Hort. Sci. 97, 235-237. Rekhi, S. S. 1966. Nagpur Coll. Agr. Mag., Spec. Res., p. 100. Rimikhanov., A. A. 1967. Sb. Tr. Aspir. Molodykh Nauch. Sotrudnikov. Vses. Nauch.-Issled. Inst. Rastenievod. 8, 171-176. Rimikhanov, A. A. 1968. Tr. inst. Genet., Selekc. Azerb. SSR 5, 50-56; Ref. 2 . ( R e f . J . 55, Abstr. 2, 424, 1968) (Russian). Roy, A., and De, D. N. 1965. Sci. Cult. 31, 93-95. Roy, R. S., and Richharia, R. H. 1948. J. Amer. Soc. Agron. 40, 479-489. RPIP. 1967. “Regional Pulse Improvement Project,” Annu. Progr. Rep. No. 5. RPIP, New Delhi, India. RPIP. 1968. “Regional Pulse Improvement Project,” Annu. Progr. Rep. No. 6. RPIP, New Delhi, India. RPIP. 1969. “Regional Pulse Improvement Project,” Annu. Progr. Rep. No. 7. RPIP, New Delhi, India. Sakai, S., and Imaseki, H. 1972. Planta 105, 165-173. Sangha, A. S., and Sandhu, R. S. 1970. J. Res. Punjab Agr. Univ. 7, 143-150. Sanjeeviah, B. S., Krishnamurthy, K., and Hedge, B. R. 1970. Mysore J. Asr. Sci. 4, 466-468. Saunders, A. R. 1952. S.A f r . J . Sci. 48, 195-197. Saunders, A. R. 1959. S.A f r . J . A g r . Sci. 2, 285-307. Saunders, A. R. 1960. S. Afr. I . Agr. Sci. 3, 327-347. Saunders, A. R., and Laubscher, F. X. 1945. Repub. S. Afr., Dep. Agr. Tech. Serv., Sci. Bull. No. 246, p. 57. Sayanova, V . V . 1970. Nauch. Dokl. Vyssh, Shk., Biol. Nauki 5, 82-86. Schenk, R. V. 1971. Ga., Agr. Exp. Sta., Bull. 22. Schmitthenner, A. F., Hoitink, H. A. J., and Kroetz, M. E. 1971. Ohio Rep. Res. Develop. Biol., Agr., Home Econ. 56, 58-60. Sekhon, G. S., Kanwar, J. S., and Bhumbla, D. R. 1966. 1. Res. Punjab Agr. Univ. 3, 350-355. Sellschop, J. P. F. 1962. Field Crop Abst 15, 259-266. Sen, N. K., and Bhowal, J. G. 1961. Genetica 32, 247-266. Sen, N. K., and Bhowal, J. G. 1962. J. Hered. 53, 44-46. Sen, N. K., and Chedda, H. R. 1960. Indian J . Agr. Sci. 30, 250-255. Sen, S., Sur., S. C., and Gupta, K. S. 1966. Zuechfer 36, 379-380. Sen, S., Sur, S. C., and Gupta, K. S. 1968. Theor. A p p l . Genet. 38, 123-125. Sen, S. K., and Sur, S. C. 1964. Agricultura (Louvain) [2] 12, 421-426. S h e , D. 1967. Agron. Trop. (Paris) 22, 309-318. S h e , D. 1968. Agron. Trop. (Paris) 23, 1345-1351. S h e , D., and N’Diaye, S. M. 1970. “Improvement of the Cowpea (Vigna unguiculala) at the CNRA, Bambey.” Paper presented to the Ford Foundation/IRAT/IITA Seminar IV. Grain Legume Research in West Africa, University of Ibadan, Nigeria. Sepswadi, P., and Meksongsee, B. 1971. Kasikorn (Thailand) 44, 289-293. Shanker, H., and Kushwaha, R. P. S. 1971. Indian I . Agr. Res. 5, 79-82. Sharma, B. 1969. Curr. Sci. 38, 520-521. Sharma, D. C. 1966. Phytopathology 57, 127-137.

130

K. 0. RACHIE AND L. M. ROBERTS

Sharma, D. C., Puntamkar, S. S., Mehta, P. C., and Seth, S. P. 1971. Indian J . Agr. Sci. 41, 636-638. Sharma, S. K., and Shinde, V. K. R. 1970. Pest Art. N e w s S u m m . 16, 176-179. Sharon, N., and Lis, H. 1972. Science 177, 949-959. Shaw, F. J. F. 1936. Indian J . Agr. Sci. 6, 139-187. Shchori, Y., and Ashri, A. 1970. Radiat. Bot. 10, 551-555. Shear, G. M., and Miller, L. T. 1955. Agron. J . 47, 354-357. Shinde, V. K., D’Cruz, R., and Deokar, A. B. 1971. Poona Agr. Coll. Mag. 61, 53-55. Shrivastava, M. P., Singh, L., and Joshi, R. K. 1972. JNKVV Res. J . 6, 47-50. Sikdar, A. K., and De, D. N. 1967. Bull. Bot. SOC.BengaZ21,25-28. Silvestre, P. 1970a. “IRAT’s Work on Soybean.” Paper presented to Ford Foundation/IRAT/IITA Seminar IV. Grain Legume Research in West Africa, University of Ibadan, Nigeria. Silvestre, P., 1970b. “IRAT’s Work on Various Food Grain Legumes.” Paper presented to Ford Foundation/IRAT/IITA Seminar IV. Grain Legume Research in West Africa, University of Ibadan, Nigeria. Simbwa-Bunnya, M. 1972. East A f r . Agr. Forest J . 37, 341-343. Sinclair, J. B., and Walker, J. C. 1955. Phytopathology 45, 563-564. Sindagi, S. S., Rajashekhara, B. G., Gowdareddy, B. S., Sanjeeviah, B. S., and K. S. K. Sastry 1972. Mysore J. Agr. Sci. 6, 58-62. Singh, A., and Archana, P. 1964. Proc. Indian Acad. Sci., Sect. B 34, 142-152. Singh, H. B., Mital, S. P.,and Kazim, M. 1968. Indian Hort. 12, 13. Singh, K., and Virk, J. S. 1965. Zndian J . Agron. 10, 50. Singh, K. B., and Jain, R. P. 1970. Indian J . Genet. Plant Breed. 30, 251-260. Singh, K. B., and Malhotra, R. S. 1970a. Madras Agr. J . 57, 155-159. Singh, K. B., and Malhotra, R. S. 1970b. Indian J . Genet. Plant Breed. 30, 244-250. Singh, K. B., and Mehndiratta, P. D. 1969. Indian J . Genet. Plant Breed. 29, 104-109. Singh, K. B., and Mehndiratta, P. D. 1970. Indian J . Genet. Plant Breed. 30, 47 1-475. Singh, K. B., and Singh, J. K. 1970. Indian J . Hered. 2, 61-62. Singh, K. B., and Singh, J. K. 1971. Sci. Cult. 37, 583. Singh, N., Subbiah, B. V., Gupta, Y. P. 1970. Indian J . Agron. 15(1), 24-28. Singh, P., and Choubey, S. D. 1971, Indian Farming 20, 33-34. Singh, S., Singh, H. D., and Sikka, K. C. 1968. Cereal Chem. 45, 13-18. Smartt, J., and Gregory, W. C. 1967. Oleagineux 22, 455-459. Smith, J. C. 1971. J . Econ. Entomol. 64, 280-283. Smith, J. C., and Porter, D. M. 1971.1. Econ. Entomol. 64, 245-246. Solomon, S., Argikar, G. P., Salanki, M. S., and Morbad, I. R. 1957. Indian J . Genet. Plant Breed. 17, 90-95. Spence, J. A., and Williams, S. J. A. 1972. Crop Sci. 12, 121-122. Sreenivas, L., Upadhyay, U. C., and Varokar, R. T. 1968. Indian J . Agron. 13, 137-141. Sreeramulu, N., and Rao, I. M. 1970. Indian J . Agr. Sci. 40, 259-267. Sreeramulu, N., and Rao, I. M. 1971. Aust. J. Bot. 19, 273-280.

GRAIN LEGUMES OF THE LOWLAND TROPICS

131

Srivastava, K. M., Verma, G. S., and Verma, H. N. 1969. Sci. Cult. 35, 475-476. Stanton, W.R. 1966. “Grain Legumes in Africa.” Food and Agriculture Organization of the United Nations, Rome. Strand, A. B. 1943. Proc. Amer. SOC.Hort. Sci. 42, 569-573. Subramanian, S. 1963a. Proc. Indian Acad. Sci., Sect. B 57, 134-148. Subramanian, S. 1963b. Proc. Indian Acad. Sci., Sect. B 57, 178-194. Summerfield, R. J., and Huxley, P. A. 1972. “Management and Plant Husbandry Problems Associated with Growing Soyabean and Cowpea Cultivars under Artificial Conditions,” IITA Int. Commun. No. 3. Reading University, Reading, England. Summerfield, R. J., and Huxley, P. A. 1973. “Daylength and Night Temerature Sensitivity Screening of Selected Cowpea and Soyabean Cultivars,” IITA Int. Commun. No. 5. Reading University Reading, England. Summerfield, R. K., Huxley, P. A., and Steele, W. 1973. Field Crop Abstr. (in press). Tahir, W. W. 1970. “Review of the World’s Plant Protein Resources.” Paper presented to Ford Foundation/IRAT/IITA Seminar IV. Grain Legume Research in West Africa, University of Ibadan, Nigeria. Takyi, S. K. 1971. Ghana I . Agr. Sci. 4, 197-199. Tanada, T. 1972. Nature (London) 236,460-461. Tawde, S. 1961. Ann. Biochem. Exp. Med. 21, 359-366. Tewari, G.P. 1965. Exp. Agr. 1, 257-259. Tewari, M.N. 1968. Curr. Sci. 37,236-237. Tewari, M. N., and Kathju, S. 1969. Indian I . Plant Physiol. 12, 13-19. Thevasagayam, E. S., and Canagasingham, L. S. C. 1960. Trop. Agr. ( C e y l o n ) 116, 287-298. Ting, I. N., and Ho, H. K. 1971. J . Chin. Agr. Chem. SOC.9, 99-109. Todd, J . W., and Canerday, T. D. 1969. J. Econ. Entomol. 61, 1327-1329. Tomar, G. S., Singh, L., and Sharma, D. 1972. S A B R A 0 Newsl. 4, 49-52. Totawat, K. L., and Saxena, S. N. 1971. Plant Soil 34, 43-47. Trehan, K. B., Bagrecha, L. R., and Srivastava, V. K. 1970. Indian J . Hered. 2, 39-43. Truelsen, T. A. 1972. Physiol. Plant. 26, 289-295. Tse, C. C., and Hsueh, Y . S. 1965. Rep. Taiwan Sugar Exp. Station 38, 13-31. Underwood, C. V., Taylor, H. M., and Hoveland, C. S. 1971. Agron. J . 63, 953-954. UPCA (University of the Philippines College of Agriculture). 1968. Mo. Bull. Univ. Philipp. Coll. Agr. 34, 4. Uprety, D. C. 1968. Indian J . Agron. 13, 177-180. USRS. 1963-1964. “Annual Report of the Department of Agriculture,” Vol. 11, pp. 1-12. Serere Res. Sta. (Uganda). Valadon, L. R. G., and Mummery, R. S. 1969. I. Exp. Eot. 10, 732-742. Van Arsedel, E. P., and Harrison, A. L. 1972. Phytopathology 62, 794. Van Emden, J. H. 1960. “Jaarverslag 1959.” Landbouwproef Station, Surinam. Van Emden, J. H. 1962. “Jaarverslag 1961 .” Landbouwproef Station, Surinam. Van Rheenan, K. A. 1972. “Improvement of Soybeans at Mokwa, Nigeria” (personal communication of 6th March and 18th April 1972). IITA, Ibadan, Nigeria.

132

K. 0. RACHIE AND L. M. ROBERTS

Varlet-Grancher, C., and Bonhomme, B. 1972. Ann. Agron. 23, 407-417. Veeraswamy, R., Rajasekaran, V. P. A., Selvakumari, G., and Sheriff, N. M. 1972a. Madras Agr. J . 59, 304-305. Veeraswamy, R., Rathnaswamy, R., Palaniswamy, G. A., and Rajasekaran, V.P.A. 1972b. Madras Agr. J . 59, 252. Ventura, M. M., and Filho, J. X. 1967. An. Acad. Brasil. Cienc. 38, 553-566. Verdcourt, B. 1970. Kew Bull. 24, 507-569. Verhoyen, N. 1960. Parasitica 16, 95-1 17. Verma, P. R., and Patel, P. N. 1969. Zndian Phytopathol. 22, 61-66. Verma, S. N. P. 1969. Indian I . Hered. 1, 105-106. Verma, S. N. P. 1971. Mysore J . Agr. Sci. 5, 477-480. Vessal, M., and Hassid, W. Z. 1972. Plant Physiol. 49, 977-981. Wang, C. C., Chang, S. S., and Tsay, J. S. 1971. J . Agr. Ass. China [N.S.] 73, 49-66. Watt, E. E., Swindell, R. E., Yohe, J. M., Bashandi, M. H. H., Sechler, D. T., Poehlman, J. M. 1973. “Evaluation of Mungbean (Vigna radiata (L) Wilezek.) Strains at Columbia, Missouri, in 1972,” Misc. Publ. No. 73-76. Dept. of Agronomy, University of Missouri, Columbia. Wienk, J. F. 1963. Meded. Landbouwhogesch. Wageningen 63, 82. Wilsie, C. P., and Takahashi, M. 1934. J . Agr. Res. 49, 923-927. Wood, J. M. W. 1970. Aust. J . Exp. Agr. Anim. Husb. 10, 95-98. Worley, R. E., Hegwood, D. A., and Harmon, S. A. 1971. J . Amer. Soc. Hort. Sci. 96, 531-533. Worthington, R. E., and Hammons, R. 0. 1971. Oleagineux 26, 695-700. Worthington, R. E., and Hammons, R. 0. 1972. J . Agr. Food Chem. 20, 727-730. Wynne, J. C., Emery, D. A., and Rice, P. W. 1970. Crop Sci. 10, 713-715. Yadav, N. K., and Vyas, S. R. 1971. lndian J . Agr. Sci. 41, 875-881. Yohe, J. M., and Poehlman, J. 1972. Crop Sci. 12,461-464. Yohe, J. M., Swindell, R. E., Watt, E. E., Bashandi, M. M. H., Sechler, D. T., and Poehlman, J. M. 1972. “Evaluation of Mungbean (Phaseolus aureus Roxb.) Strain at Columbia, Missouri, in 1971,” Misc. Publ. No. 72-79. Dept. of Agronomy, University of Missouri, Columbia. Young, C. T., Mason, M. E., Matlock, R. S., and Waller, G. R. 1972. J . Amer. Oil Chem. Soc. 49, 314-317. Young, J. H., Moore, R. P., and Allen, W. J. 1971. Oleagineux 26, 551-558. Yunghans, H., and Jaffe, M. J. 1972. Plant Physiol. 49, 1-7

LAND TREATMENT OF WASTEWATER Herman Bouwer and R. L. Chaney US. Department of Agriculture, Agricultural Research Service, US. Water Conservation Laboratory, Phoenix, Arizona, and US. Department of Agriculture, Agricultural Research Service, Biological Waste Management Laboratory, Beltsville Agricultural Research Cenfer, Beltsville, Maryland

I. Introduction .................................................... 11. Fate of Wastewater Constituents in Soil . . . . . . . . . . . . . . . . . . . . . . . . . . A. Suspended Solids and Clogging . . . . . . . . . . . . . . . . .............. B. Organic Carbon and Oxygen Demand ............................ C. Bacteria and Viruses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . D. Nitrogen . . . . . . . . . . . . . . . . . . . ........................ E. Phosphorus . . . . . . . . . . . ................................. F. Fluorine . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . G. Boron ...................................................... H. Metals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . I. Dissolved Salts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . J. pH . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 111. Crop Response . . . . . . . . . . . . . . . ............................. A. Effects on Yield and Quality . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B. Uptake of Pollutants and Location in Plant ........................ IV. Selection and Design of System . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References ....................... ........................... I.

133 135 135 137 141 146 151 155 155 157 162 163 164 164 165 167 169

Introduction

Public awareness of the need for preserving the quality of our surface water and increasingly severe legal restrictions on the discharge of pollutants into streams and lakes have revived interest in the use of land for disposal, treatment, and utilization of sewage effluent and other liquid wastes. Such systems have great public appeal. Wastewater is not only kept out of surface water, but land treatment also implies recycling, where “pollutants” become nutrients for plant growth. The simplicity, reliability, and low energy requirements of land treatment, as contrasted with the complex technology and high energy requirements of advancedtreatment plants, are other favorable aspects. Expressions, such as cleaning waste in nature’s way, living filters, plant-soil filters, soil mantle as sewage treatment plant, green-land clean-streams, etc., abound in the literature on land treatment of waste (Kardos, 1967; McGauhey and Krone, 1967; Stevens, 1972). Conversely, proponents of in-plant treatment have labeled 133

134

HERMAN BOUWER AND R. L. CHANEY

land treatment “a giant step backward” (Egeland, 1973). Selection of a certain system for treatment of wastewater should be free fron emotionalism. The economics and environmental aspects of various alternatives should be carefully considered so that the best system can be rationally selected. Liquid wastes commonly applied to land include conventionally treated sewage; wet sewage sludge (about 95% water), liquid animal waste (including feedlot runoff and lagoon or oxidation ditch effluents); and effluents from fruit or vegetable processing plants, animal processing plants, dairies, and fiber products industries. While these wastes vary widely in their composition, they all generally contain organic material, nitrogen, phosphorus, dissolved salts, trace elements, and microorganisms. Land treatment systems can generally be divided into three types: overland flow systems, low-rate application systems, and high-rate application systems (Bouwer, 1968; Thomas, 1973a). Overland flow systems are used where the soil is too impermeable or the suspended solids content of the wastewater too high to allow significant infiltration rates, causing most of the wastewater to run off. These systems are sometimes also called grass or vegetation filtration systems, or spray-runoff systems. With low-rate application systems, all wastewater applied infiltrates into the soil, but the dosages are rather small and of the same order as the water requirements of the crop or vegetation. Typically, the amounts are 2-10 cm per week, which may be given in one or several applications. Low-rate systems include all systems where wastewater is used for crop irrigation. Other uses of wastewater in this category are for revegetation of mine spoils, greenbelts, recreation areas, etc. With high-rate application systems, all wastewater again infiltrates into the soil, but the dosage is much greater than that necessary for crop growth. Amounts may range from about 0.5 m per week to several meters per week. Infiltration periods are rotated with drying or resting periods, to allow recovery of infiltration rates (infiltration rates generally decrease during application of wastewater) and to oxygenate the upper portion of the soil profile. High-rate systems require permeable soil. Often, the main function of the land with these systems is to receive and treat wastewater for groundwater recharge and reuse for irrigation, recreation, or industrial-municipal purposes. Agricultual utilization of the infitration system is of no or secondary importance. For both low-rate and high-rate systems, wastewater may be applied with sprinklers or, if the topography permits, with furrows, borders, or basins. Plow-in systems are sometimes used for essentially one-time application of thick liquids, such as sewage sludge or slurries from processing plants. The waste is injected into the plow furrow and covered with soil,

LAND TREATMENT OF WASTEWATER

II.

Fate of Wastewater Constituents in

135

Soil

A. SUSPENDED, SOLIDSAND CLOGGING Wastewater is usually screened, settled, or comminuted before it is applied to land. Thus, suspended solids received by the soil are usually rather fine and mainly in the organic form (sewage sludge, bacteria flocs, fibrous materials, fruit and vegetable peelings, straw or other roughage, algae cells, etc.) . These solids accumulate on the soil, forming a layer of high hydraulic impedance. This layer reduces the infiltration rate and, because it consists of biodegradable organic material, also constitutes an oxygen sink. This sink can cause small plants and seedlings to die, and it may diminish the movement of oxygen in the soil during drying. When worked into the soil, the solids initially could immobilize nitrogen if the nitrogen content is less than 1.3% on a dry-weight basis (Viets, 1973). Fine suspended material, such as colloidal clay particles, may move deeper into the soil (Goss and Jones, 1973). Movement of algal cells into dune sand was reported by Folkman and Wachs (1970). The soil, however, is a very effective filter, and suspended solids will be essentially completely removed from the wastewater after about 1 m of percolation. Since clogging at or near the surface of the soil is much easier to control and rectify than when it occurs at greater depth, it is important to know where the clogging is concentrated. The “symptoms” of clogging at the surface are decreasing water pressures (increasing tensions) and decreasing water contents in the upper portion of the soil profile, and increased effect of the water depth above the surface on the infiltration rate (Bouwer et al., 1974a). Clogging at greater depths is accompanied by increasing water pressures (decreasing tensions) and increasing water contents in the upper portion of the soil, and a decrease of the effect of depth of ponding on the infiltration rates. Clogging of the surface soil in a rapid-infiltration system receiving secondary sewage effluent was mainly a physical process due to the accumulation of suspended solids (Rice, 1974). The hydraulic impedance of the clogged layer was directly proportional to the total solids load. For a given solids load, high hydraulic gradients in the surface layer of the soil produced more compact layers of solids than did low hydraulic gradients. The more compact layers had a greater hydraulic impedance than the less compact layers for the same total solids load. Drying effectively restored the infiltration rate (Rice, 1974; Bouwer, et al., 1974a), as a result of the clogged layer decomposing. If the effluent contained suspended solids much in excess of 10 mg/liter, periodic removal of the sludge layer was

136

HERMAN BOUWER AND R. L. CHANEY

required to avoid a build-up of solids and, hence, a general decline in the infiltration rates in the basins. Biological clogging of the surface soil may also be caused by bacterial action, including production of polysaccharides and other organic compounds, if the wastewater contains a high dissolved organic matter content. Human urine, for example, with a chemical oxygen demand of 4000 mg/liter caused such clogging in fine-sand filters that the resulting infiltration rates were too low for practical application, and coarser sand had to be used (California Institute of Technology, 1969). Thomas et al. (1966) observed accelerated clogging of soil columns flooded with septic-tank effluent when the soil became anaerobic. Clogging was concentrated in the top centimeter. While sulfide accumulation could be used as an indicator of anaerobiosis, it was not a direct cause of clogging. Drying the soil caused infiltration recovery equivalent to the decrease in infiltration during anaerobic conditions. Since organic matter was the only material that declined during drying, clogging was attributed to the accumulation of polysaccharides, polyuronides, and other organic compounds during flooding. Nevo and Mitchell (1967) found that low redox potentials inhibited degradation of polysaccharides in laboratory experiments, but had little effect on the production of polysaccharides, indicating the need for regular drying or resting periods of treatment fields to avoid declines in infiltration rates. These workers also found that at temperatures below 20°C decomposition of polysaccharides was inhibited but synthesis slowly continued. Between 20 and 30"C, production and degradation rates of polysaccharides were approximately equal, and both rates increased with temperature. At 37"C, little polysaccharide was produced, but the decomposition rate continued to increase. This indicates that soil clogging caused by formation of polysaccharides may be of greater concern in cool climates than in warm climates. Regardless of climate, the optimum schedule of wastewater application and drying or resting of the soil must be evaluated by local experimentation. For the Flushing Meadows Project (Bouwer, 1973a; Bouwer et al., 1974a), maximum long-term infiltration rates were obtained with flooding periods of about 18 days, rotated with drying periods of about 10 days in the summer and 20 days in the winter. At the Whittier Narrows spreading grounds (McMichael and McKee, 1965), basins are flooded for about 9 hours and then dried for about 15 hours. With this schedule, about 2 feet per day infiltrated into the soil. Wastes containing very high solids contents may be applied only a few hours each week to allow drying and decomposition of the solids layer. Bendixen el al. (1968) reported satisfactory performance of a ridge-and-furrow system in northern latitudes where sec-

LAND TREATMENT OF WASTEWATER

137

ondary sewage effluent was applied on a 2 weeks on and 2 weeks off schedule. Clogging from excessive accumulation of suspended solids on the surface of the soil can be a problem in disposal fields with poor surface drainage. Because of reduced infiltration rates, surface runoff will develop and water will collect in the low places of the field, causing anaerobic conditions in the solids layer and the underlying soil. This will reduce the rate of decomposition of the solids, and odor and insect problems may develop. It is generally desirable to remove as much suspended material from the wastewater as possible before the water is applied to land. Overland-flow systems can effectively remove suspended solids of wastewater. Thomas (1973b) reported a suspended solids reduction from an average of 160 mg/liter (range 52-420) to 6-12 mg/liter for comminuted raw sewage applied to vegetated plots that were 36 m long and had a slope of 2-4%. The loading rates were from 7.4 to 9.8 cm/week, applied daily (except Sundays) in 8-9 hours. Law et al. (1970) found that the suspended solids content of screened cannery waste was reduced from 245 to 16 mg/liter by vegetation filtration over a distance of 45-100 m at loading rates of 0.9 cm/day applied in 6-8 hours. Other solids removal percentages are 95% for primary sewage effluent after 365 m of overland flow at the Melbourne system (Kirby, 1971), a reduction of 56.4 to 15.0 mg/liter for humus tank effluent at the high loading rate of 85 cm/day in an English study (Truesdale et al., 1964), and from 5215 to 63 mg/liter for sugar beet waste in a Nebraska study (Porges and Hopkins, 1955; Hopkins et al., 1956). CARBON AND OXYGEN DEMAND B. ORGANIC

Wastewater contains a variety of natural and synthetic organic compounds, usually not individually identified, but collectively expressed in terms of the biochemical oxygen demand (BOD, determined normally after 5 days incubation), the chemical oxygen demand (COD, usually determined with the dichromate technique), or the total organic carbon content (TOC, determined as the difference between total and inorganic carbon). The BOD and COD tests were developed primarily for oxygen regimes in aquatic environments. For land treatment, however, TOC content may be the most appropriate parameter. In addition to the carbonaceous oxygen demand, wastewater contains a nitrogenous oxygen demand for oxidation of organic or ammonia nitrogen to nitrate. The oxygen demands for other constituents are negligible, except perhaps for certain special wastes containing large amounts of sulfide, reduced iron, or other reduced compounds.

138

HERMAN BOUWER AND R. L. CHANEY

Good-quality secondary sewage effluent may have a BOD of around 10-20 mg/liter, a COD of 30-60 mg/liter, and a TOC content of 10-30 mg/liter. The relation between TOC and COD was evaluated as

TOC = 0.25 COD

+ 1.30

for secondary effluent (domestic and light industry) from the Phoenix area (Bouwer et al., 1974b). This relationship also includes measurements on renovated sewage water obtained by high-rate land treatment. The nitrogenous oxygen demand of secondary sewage effluent where most of the nitrogen is in the ammonium form, may be in the range of 100-200 mg/liter. Wastes from vegetable or fruit processing plants may have a BOD of several hundred to several tens of thousands of milligrams per liter (W. G. Knibbe, personal communication, 1973; California State Water Resources Control Board, 1968; Splittstoesser and Downing, 1969; Rose et al., 1971; Colston and Smallwood, 1973). Splittstoesser and Downing ( 1969) reported a COD/BOD ratio of 1.4-2 for vegetable processing effluents. Incompletely digested sewage sludge and liquid animal wastes have BOD’S of several hundred to several tens of thousands of milligrams per liter, depending on the density of the slurry or effluent (Loehr, 1968; Erickson et al., 1972). The COD of animal wastes may be 2 to 3 times as high as the BOD (Erickson et al., 1972). The soil with its biomass is extremely versatile and effective in decomposing natural and synthetic organic compounds. The processes can be divided into aerobic metabolisms where CO,, H,O, microbial cells, and NO,- and SO,*- are the main end products, and anaerobic metabolisms. The latter occur at a slower rate and are less complete, organic intermediates being formed. These include acids, alcohols, amines, and mercaptans. The end products of anaerobic decomposition consist of CH4,H,,NH,+, and H,S in addition to CO, and H,O (Miller, 1973). Organic carbon, whether supplied to the soil by the wastewater or produced in the soil by autotrophic bacteria, is a main factor in denitrification, since it supplies the energy for the denitrifying bacteria. Theoretically, aerobic conditions in the soil should prevail so that the total oxygen demand (sum of carbonaceous, nitrogenous, and other oxygen demands) of the waste load is balanced against the amount of oxygen entering the soil. Oxygen enters the soil (1 ) as dissolved oxygen in the wastewater applied (usually negligible), (2) as mass flow after the start of a drying or resting period, when the soil drains and air replaces the draining water in the soil, and (3) by diffusion from the atmosphere after the soil has drained. The deeper the water table and the higher the drainable pore space fraction of the soil, the more oxygen enters the soil as

LAND TREATMENT OF WASTEWATER

139

mass flow after infiltration stops. The longer the drying period, the more oxygen will enter by diffusion in relation to that which has entered the soil by mass flow. The depth to which oxygen can penetrate the soil by diffusion is limited and does not exceed a distance of about 1 meter in all but the most porous soils (Pincince and McKee, 1968; Lance et al., 1973). Lance et al. (1973) also found that the amount of oxygen entering by diffusion was 1.5 times greater than the amount entering by mass flow when laboratory soil columns were flooded with secondary sewage effluent on a 2-day wet, 5-day dry cycle, but twice that amount with a 9-day wet, 5-day dry cycle. Most of the oxygen was used to convert ammonium to nitrate and only a relatively small fraction was used to reduce COD. If wastewater is applied with sprinklers, considerable amounts of oxygen may enter the soil during the short periods between sprinkler revolutions, particularly on fast-draining soils. Some organic compounds are easier to degrade and exert a higher initial oxygen demand on the soil than others. The oxygen demand of secondary sewage effluent is sufficiently small and mostly due to readily degradable material. Thus, BOD is essentially completely removed as the effluent moves through the soil, even for high rate systems. In laboratory and field studies, prolonged flooding and obvious depletion of oxygen did not seem to affect the removal of BOD or COD (Bouwer et al., 1974b; Lance et al., 1973). Thus, anaerobic processes were also effective for BOD removal. This agrees with studies by Thomas and Bendixen (1969), who detected little or no effect of loading rate, duration of dosing, and temperature, on the organic carbon removal from septic-tank effluent passing through soil columns. Small, frequent applications, such as the 3 to 6 times per day rate recommended by Robeck et al. (1964) for best removal of COD, may be necessary if the wastewater contains high concentrations of organic compounds. Such schedules may increase the rate of biodegradation of these compounds in the soil, as was demonstrated by HaIIam and Bartholomew (1953) for plant residue. The BOD loading and removal at the Flushing Meadows Project was 100 kg/ha per day during flooding (Bouwer ef al., 1974b). At the Whittier Narrows Project, complete BOD removal was obtained from secondary sewage effluent at infiltration rates of about 0.6 m/day, or a BOD load also of about 100 kg/ha per day. In this rapid-infiltration system, 9-hour flooding periods were rotated with 15-hour drying periods. The sum of the carbonaceous and nitrogenous oxygen demands was about 750-1 000 kg/ha per day. Of this, about four-fifths was for nitrification of ammonium (McMichael and McKee, 1965). About three-fourths of the carbonaceous oxygen demand was removed in about 1.2 m of percolation of the effluent

140

HERMAN BOUWER AND R. L. CHANEY

through the soil. Erickson et al. (1972) reported BOD reductions from about 1200 mg/liter to 5 mg/liter when dairy waste was applied to the Barriered Landscape Wastewater Renovation System (BLWRS) at rates of about 2 cm/day, or a BOD load of about 240 kg/ha per day. Higher oxygen demands on the soil system and less complete removal of BOD are possible with effluents from vegetable or fruit processing plants, concentrated animal-waste slurries, or incompletely digested sewage sludges, where the BOD levels may be in the tens of thousands of milligrams per liter and the organic compounds readily biodegradable. D. M. Parmelee (personal communication, 1973) recommended that BOD loading rates not exceed 450 kg/ha per day for food processing plants. At these rates, W. G. Knibbe (personal communication, 1973) found that the COD of vegetable processing plant effluent was reduced from a range of about 500 to 2000 mg/liter to about 25 mg/liter in the first 50 cm of movement through soil (the COD of these effluents was about 1.7 times as high as the BOD). Higher loadings produced higher COD levels in the renovated water. Where soils are heavily overloaded with organic compounds in liquid wastes, solids in the wastewater and solids formed by bacterial activity in the soil may build up under the anaerobic conditions caused by the high oxygen demand. This will in turn cause a decrease in the infiltration rate, and hence in the oxygen demand exerted on the soil. Thus, soil may have some form of “self-defense” against excessive loadings of oxygen demand. Overland flow systems can also be effective in removing oxygen demand. provided the loading rate is sufficiently small and land has been sufficiently prepared to avoid channeling or short-circuiting. Thomas ( 1973b) reports a BOD reduction from an average of 150 mg/liter to a range of 8 to 12 mg/liter by flowing comminuted raw sewage over vegetated soil. Truesdale el al. (1964), using a much higher loading rate, found that BOD of humus tank effluent was reduced from a 16 to 24 mg/liter range to a 7 to 10 mg/liter range by overland flow in grassed plots. Wilson and Lehman (1967) obtained a reduction of only about 20% in the COD of primary effluent by flowing it through bermudagrass irrigation borders. For cannery wastes, the BOD was reduced from 580 mg/liter to 9 mg/liter in a Texas project (Law et al., 1970). A BOD reduction from 483 to 158 mg/liter was obtained for sugarbeet wastes in a field not very well graded and showing considerable channeling (Porges and Hopkins, 1955). Vela and Eubanks (1973) demonstrated that for land treatment of cannery wastes, soil bacteria, rather than enzymes or bacteria already present in the plant effluent, were responsible for the decomposition of organic matter. Thus, soil treatment can be expected to be more effective in reduc-

LAND TREATMENT OF WASTEWATER

141

ing the BOD of wastewater than, for example, lagooning or other treatment where the plant effluent will not be in contact with the soil. Only a small fraction of the bacteria population in the soil (16 out of 100 species) contributed directly to the decomposition of organic matter, which consisted of hydrolysis of the polymers followed by oxidation of the monomers. The other bacterial species probably contributed indirectly to the mineralization process. Because of this, bacteria in the soil did not correlate with the oxidative capacity of the soil. Shuval and Gruener’s (1973) statement that “. . . advanced wastewater renovation technology still cannot reduce COD or TOC to an absolute zero concentration . . .”, also applies to land treatment of wastewater. For example, while BOD was completely removed and COD reduced to the same level as that of the native groundwater at the Flushing Meadows Project, TOC values of the renovated water averaged 5 mg/liter after 9 m soil precolation (Bouwer et al., 1974b). The identity of this organic carbon is not very well known. Thus, it is subject to speculation regarding toxicants, teratogens, mutagens, and carcinogens. Perhaps this TOC can be reduced by treatment with a strong oxidant, such as ozone. Wastewaters, and particularly sewage effluent from industrialized communities, may contain hydrocarbons, detergents, pesticides, phenolic compounds, and other undesirable constituents. Usually, however, their concentrations are so low that with adsorption and gradual biodegradation generally occurring in the soil, few or no adverse effects are expected (Miller, 1973). Special precautions need to be taken, however, with land treatment of wastewaters containing unusually large concentrations of these compounds, or where porous soils, fissured rock, or cavernous limestones in the treatment fields offer little opportunity for appreciable renovation of the wastewater. Until further research has demonstrated that the refractory organics and other substances in renovated wastewater are harmless, direct use of such water (particularly sewage water) for domestic purposes is not recommended as a general practice (Long and Bell, 1972; American Water Works Association, Board of Directors, 1973; Ongerth et al., 1973). c.

BACTERIAAND VIRUSES

Of the numerous microorganisms possibly present in the wastewater, particularly in sewage effluents and sludges, the fate of pathogenic bacteria and viruses when the water moves through the soil is of utmost concern. The fecal coliform test is useful for indicating fecal pollution and, hence, possible presence of pathogens in surface water. For land treatment systems, low fecal coliform densities in the percolate or renovated water

142

HERMAN BOUWER AND R. L. CHANEY

probably mean absence or low levels of pathogenic bacteria or viruses. However, the absence of such organisms can be determined only by testing for specific microbial pathogens. The pathogenic bacteria commonly found in sewage effluent include Salmonella, Shigella, Mycobacterium, and Vibrio comma (Foster and Engelbrecht, 1973). Viruses include the enteroviruses and adenoviruses. The hepatitis virus is of great concern, but tests to detect its presence have not yet been developed. Other pathogens include the protozoa, such as Endamoeba histolytica, and helminth parasites, for example, ascaris and tapeworm ova. Fortunately, the soil is an effective filter and many reports indicate absence or very low levels of fecal coliforms or other organisms after water has moved one to several meters through soil (Stone and Garber, 1952; California State Water Pollution Control Board, 1953; Baars, 1964; McMichael and McKee, 1965; Drewry and Eliassen, 1968; Merrel and Ward, 1968; Romero, 1970; Young and Burbank, 1973; Bouwer et al., 1974b). On the other hand, situations have also been reported where appreciable numbers of microorganisms were detected in the renovated water after considerable distance of underground movement (Romero, 1970; Randall, 1970; Allen and Morrison, 1973). Such long underground travel distances of microorganisms are usually associated with macropores, as may be found in gravels, coarse-textured soils, structured clay soils, fractured rock, cavernous limestones, etc. The retention of microorganisms in the soil is largely due to physical entrapment for the larger organisms and to adsorption to clay and organic matter for viruses and other amphoteric organisms (McGauhey and Krone, 1967; Krone, 1968). Drewry and Eliassen (1968) found that virus adsorption was more rapid when the pH was below 7-7.5 than when the pH was higher. An increase in the cation concentration of the liquid phase in the soil also increased the adsorption of viruses. Young and Burbank (1973) reported virus removal in soil as a pH-dependent adsorption process. Cookson (1967) found that the adsorption of viruses by activated carbon could be described by a diffusion equation with a Langmuir adsorption boundary condition. Virus removal due to adsorption during phosphate precipitation was described by a pH-dependent Freundlich isotherm by Brunner and Sproul ( 1970). Microorganisms retained in the soil are subject to normal die-off, which usually takes several weeks to several months (Van Donsel et al., 1967). This is about the same as the die-off times in surface waters (Andre et al., 1967). Much longer survival times in soil have also been reported, however, such as 6 months to l year for salmonella (Rudolfs et al., 1950) and up to 4 years for Escherichia coli (Mallman and Mack, 1961). Miller

LAND TREATMENT OF WASTEWATER

143

(1973) found that fecal streptococci from sewage sludge survived up to 6 months in a clay soil, but not as long in coarser soils. The die-off of pathogens and other foreign microorganisms brought into the soil with the wastewater is due to the “homeostatic” reaction of the existing microbiological community in the soil (Alexander, 1971). This rejection of foreign organisms may result from production of toxins, lysis by enzymes, consumption by predatory protozoa, parasitic organisms, competition, and the general hostility of the soil environment to pathogenic organisms that are more at home in men and other warm-blooded creatures. Normally, fecal coliform bacteria are essentially completely removed after the water has traveled 1 m or at most 2 or 3 m through the soil. However, Bouwer et al. (1974b) found much deeper penetration of fecal coliforms below rapid-infiltration sewage basins after the basins were flooded following an extended drying or resting period. This was probably due to reduced entrapment of E. coli on the surface of the soil. The clogging layer of organic fines that had accumulated on the soil during flooding, forming an effective filter, was dry and partially decomposed after drying, thus yielding a more open surface of the soil and a less effective filter when flooding was resumed. Also, the bacteria population in the soil undoubtedly declined during drying because the nutrient supply was discontinued. Consequently, there was less competition from the native soil bacteria, and hence greater survival of the fecal coliforms when flooding was resumed. As flooding continued, however, fine suspended solids accumulated again on the surface of the soil and the bacteria population also increased, both resulting in increased retention of E. coli and return of the fecal coliform levels to essentially zero in renovated water sampled from a depth of 9 m. Almost all the removal of the fecal coliforms took place in the first 1 m of soil. Pathogenic and other foreign microorganisms may survive for some time in the soil, but they do not multiply (Benarde, 1973). The same has been observed for surface water (Deaner and Kerri, 1969). McMichael and McKee (1965) observed increased coliform counts in the soil with depth below spreading basins. They attributed this to a growth in Aerobacter aerogenes, which is a common soil bacterium of the coliform group, rather than to E. coEi. However, Masinova and Cledova (1957) reported that E. coli can sufficiently change in soil or water to give the biochemical tests more typical of the intermediate coliform types, including A . aerogenes. Cohen and Shuval (1973) studied the survival of coliforms, fecal coliforms, and fecal streptococci in surface water and sewage treatment plants. Fecal streptococci were generally more resistant than the other indicator organisms. In two systems, the survival of fecal streptococci paralleled the survival of enteric viruses better than the survival of coliforms.

144

HERMAN BOUWER AND R. L. CHANEY

The best insurance against contamination of groundwater by pathogenic microorganisms due to land treatment of wastewater is to allow sufficient distance between the land treatment facility and the point where groundwater leaves the aquifer for human consumption. Recommendations for this distance vary from about 10 m to 100 m (Romero, 1970; Drewry and Eliassen, 1968), depending on the soil type. Very coarse soils, wellstructured soils, and fractured or cavernous rocks cannot be expected to effectively retain microorganisms, and they should be avoided. In addition to moving underground, pathogenic organisms can spread from a land treatment site through the air, particularly if the wastewater is applied by sprinklers. Adams and Spendlove (1970) found that trickling filters of sewage plants emitted coliform bacteria into the air, and that E. coli could be sampled from the air as far as 1.2 km downwind. No matter what precautions are taken and how failsafe a land treatment system may be, some contamination and some survival of microorganisms may still take place. The simplest precaution against the possibility of infectious disease may be to chlorinate or otherwise disinfect all water for human consumption that is pumped from wells within underground traveling distance from land treatment sites or other possible sources of groundwater contamination. Most waterborne disease outbreaks are due to consumption of undisinfected groundwater (Craun and McCabe, 1973). These authors also recommend disinfection of groundwater as an easy and simple means to reduce the incidence of water-borne disease. Chlorination for virus control in wastewater is not effective if the water has a high suspended solids content. Thus, virus survival in chlorinated secondary sewage effluent is often observed (Mack, 1973). Culp et al. (1973) reported that disinfection for virus removal is most effective in water having a turbidity below 1 JTU (Jackson Turbidity Units) and as near as 0.1 JTU as possible. Chlorination to a free residual of 1 mg/liter with a contact time of 30 minutes is normally adequate to completely remove or inactivate all viruses. Since soil filtration of wastewater removes essentially all suspended solids, chlorination of the percolate or renovated water for virus and bacteria removal should be much more effective than chlorination of the wastewater prior to land treatment. In overland flow systems bacteria and viruses are removed primarily by settling and entrapment of suspended solids harboring the microorganisms. Detention times in overland flow systems normally are too short to reduce bacteria and viruses substantially by normal die-back, as usually happens in ponds or streams (Andre et al., 1967; Cohen and Shuval, 1973). Seidel (1966) reports much faster die-back of fecal coliforms in shallow impoundments where rushes (Scirpus lucustris and Spartina Townsendii) were growing than in impoundments without such vegetation.

LAND TREATMENT OF WASTEWATER

145

The removal of microorganisms in overland flow systems may possibly be improved if a flocculant such as alum or lime is added to the wastewater prior to land application. Viruses and other microorganisms may then become attached to the flocs and be detained on the treatment field. Excellent virus reductions, for example, have been obtained by flocculation and sand filtration of secondary sewage effluent (Berg et al., 1968). The addition of flocculants also helps to precipitate phosphates (Brunner and Sproul, 1970), and hence, may increase the phosphate removal in overland flow systems. Bacteria and viruses in the wastewater restrict the type of crop that can be grown on the land treatment fields. While entry of certain viruses into the plant through the root system has been observed (Murphy et al., 1958; Murphy and Syverton, 1958), normally the main concern is with pathogenic organisms that could collect on the surfaces of fruits and vegetables consumed raw (National Technical Advisory Committee, 1968). This committee suggests an interim guideline of not more than 5000 total coliform bacteria per 100 ml and not more than 1000 fecal coliforms per 100 ml, for irrigation water of crops where tops or roots are directly consumed by man or livestock. More conservative health guidelines were presented by Krishnaswami ( 1971 ) . A number of states have adopted quality criteria for irrigation with sewage effluent, sometimes based on what is theoretically desirable and practically achievable while avoiding criteria that are so stringent that they could not be met by normal irrigation water. As an example, the Arizona State Health Department ( 1972) requires secondary treatment, or its equivalent, if the sewage is used for irrigation of fibrous or forage crops not intended for human consumption, or orchard crops where the water does not come in contact with fruit or foliage. Secondary treatment and disinfection or equivalent treatment to reduce the total coliform density to 5000 per 100 m1 and the fecal coliform density to 1000 per 100 ml are required for higation of food crops that are sufficiently processed to destroy pathogens, or for orchard crops where the irrigation water does come in contact with fruit and foliage, or golf courses, cemetaries, etc. Tertiary treatment to produce a BOD and suspended solids content both of less than 10 mg/liter and disinfection or equivalent treatment to reduce the fecal coliform count to less than 200 per 100 ml are required if the effluent is to be used for irrigation of food crops that are consumed raw by man, or of play grounds, lawns, parks, etc., where children can be expected to play. One of the biggest questions with respect to the health hazards of land treatment of sewage effluent and other wastewaters is: What are acceptable levels of microorganisms, and particularly pathogens, in the renovated water or crops produced by such systems? Some persons may advocate

146

HERMAN BOUWER AND R. L. CHANEY

complete sterility, but this may not be necessary even if it were achievable. The environment as a whole is not sterile. Bacterial pathogens have been recovered from pristine mountain streams (Fair and Morrison, 1967). While some people may be alarmed to hear that fecal coliforms and, hence, possibly pathogenic bacteria, can travel through the air for long distances around sewage treatment plants (Adams and Spendlove, 1970), sewage treatment plant workers apparently do not have poorer health than people in other occupation groups. As a matter of fact, sewage plant workers were found to have the lowest absenteeism rate among a group of occupations studied, and this was attributed to the fact that “sewage workers were regularly immunized by their exposure to small amounts of infected material” (J. L. Melnick, as quoted by Benarde, 1973). Benarde (1973) also states that “one must be chary of the type of microbiological thinking that equates the presence of microbes with the potential for illness. The fact is that illness is an unusually complex phenomenon that does not have a 1 to 1 relationship to microbes.” Little is known about minimum infecting doses of pathogenic organisms and the combination of factors necessary to produce illness (Dunlop, 1968; Benarde, 1973). From a communicable disease standpoint, however, land treatment is far less hazardous than disposal of sewage effluent and other liquid wastes into rivers and streams (Benarde, 1973 ) . D.

NITROGEN

The nitrogen content of liquid waste may be as low as essentially zero for some cannery wastes and as high as 700 mg/liter for slurries of fresh swine waste (Erickson et al., 1972). Secondary sewage effluent generally contains 20-40 mg of nitrogen per liter (California Department of Water Resources, 1961) and sewage sludge 3-5% nitrogen (on a dry weight basis). Winery wastewaters may have 4-10 times as much nitrogen as domestic sewage (Schmidt, 1972). Wet sewage sludge (95% water) generally contains 1500-2500 mg of nitrogen per liter (Hinesly, 1973; Peterson et al., 1973). For cannery wastes, where the organic material consists essentially of cellulose and other carbonaceous materials, nonleguminous crops may actually become nitrogen deficient at high waste loadings in the same way that nitrogen deficiency may occur after application of crop residue containing less than about 1.3% nitrogen. The C/N ratio of these materials is usually about 35. For such wastes, release of significant amounts of nitrogen cannot, be expected unless the nitrogen content exceeds about 1.8 % on a dry weight basis (Wets, 1973). For secondary sewage effluent and similar liquid wastes, a significant

LAND TREATMENT OF WASTEWATER

147

amount of nitrogen can be removed by crop uptake if the nitrogen loading rates are not much more than the fertilizer requirement of the crop, which generally ranges from 50 to 600 kg/ha. For example, wheat removed 92 and 60% of the nitrogen applied with sewage effluent at the Pennsylvania State Project, using applications of 2.5 and 5 cm per week, respectively (Kardos, 1967). Sewage sludge should be applied so that the nitrogen load is about the same as the nitrogen requirements of the crop (Hinesly, 1973). When sewage effluent is applied in small amounts, the soil is predominantly aerobic and the nitrogen in the effluent (which is mostly in the ammonium form) will be converted to nitrate. The fate of this nitrogen will probably be about the same as that of fertilizer nitrogen; i.e., about 50% will be used by the plants, 25% will be lost by denitrification, and the remaining 25% will be lost by other processes, such as ammonia volatilization (Woldendorp, 1963). Since soils where wastewater is frequently applied may have high water contents, denitrification losses may be higher in land treatment fields than in normal agricultural fields, particularly if the wastewater contains organic carbon that can be used as an energy source by the denitrifying bacteria. However, as long as the wastewater is applied in normal irrigation schedules (for example, once every 1 to 3 weeks), nitrogen entering the soil in excess of fertilizer requirements tends to be converted to nitrate and moved down to the groundwater. If sewage effluent is used as the sole water source for irrigation in warm, arid regions, nitrogen loading may exceed crop uptake and normal denitrification and other losses. The excess nitrogen will then move down as nitrate to the groundwater. Thus, increases in the nitrate content of the groundwater below sewage irrigated fields are frequently observed (Matlock et al., 1972; Schmidt, 1972; Wells and Sweazy, 1973). Because the salt concentration of the Ieachate from the root zone of an irrigated crop may be 3 to 10 times as high as that of the irrigation water (Bouwer, 1969 ) , nitrate levels in the groundwater below these fields could exceed those in the sewage effluent. For high-rate systems, complete conversion of the nitrogen to the nitrate form is commonly observed if the wastewater applications are relatively short and frequent. This frequency may range from 3 to 6 short applications per day (Robeck et al., 1964), or about 8 hours flooding per day and 16 hours drying (McMichael and McKee, 1965) to 2 or 3 days flooding alternated with about 5 days drying (Bouwer et al., 1974b). For secondary sewage effluent or similar wastes with a relatively low organic carbon content, most of the organic carbon will also be oxidized under the predominantly aerobic soil conditions with these frequencies, leaving insufficient carbon for subsequent denitrification. As shown in Fig. 1 for July and August, the nitrate nitrogen concentrations in the renovated

148

HERMAN BOUWER AND R. L. CHANEY

30

20

10

0

FIG.1. 'Total nitrogen in effluent (0-0)

and nitrate ( Q - - - A ) and ammonium content of renovated water samples at 9.1 meters below the basins of the Flushing Meadows Project (Bouwer et al., 1974b).

(0-0)

wastewater will then be about the same as the total nitrogen concentrations in the wastewater (Lance and Whisler, 1972; Bouwer et al., 1974b). Denitrification is the most important process whereby nitrogen applied with wastewater in excess of crop requirements can be removed from the soil-water system (Lance, 1972). This requires the presence of nitrates and organic carbon under anaerobic conditions (Broadbent and Clark, 1965; Lance, 1972; Bouwer, 1973b). About 1 mg of organic carbon is required for each milligram of nitrate nitrogen to be denitrified. Denitrification in land treatment systems should be easiest to accomplish if the nitrogen in the wastewater is already in the nitrate form, and the wastewater contains sufficient organic carbon. Then all that is necessary to stimulate denitrification is to maintain anaerobic conditions in the soil by flooding for long periods (assuming that other factors, such as pH and temperature, are favorable for denitrifying bacteria). If organic carbon is limiting, it may be added by incorporating crop residues into the soil or by adding carbon sources to the wastewater. If the nitrogen in the wastewater is predominantly in the organic or ammonium form, as is usually the case with sewage water, an aerobic phase in the soil is necessary first, to convert the nitrogen to nitrate, before denitrification can take place. During this aerobic phase, organic carbon in the wastewater also will be oxidized by the numerous heterotrophic aerobic bacteria in the soil, leaving less organic carbon for denitrification

LAND TREATMENT OF WASTEWATER

149

when the wastewater moves into anaerobic zones. This could limit subsequent denitrification for secondary effluent and similar wastewaters which already contain relatively low organic carbon levels. The C/N ratio for secondary sewage effluent, for example, is of the order of 0.7. Denitrification following nitrification was successfully achieved by Erickson et al. (1972) for fresh swine and dairy waste slurries in the Barriered Landscape Wastewater Renovation System (BLWRS) . This is a specially constructed soil filter with an artificial barrier at a depth of about 2 m to create a perched groundwater table below which anaerobic conditions can prevail. Drains along both sides of the barrier collect the wastewater in renovated form. By applying the wastewater in frequent, small amounts (for example less than 2 cm per day), the upper portion of the soil is sufficiently aerobic to convert the nitrogen in the wastewater (concentration 3 10-660 mg/liter, mostly as organic nitrogen and ammonium) to nitrate. Because the organic carbon of the wastewater is high ( a COD of 2000-3000 mg/liter) , sufficient organic carbon is left for denitrification when the waste liquid moves from the upper aerobic zone into the lower anaerobic zone below the perched water table. This system removed 96-99% of the nitrogen from the wastewater. Additional nitrogen removal was obtained by mixing organic carbon as corn cobs, molasses, etc. in the soil above the barrier during construction. For the summer period, denitrification removed about 700 kg of N per ha per month. This is much higher than denitrification rates in normal agricultural fields, which may be about 25 kg/ha per growing season. Denitrification in secondary sewage effluent was achieved below the high-rate infiltration basins of the Flushing Meadows Project when relatively long flooding and drying periods were used; for example, 2 weeks flooding alternated with 10 days drying in summer and 20 days drying in winter (Bouwer et al., 1974b). With these schedules, oxygen became depleted in the soil below the basins shortly after flooding was started, so that nitrification could no longer occur. This left the nitrogen in the ammonium form, which was then adsorbed by the cation exchange complex of the soil, yielding both low nitrate and ammonium levels in the renovated water (Fig. 1). Flooding had to be stopped before the cation exchange complex became saturated with ammonium; otherwise, the ammonium content of the renovated water increased (Lance and Whisler, 1972). The oxygen entering the soil during subsequent drying caused bio-oxidation of the adsorbed ammonium to nitrate, part of which was then denitrified in anaerobic microenvironments. Such microenvironments could exist even in predominantly aerobic zones due to locally low oxygen diffusion rates and oxygen sinks caused by nitrification or decomposition of organic material. Nitrate not denitrified in this way was then

150

HERMAN BOUWER AND R. L. CHANEY

leached out by the newly infiltrating effluent when flooding was resumed. Some of this nitrate could be denitrified as it moved to deeper anaerobic zones. The rest of the nitrates stayed in the water and caused a nitrate peak in the renovated water collected from wells in the area upon arrival of the newly infiltrated water (Fig. 1 ). Flooding and drying should be scheduled so that the amount of ammonium adsorbed during flooding is not more than can be nitrified during drying (Lance er al., 1973). Otherwise, some adsorbed ammonium will not be oxidized, causing less ammonium to be adsorbed during subsequent flooding and hence an increase in the ammonium content of the renovated water. When this is observed, a sequence of short, frequent flooding periods or several long drying periods should be used to nitrify the adsorbed ammonium (Bouwer et al., 1974b). The total nitrogen concentration in the renovated water between NO, peaks was sometimes 80% less than that of the secondary effluent (Fig. 1) , During NO, peaks, the renovated water often contained as much total nitrogen as the effluent, and sometimes even more. The total nitrogen removal for sequences of sufficiently long flooding and drying periods to yield NO, peaks in the renovated water was about 30%. This figure was obtained by combining nitrogen relations in effluent and renovated water with infiltration rates in the basins (Bouwer et al., 1974b). The 30% removal agreed with the percentage obtained from the average total nitrogen concentration in the renovated water from the more distant wells, where the NO, peaks were attenuated by mixing and dispersion (Bouwer er al., 1974b). It also agreed with results from laboratory studies (Lance and Whisler, 1972). Since the annual nitrogen load was about 25,000 kg/ha, the 30% removal rate corresponded to a nitrogen loss of 7500 kg/ha per year, or about 625 kg/ha per month. This is close to the 700 kg/ha per month removed by denitrification in the BLWRS (Erickson et al., 1972). Most of the 70% of the nitrogen not removed in the Flushing Meadows Project is concentrated in the NOs peaks (Fig. 1) . Laboratory studies have indicated that if the portions of the renovated water containing the NOa peaks are pumped back into the basins, where they can mix with the effluent and pass once more through the soil, the total nitrogen removal can be increased to almost 80% (Lance and Whisler, 1973). These authors also increased nitrogen removal by adding organic carbon to the effluent prior to infiltration, or by reducing the infiltration rate. The latter was accomplished by decreasing the depth of ponding above the soil. At nitrogen loadings of 25,000 kg/ha per year, crop uptake of nitrogen is insignificant. However, crops may increase the nitrogen removal by stimulating denitrification in the root zone due to exudation of organic carbon and the creation of low oxygen levels, as reported by Woldendorp (1963) and Stefanson (1973). Some evidence of lower nitrate contents in the reno-

LAND TREAWENT OF WASTEWATER

151

vated water below grass-covered basins as compared to that below nonvegetated basins, was obtained at the Flushing Meadows Project (Bouwer et al., 1974b). However, these lower nitrate contents could also be the result of inhibitatory effects of roots on nitrification, as reported by Moore and Waid (1971). The slower release of nitrate resulting from this action could lead to reduced nitrate leaching during the initial stages of a new flooding period, and to more denitrification in the biologically active upper soil layers. Nitrogen from wastewater treated by overland flow or spray runoff systems can be removed by adsorption of ammonium to the soil and by denitrification in the biologically active surface layer of the soil. Organic or ammonium nitrogen in the wastewater can be converted to nitrate in the overland flow sheet, which is in direct contact with atmospheric oxygen. Shallow flow and relatively long detention times are required for significant nitrogen removal. Thus, while high loading rates yielded little or no nitrogen removal in overland flow systems (Truesdale et al., 1964; Wilson and Lehman, 1966), lower rates showed nitrogen reductions from an average of 23.6 mg/liter in the raw sewage to a range of 2.2 to 7.2 mg/liter in the runoff, depending on loading rate and age of the system (Thomas, 1973b). Law et al. (1970) reported nitrogen reductions from 17.2 to 2.8 mg/liter in an overland flow system for treatment of cannery waste.

E. PHOSPHORUS The phosphate content of secondary effluent varies widely among municipalities (Pound and Crites, 1973a,b). The observed range is about 0.5 to 40 mg of phosphorus per liter. The EPA “theoretical effluent” contained . waste dis10 mg of phosphorus per liter (Thomas, 1 9 7 3 ~ ) Industrial charges can reduce the phosphate concentration in municipal sewers, or greatly increase it. The phosphate content of a municipality’s wastewater may vary with time. The phosphate in the wastewater used at the Pennsylvania State University project fell steadily from 9.7 mg of phosphorus per liter in 1963 to 4.2 mg in 1970 (Sopper and Kardos). A similar decrease was reported by Bouwer et al. ( 1974b). Technology has been developed to minimize effluent phosphate by additions of phosphate precipitant chemicals (lime, aluminum sulfate, ferric chloride) during sewage treatment (Barth and Ettinger, 1967). The effluents from these processes contain low levels of phosphate. Also, alternative biological technology has been developed to reduce effluent phosphate to as low as 0.55 mg of phosphorus per liter (Levin et al., 1972). Total sewage phosphate removed by conventional treatment processes ranges from 20 to 90%. This variation led to the search for the improved biological technology to remove phosphate (Levin et al., 1972). The treat-

152

HERMAN BOUWER AND R. L. CHANEY

ment processes generally lead to hydrolysis of sewage polyphosphates to orthophosphate (Bunch et al., 1961). Polyphosphates are also rapidly hydrolyzed in soil (Gilliam and Sample, 1968). The reactions of wastewater phosphate in soils recently have been described by Ellis (1973), Ellis and Erickson (1969), Lindsay (1973), and Thomas ( 1 9 7 3 ~ ) The . Langmuir adsorption isotherm has been applied to the adsorption of phosphate in soils by numerous authors (Griffin and Jurinak, 1973 ) . Schneider and Erickson ( 1972), Ellis ( 1972), and Ellis and Erickson (1969) described the use of Langmuir constants to estimate the phosphate adsorption capacity of particular soils from a solution containing 10 mg of phosphorus per liter, The adsorbing capacity of the soils seemed to be related to the iron and aluminum contents. For example, the phosphate absorbing capacity of some highly weathered soils was much higher in the B-horizon than in the A-horizon, presumably because iron and aluminum oxides had accumulated in the B-horizon. The calcareous soil used in the study had a low phosphate adsorbing capacity. Such soils contain little iron and aluminum oxides, and phosphate removal may be due to precipitation of calcium phosphates. Ellis ( 1973) noted that the adsorption capacity of phosphorus-saturated soil was regenerated during 3 months’ incubation. The regeneration was probably due to crystallization of adsorbed phosphate into less soluble compounds and to the production of more iron and aluminum oxides by weathering. Thus, use of Langmuir constants to calculate the potential life of a land treatment site can lead to serious underestimation. On the other hand, presumption that all the hydrous oxides of iron and aluminum will be available to adsorb phosphate (Bauer and Matsche, 1973) can lead to overestimation of the life of a site. Schneider and Erickson (1972) compiled phosphate adsorption capacities for Michigan soils based on Langmuir constants. Griffin and Jurinak (1973) modified Langmuir adsorption isotherms to account for two simultaneous adsorption reactions. A convenient one-point method has been developed by Bache and Williams (1971) to determine Langmuir constants. Various phosphorus compounds also precipitate in soils depending on concentrations of phosphate, Fe3+, Al”’, Ca’+, F-, CO,“, and on pH. Lindsay and Moreno (1960) developed a solubility vs pH diagram for variscite, strengite, fluoroapatite, hydroxyapatite, octacalcium phosphate, and dicalcium phosphate dihydrate as end products of the adsorption-precipitation sequence. The kinetics of some of these phosphate precipitation reactions are relatively slow, and equilibrium with the predicted crystalline precipitates should not be expected. However, phosphate precipitation may be the main mechanism for phosphate removal from wastewater in calcareous soils.

LAND TREATMENT OF WASTEWATER

153

With low-rate systems, so little phosphate can be applied that crop removal balances phosphorus additions with wastewater. Sometimes, phosphorus fertilizer may have to be added to maintain fertility. Kardos and Sopper (1973) described renovation of secondary sewage effluent by sampling from porous cups installed 15, 60, and 120 cm deep in a soil cropped to corn and Reed canarygrass, and in two forested soils. Although the phosphorus level in the soil water at the 15-cm depth was increased by wastewater application, the phosphorus concentration at 120 cm was only slightly affected. Areas covered by Reed canarygrass received about twice as much phosphorus as areas in corn, but the phosphorus concentration in the soil water at the 120-cm depth was lower in the Reed canarygrass areas than in the corn areas. The phosphorus concentration in the soil solution was higher where effluent was applied at 5 cm/week than at 2.5 cm/week. Through 1970, the removal of phosphorus from the wastewater was about equal on Hubersburg silt loam and Morrison sandy loam. Sopper and Kardos (1973) reported the crop responses to wastewater application, and crop removal of phosphorus. In the early years of their project, wastewater phosphorus at 5 cm/week application supplied as much as 134 kg of phosphorus per hectare per year, clearly in excess of crop removal. However, by 1971, the wastewater phosphorus application had dropped considerably, and corn silage or Reed canarygrass removed more phosphorus than was added with wastewater. Forest crops did not remove nearly as much phosphorus. In 1970, only 19% of the applied phosphorus was removed. Hook et al. (1973) reported the soil phosphorus relations for these same plots. The Hubersburg silt loam showed considerable increase in Bray-extractable phosphorus in the surface 30 cm of soil, but little change in the second 30 cm. Morrison sandy loam soil showed increased extractable phosphorus as deep at 120 cm. They considered three bases for the deeper penetration of phosphorus in the Morrison soil: (1 ) crops had not been removed; ( 2 ) the sandy loam has a greater hydraulic conductivity, thus phosphorus in percolating solution has less time to react with particle surfaces; and ( 3 ) the concentrations of free iron and aluminum oxides are much lower in the Morrison soil. Day et al. (1972) found that the available phosphorus was increased in the Ap-horizon after 14 years of irrigation of calcareous soils with sewage effluent; available phosphorus was not significantly increased in the C-horizon. Relatively low application rates were used, and no data have been reported on soil solution levels of phosphate at different depths. Kirby (1971) reported that the moderately acid soils of the infiltration system at Werribee, Australia, removed about 80% of the phosphorus in the (settled) sewage. By 1958 (after 70 years of operation) as much as

154

HERMAN BOUWER AND R. L. CHANEY

3200 mg of phosphorous per kilogram of soil (surface 10 cm) had accumulated in the irrigated areas, with considerable movement below 30 cm (Khin and Leeper, 1960). The 3200 mg of phosphorus per kilogram of soil is about 10 times the adsorption maximum for high adsorbing soils determined by Ellis and Erickson (1969) using Langmuir constants. More recently, R. D. Johnson, R. L. Jones, T. D. Hinesly, and D. J. David (personal communication, 1974) found greater accumulation and deeper penetration of phosphorus than did Khin and Leeper. In characterizing the soil phosphate in irrigated and control areas, .Khin and Leeper (1960) found that one-third of the phosphorus was organic bound. Crop removal accounted for little phosphorus removal because grazing cattle and sheep returned about 85% of dietary phosphorus to the soil. With high-rate systems, the phosphate applied to the soil greatly exceeds crop uptake. At the Flushing Meadows Project (Bouwer et al., 1974b), the annual application was about 10,000 kg of phosphorus per hectare. The PO,-phosphorus concentration of the renovated water 9 m below the basins was 30-70% less than in the sewage effluent, depending on hydraulic loading and PO,-phosphorus content of the effluent. Further underground travel through the predominantly sandy and gravely materials resulted in additional PO, reduction. Wells 6 m deep and 30 m away from the basins yielded renovated water with phosphorus concentrations of 1-3 ppm, or removal percentages of 70-90%. After 5 years of operation of the project, during which a total of almost 50,000 kg of PO,-phosphorus was applied per hectare, the phosphorus removal efficiency of the system was still stable. Since the soils were calcareous sands and gravels, which contained little or no iron and aluminum oxides and less than 2% clay, the phosphorus was probably removed by precipitation of calcium phosphates. Larson (1960) found that 75% of the 2700 kg of phosphorus per hectare per year applied with wastewater was removed after 9 m of movement through coarse soil. Significant reductions in phosphorus concentrations of wastewater have also been observed in overland-flow systems. Kirby ( 197 1 ) observed 35 % removal from sewage effluent at the Werribee, Australia, system. Law et al. ( 1970) reported reductions in phosphorus-concentrations of cannery waste from 7.4 to 4.3 mg/liter due to overland flow, with daily applications. The phosphorus removal was essentially doubled when the frequency of application was reduced to three times per week. Thomas (1973b) reported phosphorus reduction in raw sewage from an average of 10 mg/liter to an average of 4.0 to 5.4 mg/liter, depending on age of the overland-flow system and loading rate. Most work on phosphorus removal from wastewater applied to soil has consisted of determining phosphorus concentrations in the renovated water.

LAND TREATMENT O F WASTEWATER

155

More knowledge of the reaction kinetics of phosphorus precipitation and adsorption is needed before the phosphorus removal capacity, and hence the useful life, of a land treatment system can be accurately assessed.

F. FLUORINE Wastewater is enriched in fluoride by industrial and domestic additions. Many cities now add fluoride to the drinking water so that it contains about 1 mg of fluoride per liter. Fluoride is adsorbed by various soil components, especially hydrous aluminum oxides, according to the Langmuir adsorption equation (Bower and Hatcher, 1967). The adsorption of wastewater F by soil and its subsequent equilibration with fluorite (CaF,) and fluoroapatite leads to both retention of fluoride in the soil and control of injury to plants and food chain. The Ca2+added with wastewater maintains the soil Ca level high enough to prevent fluoride injury. Injury from added NaF has been demonstrated in acidic soils low in Ca, but not in well-limed soils (Prince et al., 1949). Crops raised on fluoride-enriched soils show little increased F uptake as long as the soil is near neutral pH. A recent review by Brewer (1966) summarizes plant and soil relationships of fluoride. Larsen and Widdowson ( 1971 ) have examined “labile” fluoride in soils. The maximum limit of fluoride in irrigation water for continuous use on all soils is 2 mg of fluorine per liter (National Academy of Science-National Academy of Engineering, 1972). Few surface waters exceed 1 mg of fluorine per liter. Very little study of the fate of fluoride during wastewater irrigation has been reported. Bouwer et d.(1974b) reported that the fluorine content of secondary effluent was reduced from 4.1 to 2.6 mg/liter after 9 m of movement through sandy material in a high-rate system, and reduction continued with further movement through the coarse textured soil. The fluoride removal somewhat paralleled the phosphate removal, suggesting precipitation of fluorapatite and fluorite. Soil retention of fluorine should be related somewhat to kinetics of water movement and length of path (Bower and Hatcher, 1967). Possibly, irrigation water containing higher levels of fluorine will lead to slightly higher fluorine content of plants (Rand and Schmidt, 1952), perhaps because of the temporary surface adsorption of fluorine in forms more available to plants than fluorite and fluorapatite. G . BORON

As borates are substituted for phosphate in household detergents the boron content of sewage effluent may increase. Bouwer et al. (1974b)

156

HERMAN BOUWER AND R. L. CHANEY

found that boron in the secondary effluent from the city of Phoenix, Arizona, had risen from 0.4 mg/liter in 1969 to 0.9 mg/liter in 1971. The effects of boron on crops and soils have been studied for many years because they are a hazard in some natural irrigation waters in arid areas. Differences in crop sensitivity to boron in irrigation water have been identified (Eaton, 1944; Richards, 1954). Sensitive crops showed toxicity at 0.5-1 mg of boron per liter, semitolerant crops at 1-2 mg of boron per liter, and tolerant crops at 2-4 mg of boron per liter. The maximum level of boron in irrigation water for continuous use on all soils is 0.75 mg of boron per liter (National Academy of Science-National Academy of Engineering, 1972); this level is based on studies of the boron-sensitive citrus crops. Ellis and Knezek (1972) summarized the reactions of boron with soils. Boron adsorption appears to occur on: (1 ) iron and aluminum-hydrous oxide coatings on clay minerals; ( 2 ) iron and aluminum oxides; ( 3 ) clay minerals, particularly micaceous-type clay minerals; and (4) magnesiumhydroxy clusters or coatings that exist on the weathering surface of ferromagnesian minerals. Several authors have found that boron adsorption can be described by the Langmuir adsorption equation, at least over a limited range of concentration. Studies of soil adsorption of boron particularly relevant to irrigation have been made using soil columns (Biggar and Fireman, 1960; Hatcher and Bower, 1958; Okazaki and Chao, 1968; Rhoades et al., 1970; Tanji, 1970). Rhoades et al. (1970), studying leaching of soils to remove naturally occurring excess boron, found that weatherable boron can be released during incubation after the leachable adsorbed boron has been removed. Thus, wastewater boron will be retained until its concentration reaches equilibrium with the soil solution boron. Wastewater irrigation effects on plant, soil, and percolating water boron have been reported in only a few studies. Bouwer et al. (1974b) found essentially no boron removal in the sandy and gravelly soils below their infiltration basins. On the other hand, Sopper and Kardos (1973) found that the heavier soils were still retaining up to 90% of the added boron (as measured by soil solution extracted at 1.2 m) . The soil solution boron was higher where 5 cm of wastewater were added per week (ca. 0.09 mg/liter) than where 2.5 cm was added per week (ca. 0.06 mg/liter) (control was ca. 0.03 mg/liter) . The added wastewater contained about 0.29 mg of boron per liter. Analysis of several crops growing in the different experiments showed only a slight increase in foliar boron content. In humid areas rainfall will leach some of the boron adsorbed from wastewater. Clearly, however, questions remain about the safety of boron additions with wastewater added in excess of crop requirements, especially when some wastewaters already contain boron in excess of the recommended limit for irrigation water (Pound and Crites, 1973a).

LAND TREATMENT OF WASTEWATER

157

H. METALS The concentration of heavy metals in various wastewaters is generally quite low if there is no specific metal pollution. Most of the metals in sewage end up in the sludge. Brown et al. ( 1973), found that the higher the influent metal level or the higher the suspended solids removal, the higher the observed metal removal efficiency. Cadmium removal was poor (averaging 16% ), apparently because of its low concentration. Argo and Culp (1972) and Nilsson (1971 ) summarized metal removal by different sewage treatment practices. Mytelka et al. (1973) reported the contents of silver, cadmium, cobalt, chromium, copper, iron, mercury, manganese, nickel, lead, and zinc in raw and treated sewage collected from treatment plants in the Interstate Sanitation District (New York, New Jersey, and Connecticut). The range and median values for selected elements are presented in Table I. Blakeslee (1973) reported the total and dissolved cadmium, chromium, copper, mercury, nickel, lead, and zinc of wastewater effluents from 5 8 treatment plants in Michigan. The range and median values are shown in Table 11. The amount oi metals that would enter the soil with the wastewater could be considerably lower than permitted under the 1972 Irrigation Water Standard (National Academy of Science-National Academy of Engineering, 1972), as shown in Table 111. The reactions of heavy metals with soils and uptake by plants have recently been reviewed by several authors (Allaway, 1968; Chaney, 1973; Ellis and Knezek, 1972; Hodgson, 1963; Jenne, 1968; Knezek, 1972; TABLE I Range and Median Heavy Metal Contents of Wnstewater Treatment Plant Effluents in t h e Tnterstate Sanitation Districtn Range

rdow Elemerit

(Ing/liter)

High (mg/liter)

Median (ing/liter)’


6.4



05 <0.05

0.05

<0.05 < 0 , 0001

5 , !I 0.1%

<0.05 <0.05 0 .1 0 n . QOOB

<0.1

1.5 G.O


6.8

50.0

From Mytelka el al. (1973).

<0.1

< 0 . 05 Q,l5

158

HERMAN BOUWER AND R. L. CHANEY

TABLE I1 Range and Median Heavy Metal Contents of 58 Wastewater Treatment Plant Effluents in Michigan" Range

Element Cd Cr cu Hg Ni Pb Zn a

Low (mg/liter)

High (mg/liter)

Median (mg/liter)

10.005

<0.005

0.01
0.15 1.46 1.3 0,001


5.4



1.3

0.03

4.7

0.025 0.04 0.2 0.02 0.05 0.19

From Blakeslee (1973).

TABLE I11 Comparison of the Amount of Toxic Metals That Would Be Applied with a Typical Effluent vs That Sdded under the 1972 Irrigation Water Standards@ 1972 Irrigation Water

Standard

Element

Concentration Amountb (mg/liter) (kg/ha/year)

Cd cu Ni

0.01

Pb

5.0 2.0

Zn a

0.20

0.20

0.2 4.0 4.0 100 40

National Academy of Science-National

Typical effluent Concentration (mg/liter)

Amountb (kg/ha/year)

< O . 005

0.1

0.10 0.02 0.05 0.15

2.0 0.4

1 .o 3.0

Academy of Engineering

(1974).

Applied at 5 cm/week for 40 weeks/year.

Leeper, 1972; Lindsay, 1973; Lisk, 1972; Mitchell, 1964; Murrmann and Koutz, 1972; Page, 1973). The metal ions are bound by clay, organic matter, and hydrous oxides components of the soil. A high pH favors immobilization of some metals in the soil. Jenne (1968) suggested that the hydrous oxides of manganese and iron were very important in adsorption of heavy

LAND TREATMENT OF WASTEWATER

159

metals. Stable organic matter in the soil may significantly contribute to the binding of metals (Ellis and Knezek, 1972); the chelation of copper by organic matter is especially important. Bondietti and Sweeton (1973) observed that the apparent stability constant of cadmium with soil organic matter is dependent on the percentage of the cadmium binding capacity filled. Apparently, small amounts are bound to relatively specific sites of the organic matter which form higher stability chelates with the metal ion. Large amounts of metals (>50% saturation) are held much less firmly, more nearly like cation exchange than chelation. The heavy metals do enter into the general cation exchange reactions with clays and organic matter, in addition to the chelation reactions with organic matter. The metal ions in wastewater should occur largely as low molecular weight soluble chelates which could affect the kinetics and extent of metal reactions and movement in the soil. Norvell ( 1972) recently summarized the reactions of metal chelates in soils. Volk (1970) found that zinc EDTA moved rapidly through soil. Thus the presence of strong chelating agents in wastewater could lead to deeper penetration of heavy metals into the soil. On the other hand, the physical filtering activity of soils could remove high molecular weight metal complexes from the wastewater. The ability of soils and plants to remove heavy metals from wastewater is generally considered to be a benefit of wastewater irrigation. In the past, raw sewage was often applied and industrial release of heavy metals to the sewers was not controlled. Thus, irrigation with sewage sometimes has led to substantial accumulations of metals in soils. Phytotoxicity to metalsensitive crops has sometimes been observed. Blood (1963) reported that “on a sand land farm with a history of annual applications of [raw] sewage effluent over 80 years, the soil now contains 500 ppm of Zn down to 25 cm, whilst soil from adjacent land receiving no effluent contains less than 20 ppm.” Sugar beets failed where the soil pH fell below 6.4. Rohde (1962) reported on heavy metals accumulated in the soils of the (raw) sewage farms of Berlin and Paris. He found that copper and zinc were significantly increased in the irrigated soil, and that local areas where “exhaustion” was observed contained higher levels of metals than nearby areas where plants looked healthy. The problems appear to have been corrected by more careful management of soil pH and use of crops less sensitive to toxic metals. The Paris farm may have experienced zinc-induced manganese deficiency; foliar sprays of manganese can be used to correct this deficiency (Trocme ef al., 1950). A more recent study of the Werribee Sewage Farm at Melbourne, Australia reported by R. D. Johnson, R. L, Jones, T. D. Hinesly, and D. J. David (personal communication, 1974) has revealed that substantial levels

160

HERMAN BOUWER AND R. L. CHANEY

of metals have accumulated during 70 years of application of raw and settled sewage. Six irrigated locations and one unirrigated location were sampled at three depths. The samples were taken from the center of the irrigation paddocks; the accumulation of metals, etc., would have been much higher nearer to the sewage outfall because flood irrigation would lead to grass filtration near the outfall. The samples were analyzed for total and 0.1 N HCI extractable metals (Table IV). It is clear that these metals have accumulated in the soil, and that considerable leaching of the metals has occurred. The extent of leaching was higher at locations with higher total metal accumulation. The zinc content of the largely perennial rye grass crop was 50 mg/kg on unirrigated soil and 104 mg/kg on irrigated soil (mean of four irrigated sites). The inactivation or reversion of non-wastewater-related metals observed by other workers (Follett and Lindsay, 1971) was not observed at Werribee. Nearly 80% of all zinc added remained in a form extractable by 0.1 N HC1. It is not clear whether the breakdown of organic matter, cycling of redox potential, or other factors related to the addition of sewage effluent led to this result. Perhaps, the observations of Follett and Lindsay have limited application when the metals are applied with sewage effluent. The existence of some of these older wastewater irrigation sites with accumulated metals, phosphate, etc., provide a valuable opportunity to determine the environmental impact of land disposal of wastewater. Several TABLE IV Total Metals and 0.1 N HC1 Extractable Heavy Metals in Soils at Werribee

Soil Total metals Irrigated (6)

Unirrigated (1)

320

0-2.5 2.5-18 15-45 0-2.5 2.5-18 25-45

82 60 51

144 91 70 60 84

47 42 56

23

30

21 47

30 58

-

-

-

0.1 N HCl extractable metals ~~

Irrigated (6)

Unirrigated (1)

0-2.5 2.5-15 25-45 0-2.5 2.5-15 15-45

~~~

210 70 17 16 1.8 0.9

~

2.0 4.0 2.9 1.1 0.8 2.2

13.0 4.8 4.7 1.9 1.0 3.0

1.78 0.57

0.%3 0.17 <0.13 <0.13

LAND TREATMENT OF WASTEWATER

161

recent surveys have been conducted to identify these sites (Sullivan et al., 19?3; Pound and Crites, 1973b). Wheatland and Borne (1961) found considerable removal of chromium, copper, manganese, nickel, lead, and zinc from river water but did not examine soil parameters involved. Lehman and Wilson (1971 ) studied removal of metals from wastewater by filtration through mostly sandy calcareous soils. The concentrations of iron, manganese, nickel, copper, zinc, lead, and cadmium were effectively reduced. Aerobic conditions in the soil resulted in greater immobilization of the metals than anaerobic conditions. Metal concentrations in the secondary effluent and in the renovated water from a well 27 m from the basins of the Flushing Meadows project showed considerable removal of copper and zinc, but not of cadmium and lead (Table V ) . The maximum permissible limits for these metals in raw public water supplies, listed by the National Technical Advisory Committee (1968), are shown for comparison. Metals did not accumulate in the surface 1.5 m of the soil (R. L. Chaney, R. C. Rice, and H. Bouwer, unpublished, 1972) probably because of the low organic matter and clay content of the basin soils, the low retention times of the water in the surface soils, and the low metal concentrations in the effluent. A study of metal accumulation in infiltration basins at Ft. Devens, Massachusetts, by E. P. Meier and S . A. Schaub (personal communication, 1973) revealed a peak of heavy metals which coincided with an organic matter accumulation zone at 45 cm. The organic matter in this zone and its metal content appeared to increase during winter and decrease during summer. K. W. Brown, C. E. Woods, and J. F. Slowey (personal communication, 1973) at Texas A&M are studying soil retention of metals from wastewater enriched to 1 mg/liter each of cadmium, copper, nickel, lead, and TABLE V Metal Concentrations in Secondary Effluent and Renovated Water at Flushing Meadows Project" and Maximum Limits in Raw Municipal Water Suppliesb

Element Zn cu Cd

Pb Hg a

Secondary effluent Renovated water Maximum limits (mg/iiter) (mg/liter) (mg/liter) 0.193 0.143 0.008 0.084 0,002

0.037 0.017 0.007 0.066 0.001

5 1 0.01 0.05 Not given

From Bouwer d al. (1974b). From the National Technical Advisory Committee (1968).

162

HERMAN BOUWER AND R. L. CHANEY

zinc. This retention is studied in relation to depth, variations in soil pH, and clay and organic matter contents. In general, most studies of metal leaching have found that the pH, organic matter, and ion exchange capacity are the dominant factors in metal movement through soils. Jones and Belling ( 1967) and Jones et al. (1957) examined the effect of water, C0,-saturated water, superphosphate, and an aqueous extract of lucerne on leaching zinc, copper, and cobalt in several soils. They found appreciable movement only in a sand with low organic matter. Smith et al. (1962) also observed movement of copper and zinc only in soils of low cation exchange capacity and low organic matter; low pH promoted leaching. Peterson and Geschwind (1972) found leaching of heavy metals from acidic strip mine spoils amended with sewage sludge only at low pH. Ng and Bloomfield (1962) suggest that under reducing conditions (unstable organic amendments and waterlogging) heavy metals can be mobilized and subsequently leached. In contrast, when a method to obtain zinc leaching through soils was desired in order to fertilize apple trees, Benson (1966) found that this could be achieved by applying an excess of potassium, calcium, or magnesium salts with the zinc. A comparison of amounts of nitrogen and heavy metals added to a given soil with sludge vs effluent shows that, at application rates that avoid nitrate pollution, more heavy metals can be added with sludge in one year than are added in a century of effluent irrigation. It is generally considered that the amount of plant-available nitrogen limits the yearly application of sludge. However, some sludges contain large amounts of metals (Chaney, 1973; Page, 1973), and metal toxicity to sensitive crops could result from sludge applications that do not contain more nitrogen than can be removed by a crop. The level of metals in municipal wastewater should soon be sufficiently low that heavy metals will not be a limiting factor in long-term wastewater irrigation practices. New effluent and pretreatment guidelines (Environmental Protection Agency, 1973) should lead to control of metal release from industrial sources. Although other pretreatment technologies to remove metals from industrial wastewaters are available (precipitation, ion exchange resins, and reverse osmosis), Wentink and Etzel (1972) found that soil filtration was an adequate pretreatment to remove metals. I. DISSOLVED SALTS

Sewage effluent usually contains about 100-300 mg more salt (predominantly NaCl) per liter than the water going into the municipal water supply system (California Department of Water Resources, 1961). Cattle manures may contain high NaCl levels, whereas effluent from vegetable processing

LAND TREATMENT OF WASTEWATER

163

plants may contain considerable amounts of NaOH if the lye-peeling process is used. Dissolved salts in the wastewater react with the soil through ion exchange and dispersion or flocculation of clay. While ion exchange may initially affect the quality of the percolate or renovated water, the ionic composition of the renovated water will eventually be the same as that of the wastewater as equilibrium between the cation exchange complex and the soil solution is reached. Much higher salt concentrations in the percolate than in the original wastewater can be expected if the amount of wastewater applied (plus rainfall) is not much larger than the evapotranspiration of the crop. For normally irrigated fields in arid regions, the salt content has increased 3- to 10-fold as the irrigation water moved through the root zone (Bouwer, 1969). The salt balance equation (Bouwer, 1969) shows that the ratio of the salt concentration in the percolate, Cd, to that in the wastewater, Ci,can be calculated as

C d C i = Di/(Dt - DJ where Di is the amount of water applied and D, is the evapotranspiration. Where rainfall or precipitation of salts in the soil are significant, appropriate corrections should be made in this equation to take these effects into account. The equation shows that even when three times as much water is applied as is needed by the crop, the salt concentration in the percolate will be 1.5 times that in the wastewater. Thus, overirrigation with wastewater in arid regions tends to produce renovated water with too high salt concentration. Only when D iis much greater than D, (high-rate systems) can renovated water with essentially the same salt content as the wastewater be obtained. At the Flushing Meadows Project, for example, D , = 1.8 m/year and Di is about 100 m (Bouwer et al., 1974b). This gives renovated water with only about 2 % more salt than in the sewage effluent.

J. PH The pH of soil and the pH of wastewater can be modified by the chemical reactions in the soil. Wastewater containing organic acids may show an increase in pH as it moves through the soil because of biodegradation of the acids. For biodegradable material with a near neutral pH, such as sewage effluent, the pH may decrease because soil microbial activity produces CO, and organic acids. At the Flushing Meadows project, for example, the p H of sewage effluent was about 8, whereas that of the renovated water was about 7 (Bouwer et al., 1974b). Such a decrease in pH could result in increased solubility of CaCO, which may occur in the soil as a precipitate if the original p H was in the 7.5 to 8.2 range.

164

HERMAN BOUWER AND R. L. CHANEY

Ill.

A.

Crop Response

EFFECTSON YIELDAND QUALITY

Experimental examination of the effect of wastewater irrigation on crop yield is made difficult by the need to compare unirrigated unfertilized control, unirrigated fertilized control, water irrigated control, and water plus fertilizers equal to wastewater treatment control with the wastewater treatment. Where rate of wastewater application is varied, comparable treatments for each rate would also be needed. Without this group of treatments, it is difficult to evaluate the role of water, plant macronutrients and micronutrients, etc., on crop yield and quality. Major studies to date, while finding a few negative and many positive yield responses of crops to wastewater, have not included this range of treatments. For arid regions, the unirrigated controls are not necessary. Research at Pennsylvania State University has compared unirrigated, unfertilized control (for forest crops) or unirrigated, fertilized control (for field crops) with one or several rates of wastewater application. Sopper and Kardos (1973) report 9 years’ experience: “Effluent irrigation at 2 inches per week resulted in annual yield increases ranging from -8 to +346% for corn grain, 5 to 130% for corn silage, 85 to 191% for red clover, and 79 to 139% for alfalfa.” Precipitation during the growing season greatly influenced yield differences, but the experimental design prevented study of this factor. Since the major theme of the studies was the ability of crops and soils to remove nutrients from percolating wastewater, wastewater application rates exceeded the water requirements of the crops. In the University of Arizona studies, wastewater is considered an alternative water source for otherwise necessary irrigation; thus, one wastewater application rate was used: the amount of water required to meet recommended irrigation practices. Day and co-workers (1962, 1963; Day and Tucker, 1959, 1960; Day and Kirkpatrick, 1973; Day, 1973) in Arizona studied growth of several crops (barley, oats, and wheat as forage and grain) irrigated with water; water plus recommended N, P, and K; water plus N, P, and K equal to contents of these elements in secondary sewage effluent; and secondary effluent. The application rates were governed by the irrigation requirements of the crops. Wastewater generally produced equal or somewhat higher yields of grain or forage than well water with N, P, and K added equal to the N, P, and K of the wastewater, although barley was more easily injured by wastewater than were oats and wheat. Effects of wastewater irrigation on crop quality were included in both

LAND TREATMENT OF WASTEWATER

165

the Arizona and Pennsylvania studies. Day and co-workers ( 1962, 1963; Day and Tucker, 1959, 1960; Day and Kirkpatrick, 1973; Day, 1973) found that small-grain forage contained similar amounts of protein and digestible laboratory nutrients when irrigated with wastewater or water plus N, P, and K equal to wastewater. The danger of nitrate poisoning from forage grown with wastewater was no greater than the danger from forage grown with well water plus N, P, and K (Day et al., 1961). Although the yields and quality of wheat grain for livestock feed were not impaired by wastewater irrigation, the milling and baking qualities of wheat grain produced with wastewater were lower than those of grains grown with well water plus N, P, and K (Day, 1965). No explanation is available for this observation. The quality of other crops as food or fiber could be modified by wastewater irrigation. Slightly higher N, P, and K in wastewater irrigated forage crops than in chemically fertilized, unirrigated crops were found in the Pennsylvania State University studies. No reduction in feed quality of these forages was observed. Although harvest of corn silage and Reed canarygrass forages can remove as much N, P, and K as is added with some wastewaters, cycling the forages through animals will generate manures that will contain much of the N, P, and K removed from the wastewater irrigation site. The properties of red pine and red oak necessary for pulpwood were enhanced by wastewater irrigation (Murphey et al., 1973). Sopper and Kardos (1973) report considerable variation in response of different tree species to wasterwater irrigation; white spruce was quite responsive to wastewater, whereas red pine showed yield response for 2.5 cm per week application but none to 5 cm per week. In one of their experiments, unusual weather conditions led to windthrow of all trees on the irrigated treatment apparently because irrigation decreased the rooting depth of red pine. While nitrogen in sewage effluent or other wastewater has fertilizer value, nitrogen in irrigation water is not always desirable because it may unfavorably affect yield and quality of some crops. Baier and Fryer (1973) report reduced yield, fruit size, and/or fruit quality of certain fruit crops. Too much nitrogen at the end of the growing season may delay the maturity of the crop (cotton, for example). It can also reduce the sugar content of sugar beets and the starch content of potatoes. Lodging may be a problem in grain crops.

B. UPTAKEOF POLLUTANTS AND LOCATION IN PLANT It is important to assess the absorption by plants of pollutants in wastewater even if they do not appear to affect the plants themselves, because

166

HERMAN BOUWER AND R. L. CHANEY

of the potential for injury to organisms higher in the food chain. The pollutants of major interest include toxic elements, chlorinated hydrocarbons, and other pesticides. Many ill-defined organic chemicals enter the wastewater of municipalities because of the complexity of our industrial and domestic effluents. Little information has been obtained on the reactions of the organic chemicals in wastewater with soils, or their uptake by plants. “Only a small number of pesticides have been investigated for uptake by plants,” is the way Nash (1974) summarized the state of knowledge in this area. In most cases, the amounts of persistent chlorinated hydrocarbons added with wastewater will be considerably lower than those added during normal agricultural operations. Some specific industrial compounds such as polychlorinated biphenyls (PCB’s) have no agricultural use. Several authors have recently reviewed plant accumulation of pesticides (Caro, 1969; Foy et al., 1971; Nash; 1974). Considerable variation in uptake was observed among plant species. External surfaces of root crops were heavily laden with organochlorine compounds when grown in soils containing such compounds, but leaves and root interiors had only very low amounts. Nash and Harris ( 1973) found that soybeans transported several organochlorine pesticides to the grain; other crops differed sharply in uptake and translocation (corn, oats, wheat). Little is known about the uptake of PCB’s by plants. G. B. Jones and T. D. Hinesly (personal communication, 1974) observed no increase in the PCB content of corn grain due to application of sewage sludge containing these compounds. The impact of land application of wastewater on soil and plant levels of pesticides has remained essentially unassessed. The toxic trace elements of major interest include arsenic, cadmium, copper, mercury, molybdenum, lead, selenium, and zinc. Plant absorption of these metals has been reviewed by Allaway (1968), Chaney (1973), Leeper (1972), Lisk (1972), Page (1973), and Tiffin et al. (1973). Again, plant species differ markedly in accumulation of these elements. The beet family accumulates large amounts of many of these elements in its leaves. Most plants exclude toxic trace elements from their seeds and fruits; most root crops exclude these elements from the edible root, although carrot roots accumulate considerable amounts of cadmium, lead, etc., from rnetal-enriched soils. The data shown in Table III suggest that the accumulation of toxic trace elements need not be a food chain hazard where “normal” domestic effluents are applied to land. Although foliar uptake or fixation of toxic elements from spray irrigated effluents could occur, this topic is similarly unresearched. In most cases the hazard will build up over decades in contrast to sludge application where a single application of some sludges can lead to potential food chain hazards.

LAND TREATMENT OF WASTEWATER

IV.

167

Selection and Design of System

Minimum impact on the environment and minimum total cost of operation are the two main design criteria for land treatment of liquid waste. The choice of system is largely controlled by soil and hydrogeologic conditions, and by the availability of land. If the available soils in a certain region have very low infiltration rates, about the only choice is an overland flow system (this term is preferred over spray-runoff systems because the wastewater does not necessarily have to be sprayed on the field but can also be applied with ditches, hydrants, or other techniques normally used in surface irrigation). Low-rate systems are the only possibility for soils that are not permeable enough for high-rate systems but too permeable for overland flow systems. Sandy loams and coarser-textured soils generally have sufficient permeability to permit high-rate systems, so that a choice between low-rate and high-rate systems must be made. Coarse sands, gravels, or shallow soils underlain by coarse material or by fractured or cavernous rock are usually not suitable for land treatment because of the hazards of groundwater contamination. While a deep water table ( 1 m or more) is desirable for land treatment fields (other than overland flow systems) , temporarily higher water tables can be tolerated, especially during infiltration, provided that the soil drains rapidly after wastewater infiltration stops. Even with very deep water tables, aerobic conditions may be restricted to the top meter or so of the soil profile (Pincince and McKee, 1968; Lance et al., 1973). Thus, water table depths of several meters as have sometimes been recommended (Robeck et al., 1964) may not be required under those conditions. The greater the depth to which the water table drops after the start of a resting period, the greater will be the amount of oxygen which enters the soil as air replaces the water draining from the soil. Entry of greater amounts of oxygen into the soil during drying would permit higher loadings of oxygen-demanding materials during wastewater application. The same would be true for porous soils, which have a higher oxygen diffusion rate than finer soils. Thus, the optimum schedule of infiltration and resting periods of a treatment field can be influenced by the drainage and diffusion parameters of the soil. In soils with restricted internal drainage, subsurface drains may be necessary to assure sufficiently rapid drainage of the soil profile after the start of a resting period. Overland flow systems should be designed to yield uniform, shallow depths of the flow above the soil and sufficient detention times. The slope of the fields should be high enough to give small depths of water on the soil but low enough to prevent channeling. The slopes in the system for

168

HERMAN BOUWER AND R. L. CHANEY

cannery wastes at Paris, Texas, were 2-6% and the overland flow distances were 60-90 m (Law et al., 1970; Gilde, 1973). The average application was about 1.3 cm/day, which was applied in 6-8 hours each day. Thomas (1973b) used slopes of 2-4%, overland flow distances of 36 m, and applications of 1-1.4 cm/day applied in about 8 hours, in his study with raw sewage. The loading rates and hence the land requirements for overland flow systems are approximately in the same category as those for low-rate infiltration systems, Overland flow systems require smooth topographies and uniform application of the wastewater at the upper end of the reach to prevent channeling. The vegetation may consist of Reed canarygrass in temperate climates and of bermudagrass in warm climates. Other grasses have also been used. Certain rushes (Scirpus Zacustris) have been reported to be very effective in removing pollutants from wastewater (Seidel, 1966). Low-rate systems are best suited for humid climates with relatively low evapotranspiration rates. Here, the salt content of the percolate will not be much higher than that of the original wastewater. Low-rate systems often permit normal agricultural use of the receiving fields. Compared to other land treatment systems, low rate systems normally will yield the best quality renovated water. If used on a large scale, however, native groundwater supplies may be affected over a large area. The spread of contamination may be difficult to control, raising concern over the long-term effects of diffuse sources (Walker, 1973). Wastewater may be applied with sprinklers, basins, or furrows, depending on the topography. Bendixen et al. (1968) reported no significant differences between these application techniques with respect to the performance of the system. Sands and other permeable soil may drain so rapidly that, when the wastewater is applied with sprinklers, sufficient air may enter the soil between sprinkler-head rotations to maintain aerobic conditions in the upper .,ortion of the soil. This will restrict denitrification and complete nitrificaion of the nitrogen in the wastewater can be expected, even after prolonbed application (Smith, 1971 ) . If wastewater is used for irrigation or similar low-rate systems in arid regions with high evapotranspiration rates, the salt content of the percolate will be much higher than that of the wastewater and the percolate will be unsuitable for groundwater recharge (Bouwer, 1974). Thus, artificial drainage may be required to remove the salty deep percolation water from the soil, as commonly practiced in irrigated agriculture. The large land requirements of low-rate systems may pose a problem when large volumes of effluent are to be applied. At an application rate of 2.5 cm/week for example, a city of 100,000 people would require some 1200 ha to dispose of its effluent. Muskegon County in Michigan has acquired 4000 ha of land northeast of the city of Muskegon to handle

LAND TREATMENT OF WASTEWATER

169

a domestic and industrial wastewater flow of 164,000 m3/day (Chaiken et al., 1973). Where permeable soils (loams, sandy loams, and fine sands) are available, high-rate systems are possible. The land requirements of high-rate systems may only be a few percent of those for low-rate systems. However, normal agricultural utilization of the land is often not feasible, requiring that the land be dedicated to wastewater renovation. High-rate systems are the only land treatment systems that can be used in warm, dry climates to yield renovated water that has about the same salt content as the original wastewater (Bouwer, 1974). This is important if the renovated water is to be reused. The spread of renovated water into the groundwater basin below highrate systems can be controlled by collecting the renovated water with drains if the aquifer is shallow, or with wells if the aquifer is deep (Bouwer, 1970, 1973c, 1974). The drains or wells must be located far enough from the infiltration system to allow sufficient time and distance of underground travel for the wastewater. The system can be so designed and operated that theoretically all the wastewater infiltrating into the soil can be collected by the wells or drains, without any renovated water moving into the groundwater outside the system of infiltration fields and collection facilities. This means that the portion of the aquifer between the infiltration and collection facilities is dedicated to renovation of wastewater. After collection, the renovated water can be used for unrestricted irrigation, recreation, industrial purposes, or it can be discharged into surface water. Domestic use of this water is not recommended until it is proven safe (Long and Bell, 1972; American Water Works Association Board of Directors, 1973). High-rate systems lend themselves for pre- or posttreatment of water in connection with advanced in-plant treatment of wastewater. This is done in various countries where low-quality surface water is used for municipal water supplies. Since the performance of a land-treatment system depends so much on the local soil, hydrogeology, and climate, as well as on the waste characteristics themselves, local experimentation and pilot projects are usually needed if land treatment is considered and local experience with such systems is not available. After installation of the full-scale project, good management, and monitoring of the system so that undesirable performance can be corrected before too much damage is done, are essential. REFERENCES Adams, A. P., and Spendlove, J. C. 1970. Science 169, 1218-1220. Alexander, M. 1971. “Microbial Ecology.” Wiley, New York. Allaway, W. H. 1968. Advan. Agron. 20, 235-274.

170

HERMAN BOUWER AND R . L. CHANEY

Allen, M. J., and Morrison, S. M. 1973. Ground Water 11, 6-10. American Water Works Association Board of Directors. 1973. J . Amer. Wafer Works Ass. 65, Part 11, 64. Andre, D. A., Weiser, H. H., and Maloney, G. W. 1967. J. Amer. Water Works Ass. 59, 503-508. Argo, D. G., and Culp, G. L. 1972. Water Sewage Works 119 (8), 62-65; (9), 128-132. Arizona State Health Department. 1972. “Proposed Rules and Regulations for Reclaimed Wastes.” Phoenix, Arizona. Baars, J. K. 1964. “Principles and Applications in Aquatic Microbiology,” Rudolfs Res. Conf. Proc., 1963, Chapter 17. Wiley, New York. Bache, B. W., and Williams, E. G. 1971. J . Soil Sci. 22, 289-301. Baier, D. C., and Fryer, W. B. 1973. J . Irrig. Drain. Div., Amcr. SOC. Civil Eng. 99(IR2), 133-141. Barth, E. F., and Ettinger, M. B. 1967. J . Water Pollut. Contr. Fed. 39, 1362-1368. Bauer, W. J., and Matsche, D. E. 1973. In “Recycling Treated Municipal Wastewater and Sludge Through Forest and Cropland” (W. E. Sopper and L. T. Kardos, eds.), pp. 345-363. Penn. State Univ. Press, University Park, Pennsylvania. Benarde, M. A. 1973. J . Amer. Warer Works Ass. 65, 432-439. Bendixen, T. W., Hill, R. D., Schwartz, W. A., and Robeck, G. G. 1968. J . Sanit. Eng. Div., Amer. SOC. Cicil Eng. 94, 147-157. Benson, N. R. 1966. Soil Sci. 101, 171-179. Berg, G., Dean, R. B., and Dahling, D. R. 1968. J . Amer. Water Works Ass. 160, 193-198. Biggar, J. W., and Fireman, M. 1960. Soil Sci. SOC.Amer., Proc. 24, 115-120. Blakeslee, P. A. 1973. I n “Recycling Municipal Sludges and Effluents on Land,” pp. 183-198. Nat. Ass. State Univ. and Land-Grant Colleges, Washington, D.C. Blood, J. W. 1963. N.A.A.S. Quart. Rev. 14, 97-100. Bondietti, E. A., and Sweeton, F. H. 1973. Agron. Abstr. p. 89. Bouwer, H. 1968. J . Soil Water Conserv. 23, 164-168. Bouwer, H. 1969. J. Irrig. Drain Div., Amer. SOC. Civil Eng. 95(IR1), 153-170. Bouwer, H. 1970. J . Sanit. Eng. Div., Amer. SOC. Civil Eng. 96(SAl), 59-74. Bouwer, H. 1973a. In “Recycling Treated Municipal Wastewater and Sludge Through Forest and Cropland” (W. E. Sopper and L. T. Kardos, eds.), pp. 164-175. Penn. State Univ. Press, University Park, Pennsylvania. Bouwer, H. 1973b. “Nitrification-denitrscation in the Soil.” Water Res. Center, University of Massachusetts, Boston. Bouwer, H. 1973c. In “Underground Waste Management and Artificial Recharge” (J. Braunstein, ed.), Vol. I, pp. 23-33. Amer. Assoc. Petrol. Geol., Tulsa, Oklahoma. Bouwer, H. 1974. Ground Water 12, 140-147. Bouwer, H., Rice, R. C., and Escarcega, E. D. 1974a. J . Water Pollitf. Contr. Fed. 46, 834-843. Bouwer, H., Lance, J. C., and Riggs, M. S . 1974b. J . Wuter Pollut. Contr. Fed. 46, 844-859. Bower, C. A., and Hatcher, J. T. 1967. Soil Sci. 103, 151-154. Brewer, R. F. 1966. In “Diagnostic Criteria for Plants and Soils” (H. D. Chapman, ed.), pp. 180-196. Div. Agr. Sci., University of California, Berkeley. Broadbent, F. E., and Clark, F. E. 1965. I n “Soil Nitrogen” (W. V. Bartholomew and F. E. Clark, eds.), Agron. Monogr. No. 10, pp. 344-359. Amer. SOC.Agr., Madison, Wisconsin.

LAND TREATMENT OF WASTEWATER

171

Brown, H. G., Hensley, C. P., McKinney, G. L., and Robinson, J. L. 1973. Environ. Lett. 5, 103-114. Brunner, D. R., and Sproul, 0. J. 1970. J . Sanit. Eng. Div., Amer. SOC. Civil Eng. 96(SA2), 365-379. Bunch, R., Barth, E., and Ettinger, M. 1961. J. Wafer Pollut. Contr. Fed. 33, 122-126. California Department of Water Resources. 196 1. “Feasibility of Reclamation of Water from Wastes in the Los Angeles Metropolitan Area,” Bull. No. 80. Calif. Dept. Water Resour., Sacramento. California Institute of Technology. 1969. “Investigation of Biochemical Stabilization of Aqueous Solutions of Organic Compounds by Unsaturated Flow through Porous Media.” J. E. McKee, Principal investigator. W. M. Keck Laboratory, CIT, Pasadena. California State Water Pollution Control Board. 1953. “Field Investigation of Water Reclamation in Relation to Groundwater Pollution,” Publ. No. 6. California State Water Resources Control Board. 1968. “Cannery Waste Treatment, Utilization and Disposal,” Publ. No. 39. Caro, J. H. 1969. Phyropathology 59, 1191-1197. Chaiken, E. I., Polonesik, S., and Wilson, C. D. 1973. Civil Eng. 43, 49-53. Chaney, R. L. 1973. In “Recycling Municipal Sludges and Effluents on Land,” pp. 129-141. Nat. Ass. State Univ. and Land-Grant Colleges, Washington, D.C. Cohen, J., and Shuval, H. I. 1973. Water, Air, Soil Pollut. 2, 85-95. Colston, N. V., and Smallwood, C., Jr. 1973. J. Water Pollut. Contr. Fed. 45, 1972-1977. Cookson, J. T., Jr. 1967. Sci. Technol. 1, 157-160. Cosgrove, D. T. 1973. J. Environ. Quai. 2,483-485. Craun, G. F., and McCabe, L. I. 1973. J . Amer. Water Works Ass. 65, 74-84. Culp, G. L., Culp, R. L., and Homann, C. L. 1973. J. Amer. Water Works Ass. 65, 641-646. Day, A. D. 1965. 1964 Wheat Newsleft. 11, 39-40. Day, A. D. 1973. Proc. Nut. Workshop Land Appl. Municipal Sludge Efluent, pp. 155-160. Day, A. D., and Kirkpatrick, R. M. 1973. J . Environ. Qual. 2, 282284. Day, A. D., and Tucker, T. C. 1959. Agron. J. 51, 569-572. Day, A. D., and Tucker, T. C. 1960. Agron. J . 52, 238-239. Day, A. D., Tucker, T. C., and Vavich, M. G. 1961. 1960 Oat Newsleft. 11, 29-3 1. Day, A. D., Tucker, T. C., and Vavich, M. G. 1962. Agron. J . 54, 133-135. Day, A. D., Dickson, A. D., and Tucker, T. C. 1963. Agron. J . 55, 317-318. Day, A. D., Stroehlein, J. L., and Tucker, T. C. 1972. J . Water Pollut. Contr. Fed. 44, 372-375. Deaner, D. G., and Kerri, K. D. 1969. J . Amer. Wafer Works Ass. 61, 465-468. Drewry, W. A., and Eliassen, R. 1968. J . Water Pollut. Contr. Fed. 40, R257-R271. Dunlop, S . G. 1968. In “Municipal Sewage Effluent for Irrigation” ( C . W. Wilson and F. E. Beckett, eds.), pp. 107-121. Louisiana Polytech. Inst., Ruston. Eaton, F. M. 1944. J. Agr. Res. 69, 237-277. Egeland, D. R. 1973. J. Water Polluf. Contr. Fed. 45, 1465-1475. Ellis, B. G. 1973. In “Recycling Treated Municipal Wastewater and Sludge Through

172

HERMAN BOUWER AND R. L. CHANEY

Forest and Cropland” (W. E. Sopper and L. T. Kardos, eds.), pp. 46-70. Penn. State Univ. Press, University Park, Pennsylvania. Ellis, B. G., and Erickson, A. E. 1969. “Movement of and Transformation of Various Phosphorus Compounds in Soil” (mimeo.). Michigan State University, East Lansing. Ellis, B. G., and Knezek, B. D. 1972. In “Micronutrients in Agriculture” (J. J. Mortvedt et al., eds.), pp. 59-78. Soil Sci. SOC.Amer., Madison, Wisconsin. Environmental Protection Agency. 1973. “Pretreatment of Discharges to Publicly Owned Treatment Works (40 CFR 128).” U.S. Environ. Protect. Agency, Off. Water Programs Operations, Washington, D.C. Erickson, A. E., Tiedje, J. M., Ellis, B. G., and Hansen, C. M. 1972. Proc. Cornell Agr. Waste Manage. Conf., pp. 405-410. Fair, J. F., and Morrison, S . M. 1967. Water Resour. Res. 3, 799-803. Folkman, Y.,and Wachs, A. M. 1970. Proc. Amer. SOC.Civil Eng. 96, 675-690. Follett, R. H., and Lindsay, W. L. 1971. Soil Sci. SOC.Amer., Proc. 35, 600-602. Foster, D. H., and Engelbrecht, R. S. 1973. In “Recycling Treated Municipal Wastewater and Sludge Through Forest and Cropland” (W. E. Sopper and L. T. Kardos, eds.), pp. 247-270. Penn. State Univ. Press, University Park, Pennsylvania. Foy, C. L., Coats, G. E., and Jones, D. W. 1971. In “Respiration and Circulation” (P. L. Altman and D. S. Dittmer, eds.), Biol. Handb., pp. 743-791. Fed. Amer. SOC.Exp. Biol., Bethesda, Maryland. Gilde, L. C. 1973. J . lrrig. Drain. Div., Amer. SOC.Civil Eng. 99(IR3), 339-352. Gilliam, J. W., and Sample, E. C. 1968. Soil Sci. 106, 352-357. Goss, D. W., and Jones, 0. R. 1973. In “Underground Waste Management and Artificial Recharge” (J. Braunstein, ed.), Vol. I, pp. 468-480. Amer. Ass. Petrol. Geol., Tulsa, Oklahoma. Griffin, R. A., and Jurinak, J. J. 1973. Soil Sci. SOC. Amer., Proc. 37, 847-850. Hallam, M. J., and Bartholomew, W. V. 1953. Soil Sci. SOC. Amer., Proc. 17, 365-368. Hatcher, J. T.,and Bower, C. A. 1958. Soil Sci. 85, 319-323. Hinesly, T. D. 1973. Water Spectrum 5, 1-8. Hodgson, J. F. 1963. Advan. Agron. 15, 119-159. Hook, J. E., Kardos, L. T., and Sopper, W. E. 1973. In “Recycling Treated Municipal Wastewater and Sludge Through Forest and Cropland” (W. E. Sopper and L. T. Kardos, eds.), pp. 200-219. Penn. State Univ. Press, University Park, Pennsylvania. Hopkins, G. J., Neel, J. K., and Nelson, F. L. 1956. Sewage Inn. Wastes 28, 1466-1474. Jenne, E. A. 1968. Advan. Chem. Ser. 73, 337-387. Jones, G.B., and Belling, G. B. 1967. Aust. J . Agr. Res. 18, 733-740. Jones, G. B., Riceman, D. S., and McKenzie, J. 0. 1957. Aust. J . Agr. Res. 8, 190-20 1. Kardos, L. T. 1967. In “Agriculture and the Quality of our Environment,” Publ. NO. 85, pp. 241-250. Amer. Ass. Advan. Sci., Washington, D.C. Kardos, L. T., and Sopper, W. E. 1973. In “Recycling Treated Municipal Wastewater and Sludge Through Forest and Cropland” (W. E. Sopper and L. T. Kardos, eds.), pp. 148-163. Penn. State Univ. Press, University Park, Pennsylvania. Khin, A., and Leeper, G. W. 1960. Agrochimica 4, 246-254. Kirby, C. F. 1971. “Sewage Treatment Farms” (Post graduate course in Public

LAND TREATMENT OF WASTEWATER

173

Health Engineering), Sess. No. 12. Dept. Civil Eng., University of Melbourne, Australia. Knezek, B. D. 1972. Mich., Univ., Inst. Water Res. [Tech. Rep.] 30, 27-43. Krishnaswami, S. K. 1971. Can. J. Pub. Health 62, 36-41. Krone, R. B. 1968. In “Municipal Sewage Effluent for Irrigation” (C. W. Wilson and F. E. Beckett, eds.), pp. 75-106. Louisiana Polytech. Inst., Ruston. Lance, J. C. 1972. J. Water Pollut. Contr. Fed. 44, 1352-1361. Lance, J. C., and Whisler, F. D. 1972. J . Environ. Qual. 1, 180-186. Lance, J. C., and Whisler, F. D. 1973. Proc. Int. Conf. Land Waste Manage., 1973 (in press), Lance, J. C., Whisler, F. D., and Bouwer, H. 1973. J. Environ. Qual. 2, 345-350. Larsen, S., and Widdowson, A. E. 1971. J. Soil Sci. 22, 210-221. Larson, W. E. 1960. I n “Algae and Metropolitan Wastes, Transactions 1960 Seminar,” pp. 125-129. U.S. Dept. of Health, Education and Welfare, Washington, D.C. Law, J. P., Jr. 1968. “Agricultural Utilization of Sewage Effluent and Sludge,” An annotated bibliography. Fed. Water Pollut. Contr. Admin., U.S. Dept. of Interior. Law, J. P., Jr., Thomas, R. E., and Myers, L. H. 1970. J. Water Pollrit. Contr. Fed. 42, 1621-1631. Leeper, G. W. 1972. “Reactions of Heavy Metals with Soils with Special Regard to their Application in Sewage Wastes,” Contract No. DACW73-73-C-0026. Dept. of Army, Corps Engineers, Washington, D.C. Lehman, G. S., and Wilson, L. G. 1971. Water Resour. Res. 7 , 90-99, Levin, G . V., Topol, G. .I. Tarnay, , A. G., and Samworth, R. B. 1972. 1. Water Pollut. Contr. Fed. 44, 1940-1954. Lindsay, W. L. 1973. I n “Recycling Municipal Sludges and Effluents on Land,” pp. 91-96. Nat. Ass. State Univ. and Land-Grant Colleges, Washington, D.C. Lindsay, W. L., and Moreno, E. C. 1960. Soil Sci. SOC. Amer., Proc. 24, 177-182. Lisk, D. J. 1972. Advan. Agron. 24, 267-325. Loehr, R. C. 1968. “Pollution Implications of Animal Wastes-A Forward Oriented Review.” Robert S. Kerr Water Res. Cent., Environmental Protection Agency, Ada, Oklahoma. Long, W. H., and Bell, F. A., Jr. 1972. J . Amer. Wafer Works Ass. 64, 220-225. McGauhey, P. H., and Krone, R. B. 1967. “Soil Mantle as a Wastewater Treatment System,” SERL Rep. No. 67-1 1. University of California, Berkeley. Mack, W. H. 1973. J. Amer. Water Works Ass. 65, 347-348. McMichael, F. C., and McKee, J. E. 1965. “Research on Waste Water Reclamation at Whittier Narrows.” W. M. Keck Lab. Environ. Health Eng., California Institute of Technology, Pasadena. Mallman, W. L., and Mack, W. H. 1961. “Groundwater Contamination,” Tech. Rep. W61-5, pp. 35-43. U S . Pub. Health Serv., Washington, D.C. Masinova, L., and Cledova, V. 1957. “Evaluation of the Positive Test for Coliform Bacteria in Sanitary Tests of Water and Soil,” Rep. No. 13-14. Inst. Hyg., Prague. Matlock, W. G. et a/. 1972. “Sewage Effluent Pollution of a Groundwater Aquifier.” Amer. SOC.Agr. Eng., Chicago, Illinois. Merrell, J. C., Jr., and Ward, P. C. 1968. J . Ainer. Water Works Ass. 60, 145-153. Miller, R. H. 1973. In “Recycling Treated Municipal Wastewater and Sludge Through Forest and Cropland” (W. E. Sopper and L. T. Kardos, eds.), pp. 71-94. Penn. State Univ. Press, University Park, Pennsylvania.

174

HERMAN BOUWER AND R. L. CHANEY

Mitchell, R. L. 1964. In “Chemistry of the Soil” (F. E. Bears, ed.), 2nd ed., pp. 320-368. Van Nostrand-Reinhold, New York. Moore, D. R. E., and Waid, J. S. 1971. Soil Biol. & Biochem. 3, 69-83. Murphey, W. K., Brisbin, R. L., Young, W. J., and Cutter, B. E. 1973. In “Recycling Treated Municipal Wastewater and Sludge Through Forest and Cropland” ( W. E. Sopper and L. T. Kardos, eds.), pp. 295-310. Penn. State Univ. Press, University Park, Pennsylvania. Murphy, W. H., and Syverton, J. T. 1958. Virology 6, 623-636. Murphy, W. H., Eylar, 0. R., Schmidt, E. L., and Syverton, J. T. 1958. Virology 6, 612-622. Murrmann, R. P., and Koutz, F. R. 1972. In “Wastewater Management by Disposal on the Land” (S.C. Reed, ed.), CRREL Spec. Rep. No. 171, pp. 48-76. Mytelka, A. I., Czachor, J. S., Guggino, W. R., and Golub, H. 1973. J . Water Pollut. Contr. Fed. 45, 1859-1864. Nash, R. G. 1974. In “Pesticides and their Effects on Soils and Water,” SpecPubl. No. 8 (revised), Chapter 11. Soil Sci. SOC. Amer., Madison, Wisconsin (in press). Nash, R. G., and Harris, W. G. 1973. J. Environ. Qual. 2, 269-273. National Academy of Science-National Academy of Engineering. 1974. “Water Quality Criteria 1972” (in press). National Technical Advisory Committee to the Secretary of Interior. 1968. “Water Quality Criteria.” Environmental Protection Agency, Washington, D.C. Nevo, Z., and Mitchell, K. 1967. Water Res. 1, 231-236. Ng, S. K., and Bloomfield, C. 1962. Plant Soil 16, 108-135. Nilsson, R. 1971. Water Res. 5, 51-60. Norvell, W. A. 1972. In “Micronutrients in Agriculture” (J. J. Mortvedt et al., eds.), pp. 115-138. Soil Sci. SOC.Amer., Madison, Wisconsin. Okazaki, E., and Chao, T. T. 1968. Soil Sci. 105, 255-259. Ongerth, H. J., Spath, D. P., Crook, J., and Greenberg, A. E. 1973. J. Amer. Water Works Ass. 65, 495-498. Page, A. L. 1973. “Fate and Effects of Trace Elements in Sewage Sludge when Applied to Agricultural Lands. Literature Review Study for Ultimate Disposal Research Program.” Environmental Protection Agency, Cincinnati, Ohio. Peterson, J. R., and Geschwind, J. 1972. J . Environ. Qual. 1, 410-412. Peterson, J. R., Lue-Hing, C., and Zenz, D. R. 1973. In “Recycling Treated Municipal Wastewater and Sludge Through Forest and Cropland” (W. E. Sopper and L. T. Kardos, eds.), pp. 26-37. Penn. State Univ. Press, University Park, Pennsylvania. Pincince, A. B., and McKee, J. E. 1968. J . Sanif. Eng. Div., Amer. SOC. Civil Eng. 94, 1093-1 119. Porges, R., and Hopkins, G. J. 1955. Sewage Znd. Wastes 27, 1160-1170. Pound, C. E., and Crites, R. W. 1973a. In “Recycling Municipal Sludges and EWuents on Land,” pp. 49-61. Nat. Ass. State Univ. and Land-Grant Colleges, Washington, D.C. Pound, C. E., and Crites, R. W. 1973b. “Wastewater Treatment and Reuse by Land Application,” Vols. I and 11, EPA-660/2-73-006a and b. U.S. Govt. Printing Office, Washington, D.C. Prince, A. L., Bear, F. E., Brennan, E. G., Leone, I. A., and Daines, R. H. 1949. Soil Sci. 67, 269-277. Rand, W. E., and Schmidt, H. J. 1952. Amer. J . Vet. Res. 13, 50-61.

LAND TREATMENT OF WASTEWATER

175

Randall, A. D. 1970. J. Amer. Water Works Ass. 62, 716-720. Rhoades, E. D. 1964. Trans. ASAE (Amer. SOC.Agr. Eng.) 7, 165-166. Rhoades, J. D., Ingvalson, R. D., and Hatcher, J. T. 1970. Soil Sci. Sco. Amer., Proc. 34, 871-875. Rice, R. C. 1974. J . Water Pollut. Contr. Fed. 46, 708-716. Richards, L. A., ed. 1954. “Diagnosis and Improvement of Saline and Alkali Soils,” Handb. No. 60. U.S.Dept. of Agriculture, Washington, D.C. Robeck, G. G., Bendixen, T. W., Schwartz, W. A., and Woodward, R. L. 1964. J . Water Pollut. Contr. Fed. 36, 971-983. Rohde, G. 1962. J . Inst. Sewage Purification 1962 pp. 581-585. Romero, J. C. 1970. Ground Water 8, 37-49. Rose, W. W., Merier, W. A., Katsuyuma, A., Sternberg, R. W., Brauner, G. V., Olsen, N. A., and Weckel, K. G. 1971. Proc. Nut. Symp. Food Process. Wastes, Pac. Northwest Water Lab., EPA Nat. Canners’ Ass., 2nd. pp. 109-127. Rudolfs, W., Falk, L. L., and Ragotzkie, R. A. 1950. Sewage Ind. Waste 22, 1261-1 28 1. Schmidt, K. D. 1972. Ground Water 10, 50-61. Schneider, I. F., and Erickson, A. E. 1972. Mich., Agr. Exp. Sta., Res. Rep. 195. Seidel, K. 1966. Naturwissenschaften 53, 289-297. Shuval, H. I., and Gruener, N. 1973. Environ. Sci. Technol. 7 , 600-604. Smith, P. F., Rasmussen, G. K., and Hrnciar, G. 1962. Soil Sci. 94, 235-238. Smith, T. P. 1971. “Actual Spray Field Operations,” Proc. Land Spreading Conf., Pap. No. 8. East Central Florida Regional Planning Council, Orlando, Florida. Sopper, W. E., and Kardos, L. T. 1973. In “Recycling Treated Municipal Wastewater and Sludge Through Forest and Cropland” (W. E. Sopper and L. T. Kardos, eds), pp. 271-294. Penn. State Univ. Press, University Park, Pennsylvania. Splittstoesser, D. F., Downing, D. L. 1969. N. Y., Agr. Exp. Sta., Geneva, Res. Circ. 17, 1-15. Stefanson, R. C. 1973. Aust. J. Soil Res. 10, 183-195. Stevens, R. M. 1972. “Green Land-Clean Streams,” Report by Center for the Study of Federalism. Temple University, Philadelphia, Pennsylvania. Stone, R., and Garber, W. F. 1952. Trans. Amer. SOC.Civil Eng. 117, 1189. Sullivan, R. H., Cohn, M. M., and Baxter, S. S. 1973. “Survey of Facilities Using Land Application of Wastewater,” EPA-430/9-73-006. U.S. Govt. Printing Office, Washington, D.C. Tanji, K. K. 1970. Soil Sci. 110, 44-51. Thomas, R. E. 1973a. I . Water Pollut. Contr. Fed. 45, 1476-1484. Thomas, R. E. 1973b. Proc. Int. Conf. Land Waste Manage., 1973 (in press). Thomas, R. E. 1973c. Proc. Conf. Land Disposal Municipal EfPuents Sludges EPA-902/9-73-001, pp. 181-200. Thomas, R. E., and Bendixen, T. W. 1969. J. Water Pollut. Contr. Fed. 41, 808-813. Thomas, R. E., Schwartz, W. A., and Bendixen, T. W. 1966. Soil Sci. SOC. Amer., Proc. 30, 641-646. Tiffin, L. O., Lagerwerff, J. V., and Taylor, A. W. 1973. “Heavy Metal and Radionuclide Behavior in Soils and Crops. A Review,” AEC Res. Contract AT(49-7)-1, 182 pp. Trocme, S., Barbier, G., and Chabannes, J. 1950. Ann. Agron., Ser. A 1, 663-685. Truesdale, G. A., Birkbeck, A. E., and Shaw. D. 1964. In “Proceedings of The Institute of Sewage Purification,” Part I, pp. 3-23. Water Pollut. Res. Lab., Stevenage, U. K.

176

HERMAN BOUWER AND R. L. CHANEY

Van Donsel, D. J., Geldreich, E. E., and Clarke, N. A. 1967. Appl. Microbiol 15, 1362-1370. Vela, G. R., and Eubanks, E. R. 1973. J . Water Pollut. Contr. Fed. 45, 1789-1794. Viets, F. G., Jr. 1973. “Nitrogen Transformations in Organic Waste Applied to Land,” Int. Rep., Agr. Res. Serv., US. Department of Agriculture, Washington, D.C. Volk, B. G. 1970. Ph.D. Thesis, Michigan State University, Ann Arbor (Diss. Abstr. B 31, 63-79). Walker, W. H . 1973. Ground Water 11, 11-20. Wells, D. M., and Sweazy, R. M. 1973. Proc. Znt. Conf. Land Waste Manage., 1973 (in press). Wentink, G. R., and Etzel, J. E. 1972. J. Water Pollut. Contr. Fed. 44, 1561-1574. Wheatland, A. B., and Borne, B. J. 1961. Water Waste Treat. I . 8, 330-335. Wilson, L. A., and Lehman, G. S. 1966. “Grass Filtration of Sewage Effluent for Quality Improvement Prior to Artificial Recharge.” Amer. SOC.Agr. Eng., Chicago, Illinois. Wilson, L. G., and Lehrnan, G. S. 1967. “Progressive Agriculture in Arizona,” July-August issue, pp. 22-24. College of Agriculture, University of Arizona, Tucson. Woldendorp, J. W. 1963. “The Influence of Living Plants on Denitriftcation,” Bull. No. 63( 13). Agr. Univ., Wageningen, The Netherlands. Young, R. H. F., and Burbank, N. C., Jr. 1973. J . Amer. Water Works Ass. 65, 598-603.

BIOMASS PRODUCTIVITY OF MIXTURES 6. R. Trenbath' Waite Agricultural Research Institute, University of Adelaide, Adelaide, South Australia

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Comparison of Yields of Mixtures and Monocultures . . . . . . . . . . . . . . . . . Theoretical Considerations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Types of Interaction Causing Nontransgressive Deviations of Mixture Yields from Mid-Monoculture Values .............................. V. Mechanisms Capable of Causing Transgressive Yielding by Mixtures . . . . VI. Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

I. 11. 111. IV.

I.

177 179 183 186 196 205 206

Introduction

Mixtures of field crops are still extensively grown in primitive agriculture, but where more advanced methods are used, monocultures are usual. Claims have frequently been made, however, that communities with some degree of genotypic heterogeneity have advantages over pure stands. The alleged advantages have included one or more of the following: higher yields, lower variability of yield from season to season, a better spreading of production over the growth period, less susceptibility to disease or lodging, and an improved quality of the crop product. The yield advantage that would justify the growing of a mixture has commonly been thought to be a higher yield from the mixture than from an equal area divided between monocultures of the components in the same proportion as that in which they occur in the mixture (e.g., Jensen, 1952). Numerous reports of this kind of mixture advantage [reviewed in Jensen (1952) and in Simmonds ( 1962)] gave the impression that mixed cropping was highly desirable. More recently the advantage required to justify mixed culture has been realized to be a superiority of the yield of the mixture over that of the better (or best) of its components grown in monoculture over the whole of the same area (Donald, 1963; Frey and Maldonado, Present address: Research School of Biological Sciences, Australian National University, Canberra City, Australia.

177

178

B. R . TRENBATH

1967). Judged from this standpoint, the performance of mixtures seems much less encouraging. Indeed, mixtures occasionally yield less than the less (or least) productive monoculture, a type of result which tends to receive little emphasis. Although a yield and quality advantage has been firmly established for legume-grass mixtures grown for forage, the experimental substantiation of claims for other combinations, such as varieties of field crops, has proved difficult. This has been due to the usually small size of the advantage and its apparent dependence on relatively specific co.mbinations of biological material and environment. However, the possibility of gaining benefits from the simple mixing of seeds holds such attractions for agriculturalists that research into the productivity of mixtures continues. There are three main objectives in this research: the first is to screen mixtures composed of more-or-less randomly selected genotypes for “lucky,” high-yielding combinations. The second objective is to test the alleged advantages of traditionally grown mixtures, particularly those of some less-developed countries. The third objective is to gain an understanding of the processes which lead to mixture advantages so that in a specific environment, a rational choice of components may produce mixtures showing benefits unobtainable from pure cultures. Since there is much literature concerning the yields of mixtures (reviewed by Aiyer, 1949; Simmonds, 1962; Baldy, 1963; Donald, 1963), the present survey treats only a limited part of what has been written on this subject. The first limitation refers to sources: the reports considered have mostly appeared in English-language and West European journals and dissertations. The second limitation is that grazed crops are not considered although crops cut for fodder are included. A third restriction refers to experimental design: in order to be included in the section concerned with yields, reports must have described experiments in which biomass (dry-matter yield of shoots) was measured in mixtures of pairs of components planted simultaneously in 1 : 1 proportions, and in which monocultures of the components were grown at the same total density within the same experiment. The numerous reports of experiments involving mixtures of cereal varieties (see Simmonds, 1962) mostly consider only grain yield; since the degree of correlation between grain yield and shoot weight is uncertain, reference will be made to such reports only where a point cannot be adequately illustrated by data on biomass. The subject of mixtures of legumes and nonleguminous species is not considered because a special nutrient relationship is involved in the interaction between the components; this type of interaction seems to be already much better understood than many of the types of interaction occurring in the mixtures to be considered in the present work.

BIOMASS PRODUCTIVITY OF MIXTURES

179

Apart from a section on underlying theory, this review will consider evidence concerning the following topics : the biomass yields of mixtures compared with those of their components’ monocultures, mechanisms causing minor (nontransgressive) deviations from the “mid-monoculture” yield midway between the yields of the components’ monocultures, and mechanisms causing transgressive deviation, so that the yield of the mixture is either less than that of the lower-yielding monoculture or more than that of the higher-yielding monoculture.

II.

Comparison of Yields of Mixtures and Monocultures

To facilitate description, a number of terms and symbols are introduced. The mid-monoculture yield is denoted by P. A mixture will be said to have “overyielded” when the mixture biomass, M , has exceeded that of the more productive pure culture, P I ; i.e., when M > P,. The mixture will be said to have “underyielded” when the mixture biomass has fallen below that of the less productive pure culture, P,, i.e., when M < P,. These two “transgressive” situations will be shown to have occurred with a relatively low frequency compared with the cases where P , > M > P,. Since an understanding of the reasons for transgressive yielding by mixtures may enable the agriculturalist to produce overyielding at will, attention has been particularly focused on reports of its occurrence. In his review of the performance of mixtures, Simmonds (1962) was concerned almost entirely with the grain yield of cereals. His conclusions are of interest, however, because they are so similar to those of Donald ( 1963), who considered yields of biomass. Simmonds found that often M = P, sometimes P, > M > P, and occasionally M > P,. He noted that “negative” interactions (i.e., M < P ) seemed to be rare. Reviewing evidence concerning the possibility that mixtures might overyield in biomass, Donald (1963) concluded that the yields of mixtures usually lay between PI and P,, and that there was no substantial evidence to show that a mixture of two genotypes can fix more carbon than the more productive of the two genotypes grown in monoculture. Van den Bergh (1968) similarly concluded that in almost all experiments, the mixed culture yielded less than the monoculture of the more productive species. In the same way neither Donald (1963) nor van den Bergh (1968) found strong evidence of underyielding. To show the relationship between the observed yields of binary mixtures and the yields of their components grown in monoculture, the results of 344 mixtures are summarized in Table I. The data are taken from published reports of experiments in which the experimental design known as

180

B. R. TRENBATH

TABLE I The Distribution of Biomass Yields of Varietal or Interspecific Mixtures Compared with Yields of Their Components’ Monocultures, Based on Published Data of 344 Mixtures” Below

Pz

Crop Grasses

PftoP P t o p 1

Above

PI

3

Either grasses or leguniesc

9 1 1 9 12 1

7 38 4

1 9 2 11 10 1 8 42 5

Grasses“

5

9

11

15

Grasses Grasses“ Grasses Rye

2 2 1 4

2 7 4 4

1 3 5 16

1 4 7

-

-

-

-

45

92

124

83

3 1

1

Wheat Rice A series of nonlegumes~ Subterranean clover Flax and linseed Grasses“ Barley

12 1

2 3 1 4 3 1 12 a8

a

Author Ahlgren and Aamodt (1939) Aberg et al. (1943) Donald (1946) Sakai (1953) Sakai (1955) Williams (1962) Williams (1963) Harper (1965) England (1965) Norrington-Davies (1967) Norrington-Davies (1968) Rhodes (1968b) Thomson (1969) Rhodes (1970a) nor rington-Davies and Hutto (1972) = 344

62.8% c _ _ _

39.8% a

_T_I

60.2%

P I and PZ are the yields of the higher- and lower-yielding monocultures, respectively.

13 is the mid-monoculture yield.

” Data derived from a series of cuts. c

Mixtures of leguminous and nonleguminous species have been omitted.

a “mixture diallel” was used. Experiments conforming to this design contain a series of genotypes grown in monoculture and all possible binary combinations. All plots (or containers) are sown with a constant overall density of plants, and 1 : 1 proportions are used in the mixtures. Such mixture experiments satisfy the conditions specified earlier for consideration in this part of the review. Although replication is not specifically mentioned by Harper (1965), it is believed that all data used in Table I were based on replicated experiments. The data of Jacquard and Caputa (1970) were reported as being unreplicated and are therefore not included. Taken together, the data show that mixture yields tend to lie above the mean yields

BIOMASS PRODUCTIVITY OF MIXTURES

181

of the monocultures; the difference between 60.2% and 39.8% is highly significant ( P < 0.001 ). Presumably reflecting this difference, the reported frequency of overyielding is significantly ( P < 0.001) greater than that of underyielding. According to these results, apparently transgressive yielding is by no means uncommon, comprising 37% of all the mixtures considered. Although differences between overyielding mixtures and their higher-yielding monocultures have often not been examined for statistical significance, of those so examined, very few have reached even the 5 % probability level. As far as the author is aware, there are only four investigations where overyielding has been reported as being statistically significant. The first case (Whittington and O’Brien, 1968) concerns the “significant” overyielding by three mixtures of grasses in the third year of a field experiment. This overyielding affected only the treatment with frequent cutting, but there was also a tendency toward it in the less-frequent cutting treatment. It should be noted, however, that the statistical significance of the overyielding was tested using selected data (6 out of 15 harvest periods); van den Bergh ( 1968) has observed that apparently highly productive mixtures in some cuts may be quite the reverse in other cuts. Since the mixtures of Whittington and O’Brien usually yielded below PI in the first two years of the experiment, there may have been extra reserves of nutrients in the soil to provide the high yields of the third year. It must also be realized that the apparent degrees of overyielding of several mixtures in a diallel may be simultaneously affected if, owing to a chance effect, the monoculture yield of one high-yielding genotype is considerably lower (or higher) than its potential as expressed in other plots. Since the measures of overyielding of mixtures in this diallel were not independent at any one harvest (or between harvests), the multiple instances of overyielding reported by Whittington and O’Brien do not provide any especially strong grounds for believing that the overyielding was a real effect. The second case in which overyielding achieved significance ( P << 0.01 in two independent comparisons) was reported by Rhodes ( 1968b). In a diallel of ryegrass varieties, under two regimes of cutting (high and intermediate frequency), a particular mixture overyielded by 12% and 15%. In the infrequent-cutting treatment, this mixture’s yield was below P. The plots with the different cutting regimes were independent, and so there seems to be reason to accept this result as due to a real effect. The third case of significant overyielding was also reported by Rhodes (1970a) in two mixtures of ryegrass genotypes grown under near-optimal soil conditions. These mixtures overyielded by 11% ( P < 0.01) and 9% (P < 0.05) under a regime of infrequent cutting; with frequent cutting, the same mixtures did not yield transgressively.

182

B. R. TRENBATH

A fourth instance of “significant” overyielding has been reported which, although lacking clear statistical treatment, seems also to be due to a real effect. Data of Khan (in Harper, 1965) and data of an independent experiment by Harper (1965) both show that mixtures of various flax and/or linseed varieties overyielded. Khan’s data show two flax-linseed mixtures as overyielding (significantly,” Harper, 1965, p. 471) by 13% and 14%. Harper’s own experiment included the same two mixtures. At low density, one overyielded by 38% while the yield of the other was nontransgressive; at high density, neither mixture yielded transgressively. Since, at low density, only two plants were present in each pot, the experimental errors must have been very large. Furthermore, Harper mentioned neither replication nor significance levels, and so the value of this experiment as an independent confirmation of Khan’s results is uncertain. If Harper’s experiment was in fact adequately replicated, the high proportion of overyielding mixtures (40% ) would suggest that his mixtures were worth further investigation. While not based on the same varieties, data of Obeid and Harper (in Harper, 1968) again show a flax-linseed mixture (planted this time with a range of proportions) which overyielded by a maximum of l o % , 29%, and 31% at the three densities studied. The flax-linseed (WEIRA-VALUTA) mixture of Khan remains, however, the only mixture which has, to this author’s knowledge, ever been reported to overyield in two separate experimenkl Little consideration has been given by authors to the possibility that mixtures may underyield. No cases of significant underyielding appear to have been reported, although Ahlgren and Aamodt (1939) found that all three mixtures in a incomplete diallel of grass species underyielded (nonsignificantly, by 11 % , 23 %, and 26% ) . Donald ( 1946) repeated this experiment using the same species, but none of the mixtures underyielded. As mentioned previously, Donald (1963) and van den Bergh (1968) have each concluded that there is no firm evidence that a mixture can have an advantage over the higher-yielding component monoculture (mixtures of legume and nonlegume are excepted), Also, Woodford (1 966) has expressed the same opinion. The additional evidence presented in this review is consistent with these conclusions. However, it seems that “firm evidence” has rarely been sought. Experiments showing transgressive yielding by mixtures have either seldom been repeated, or if they have, the results have not been published. Until the transgressive yielding of more mixtures can be shown to be repeatable, the addition of further data concerning once‘There are only two analogous cases in the production of grain. One is the mixture of rice varieties (BK and 2A) of Roy (1960), and the other is a mixture of wheat varieties (RAMONA and BAART) of Chapman ef al. (1969) and Allard and Adams (1969).

BIOMASS PRODUCTIVITY OF MIXTURES

183

performed experiments can add little to what is already known about mixture performance. 111.

Theoretical Considerations

Let us assume that a 1 : l mixture of the seeds of two genotypes has been sown along with monocultures of the components. If plants of each component yield the same as they do in their respective monoculture, the yield of the mixture will be the mid-monoculture yield ( P ) . The plants of the components of the mixture will be said to have given their “expected yields” (Alcock and Morgan, 1966) since they have yielded in accordance with their genotypic potential, as expressed in monoculture. The midmonoculture yield will be the corresponding “expected yield” of the mixture. Writing the per-plant yield of genotype i in mixture with genotype j as Y i j ,the average per-plant yield, M i j , of the i, j mixture will be: Mij

=

1 -(Yij

a

+

Yji)

If per-plant yields of i and j in the mixture are the same as in their monocultures, i.e., if Y i j = Yii and Y j i = Y j j , the mixture yield is the “expected yield,” i.e., M =: P. In such a mixture, the average per-plant yield would be given by Mij

=

1

-(Yii

a

+ Y,jj)

If the density of plants is n plants per plot of unit area, then M ,P,,and P, (the yields per unit area of the mixture and of the higher and lower

yielding monocultures, respectively) are given by

M

=

nM;,

PI = nYii (for Yii > Pz = nYjj (for Y ; ; >

Yjj) Yjj)

However, as many early experiments showed (e.g., Montgomery, 1912; Tansley, 191 7; Clements and Weaver, 1924; Sukatschew, 1928), per-plant yields of genotypes in mixture and monoculture are seldom the same. From this springs the tendency for mixture yields to deviate from the “expected yield.” The biological processes responsible for the deviation of component performance from that expected are complex and varied. A general term which has been applied to them is “interference” (Crombie, 1947; Harper,

184

B. R. TRENBATH

1961). They result in what may be called either “interference effects” or “neighbor effects” (“Nachbarwirkungen,” Lampeter, 1960). The latter is a more objective and neutral term and, hence, perhaps preferable (Trenbath and Harper, 1973). The best understood, and perhaps the most important, mechanism that can cause the biomasses of plants of a genotype to differ between mixture and monoculture is the process of “competitionyy(Clements et al., 1929; de Wit, 1960; Donald, 1963). Plants are conceived as “competing” for the limited supplies of environmental resources necessary for growth. Clements (1904) showed how the anthropomorphic overtones of this concept could be avoided when he asserted that the reaction of a plant to neighboring individuals is not a direct response to the neighbors themselves, but to the plant’s own microenvironment insofar as it has been altered by the presence of the neighbors. In a mixture, differences of morphology and physiology between the mixture components cause their individuals to experience different microenvironments and hence different resource availabilities from those experienced by plants of the same genotype in monoculture. As a consequence, resources are unevenly shared between the components and per-plant mixture yields deviate from “expected yields.” Such effects are said to be due to competition. De Wit ( 1960) has presented a model of intergenotypic competition based on the simple assumption that the biomass yield of each component is strictly proportional to the share of environmental resources it can acquire, According to this model, if the sharing is uneven, plants of one genotype, say i, will be larger in mixture than in monoculture while plants of the other component, genotype 1, will be correspondingly smaller. In such a case, genotype i is termed the “aggressor” (Donald, 1946) and genotype j may be termed the “subordinate.” Hence, according to this simple model of unequal sharing of environmental resources, Yij > Yii and Yji < Y j j .This appears to be the commonest situation in mixtures, for of 70 results reviewed by Donald* (1963), 51 (74%) were of this type. Among the 344 results reviewed in Section 11, there are 326 cases where data are available of the performance of the components within the mixtures; of these 326 mixtures, 255 (78%) showed the same pattern. The more specific predictions of the de Wit (1960) model have also been shown to be fulfilled in many field and pot experiments using gramineous species (de Wit, 1960; van den Bergh, 1968). When the per-plant yield of one genotype is higher in mixture than in monoculture and the per-plant yield of the other genotype is correspondingly lower, then the behavior of the mixture components is said to be Of the 70 mixtures, 34 were of a grass and a legume; in 10 mixtures the density was not closely controlled.

BIOMASS PRODUCTIVITY OF MIXTURES

185

of a “compensating” type (Aberg et al., 1943; Donald, 1963). If the plant relative yield (PRY) (based on the per area relative yield of de Wit and van den Bergh, 1965) of a component is defined as the ratio of the perplant yield in mixture to that in monoculture, then in such a mixture, the PRY of the aggressor will be greater than unity; that of the subordinate will be less than unity. Thus, if genotype i is the aggressor, Yii/Yi,i> 1 and Y i i / Y i j < 1. The terminology of this relationship has unfortunately been confused by Schutz and Brim (1967), who applied the term “complementary” to mixtures in which, as above, deviations of PRY from unity are of the type (+,-) . “Complementary” as applied to mixture components had already been used in the botanical literature to refer to something rather different (Salisbury, 1929; and see Section V ) . Among other terms which they introduced, Schutz and Brim used “neutral” for cases where both components give their “expected yields,” i.e., cases of the type (0,O). Mixtures of the type (+,O) and (-70) were described by Schutz and Brim as showing “over-compensation” and “under-compensation,” respectively. While “neutral” and “over-compensation” and “under-compensation” seem to be useful terms, the original term “compensating” (noun “compensation”) will here be retained for cases of the type (+,-) . With respect to mixture productivity, the common occurrence of compensation tends to keep mixture yields between the monoculture values. According to the de Wit (1960) model, when components in a 1 : 1 mixture are competing for the same supplies of environmental resources, the proportional increase of per-plant yield of one component will tend to equal the proportional decrease of per-plant yield of the other (see Fig. 1) . This implies that the mean of the PRY’S will have a value close to unity. In terms of the per-area relative yields of de Wit and van den Bergh ( 19651, this value is a total which they have called the relative yield total (RYT). The RYT is given in the present notation by

-+-

R Y T = -1 (Yii 2 Yii

G;)

Van den Bergh ( 1 968) showed that if RYT = 1, the mixture yield must lie between the yields of the pure cultures (strictly P , 2 M 2 P 2 ) . The general scatter of observed RYT’s around the value of unity is shown for 572 mixtures’in Table 11. These data seem to provide ample basis for expecting RYT’s to lie close to unity, and incidentally provide support for the wide applicability of de Wit’s (1960) model. The asymmetry of the low and the high deviations from unity (13.6% compared with 20.3%) is significant at the 1% probability level.

186

B. R. TRENBATH

Quontity of resource acquired

/

plant

FIG. 1. Graphical interpretation of de Wit's (1960) model when components of a 1:l mixture are competing for the same environmental resource. Compared with the unit quantity of resource acquired in monoculture, each plant of the aggressor gains an extra amount, F, in mixture. The quantity of resource acquired by each plant of the subordinate is correspondingly reduced by F in mixture. The per-plant acquisition of resource is assumed to be the s_ame in the two monocultures. M,, is the average per-plant yield of the mixture; P/rt is the mid-monoculture yield in per-plant terms; Y,,(open circle) and Y i , (filled circle) are per-plant yields of genotype i in monoculture and mixture respectively; Y,,(open square) and Yji (filled square) are corresponding per-plant yields of genotype j . Arrows by the monoculture points lie in the direction of the points of the same genotype grown in the mixture. The numerical values shown for biomasses indicate that in such a system, the proportional increase of Yl, over Yii equals the proportional decrease of Y , c below Y j , .

IV.

Types of Interaction Causing Nontransgressive Deviations

of Mixture Yields from Mid-Monoculture Values

Since the concept of relative yield total (RYT) has been introduced in the preceding section, it can be immediately explained that the present section treats only those mechanisms (chiefly competition) which seem able to cause deviations of M from I' without any deviation of RYT from unity. Such deviations of mixture yield are necessarily nontransgressive. (Other mechanisms of interaction exist which can cause nontransgressive deviations from I' but with RYT not equal to unity. Since these mechanisms are also potentially capable of giving rise to more extreme deviations, they are treated in Section V.) We examine first the nature of each mecha-

187

BIOMASS PRODUCTIVITY OF MIXTURES

TABLE I1 Distribution of the Relative Yield Totals of Mixtures Based on Published Data of Biomass of Components in 572 Mixtures ~

~~~

RYT value 0.5 to 0.7

0.7 to 0.9

1

2

Either grasses or legumesb Wheat Rice Grasses" A series of nonlegumesb Subterranean clover Flax and linseed 2 Grassesa Barley

1 3

Crop Grasses

0.9 to 1.1

1.1 to 1.3

6 5

4 9 2 26 41

11

4

8

3

1.3 to 1.5

21.7

4 1

Williams (1962)

3

5 3

Williams (1963) 9 22

5

95 10

3

7

20

9

Grasses.

3

106

4

21

26

21

4

6 6

5 18

2

3

6

-

-

3

Grasses. Rye -

6 1.0%

3

3

' L - 2

-

72 378 12.6% 66.1%

13.6%

Author Ahlgen and Aamodt (1939) Aberg et al. (1943) Donald (1946) Sakai (1953) Sakai (1955) Lampeter (1960)

1 2 1

G rassesa

Grassesa

1.5 to 1.7

95 13 16.6% 2 . 3 %

5

0.9%

3 0.5%

Harper (1965) England (1965) NorringtonDavies (1967) NorringtonDavies (1968) van den Bergh (1968) Whittington and O'Brien (1968) Tbomson (1969) NorringtonDnvies and Hutto (1972) = 572

20.3% ~~

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

Data derived from a series of cuts. Mixtures of leguminous and nonleguminous species have been omitted.

nism and then consider how it may influence the relationship between the yield of a mixture and the yields of its component monocultures. The environmental resources for which plants compete are principally the light, water, and soil-nutrient supplies necessary for growth (Clements et al., 1929; Harper, 1961; Donald, 1963; Risser, 1969; Rhodes, 1970b). Although carbon dioxide is required for shoot photosynthesis and oxygen

188

B. R. TRENBATH

is needed for the respiration of the roots, competition for these seems normally unlikely to occur. Competition for C02 is probably absent because the air within the canopy is generally well mixed (Monteith, 1963; Impens et al., 1967) ; all leaves are surrounded by air of approximately the same CO, concentration. Similarly, except in almost waterlogged soils, the diffusion of oxygen in soil is usually fast enough to maintain adequate supplies to all roots (Greenwood, 1969). Apart from the relatively small number of instances where chemical secretions by one species are thought to have influenced the growth of another, it seems that most investigators who have reported results of mixture experiments in English-language journals appear satisfied that the neighbor effects which they observed were due to competition for light, water, or one of the major nutrients (nitrogen, phosphate, and potassium), or a combination of these factors. Since it is relatively easy to measure and explain the unevenness of the sharing of light between the foliage of two components in a mixture, the studies involving competition for light have generally been the most conclusive. Clements and many others have related differences in success in mixture to differences in height between the leaves of the components of mixtures, simply inferring that leaves of the shorter component must be experiencing some degree of harmful shading. More recent analytical approaches have been based on methods suggested by Monsi and Saeki (1953). Working with natural communities, Monsi and Saeki introduced techniques for the measurement of profiles of light intensity and leaf area index (LAI) . Since then, these techniques have been used very successfully in experimental mixtures to relate the differences in growth rates of the two components to differences in the proportions of the total incident light intercepted by leaves of the two components (Black, 1958; Iwaki, 1959; Stern and Donald, 1962a,b; Williams, 1963). The general conclusion from all experiments involving competition for light is that the component with its leaf area higher in the canopy is at an advantage. It is also likely (Stern and Donald, 1962a) that, if the leaves are horizontal, the advantage is greater than if they are erect (this is because horizontal leaves intercept more of the total downward light flux per unit area of leaf than do erect leaves). If the taller component has a greater leaf area, its advantage is again correspondingly greater (Iwaki, 1959). In most of the experiments cited above, the investigators have attempted to exclude the possibility of competition for soil factors. By providing optimal soil conditions or by separating the roots of the mixture components, it was intended that neighbor effects would be interpretable in terms of competition only for light. In most agricultural environments, however, soil conditions are suboptimal and since root systems usually interpenetrate

BIOMASS PRODUCTIVITY OF MIXTURES

189

each other freely, there is a possibility of competition for water and/or nutrients. Where the soil is very infertile and the density of planting low, competition between root systems for soil factors is likely to decide which component becomes the aggressor because the leaf area produced may never become great enough to lead to significant competition for light. Indeed, large neighbor effects have been reported in cases where shading was claimed to have been absent (Pavlychenko and Harrington, 1935; Myers and Lipsett, 1958). In experiments in containers where competition for light has been prevented by partitions, competition for soil factors has similarly been shown to produce large effects (Donald, 1958; Aspinall, 1960; Rhodes, 1 9 6 8 ~Snaydon, ; 1971; Eagles, 1972). The principles involved in competition between root systems have not been as well worked out as in the case of competition between shoots, but Bray (1954) has outlined a general theoretical approach. In contrast with supplies of light, nutrient supplies are usually not greatly added to after the beginning of growth of an annual crop although subsequent rainfali or irrigation normally supplements the initial store of soil water. Whereas success of a species in competition for light depends on it having a large absorptive area closer to the light source than the leaves of the other component, the variable and complex geometry of root systems and the sources of water and nutrients makes it less easy to define the plant characteristics likely to confer competitive success. Bray (1954) noted that competition between root systems for nitrogen is likely to start at lower root densities, i.e. earlier in growth, than competition for phosphorus or potassium. This is because nitrogen is much more mobile in soil than phosphorus and potassium. Zones of nitrogen depletion round individual roots will therefore begin to overlap relatively early. To test this proposition, Andrews and Newman (1970) grew pure and mixed communities of young wheat plants in soil deficient in nitrogen and phosphorus with root densities between 1.2 and 8 cm/cm3; their resuIts showed that whereas competition for nitrogen was intense, competition for phosphorus was indeed slight. Since the mobilities of nitrogen and water in soil are similar (Fried and Broeshart, 1967), the soil resources most likely to be subject to competition are nitrogen and/or water. If Bray’s propositions are correct, they allow the identification of the characteristics that will determine the relative aggressiveness of the components of a crop mixture growing on infertile soil. For one component to gain an advantage over the other in the early competition for nitrogen and water, a faster growing root system (i.e., generally a greater length of root) is required. The limited evidence so far available does indeed suggest a possible correlation between seedling competitive ability and early root production (see review in Rhodes, 1970b). However, it also seems clear that

190

B. R. TRENBATH

if root lengths are similar, the genotype having the more widely spreading, less-branched root system will be at an advantage. Theoretical considerations suggest that there will probably also be an advantage in having roots as thin as possible, thus allowing the available root material to be present as the maximum length of root (Olsen et al., 1962). Abundant, long root hairs (see Olsen and Kemper, 1968) and a high root “demand” factor (Drew et al., 1969) are also likely to contribute to competitive success. This theoretical consideration of competition for soil factors has treated competition between root systems (root competition) as independent of competition for light between shoots (shoot competition). However, Donald (1958) has suggested that in well developed agricultural crops of uniform genotype, both shoot and root competition are usually occurring. The relative importance and time of onset of root and shoot competition will depend on environmental conditions (Aspinall, 1960) ; the nature of the genotypes involved will also have an effect. To compare, for a particular set of conditions and genotypes, the contributions of the effects of root and shoot competition with the overall effect of being grown in mixture, pot experiments have been designed in which partitions separated either roots or shoots, or both, or none (Donald, 1958; Griimmer, 1958; Aspinall, 1960; Rhodes, 1968c; Snaydon, 1971; Eagles, 1972). Of the results of these experiments, those of Snaydon (1971 ) are the easiest to interpret, for in his experiment alone were there constant numbers of partitions per treatment and plants per compartment. Using ecotypes of white clover, Snaydon confirmed the results of the other investigators who had shown that root competition had a greater effect (and probably began earlier) than shoot competition. Although this finding suggests that dominance-suppression relationships in crop mixtures may depend more on root characters than on shoot characters, the concentration of roots at the surfaces of the containers in such experiments must have hastened the overlapping of zones of depletion around individual roots, and hence led to an overestimation of the agronomic effect of competition for soil factors. Considering a possible interaction between the two types of Competition, Donald (1958) showed that the proportional reduction in the subordinate’s yield due to shoot competition was greater when root competition was occurring than when it was absent. Similarly, the proportional reduction due to root competition was greater when shoots were competing for light than when they were separated. Donald suggested that the failure of, say, the root to acquire sufficient nitrogen caused leaf development to suffer. This in turn reduced the supplies of assimilate to the roots which grew less and so were less able to compete for nitrogen. Donald proposed a similar set of causes and effects for a component of a mixture which was unsuccessful

BIOMASS PRODUCTIVITY OF MIXTURES

191

in competition for light. The important conclusion was that the effects of Competition, once initiated, tend to be magnified by a system of positive feedbacks if simultaneous shoot and root competition are occurring. Milthorpe (1961) has discussed the effects of the onset of competition for soil factors in a mixture where competition for light has already caused some depression of the growth of one component. Arguing from the depressive effect of shading on root:shoot ratio (e.g., Brouwer, 1966), Milthorpe considered that any deficiency of soil water or nutrients would cause the accelerated suppression of the subordinate if any previous shading had reduced its root’s uptake capacity relative to the size of its shoot. Similarly, the writer has presented results (Trenbath and Harper, 1973) which suggest that the development of a slight deficiency of soil factors where competition for light was already occurring caused an approximately 4-fold increase in the depression shown by a series of subordinates. While competition for resources has usually seemed able to explain observed differences between per-plant yields in mixture and monoculture, a number of workers believe that these differences (neighbor effects) are sometimes caused by other processes. This belief has been expressed by workers well acquainted with the mechanisms of competition (Warne, 1953), has been implied by other investigators (Sakai, 1955), and has been used as a convenient explanation of unexpected results by others (Ahlgren and Aamodt, 1939; Went, 1942). As an alternative explanation of neighbor effects, allelopathy (Grummer, 1955) has been frequently proposed. The type of allelopathy most likely to occur in a field crop is that described by Winter ( 1961), where one or more biologically active substances are liberated from the root or shoot of a plant, enter a neighboring plant, and there cause a depression of that plant’s growth. T o allow for cases of stimulation of growth in the affected plant, the general term “allelochemical” effect (Anonymous, 1971 ) has been introduced to include both positive and negative effects on growth. Since, however, the term “allelopathy” is both more convenient and more familiar, in what follows it will be used in a general sense, i.e., allowing the possibility of either positive or negative effects. Allelopathy in plant communities has been studied most actively in the Soviet Union and Germany; the rapidly expanding literature has been reviewed by Griimmer (1955), Rademacher (1959), Grodzinsky (1965, 1971), and Borner (1971). On the other hand, in some other European countries and in the United States, research on allelopathy has had an uneven history. Several instances of apparent allelopathy have been shown (sometimes by the original investigators) either to be due to other causes (Went, 1942; Gray and Bonner, 1948; but see Muller, 1953; Grummer, 1958; but see de Wit, 1960; Sandfaer, 1968; but see Sandfaer, 1970a,b)

192

B. R. TRENBATH

or to be explained by alternative mechanisms (Muller, 1966, but see Bartholomew, 1970). These cases have tended to make investigators cautious of suggesting allelopathy as the cause of neighbor effects, indeed, they have inclined the present author to agree with Harper’s (1965) opinion that research into the role of allelopathic substances in mixtures is beset with extreme technical difficulties. To avoid greatly extending this review no detailed discussion will be attempted, but three general comments will be made. First, as Borner (1960) pointed out, although many investigators have demonstrated allelopathic effects between plants growing in solution culture or sand culture, instances where the same effects have been conclusively proved to be significant under field conditions are rather few. Among such cases reported in English-language and western European literature, the most convincing seem to be the existence of inhibitory effects on associated plants due to walnut (Massey, 1925; Bode, 1958) and to Arctostaphylos species (Hanawalt, 1971) . Also, the demonstration of autotoxicity among individuals of Grevillea robusta (Webb et al., 1967) grown in both the field and glasshouse seems conclusive. It should be noted that although Bonner (1946) demonstrated the autotoxic activity of root exudates of guayule plants in gravel culture, the active substance was degraded so fast in field soil that no autotoxic effect was likely under normal agricultural conditions. A rather different case is that of Hirano and Kira (1965), who demonstrated apparent autotoxicity among densely planted peach saplings in an experiment planted in field soil (clay loam) under glass; this result accorded well with expectations (Hirano and Morioka, 1964; Proebsting and Gilmore, 1940). However, a similar experiment carried out on a sandy soil in the open gave no indication of an autotoxicity effect (Hirano and Kira, 1965). Second, since crops are normally grown at densities high enough for competition between neighbors for resources to be intense (Donald, 1963) , any allelopathic effect would be either exaggerated (or reversed) through competition. To emphasize the importance of taking into account both aspects of the total interaction, we may consider an hypothetical example of a crop mixture growing on a soil poor in nitrogen. If the suppression of one of two otherwise evenly-matched components was initiated through some slight allelopathic inhibition of its nitrogen uptake, later competition for nutrients might result in plants of this component being nitrogen deficient and heavily suppressed. The agronomist would “prove” the suppression to be due to competition for nitrogen since he would find, in an experiment with separated shoot systems, that neighbor effects were strongest on soils with the lowest nitrogen level; in the meanwhile, the specialized investigator of allelopathic effects might have discounted some slight inhibitory effect as insignificant in view of the great intensity of the suppression in the field. Through the lack of an integrated approach, both workers

BIOMASS PRODUCTIVITY OF MIXTURES

193

would have drawn false conclusions. The experimental separation of the effects of competition and allelopathy is thus an important objective, although it is certain to be difficult to achieve (Welbank, 196l ) . Third, providing (a) that an allelopathic inhibition takes effect early enough in the life of the crop for compensatory growth to occur in the other component (see Section V ) , and (b) that the inhibition reduces the uptake of growth factors rather than the efficiency with which they are used, the total quantity of environmental resources intercepted by the mixture may not be much affected. Consequently, the yield may scarcely be less than that of another mixture with the same values of P, and P , in which a similar degree of suppression has been brought about by competition acting alone. Where, however, the allelopathic effect reduces the efficiency with which growth factors are used in dry-matter production, the RYT will probably be lowered and the mixture could underyield (see Section V) . Whether caused by competition alone or by a combination of competition and allelopathic effect, the unequal sharing of resources between components generally affects the yields of mixtures. If the higher-yielding component in monoculture is the aggressor, then with RYT = 1 the mixture yield will lie between f' and P I . The greater the depression of the loweryielding component, the closer the approach of M to P,; with its complete suppression, M = PI. Similarly, if the lower-yielding component in monoculture is the aggressor in the mixture (the so-called Montgomery effect, Gustaffson, 1951), then M lies between P and P,. With complete suppression of the subordinate, M = P,. The tendency for mixture yields to lie above P (Table I) combined with the closeness of RYT's to unity (Table 11) suggests that there is a positive correlation between aggressiveness in mixtures and biomass production in pure stands. The association is not strong, however, for the value of the correlation coefficient has been estimated from published data to be only about 0.3 (Trenbath, 1972). The existence of a positive correlation is nevertheless in agreement with the ideas discussed earlier, namely, that a large leaf area displayed at a sufficient height gives an advantage in competition for light (e.g., Donald, 1961) ; the close dependence of productiv; ity in monoculture on LA1 is well established (Watson, 1952). A possible reason for the looseness of the correlation between aggressiveness and monoculture yield is found in a suggestion by Iwaki (1959). Iwaki postulated that a species would be aggressive in mixture if it diverted a particularly large share of photosynthetic product into building tall stems. Associated species would be shaded and suppressed. In monoculture, however, the tall species would grow relatively slowly because of the low proportion of its dry matter invested in productive leaves. A negative correlation between aggressiveness and monoculture grain yield in rice (Jennings

194

B. R. TRENBATH

and Aquino, 1968) was similarly related to the tallness of the stems of one type of rice, but in this case the low monoculture yield of the tall type was mainly the result of its tendency to lodge. The bending over of the tips of long grass leaves (“flagging”) will also lead to greater aggressiveness (see above) and lower growth rates in monoculture (Alberda, 1966). The aggressiveness of a species is well known to depend on environmental conditions and such a dependence will affect the relative values of M ,P I , and P,. The dominance-suppression relationship between genotypes has been reversed by changing the temperature regime (Eagles, 1972) and by aItering the soil conditions (van Dobben, 1955; Sakai and Iyama, 1959; Stern and Donald, 1962a; van den Bergh and Elberse, 1962; Snaydon, 1971). Aggressiveness often depends on the stage of growth of the plants (e.g., de Wit and van den Bergh, 1965; van deli Bergh, 1968; Rhodes, 1968a; Nguyen Van, 1968), but in field experiments, the concurrent changes in developmental stage and in meteorological factors usually make it difficult to identify the factor responsible for any change of aggressiveness with time. The hard-to-define differences between growing seasons have marked effects on aggressiveness at least where the index used to measure it is based on seed yields (Laude and Swanson, 1942; Sakai and Oka, 1955; Allard and Workman, 1963; Workman and Allard, 1964; Lin and Torrie, 1968). Allard et al. (in Edwards and Allard, 1963) showed from the data of Suneson (1949) that taking the value of ATLAS as 100, the selective advantage of VAUGHN ranged from 40 to 173 in 13 seasons. Transfer from the field to the greenhouse may also cause reversals of dominance (Aberg e f al., 1943; Syme and Bremner, 1968). If aggressiveness is so sensitive to environmental conditions, it might be asked how this sensitivity affects mixture yields over a series of sites or seasons. Unfortunately, although diallel experiments have been carried out using a range of environments or treatments (Harper, 1965; England; 1965; Whitehouse et al., 1967; Bell et al., 1968; Norrington-Davies, 1968; Sandfaer, 1970b; Schutz and Brim, 1971), no systematic attempt has yet been made to study the effect of associated changes of aggressiveness on mixture yields. Presumably connected in some way with such changes, a greater stability of seed yields has often been found in varietal mixtures of cereals and soybeans grown over a range of environments [Allard, 1961; Simmonds, 1962; Pfahler, 1965; Frey and Maldonado, 1967; Byth and Weber, 1968; Qualset and Granger, 1970; Schutz and Brim, 1971; but see Rasmusson (1968) and Clay and AIIard (1969)l. Taking together the results of three reports (Allard, 1961; Pfahler, 1965; Qualset and Granger, 1970) in which an index of yield stability was given for each mixture and monoculture, 5 out of a total of 12 binary mixtures were more

BIOMASS PRODUCTIVITY OF MIXTURES

195

stable than their more stable component monoculture; the remaining mixtures showed stabilities between those of their component monocultures. In a comprehensive study involving 16 environments, Schutz and Brim (1971) calculated, for each type of stand, indices of stability with respect to several types of environmental variation (seasons, locations, replicates, etc.) . The lack of agreement between the several stability indices of individual types of stand suggests a complex situation with no clear-cut stability advantage for mixtures. To add to difficulties of interpretation, stability parameters are much less accurately estimated than parameters related directly to yield (Allard, 1961;Marshall and Brown, 1973) . With reference to biomass rather than to seed yield, very few data are available that allow the comparison of any index of yield stability of individual mixtures with those of their component monocultures. While two studies (England, 1968; Sechler and Chapman, 1967) suggested that mixture yields were collectively more stable than monoculture yields, data of individual mixtures from two other studies (Pfahler, 1965; Thomson, 1969) showed stabilities to be intermediate between those of the monocultures in 9 out of a total of 10 cases. These limited data seem to indicate that the stability of mixture yields is, like yield itself, usually nontransgressive, and that in seed yield, but not in biomass yield, there is a tendency for mixture stability to be close to that of the more stable component. In a theoretical study of mixture stability, Marshall and Brown (1973) suggested that varietal mixtures were most likely to show agronomically useful stability in highly variable environments to which the available genotypes were not individually well adapted. Perhaps it is partly the lack of crop species with sufficiently wide adaptation that has led to the extensive culture of multispecific mixtures in India (Aiyer, 1949). The markedly differing adaptation characteristics of the components of such mixtures ensure that even in the worst season there will be something to harvest. The principal conclusions from this and the previous sections may now be summarized. In field-crop mixtures, competition for both light and soil resources will usually be occurring; allelopathic effects (if present) will operate in conjunction with this competition; competition for resources and/or allelopathy will usually cause per-plant yields of each genotype in mixture to differ from that in monoculture; the differences of per-plant yields are usually of a compensating type; if RYT = 1, M will lie between PIand P,;any positive correlation between aggressiveness and monoculture yield will produce a tendency for M > P; there are some indications that the biomass yield of a mixture having sufficiently different components is likely to be more stable than that of the more stable of its components in monoculture. We continue now by turning to the more varied and subtle forms of interaction that may lead to transgressive yielding.

196

B. R. TRENBATH

V.

Mechanisms Capable of Causing Transgressive Yielding by Mixtures

The results reviewed in Section I1 show that mixtures have often been recorded as apparently yielding transgressively. Furthermore, the data indicate that records of mixtures overyielding are significantly more frequent than records of underyielding. Two contrasting interpretations of this situation could be proposed. 1 . It might be suggested that if RYT could be measured without experimental error, its value might always be close to unity; the observed skewness of its distribution (Table 11) could be a consequence of basing RYT on two ratios of random normal variables (Fieller, 1932). Since there is a weak, positive correlation between aggressiveness and monoculture yield (Section IV) , the preponderance of mixture yields greater than P could be the result of this correlation. If this were the case, theoretical error-free experiments would show mixture yields lying usually between P, and p, less frequently between P and P,, and never outside the range P, to P2. The experimental error found in actual experiments would be expected to disperse the observed results about their “true” values, so that overyielding mixtures would be recorded more frequently than underyielding ones. This view implies that observed cases of transgressive yielding are due only to experimental error. 2. An alternative interpretation might be that at least some of the observed cases of transgressive yielding do represent real effects. If the correlation between aggressiveness and monoculture yield were discounted as being too weak to have appreciable effect, the tendency for mixture yields to exceed P could be attributed to apparent mutual stimulation of the mixture components, as indicated by the preponderance of RYT’s greater than unity (Table 11). These two interpretations are only preliminary attempts to explain the observations; they represent extreme, but not necessarily incompatible, approaches. Having discussed in Section I1 the evidence in favor of transgressive yielding being a real phenomenon, we consider here the mechanisms that could lead to transgressive yielding. If they appear likely to operate under the conditions normally used for mixture experiments, this would add credibility to the second interpretation given above. In connection with overyielding, the findings and theory of animal ecology may be relevant since they help to define conditions that might lead to this kind of transgressive yielding. Gause (1934) quoted Formozov’s observations of natural mixed populations of tern in which four species coexisted, apparently because they exploited the environment in different ways. In such a case, it seems likely that a mixed population would

BIOMASS PRODUCTIVITY OF MIXTURES

197

be more productive than any monospecific population. Gause (1934) summed up his own experimental results in the principle that species of very similar requirements tend not to coexist. Using this principle, Harper (1967) and others argued that the species diversity found in natural communities implies that the component species occupy differing niches. Together, therefore, they must exploit the environment more completely than a community of few species. Providing support for this view, Beardmore et ul. (1960) and Seaton and Antonovics (1967) reported overyielding of mixtures of Drosophilu strains in controlled environments. Seaton and Antonovics suggested that in their mixtures the two genotypes were avoiding competition by occupying different niches. In the botanical literature, genotypes believed to be avoiding competition while they share a habitat are said to be “complementary” (Woodhead, in Salisbury, 1929). Woodhead first used the term in connection with natural woodlands where, by the sharing of environmental resources in either time or space, the various types of species seem to escape, at least partially, the effects of competition. His concept of the complementary use of resources would seem to be applicable to crop mixtures; the possible ways in which resources may be used in such a manner will now be considered. Van den Bergh and de Wit (1960) reported an example where temporal sharing of the environment may have been responsible for a case of apparent mutual stimulation in mixture. In a mixture of two grass species which differed markedly in time of development, plants of both components had more tillers (53% and 3 6 % ) than did plants in the corresponding monocultures; although biomasses were not reported, it may be noted that the RYT based on tiller number was 1.49. Syme and Bremner ( 1968) reported a series of experiments involving oats and barley varieties chosen to differ in flowering time; the results of an experiment performed under glass (data given in Syme, 1963) showed that in all four oats-barley mixtures, both components showed higher per-plant shoot weights than in monoculture. While overyielding of dry matter did not occur, the RYT based on shoot weights averaged over the four (replicated) mixtures and two densities was 1.12. Such an RYT value in a mixture of which the two monoculture yields were very similar would be associated with the mixture outyielding the monocultures by 12%. Sechler and Chapman (1967) also reported an experiment with cereal genotypes (oats) which differed greatly in flowering date. Again no mixture overyielded. Unfortunately, the data given were insufficient to allow the calculation of RYT’s and so no detailed comparison is possible with the experiment of Syme and Bremner ( 1968). In corn-rice mixtures in the Philippines, the corn flowers and matures before the rice begins to flower; this phenological difference probably explains the reported overyielding of grain by such a mixture (International

198

B. R. TRENBATH

Rice Research Institute, 1974). Flax and linseed differ in maturity date and their overyielding of biomass in several mixtures was attributed to this by Obeid ( 1965) (see Section 11). Differing temporal patterns of growth or development sometimes result in a reversal of dominance during the growth of a mixture (Harper and Clatworthy, 1963; England, 1965). Even if the biomass increments of the mixture over the periods before and after the reversal are not transgressive compared with the increments in the monocultures over the same intervals, the biomass accumulated by the mixture over the whole growing season may yet be transgressive. Van den Bergh (1968) gave a hypothetical example of this in which the more aggressive component in each phase was the component with the greater biomass-increment in its monoculture during that phase. In this example, the total accumulation of biomass over the two phases was the same in each component monoculture; the association of greater aggressiveness and faster biomass accumulation within each phase resulted in overyielding by the mixture at the final harvest. If the direction of this association had been reversed, the result would have been an underyielding mixture. Examples of such effects have not yet been found, and too few suitable sets of data exist to judge their plausibility. Van den Bergh (1968), failing to find this kind of effect in the data of England ( 1965), suggested that the originally suppressed species had been unable to take advantage of the sharp reduction in the vigor of the other component of the mixture. The components of a mixture may be complementary in a spatial sense by exploiting different layers of the soil with their root systems. Although Gustaffson’s ( 1954) discussion and vague reference to such a case cannot be taken as more than an indication, he claims the possibility of increased grain yields in cereal mixtures through the stratification of root systems. Cable (1969) found that desert plants were affected least by the proximity of plants of other species when the root systems of these neighbors did not overlap the depth of their own. Mutual avoidance^' by adjacent root systems (Raper and Barber, 1970; Baldwin and Tinker, 1972) could lead to a late-developing root system occupying deeper soil horizons in mixture than in monoculture. The significant overyielding of grass mixtures reported by Whittington and O’Brien (1968) was accompanied by phosphorus uptake from greater depths in the mixtures than in the monoculture (O’BrienBet al., 1967) and so might be due to such an effect. However, the different pattern of uptake in mixture could also have been the result rather than the cause of enhanced growth. The writer (Trenbath, 1970), measuring panicle weights in a field experiment involving 5 mixtures of oat species, found in one replicate that 5 out of 5 mixtures overyielded; in the succeeding two replicates in a linear sequence of three, the numbers

BIOMASS PRODUCTIVITY OF MIXTURES

199

of overyielding mixtures were, respectively, 2 and 1 out of 5. The reality of this trend was supported by the significance (P < 0.01) of the regression of panicle-weight RYT’s of the 5 mixtures on the position of the replicate in the linear sequence across the field. The trend in mixture yields was tentatively related to an observed soil-depth gradient. It was suggested that stratification of root systems had occurred on the deep soil, leading to high RYT’s and overyielding, but that this stratification had been prevented on the shallow soil. The two species making up the mixture with the highest RYT were suspected to differ the most in depth of rooting. Ellern et al. (1970) showed that indeed these two species had rooting depths significantly different; the other species were unfortunately not tested. Components of a mixture may complement each other nutritionally; one component may require much of an element of which the other component needs little (see Kolb, 1962; Davies and Snaydon, 1973). Considering a particular element, one component may be able to utilize a form that is unavailable to the other. Although mixtures of leguminous and nonleguminous species are not formally treated in this review, it is such mixtures which provide the most striking and repeatable examples of overyielding due to nutritional complementation. Since the roots of the components are drawing on different supplies of nitrogen (soil and atmospheric sources), on nitrogen-deficient soils, RYT values regularly exceed unity and overyielding is often recorded (e.g., de Wit et af.,1966; Ennik, 1969). When nodulation is prevented, the species compete for the same supply of nitrogen, the RYT falls to unity, and overyielding does not occur (de Wit et a!., 1966). Since phosphorus is present in the soil in several forms of differing availability to different species (Richardson et al., 1931; Schander, 1941 ) , nutritional complementation with respect to phosphorus could also occur but has not yet been reported. Overyielding by mixtures has in some instances been attributed to a more efficient utilization of light by their canopies. The use of mathematical models has suggested that the highest photosynthetic rate might be obtained from a canopy in which the steepness of the inclination of the leaves decreases with depth (Warren Wilson, 1960; Verhagen et al., 1963; Duncan et al., 1967; Nilson, 1968). This “ideal” leaf arrangement could be approached by a mixture of a tall erect-leaved genotype and a short, prostrateleaved one. A mixture of such components might possibly overyield. In addition, Verhagen et al. (1963) have used a simple model to show theoretically that to maximize photosynthetic production, as the leaf area of a crop increases, the average inclination of the leaves should also increase. I n the “leaf-inclination” mixture just described, the shorter, prostrateleaved form would tend to be progressively suppressed through shading by

200

B. R. TRENBATH

the taller form, and hence the average inclination of leaves in the canopy as a whole would indeed increase. To examine further the possibility of mixtures overyielding due to favorable canopy configurations, the author (Trenbath, 1972) used a computer model based on that of Duncan et al. (1967) to simulate daily gross photosynthesis of mixtures and monocultures of wheat varieties with contrasting leaf inclinations. In 3 of the 4 sets of conditions used, mixture advantages in photosynthetic rate were predicted to appear between LA1 2 and 7 (approximately) ; such advantages depended on the latitude, cloud cover, and planting time being such that most of the incident light arrived from relatively high elevations. In a mixture with LA1 4, given the radiation conditions of latitude Oo at the equinox (sunny, with a solar elevation at noon of 9 0 ” ) , the RYT based on simulated daily photosynthesis was 1.087. This RYT corresponded to the unreal, but extremely favorable, situation of a mixture of nonoverlapping, stemless canopies, the upper canopy with a leaf inclination of 75O and the lower with an inclination of 1 5 O . The introduction of stems and the inclusion of a moderate degree of overlapping of the canopies reduced the RYT considerably. If the components were to have less extremely erectophile and planophile leaf canopies, the RYT would be reduced still further and any predicted mixture advantage in photosynthesis would be very small. Rhodes (1968b) reported the yield of grass in a mixture whose canopy structure approached Warren Wilson’s ( 1960) “ideal” configuration mentioned above. The mixture overyielded significantly in two treatments with frequent cutting, but not in the treatment where plots were left to grow until their canopies intercepted 95% of the incident light. It is unlikely that this overyielding was due to any similarity between the mixture canopy and Warren Wilson’s (1960) ideal type since, as already mentioned, theoretical experiments (Trenbath, 1972) suggest that the advantage of a “leaf inclination” mixture (erect-leaved canopy above a prostrate-leaved canopy) will be greatest in stands with LAI’s between 2 and 7. In a similar way to Rhodes, van den Bergh (1968) and Alcock and Morgan (1966) found (nonsignificant) overyielding in grass mixtures where repeated cutting probably prevented the LA1 from exceeding 2 for any extended period. Although van den Bergh appealed to possibilities of better use of light by the sods of such mixtures, he gave no measurements or theoretical basis for such an explanation. Of extreme interest is Rhodes’ (1970a) finding that two mixtures, of which the components had contrasting leaf inclinations, overyielded significantly (see Section 11) under infrequent cutting, whereas under frequent cutting they yielded between PI and P,. Such observations accord much better with theory than do Rhodes’ (1968b) results, but it should be noted

BIOMASS PRODUCTIVITY OF MIXTURES

20 1

( a ) that canopy structures were not actually measured in any of these studies, and ( b ) that other mixtures which contained apparently similar combinations of tall-erect with short-prostrate forms (two in Rhodes’ 1970a experiment and three in his 1968b experiment) did not overyield. If yields of individual components had been measured in Rhodes’ studies, it would have been possible to use RYT values to detect and quantify positive interactions between components in a manner independent of overall yields. It is possible that positive interactions occurred in all Rhodes’ “leaf-inclination” mixtures, but that differences of monoculture yields were sufficient to prevent their expression as overyielding. Before leaving the topic of light utilization, it is pertinent to point out that Duncan et al. (1967) simulated the photosynthesis of a canopy of which the leaves were horizontal above, becoming more inclined with depth. According to their result, it can be expected that a mixture of a tall, prostrate-leaved form and a short, erect-leaved form might underyield. No detailed models of competition for water appear to have been published and hence there are no quantitative predictions of the effect of water shortage on mixture yields. Nevertheless, considering the growth of mesophytic crops under conditions of high radiation and water shortage, Aiyer (1949) has emphasized the beneficial effect of shade trees or other tallergrowing mixture components; also, Baldy (1963) has argued that the many-layered mixed communities traditionally grown in desert oases (e.g., date palm apricot vegetables) may use applied water more efficiently in biomass production than pure stands. In such oasis communities, the shading- and windbreak-effect of the upper layer(s) creates a favorable microclimate for the layer below; the component chosen for each successively lower layer is more mesophytic and less light-demanding than the one above. The shortest component (the vegetable crop) escapes suppression possibly because soil cultivations tend to prevent intermingling of the root systems. Where the shorter component grows better under shade than in the open (Aiyer, 1949) and where the taller component benefits from the presence of the shorter one (Baldy, 1963, p. 339), water-use efficiency may be superior in the mixture and overyielding of biomass may occur. To the author’s knowledge, this possibility has not been rigorously tested. Allelopathic effects can theoretically cause transgressive yielding. If an allelopathic substance produced by one component affects the growth rate of the other component by changing only the rate of uptake of some limiting growth factor, the apparent relative competitive abilities of the mixture components will change but the total quantity of the factor taken up may not be much different from that in the absence of allelopathy. If this is so, RYT will be close to unity (Section IV). If, however, the substance changes the efficiency with which the growth factor is utilized, RYT will

+

+

202

B. R . TRENBATH

deviate from unity and transgressive yielding is possible. Roy (1960) discovered a binary mixture of rice varieties which overyielded in grain weight in several independent trials. By manipulating the cultural conditions, he found that the growth stimulation appeared to be caused by some agent carried between components in the irrigation water. Although the importance of this result ought to have led to an immediate repetition of the work, no further reports appear to have been published. Allelopathic stimulation has not so far been invoked as an explanation of overyielding in experiments where the yield was dry matter. Another way in which allelopathy could lead to overyielding is suggested by the experience of foresters in New South Wales and Queensland. Webb et al. (1967) have reported that six rain-forest trees which do not form natural pure stands show unexpectedly poor growth in commercial monocultures. Detailed experimentation using one of the species (Grevillea robustu) indicated that the growth of young individuals was inhibited by a water-soluble substance apparently produced by the roots of adjacent G . robusta plants. If the same mechanism is causing the autoinhibition of the other five species, mixtures of them might overyield. It is to be hoped that experiments will soon be undertaken to confirm and clarify these and other possible effects of allelopathy on mixture yields. Mechanical factors could, again theoretically, lead to transgressive yielding by a mixture. For example, let us suppose first that the component with the potentially higher yield in monoculture is susceptible to lodging, and second that the other component resists lodging strongly enough to cause the mixture to stand while the susceptible monoculture lodges. If the lodged monoculture yields less than the unlodged mixture, the mixture is expected to overyield. Such a situation is not unlikely since lodging in mixtures commonly follows the behavior of the more resistant component (Atkinson, 1900; Tsedik-Tomashevich, 195 1 ; StringEeld, 1959; de Wit, 1960; B. R. Trenbath, unpublished data; but see Probst, 1957; Qualset and Granger, 1970). Although very few critical experiments appear to have been performed to test the point, a consensus of opinion is developing which maintains that mixtures have a generally greater tolerance of disease and pest attack (Schwerdtfeger, 1954; Borlaug, 1959; Simmonds, 1962; Browning and Frey, 1969; Adams et al., 1971; Cherrett et a]., 1971). When a stand of susceptible plants is “diluted” with resistant plants, the level of infestation or damage of individual susceptible plants may be reduced. (Suneson, 1960; Browning, 1966). Since the foliage of the resistant plants acts as a spore trap, the growth rate of rust in an epidemic is reduced (Browning, 1966; Leonard, 1969). Similarly, resistant plants can act as a barrier to the transmission of a virus among plants of a susceptible component in a

BIOMASS PRODUCTIVITY OF MIXTURES

203

mixture (Broadbent, 1957; Sandfaer, 1970b). The effects of admixed nonhost plants on insect infestations may however be more subtle. The nonhost plants may interfere with the visual cues by which the insect finds its host (International Rice Research Institute, 1974), or with the olfactory cues for host finding (Tahvanainen and Root, 1972); the nonhost plants may interfere with feeding behavior (Tahvanainen and Root, 1972), may attract predators (International Rice Research Institute, 1974), or provide a more favorable biotic environment for the growth of predator populations (Muller, in Lampeter, 1960). If, as in some mixtures of cotton genotypes grown in India (Simmonds, 1962), each of the two components is attacked simultaneously by a disease to which the other is resistant, theory suggests that pathogen escape in the mixture may result in the mixture overyielding (Chilvers and Brittain, 1972). If only one component is affected, the mixture is more likely to yield intermediate between the monocultures. Whether the mixture yields close to PI (as in Suneson, 1960), or closer to P (as in Klages, 1936) is decided by several factors such as time of onset of the disease, the degree of pathogen escape due to the dilution effect mentioned above, and whether the affected component is the aggressor or the subordinate (de Wit, 1960; Sibma et al., 1964). Although the above discussion suggests that the growing of appropriate mixtures will help to minimize the effects of pathogens, it should be pointed out that mixtures have no automatic advantage in this respect. After a long series of elegant experiments, Sandfaer (1968, 1970a,b) concluded that the low RYT values (based on grain yield) in a series of varietal mixtures of barley were due to barley stripe mosaic virus being transmitted from a certain variety, a symptomless carrier, to the other, sensitive components of the mixtures. The virus caused sterility in the sensitive varieties, an effect that occurred too late in growth for the carrier variety to show a compensating increase of grain growth. In this context, an earlier study of van den Bergh and Elberse (1962) may be mentioned. Here, low values of dry-matter RYT in a series of grass mixtures were attributed to an exactly similar situation; the evidence for the involvement of the virus was, however, only circumstantial. In connection with diseases caused by fungi, Butler and Jones (1949) stated that, in general, more active carbon assimilation (due for instance to increased light levels) increases susceptibility to obligate parasites of the green parts of plants (e.g., rusts). According to Gaumann (1950), however, reduced light levels lead to increased susceptibility to “eusymbiotic” parasites (e.g., Fusarium spp.). With this in mind, it is not difficult to imagine a crop mixture in which, owing to neighbor effects, each of the components was more susceptible than in monoculture to a parasitic fungus which was attacking it. However, so far only single

B. R. TRENBATH

204

components have been reported to be worse infected in mixture, e.g., alfalfa by Rhizoctonia solani when mixed with a grass (Chamblee, 1958) and rice by blast when mixed with corn (International Rice Research Institute, 1974). Before passing to the final conclusions, the points discussed in this section may be summarized by means of Table 111. When considering the implications of an observed deviation of RYT from unity unaccompanied by transgressive yielding, or of a report of transgressive yielding without the data needed to calculate an RYT, it should be remembered that overyielding implies RYT > 1 and underyielding implies RYT < 1 but that the converses are true only if the monoculture yields are sufficiently similar. TABLE 111 Summary of the Evidence Discussed in Section V Concerning Deviations of Relative Yield Totals (RYT) from Unity and Transgressive Yielding.

>1 without overyielding RYT

Mechanism that might lead to RYT

>1

+ +"

Differing growth rhythms Differing rooting depths Nutritional complementation Enhanced light-use efficiency Enhanced water-use efficiency Allelopathy Lodging escape Pathogen escape

Mechanism that might lead t o R Y T

<1

Lowered light-use efficiency Allelopathy Induced lodgingd Disease expression through infection of a susceptible component

RYT < 1 without underyielding

Overyielding

+ +" + +b

Underyielding

(+)

A cross shows that a t least one report exists t o substantiate the operation of the given mechanism to produce the given result. I f , for lack of data o n biomass, n report concerning grain yield has been cited, the cross is shown in parentheses. Unreplicated, but 5 out of the 6 mixtures grown overyielded. nonlegume. Only in mixtures of legume If lodging in the mixture were to follow the pattern in the more susceptible monoculture.

+

BIOMASS PRODUCTIVITY OF MIXTURES

VI.

205

Conclusions

The main findings of this review are now listed: 1. Most binary mixtures have been recorded as yielding at a level between the yields of the components’ monocultures (see Table I ) . This “nontransgressive” yielding is what might be predicted on the assumption of competition between components for the same resources. Such competition would be expected, as a first approximation, to lead to equal proportional increases and decreases of plant biomass compared with per-plant performance of the components in monocultures. This implies that the relative yield totals (RYT’s) of mixtures would have values close to unity. This is found in practice (see Table 11). 2. A minority of binary mixtures has been recorded as yielding transgressively, that is to say outside the range defined by the yields of the components grown in monoculture. This suggests that the above proportional model may not always apply, but the frequent lack of repetition of experiments and the small margins by which the mixture yields transgressed the range between the monoculture yields usually make it impossible to say whether a given case of transgressive yielding was due to experimental error or to a real effect. Since a series of mechanisms can be suggested that could plausibly lead to mutually beneficial effects between mixture components, it seems likely, or at least possible, that some of the observed cases of overyielding are due to such mechanisms. The frequent absence of conclusive experimental evidence in favor of the operation of the hypothetical mechanisms may be due in part to the lack of sustained investigations of specific cases of overyielding and in part to their extreme sensitivity to variations of environmental conditions. Also, few informed attempts have yet been made to increase the likelihood of overyielding by a conscious choice of conditions and genotypes. Similar remarks apply to underyielding by mixtures, although rather fewer mechanisms have been suggested to account for it. 3. The operation of mechanisms resulting in the stimulation or inhibition of the growth of mixture components beyond expectations based on the proportional model give transgressive yields only if the monocultures are sufficiently similar. A convenient measure of such stimulation or inhibition, and thus of the mechanisms underlying them, is the relative yield total. This index takes account of the monoculture yields in such a way as to detect the operation of these mechanisms even in mixtures which do not yield transgressively. Owing to the complexity and unpredictability of many agricultural PCOsystems, the control over them which man claims he has is often only nomi-

206

B. R. TRENBATH

nal. Given a sound theoretical framework, however, he can hope for fruitful research into ways of improving this control. This survey of literature attempts to point to areas of ignorance and to suggest means of strengthening our research so that we may seek to proceed towards more effective exploitation of mixed crops. ACKNOWLEDGMENT

I am grateful to Professor C. M. Donald and Dr. I. Valentine for their critical reading and discussion of drafts of the manuscript. REFERENCES Aberg, E., Johnson, I. J., and Wilsie, C. P. 1943. J. Amer. SOC. Agron. 35, 357-369. Adams, M. W., Ellingboe, A. H., and Rossman, E. C. 1971. BioScience 21, 1067-1070. Ahlgren, H. L., and Aamodt, 0. S. 1939. J . Amer. SOC. Agron. 31, 982-985. Aiyer, A. K. Y. N. 1949. Indian J. Agr. Sci. 19,439-543. Alberda, T. 1966. In “The Growth of Cereals and Grasses” (F. L. Milthorpe and J. D. Ivins, eds.), pp. 200-212. Butterworth, London. Alcock, M. B., and Morgan, E. W. 1966. J. Brit. Grassland SOC.21, 62-64. Allard, R. W. 1961. Crop Sci. 1, 127-133. Allard, R. W., and Adams, J. 1969. Proc. Int. Congr. Genet., 12th, 1968 Vol. 3, pp. 344-370. Allard, R. W., and Workman, P. L. 1963. Evolution 17, 470-480. Andrews, R. E., and Newman, E. I. 1970. Oecol. Plant. 5, 319-334. Anonymous. 1971. I n “Biochemical Interactions among Plants,” pp. 7-9. Nat. Acad. Sci., Washington, D.C. Aspinall, D. 1960. Ann. AppJ. Biol. 48, 637-654. Atkinson, J. 1900. Iowa Agr. Exp. Sta. Bull. 45, 1-200. Baldy, C. 1963. Ann. Agron. 14, 489-534. Baldwin, J. P., and Tinker, P. B. 1972. Plant Soil 37, 209-213. Bartholomew, B. 1970. Science 170, 1210-1212. Beardmore, J. A., Dobzhansky, T., and Pavlovsky, 0. 1960. Heredity 14, 19-33. Bell, G. D. H., Whitehouse, R. N. H., Jenkins, G., Kirby, E. J. M., and Sage, G. C. M. 1968. Rep. Plant Breed. Inst. (Camb.), 1966-1967 p. 62. Black, J. N. 1958. Aust. J. Agr. Res. 9, 299-318. Black, J. N. 1966. I n “The Growth of Cereals and Grasses” (F. L. Milthorpe and J. D. Ivins, eds.), pp. 167-178. Butterworth, London. Bode, H. R. 1958. Planta 51, 440-480. Bonner, J. 1946. Bot. Gaz. (Chicago) 107, 343-351. Borlaug, N. E. 1959. Proc. Int. Wheat Genet. Symp., lst, 1958 pp. 12-26. Borner, H. 1960. Bot. Rev. 26, 393-424. Borner, H. 1971. In “Biochemical Interactions among Plants,” pp. 52-56. Nat. Acad. Sci., Washington, D.C. Bray, R. H. 1954. Soil Sci. 78, 9-22. Broadbent, L. 1957. “Investigation of Virus Diseases of Brassica Crops.” Cambridge Univ. Press, London and New York.

BIOMASS PRODUCTIVITY OF MIXTURES

207

Brouwer, R. 1966. In “The Growth of Cereals and Grasses” (F. L. Milthorpe and J. D. Ivins, eds.), pp. 153-166. Butterworth, London. Browning, J. A. 1966. In “Plant Breeding” (K. J. Frey, ed.), pp. 233-236. Iowa State Univ. Press, Ames. Browning, J. A., and Frey, K. J. 1969. Annu. Rev. Phytophathol. 7, 355-382. Butler, E. J., and Jones, S. G. 1949. “Plant Pathology.” Macmillan, New York. Byth, D. E., and Weber, C. R. 1968. Crop Sci. 8, 44-47. Cable, D. R. 1969. Ecology 50, 27-38. Chamblee, D. S. 1958. Agron. J . 50, 434-437. Chapman, S. R., Allard, R. W., and Adanis, J. 1969. Crop Sci. 9, 575-576. Cherrett, J. M., Ford, J. B., Herbert, I. V., and Probert, A. J. 1971. “The Control of Injurious Animals.” English Univ. Press, London. Chilvers, G. A., and Brittain, E. G. 1972. Aust. J . Biol. Sci. 25, 749-756. Clay, R. E., and Allard, R. W. 1969. Crop Sci. 9, 407-412. Clements, F. E. 1904. “Lincoln, Botanical Survey of Nebraska” (quoted from Black, 1966). Clements, F. E., and Weaver, J. E. 1924. “Experimental Vegetation.” Carnegie Institution of Washington, Washington, D.C. Clements, F. E., Weaver, J. E., and Hanson, H. 1929. Carnegie Inst. Wash. Publ. 398 (quoted from Donald, 1963). Crombie, A. C. 1947. J . Anim. Ecol. 16,44-73. Davies, M. S., and Snayon, R. W. 1973. J . Appl. Ecol. 10, 33-45. de Wit, C. T. 1960. Versl. Landbouwk. Onderz. Ned. 66, 1-82. de Wit, C. T., and van den Bergh, J. P. 1965. Netlz. J. Agr. Sci. 13, 212-221. de Wit, C. T., Tow, P. G., and Ennik, G. C. 1966. Versl. Landbouwk. Onderz. Ned. 687, 1-30 Donald, C. M. 1946. J. Counc. Sci. 2nd. Res. Aust. 19, 32-37. Donald, C. M. 1958. Aust. J . Agr. Res. 9, 421-435. Donald, C. M. 1961. Symp. SOC.Exp. Biol. 156, 282-313. Donald, C. M. 1963. Advan. Agron. 15, 1-118. Drew, M. C., Nye, P. H., and Vaidyanathan, L. V. 1969. Plant Soil 30, 252-270. Duncan, W. G., Loomis, R. S., Williams, W. A,, and Hanau, R. 1967. Hilgardia 38, 181-205. Eagles, C. F. 1972. J . Appl. Ecol. 9, 141-152. Edwards, K. J. R., and Allard, R. W. 1963. Amer. Natur. 97, 243-248. Ellern, S. J., Harper, J. L., and Sagar, G . R. 1970. J. Ecol. 58, 865-868. England, F. J. W. 1965. Scot. Plant Breed. Sta. Rec. pp. 125-149. England, F. 1968. J . Appl. Ecol. 5, 227-242. Ennik, G. C. 1969. Proc. White Clover Res. Symp., 1969 pp. 165-174. Fieller, E. C. 1932. Biometrika 24, 428-440. Frey, K. J., and Maldonado, U. 1967. Crop. Sci. 7, 532-535. Fried, M., and Broeshart, H. 1967. “The Soil-plant System in Relation to Inorganic Nutrition.” Academic Press, New York. Giiumann, E. 1950. “Principles of Plant Infection” (translated W. B. Brierley). Lockwood & Son, London. Gause, G. F. 1934. “The Struggle for Existence.” Williams & Wilkins, Baltimore, Maryland. Gray, R., and Bonner, J. 1948. Amer. J . Bot. 35, 52-57. Greenwood, D. J. 1969. fn “Root Growth” (W. J. Whittington, ed.). pp. 202-223. Butterworth, London.

208

B. R. TRENBATH

Grodzinsky, A. M. 1965. “Allelopathy in Life of Plants.” Naukova Duma, Kiev. Grodzinsky, A. M. 197 1. In “Biochemical Interactions among Plants,” pp. 44-51. Nat. Acad. Sci., Washington, D.C. Grummer, G. 1955. “Die gegenseitige Beeinflussung hoheren Pflanzen-Allelopathie.” Fischer, Jena. Grummer, G. 1958. Flora 146, 158-177. Gustafsson, A. 1951. Cold Spring Harbor Symp. Quant. Biol. 16, 263-280. Gustafsson, A. 1954. Acta Agr. Scand. 4, 601-632. Hanawalt, R. B. 1971. I n “Biochemical Interactions among Plants,” pp. 33-38. Nat. Acad. Sci., Washington, D.C. Harper, J. L. 1961. Symp. SOC.Exp. Biol. 15, 1-39. Harper, J. L. 1965. Genet. Today, Proc. fnt. Congr. Genet. I l i h , 1963, Vol. 2. pp. 465482. Harper, J. L. 1967. J . Ecol. 55, 247-270. Harper, J. L. 1968, I n “Population Biology and Evolution” (R. C. Lewontin, ed.), pp. 139-158. Syracuse Univ. Press, Syracuse, New York. Harper, J. L., and Clatworthy, J. N. 1963. J . Exp. Bot. 14, 172-190. Hirano, S., and Kira, T. 1965. 1. Inst. Polytech., Osaka City Univ., Ser. D 16, 27-44. Hirano, S, and Morioka, S. 1964. J . Hort. Ass. Jap. 33, 13-22. Impens, I. I., Stewart, D. W., Allen, L. N., Jr., and Lemon, E. R. 1967. Plant Physiol. 42, 99-104. International Rice Research Institute. 1974. Annu. Rep. 1973. Iwaki, H. 1959. Jap. J . Bor. 17, 120-138. Jacquard, P., and Caputa, J. 1970. Ann. Amelior. Plant. 20, 115-158. Jennings, P. R., and Aquino, R.C. 1968. Evolution 22, 529-542. Jensen, N. F. 1952. Agron. J . 44, 30-34. Klages, K. H. W. 1936. J . Amer. SOC.Agron. 28, 935-940. Kolb, F. 1962. Z . Acker- Pflanzenbau 115, 375-406. Lampeter, W. 1960. Wiss. Z . Karl-Marx-Univ. Leipzig. Math. Naturwiss. Reihe9, 611-722. Laude, H. H., and Swanson, A. F. 1942. J . Amer. SOC.Agron. 34, 270-274. Leonard, K. J. 1969. Phyropathology 59, 1845-1850. Lin, C. S., and Torrie, J. H. 1968. Crop. Sci. 8, 585-588. Marshall, D. R., and Brown, A. H. D. 1973. Euphytica 22, 405-412. Massey, A. B. 1925. Phytopathology 15, 773-784. Milthorpe, F. L. 1961. Symp. SOC. Exp. Biol. 15, 330-355. Monsi, M., and Saeki, T. 1953. Jap. J . Bot. 14, 22-52. Monteith, J. L. 1963. In “Environmental Control of Plant Growth” (L. T. Evans, ed.), pp. 95-112. Academic Press, New York. Montgomery, E. G. 1912. Nebr., Agr. Exp. Sta., 81111. 127, Vol. 24, No. 5, 3-22. Muller, C. H. 1953. Amer. J . Bot. 40, 53-60. Muller, C. H. 1966. Bull. Torrey Bot. Club 93, 332-351. Myers, L. F., and Lipsett, J. 1958. Ausr. J . Agr. Res. 9, 1-12. Nguyen Van, J. 1968. Ann. Amelior. Plant. 18, 391-403. Nilson, T. 1968. “Rejhim solnechnoi radiatsii v rastitel’nom pokrove,” pp. 112-145. Akad. Nauk Est. S.S.R., Tartu, Est. S.S.R. Norrington-Davies, J. 1967. Euphyticu 16, 391-406. Norrington-Davies, J. 1968. J . Agr. Sci. 71, 223-231. Norrington-Davies, J., and Hutto, J. M. 1972. J . Agr. Sci. 78, 251-256. Obeid, M. 1965. Ph.D. Thesis, University of Wales.

BIOMASS PRODUCTIVITY OF MIXTURES

209

O’Brien, T. A., Whittington, W. J., and Slack, P. 1967. J . Appl. Ecol. 4, 501-520. Olsen, S. R., and Kemper, W. D. 1968. Advan. Agron. 20, 91-151. Olsen, S. R., Kemper, W. D., and Jackson, R. D. 1962. Soil. Sci. SOC. Amer., Proc. 26, 222-227. Pavlychenko, T. K., and Harrington, J. B. 1935. Sci. Agr. 16, 151-160. Pfahler, L. A. 1965. Crop Sci. 5, 271-275. Probst, A. H. 1957. Agron. J. 49, 148-150. Proebsting, E L., and Gilmore, A. E. 1940. Proc. A m e r . SOC. Hort. Sci. 38, 21-26. Qualset, C. O., and Granger, R. M. 1970. Crop Sci. 10, 386-389. Rademacher, B. 1959. I n “Handbuch der Pflanzenphysiologie” (W. Ruhland, ed.), Vol. 11, pp. 655-706. Springer-Verlag, Berlin and New York. Raper, C. D., and Barber, S. A. 1970. Agron. J. 62, 581- 584. Rasmusson, D. C. 1968. Crop Sci. 8, 600-602. Rhodes, I. 1968a. J. Brit. Grassland SOC. 23, 129-136. Rhodes, I. 1968b. J . Brit. Grassland Soc. 23, 156-158. Rhodes, I. 1968c. J . Brit. Grassland SOC. 23, 330-335. Rhodes, I. 1970a. J . Brit. Grassland SOC.25, 285-288. Rhodes, I. 1970b. Herb. Abstr. 40, 115-121. Richardson, A. E. V., Trumble, H. C., and Shapter, R. E. 193 1. Corrnc. Sci. I n d . Res. Artst. Bull. No. 49. Risser, P. G. 1969. Bot. Rev. 35, 251-284. Roy, S. K. 1960. J . Genet. 57, 137-152. Sakai, K. 1953. Jap. J . Bot. 14, 161-168. Sakai, K. 1955. Cold Spring Harbor Symp. Qirant. Biol. 20, 137-157. Sakai, K., and Iyama, S. 1959. Annrr. Rep. Nut. Inst. Genet. (Jap.) 9, 63-64. Sakai, K., and Oka, H. 1955. Annrr. Rep. Nut. Inst. Genet. (Jap.) 5, 56-58. Salisbury, E. J. 1929. J . Ecol. 17, 197-222. Sandfaer, J. 1968. Nature ( L o n d o n ) 218, 241-243. Sandfaer, J. 1970a. Hereditas 64, 150-152. Sandfaer, J. 1970b. Dan. A t . Energy C o m m . , Riso Rep. No. 230. Schander, H. 194 I . Bodenk. Pflanzenernaelrr. 20, 129-1 5 1. Schutz, W. M., and Brim, C. A. 1967. Crop Sci. 7, 371-376. Schutz, W. M., and Brim, C. A. 1971. Crop Sci. 11, 684-689. Schwerdtfeger, F. 1954. Allg. Forstz. 9, 278-282 (quoted from VoOte, 1964). Seaton, A. P. C., and Antonovics, J. 1967. Heredity 22, 19-33. Sechler, D., and Chapman, W. H. 1967. Soil Crop Sci. SOC.Fla., Proc. 27, 68-72. Sibma, L., Kort, J., and de Wit, C. T. 1964. Jaarb. Inst. Biol. Sclieik. Onderz. Laridbouwgewassen, Wageningen, Meded. pp. 119-1 24. Simrnonds, N. W. 1962. Biol. R e v . 37, 442-465. Snaydon, R. W. 1971. J . Appl. Ecol. 8, 687-697. Stern, W. R., and Donald, C. M. 1962a. Aust. J . Agr. Res. 13, 599-614. Stern, W. R., and Donald, C. M. 1962b. Arrst. J. Agr. Res. 13, 615-623. Stringfield, G. H. 1959. Agron. J . 51, 472-473. Sukatschew, W. 1928. Z. Indukt. Abstamm.- Vererbungsl. 47, 54-74. Suneson, C. A. 1949. Agron. J . 41,459-461. Suneson, C. A. 1960. Agron. J. 52, 319-321. Syme, J. R. 1963. Ph.D. Thesis, University of Nottingham. Syme, J. R., and Bremner, P. M. 1968.1. Appl. Ecol. 5, 659-674. Tahvanainen, J. O., and Root, R. B. 1972. Oecologia (Berlin) 10, 321-346. Tansley, A. G. 1917. J. Ecol. 5, 173-179. Thornson, A. J. 1969. J. Agr. Sci. 73, 321-328.

210

B. R. TRENBATH

Trenbath, B. R. 1970. M.Sc. Thesis, University of Wales. Trenbath, B. R. 1972. Ph.D. Thesis, University of Adelaide. Trenbath, B. R., and Harper, J. L. 1973. J . Appl. Ecol. 10, 379-400. Tsedik-Tomashevich, Z. F. 1951. Agrobiologiya 1, 100-121. van den Bergh, J. P. 1968. Versl. Landbouwk. Onderz. Ned. 714, 1-71. van den Bergh, J. P., and de Wit, C. T. 1960. Meded. Inst. Biol. Scheik. Onderz. Landbouwgewassen, Wageningen, Meded. 121, 155-165. van den Bergh, J. P., and Elberse, W. T. 1962. J . Ecol. 50, 87-95. van Dobben, W. H. 1955. Versl. Cent. Inst. Landbouwk. Onderz. pp. 128-131 (quoted from de Wit, 1960). Verhagen, A. M. W., Wilson, J. H., and Britten, E. J. 1963. Ann. Bot. (London) [N.S.] 27, 627-640. Voiite, A. D. 1964. Int. Rev. Forest Res. 1, 325-383. Warne, L. G. G. 1953. J . Hort. Sci. 28, 152-159. Warren Wilson, J. 1960, Proc. Int. Grassland Congr., 8th, 1960. pp. 275-279. Watson, D. J. 1952. Advan. Agron. 4, 101-145. Webb, L. J., Tracey, J. G., and Haydock, K. P. 1967. J . Appl. Ecol. 4, 13-25. Welbank, P. J. 1961. Ann. Bot. (London) [N.S.] 25, 116-137. Went, F. W. 1942. Bull. Torrey Bot. Club 69, 100-114. Whitehouse, R. N. H., Kirby, E. J . M., and Sage, G. C. M. 1967. Rep. Plant Breed. Inst. ( C a m b . ) 1965-1966 p. 62. Whittington, W. J., and O’Brien, T. A. 1968. J . A p p l . Ecol. 5, 209-213. Williams, E. J. 1962. Ausr. J . Biol. Sci. 15, 509-525. Williams, W. A. 1963. Ecology 44, 475-485. Winter, A. G. 1961. Symp. Soc. Exp. Biol. 15, 229-243. Woodford, E. K. 1966. J . Brit. Grassland SOC.21, 109-115. Workman, P. L., and Allard, R. W. 1964. Heredity 19, 181-189.

AMORPHOUS CLAY CONSTITUENTS OF SOILS Koji Wada and M. E. Harward Kyushu University, Fukuoka, Japan and Soil Science Department, Oregon State University, Corvallis, Oregon

........................... ..................... ..................... ..................... ............................................. A. Opaline Silica ............................................... B. Aluminum and Iron Oxides and Hydrous Oxides . . . . . . . . . . . . . . . . . . C. Allophane, Imogolite, and Related Silicates ...................... IV. Identfiication and Quantitative Estimation . . .................. V. Formation and Transformation . . . . . . . . . . . . .................. A. Opaline Silica ....................... .................. I. Introduction

11. Definition and Scope 111. Nature of Materials

B. Aluminum and Iron Oxides, Hydraxides, a anic Complex’es . . . . C. Allophane, Imogolite, and Related Silicates ...................... VI. Relationship to Soil Properties .................................... A. Chemical Properties .......................................... B. Physical and Engineering Properties ............................ C. Current Status and Soil Systems . . . . . . . . . . . . . . . . . . VII. Summary . . . . . . . . . . .............................. References .....................................................

1.

211 212 213 213 214 217 230 233 233 234 235 242 242 249 252 253 254

Introduction

Emphasis in the study of clay components of soil has changed with time. In the 1920’s colloid chemists stressed the analogy between soil clays and the synthetic amorphous silicates mixed with hydrous oxides. They made many important contributions, such as elucidation of electrochemical interactions between clay surfaces and ions in solution. With application of X-ray diffraction and discovery of crystalline clays in soils, a shift in emphasis occurred. In the decades from 1930 to 1960, research emphasis was placed on characterization of crystalline layer silicates in soils and on interpretation of behavior in terms of crystal structure. There has been a renewal of interest in amorphous constituents during the last decade, and “the pendulum has returned to a more balanced position.” In a review on the clay fraction of soiIs, Rich and Thomas (1960) pointed out that there was a research trend toward greater emphasis on the amorphous con211

212

KOJl WADA AND M. E. HARWARD

stituents. Mitchell et al., reviewed developments to 1964 in studies of amorphous inorganic material in the clay fraction and indicated the lines of future studies. Scientists are increasingly aware of the importance of amorphous phases in soils and sediments. Amorphous clay materials are often described by the term active because of their large specific surface area and high chemical reactivity. When present in a substantial amount, they have a very marked effect on both physical and chemical properties of soils. This is particularly true for many soils derived from volcanic ashes, but their importance has also been suggested for other soils even though present in a relatively small amount. The objective of this article is to review recent progress in studies of the amorphous clay constituents of soils and to outline our present knowledge on their nature, properties, and genesis. Effort was made to collect analytical data on natural clay systems, although they often have been inconclusive, and to indicate similarities and differences between different kinds of amorphous clay materials. Many of the data obtained for synthetic amorphous hydrous oxides and aluminosilicates have not been included, and their use as models is imperative in future studies. We will try to avoid repetition of material presented in the excellent review by Mitchell et al. ( 1964). Rather, emphasis will be placed on the advances in research during the last ten years.

II.

Definition and Scope

Brindley (1969) defined the amorphous state as one which lacks order and is distinguished from the crystalline state which has order. Crystalline is a relative term. By convention, X-ray diffraction has been used as the criterion for crystallinity. Materials which are sufficiently well organized to yield X-ray diffraction patterns are said to be crystalline whereas those which do not are called amorphous. However, some materials are “amorphous” to X-rays but give diffraction phenomena when examined with an electron beam. Consequently, Brindley ( 1969) recommended a more satisfactory distinction between them as matter having long-range order (crystalline state) and short-range order (noncrystalline state). Long-range order has a unit cell repetition, generally in three dimensions, but sometimes in two, and possibly only in one, dimension. Some short-range, nonrepetitive order will normally be present in the so-called amorphous substances. Liquids and amorphous solids such as glass possess a degree of short-range order due to coordination tendencies of atoms associated with systems of electrostatic ionic packings and/or directed covalence bonds.

AMORPHOUS CLAY CONSTITUENTS OF SOILS

21 3

This article mainly deals with the clay materials which are, and have been thought to be, amorphous. The term amorphous clay materials is retained as a conventional and inclusive term. This includes materials having only short-range order and those having long-range order in only one dimension. For perspective and continuity of discussion of relationships to properties and genesis, it is desirable to include poorly crystallized materials such as imogolite and certain forms of halloysite. Chemically, the principal forms of the amorphous clay materials in soils are oxides and hydroxides of iron, aluminum, and silicon, and the silicates of aluminum and iron, all in various combinations with water. These materials commonly occur as particles smaller than 2 pm equivalent diameter as weathering products of primary minerals, and it is these which are to be considered here. Other amorphous components such as unweathered volcanic glass and biogenic opal (phytoliths) are not included in the present discussion. Both of these are X-ray amorphous, but are present mostly as particles larger than 2 pm in diameter. Most soil organic constituents are also present in an amorphous state. They are extremely important in physical and chemical reactions and properties of soils. However, limitations of space and our expertise preclude them from inclusion here. The interaction between the amorphous inorganic and organic constituents will be discussed briefly. Where appropriate, comments will be given on the nomenclature of each amorphous clay material. The AIPEA Nomenclature Committee has recommended that specific names be not given to poorly defined clay materials, such as irregular interstratified systems or imperfect structures, or to amorphous constituents (Brindley and Pedro, 1972).

Ill.

A.

Nature of Materials

OPALINE SILICA

Not all the materials called opal in the past are X-ray amorphous. Jones and Segnit (1971 ) classified opal into three well-defined structural groups: Opal C (well-ordered a-cristobalite) , Opal Ct (disordered a-cristobalite or a-tridymite) and Opal A (highly disordered, nearly amorphous). Most precious opals give X-ray powder patterns with only a broad diffuse band around 4.1 A. Electron microscopy shows that the play of colors, or “fire,” in them is caused by diffraction effects from arrays of uniformly sized silica spheres of 1700-3500 A in diameter (Jones et al., 1964). These spheres are secondary aggregates of primary silica particles up to 400 A across, often arranged in concentric shells. The perfection of shape and arrange-

KOJI WADA AND M. E. HARWARD

214

ment of the secondary aggregates is greatest in precious opal and least in varieties of natural opaline silica such as siliceous duricrust, wood opal, and silica deposited in plants. In the latter, the secondary aggregates are less than 1500 A in diameter and irregular in shape (Darragh et al., 1966). Laminar opaline silica particles were found in the 0.2-20 pm fractions, and most abundantly in the 0.4-2 pm fractions, of volcanic ash soils by Shoji and Masui (1969a,b, 1971). They distinguished four morphological types; circular, elliptical, rectangular, and rhombic. The thickness of the particles varies from one-twentieth to one-fifteenth of their diameter and they show fine-grained uneven surfaces. Weathered and alkali-treated opaline silica particles appear to be very porous suggesting that they are actually composed of extremely fine silica spheres. No data for chemical composition of the isolated opaline silica particles are available. They are soluble in hot 0.5 N NaOH. The Si0,/A1,08 ratios of the 0.5 N NaOH soluble fraction of soil clays which contain many of the particles were in the range from 2.4 to 21.7. Refractive indices and specific gravities of opaline silica particles are in the range from 1.42 to 1.43 and from 2.1 to 2.3, respectively. Shoji and Masui ( 1969a) indicated that opaline silica has two absorption maxima on the infrared spectra at 1100 and 800 cm-l which are assigned to the Si-0 vibrations. The difference spectrum of a 0.5 N NaOH soluble fraction of a volcanic ash soil clay in which opaline silica is present in a fair amount, showed absorption maxima at 1200 (shoulder), 1075, 935 (shoulder) and 790 cm-* (Tokashiki and Wada, 1972b). These frequencies are the same as those found for amorphous silica gels synthesized in the laboratory (Mitchell et al., 1964; Leonard et al., 1964).

B.

ALUMINUM AND IRONOXIDESAND HYDROUS OXIDES

The name “limonite” has often been used for amorphous, hydrated iron oxides with composition Fe,O,.nH,O (Brown, 1955) or with a molecular formula such as 2Fe20,.3H,0. Rooksby (1961) points out that most of the limonite with the latter composition are very finely divided goethite and they hold moisture over that for a monohydrate, probably by adsorption forces. Very little information about the composition, structure, and morphology of discrete amorphous oxides and hydrous oxides of aluminum and iron present in soils is available. Many of these materials exist as coatings of gel-hull5 on other particles (Jones and Uehara, 1973) rather than discrete units. Under such conditions it is difficult to separate the amorphous and crystalline components without changing the nature and properties. De Villiers (1969) reported the occurrence of an amorphous alumina of a boehmite character in certain tropical soils. The evidence provided is

AMORPHOUS CLAY CONSTITUENTS OF SOILS

215

that the amounts of gibbsite in these soil clays, as measured by differential thermal analysis, were lower than those found using the selective dissolution procedure of Hashimoto and Jackson (1960) and that the weight losses between 110” and 330°C fell short of those expected on the basis of dissolution treatments. However, there is an implied but unwarranted assumption in this that no crystalline constituents other than gibbsite are dissolved in the hot 0.5 N NaOH. The occurrence of ferric hydroxide gel in Akaka soils in Hawaii was reported by Matsusaka and Sherman (1961). The magnetic attraction of the soil increased from 2.2 mg per gram of soil when naturally moist, to 3.7 mg per gram of soil when air-dried, and to 9.2 mg per gram of soil when heated to 600°C. On the basis of X-ray and differential thermal analyses, they suggested an alteration sequence of amorphous ferric hydroxide -+ cryptocrystalline lepidocrocite + maghemite system. Sherman et aE. (1964) also observed gibbsite in the dried Akaka soils, but not in the undried soils. Microscopic examination indicated the occurrence of gels in channels and pores in the soils. They then suggested segregation and crystallization of gibbsite from the amorphous aluminum-iron gels. No mention was made, however, of the effect of moisture in detection of crystallinity by X-ray analysis in these highly hydrated materials. The most common forms of amorphous aluminum hydroxy materials in soils are probably those which exist as aluminum-hydroxy interlayer “islands” in expansible layer silicates. Their nature and occurrence has been reviewed by Rich (1968). The aluminum-hydroxy interlayers appear in a gibbsite-like monolayer structure, and disperse randomly as “islands” in the interlayer space. The hydroxy/aluminum ratio of the interlayer material present in soils is a question that is largely unresolved, but synthetic studies (Turner, 1965; Hsu, 1968) have shown that the hydroxy-aluminum species with the OH/AI ratio of 2.5 to 2.7 are favorably retained by montmorillonites. The extent of interlayering is small compared with the amount required for formation of a complete gibbsite-like monolayer, which is about 16 meq of aluminum per gram of montmorillonite. Brydon and Kodama (1966) reported that amounts up to 8 meq of aluminum per gram enters entirely into the interlayer space and that beyond this some hydroxide is present external to the montmorillonite interlayer region. That there are two kinds of OH-groups in the hydroxy-aluminum interlayers was indicated by Brydon and Kodama (1966) and Weismiller et al. (1967) using infrared spectroscopy. The latter authors assigned the 3695 cm-I pleochroic band to the inner hydroxy groups of a gibbsite-like ring [Al(OH),(H,O) ,Is6+ and the 3570 cm-1 nonpleohroic band, to the outer OH-groups. A deuteration study by Ahrlichs (1968) showed that flushing with D,O at room temperature removed water, and heating at 100°C in D,O vapor exchanges interlayer OH groups but not the clay lattice OH

216

KOJI WADA AND M. E. HARWARD

groups. These observations were made on interlayer materials prepared in the laboratory and the results of similar studies on natural counterparts are needed. Hydroxy iron interlayers exist in the phyllosilicates although they apparently are less common than hydroxy aluminum. In 1940, Bower and Truog reported that positively charged ferric dihydroxy ions were held by exchange positions; under certain conditions the clays adsorbed three times the exchange equivalent of ferric iron. Synthetic hydroxy iron interlayers in clay result in properties similar to those with hydroxy aluminum (Clark, 1964; Coleman and Thomas, 1964). Consequently, routine identification procedures based on properties of systems would not always result in recognition of the presence of iron hydroxy interlayers. The little evidence for naturally occurring interlayers composed largely of hydroxy-iron groups led Rich (1968) to suggest that occlusions of Fe(OH), in positively charged hydroxy-aluminum interlayers appear to be more likely. This indicates that iron oxides and hydrous oxides are more stable than hydroxyFe3+interlayers. Singleton and Harward ( 1971 ) have presented evidence for iron-hydroxy interlayers in two soil clays of Western Oregon. The occurrence of iron-hydroxy interlayers probably relate to moisture status, pH, and other factors which affect oxidation-reduction. Jackson (1962) suggested that iron might be involved in interlayering of clays since wetness and gleying seemed to clean up intergrade minerals. I t was implied that under anaerobic conditions the ferric dihydroxy ion would be reduced to ferrous, thus losing its capacity to be retained. Lynn and Whittig (1966) obtained evidence which suggested chlorite formations may be correlated with high ferrous ion concentrations in undrained tideland sediments. Polymeric hydroxy-aluminum may also be present in association with allophane. The negative charge of allophane is presumed to arise from substitution of aluminum for silicon in a tetrahedral silica framework. In order to account for its characteristic pH-dependent negative charge and phosphate adsorption, blocking of exchange sites by hydroxy-aluminum was inferred by de Villiers and Jackson (1967), Cloos et al. (1968; 1969), and de Villiers (1971 ). No specific information about the composition and structure of polymeric hydroxy-aluminum which may be present in allophane has, however, been available. An aluminosilica gel prepared by addition of monomeric silica to hydroxy-aluminum solution gave absorption on the infrared spectra in the region from 860 to 930 cm-l and an endotherm with maximum extending up to 200°C on the differentia1 thermal analysis curve (Wada and Kubo, 1972, 1973). These features have not been seen for either allophane or imogolite separated from soils. Available evidence indicates that hydrous oxides of aluminum and iron as well as amorphous silicates are the most important materials involved

AMORPHOUS CLAY CONSTITUENTS OF SOILS

217

in the interaction between clays and organic matter in soils (Greenland, 1971). The presence of aluminum and iron hydroxides in combination with organic matter has been indicated by extracting soils with reagents such as sodium pyrophosphate, Tamm’s acid oxalate, and citrate-dithtonite solution, which will be described in Section IV. Again, there is very little information about the composition and structure of these hydroxy compounds bound with organic matter. C. ALLOPHANE, IMOGOLITE,

AND

RELATEDSILICATES

The term allophane has been used with different meanings by different investigators. Ross and Kerr described allophane as an amorphous member of the kaolin group (Keller, 1964). This has a specific meaning and implies some chemical combination involving aluminum and silicon. Apparently Ross and Kerr also suggested that the term be used for all amorphous clay materials regardless of their composition (Grim, 1953). Fieldes (1966) and Furkert and Fieldes (1968) also proposed the use of the name allophane as any clay size material characterized by structural randomness. This concept likely results in discouraging recognition of important differences in chemical structure and surface properties of various amorphous clay constituents in soils. The problem was reviewed at the United States-Japan seminar on amorphous clay materials in 1969. A tentative definition adopted by this group will be used in this article. This defines allophanes as members of a series of naturally occurring hydrous aluminosilicates of widely varying chemical composition and which are characterized by short-range order, by the presence of Si-0-A1 bonds, and by a differential thermal analysis curve displaying a low-temperature endotherm and a high-temperature exotherm with no intermediate endotherm (van Olphen, 1971) . Alumina and silica would not be the end members of the series since allophanes are aluminosilicates containing AI-0-Si bonds. The name hisingerite has been used in two ways. Brown (1955) defined it as an iron analog of allophane with composition of Fe,O,-ZSiO,.nH,O. However, no data have been available on hisingerite in this sense. On the other hand, Gruner( 1935) described hisingerite as a mineral species which gives broad diffraction lines which coincide with those of nontronite, but which may be an “amorphous” variant. As pointed out by MacEwan ( 1961) this expression “amorphous” should perhaps be interpreted as “very finely crystalline.” Whelan and Goldich (1961) and Lindqvist and Jannson (1962) evaluated selected hisingerite samples and found three broad diffraction lines at 4.3 to 4.6, 2.57 to 2.63, and 1.54 to 1.60 A. The molecular ratios

KOJI WADA AND M. E. HARWARD

21 8

+

SiOz/(Fe,O, FeO) varied from 1.4 to 3.2 and the ratios H,0(+)/(Fe203 FeO) varied from 1.0 to 2.1. The infrared spectra exhibited a broad, featureless Si-0 absorption band with a maximum at 1000 cm-I. The diffraction data suggested development of a layer lattice but restricted crystal growth in the c axis. Whelan and Goldich (1961) considered the material as principally poorly crystallized iron saponite whereas Lindqvist and Jansson (1962) considered it to be mica mineral with extensive substitution of Fe for Si and with interlayer hydronium ions. Both investigators, however, stressed that hisingerite definitely warrants further study. Some of the more recent advances in characterization of amorphous and poorly crystallized materials involve imogolite. Imogolite is a hydrous aluminum silicate having a fine threadlike nature and unique diffraction characteristics; it was first described by Yoshinaga and Aomine (1962b). The name imogolite as a new mineral species has been approved by the AIPEA Nomenclature Committee (Brindley and Pedro, 1970). Imogolite exhibits long-range order in one dimension and is included herein because of its frequent association with allophane and the similarities in chemical and physical properties.

+

1 . Chemical Composition

Data from chemical analysis of allophane and imogolite derived from weathering of volcanic ash and pumice were provided by Yoshinaga and Aomine (1962a,b), Yoshinaga (1966, 1968), and Miyauchi and Aomine (1966b). For these analyses, the <0.2 pm fractions were collected and pretreated with dithionite-citrate (Mehra and Jackson, 1960) and 2%Na,C03 solution (Jackson, 1956). From these data, Wada and Yoshinaga (1969) showed that the SiO,/AI,O, ratio of the clays in which allophane predominates is in a range from 1.3 to 2.0, while that of the clay in which imogolite predominates is in a fairly narrow range from 1.05 to 1.15. The H,O(+)/ALO, ratio is mostly in a range from 2.5 to 3.0 without any definite difference between the two groups of clays. Higher SiO,/Al,O, ratios for imogolite, 1.5 (Russell et al., 1969) and 1.29 to 1.32 without any chemical pretreatment (Tazaki, 1971) , as well as lower values near 1 for allophane also have been reported (Russell et al., 1969; Lai and Swindale, 1969). Iimura (1969) gave Si0,/A1,03 ratios of 0.89 to 1.43 for five allophanic clays which had been dispersed at pH 4 without any chemical pretreatment. Egawa (1964) and Udagawa et al. (1969) used X-ray fluorescence spectroscopy to obtain data on the coordination number of aluminum in allophanic clay separated from a Kanuma pumice bed. Their samples were different in Si02/A120, ratios, namely, 2.37 and 1.67, but alike in the

AMORPHOUS CLAY CONSTITUENTS OF SOILS

219

coordination status of aluminum; about 50-60% of the aluminum was estimated to be in 4-fold coordination at room temperature and at 110OC. The aluminum K , line profile for samples treated with dithionite-citrate and 2% Na,CO, suggest that a considerable part of the aluminum in allophane (Si0,/AI,03 ratio close to 2.0) is in 4-fold coordination while nearly all of the aluminum in imogolite is in 6-fold coordination (T. Henmi and K. Wada, unpublished). Yoshinaga (1 968) confirmed that a small but significant amount of iron (0.3-0.9% as Fe,O,) remains in allophane even after repetition of ten dithionite-citrate treatments. Kitagawa ( 1973) interpreted electron resonance spectra of two allophanic clays as indicating the substitution of aluminum by iron similar to that in mica and montmorillonite. Gotz and Masson (1970, 1971) developed a chemical procedure for differentiating silicate anions possessing low degrees of polymerization. The procedure is based on conversion of the anion to a trimethylsilylether, and subsequent identification and quantitative determination of the volatile ether by gas chromatography. Application of this technique to imogolite gave a high yield of volatile products of which 95% was the orthosilicate ether and 5 % the pyrosilicate ether. This furnished evidence in favor of the presence of isolated orthosilicate groups (Cradwick et al., 1972). 2. Morphology As a result of improvements in instrumentation and techniques of preparation, it has become possible to observe clay particles with a resolution of about 5 A in the electron microscope. High resolution electron microscopy shows that the imogolite threads which previously appeared to be of micron length with a diameter of 100-300 A actually consist of a finer filiform unit with separations in the order of 18-22 A (Yoshinaga et al., 1968; Russell et al., 1969). More recent studies on the samples dispersed on a microgrid or cut in thin section (Wada et al., 1970) suggest that this filiform unit is a tube with the inner and outside diameters of about 10-20 A (Fig, 1 top). This tubular unit is very useful for identification and detection of imogolite even when present in a very small amount, as illustrated in Fig. 2. In this particular sample, imogolite has not been identified by other analytical techniques. Eswaran (1972) studied volcanic ash soils containing imogolite with the scanning electron microscope and found threads or ribbons forming peculiar globules. The much thicker threads up to 30,000 A in diameter led him to suggest that pretreatment for transmission electron microscopy breaks up these threads of imogolite along planes of weakness. However, neither these thicker threads nor globules have been found in the same volcanic ash soil by scanning electron microscopy. Planar nets or films

220

KOJI WADA AND M. E. HARWARD

FIG. 1. (Top): Electron micrograph of imogolite formed in saprolite of basalt obtained by N. Yoshinaga. This sample was studied by Patterson (1964) and Wada et al. (1972). Scalemarker: 1000 A. (Bottom): Scanning electron micrograph of imogolite in a glassy volcanic ash soil “Irnogo” (Lab. No. 905) obtained by courtesy of H. Yotsumoto of Japan Electron Optics Laboratory Company Ltd. Scalemarker: 10 pm.

AMORPHOUS CLAY CONSTITUENTS OF SOILS

22 1

consisting of twisting threads with diameter about 1000 A have been observed with unweathered glass shards [Fig. 1 (bottom)]. Similar objects were previously observed in the carbon replica of the gel film which consists exclusively of imogolite separated from pumice beds (Wada and Matsubara, 1968; Yoshinaga et al., 1968; Yoshinaga and Yamaguchi, 1970b). Kitagawa (1971 ) obtained high resolution electron micrographs of several allophanic clays separated from weathered pumice beds in Japan. He interpreted them as indicating that allophane has also a structural unit, which is probably a hollow sphere of about 55 A in diameter. The presence of similar objects described two-dimensionally as ringlets had been observed as an admixture of imogolite in the gel films from weathered pumice beds (Yoshinaga, 1968; Wada and Yoshinaga, 1969). Figure 2 shows an electron micrograph of allophane with a Si0,/A120, ratio 2.0 formed in weathered volcanic ash. It seems to consist of “hollow” spherical particles as described by Kitagawa (1971) though their outside diameter varies from 30 to 55 A. It is worth noting here that a prolonged exposure of both allophane and imogolite to the electron beam likely results in loss of much

FIG.2. Electron micrograph of allophane formed in weathered volcanic ash obtained by T. Henmi. This sample (VA) was studied by Aoniine and Wada (1962) and Wada and Tokashiki (1972). Scalemarker: 1000 A.

222

KOJI WADA AND M. E. HARWARD

FIG.3. Electron micrograph of halloysite formed in weathered volcanic ash obtained by courtesy of H. Yotsumoto and S.Aida of Japan Electron Optics Laboratory Company Ltd. This sample (VH) was studied by Aomine and Wada (1962) and Wada and Tokashiki (1972).

AMORPHOUS CLAY CONSTITUENTS OF SOILS

22 3

finer detail in electron micrographs as shown in Figs. 1 (top) and 2. Jones and Uehara (1973) have also stressed the danger of prolonged exposure under the electron beam. Figure 3 shows an electron micrograph of halloysite formed in the same volcanic ash bed as allophane shown in Fig. 2 (Aomine and Wada, 1962). It shows a peculiar feature of morphology common to the volcanic ashderived halloysite, possibly through allophane. Developments of long-range order along the c axis and within the ab plane are evidenced by appearance of the lattice image with separations of 7 A ( a ) and 4.4 A ( b ) on the flakes of different orientations. 3. Surface Area

Specific surface areas of allophanic clays separated from soils developed on volcanic ash were measured by Aomine and Otsuka ( 1968). Adsorption of polar liquids such as glycerol, ethylene glycol monoethyl ether (EGME) , or water gave surface areas of about 500 mz/g. Cationic compounds such as cetylpyridinium bromide and o-phenanthroline gave zero or very small values for surface area, probably an effect of predominating positive charge on the allophane surface. The surface area derived from low-temperature nitrogen adsorption was almost one-third of that measured with polar liquids. From these data, they concluded that the size of allophane particles should be less than 50 A in diameter if they are spherical and that aggregation upon drying and heating reduces the accessibility of nitrogen gas molecules to the surface. Kitagawa (1971 ) also obtained agreement between the surface area of allophanic clays calculated from the particle size as measured by electron microscopy ( 5 5 A in diameter) and specific gravity ( 1.9) and that measured with glycerol (600 mz/g). In these determinations, the cross-sectional area of the polar molecule adsorbed on allophane was assumed to be equal with that adsorbed on 2: 1 expanding clay minerals. The formation of a monolayer with such adsorbates as glycol and glycerol was also tacitly assumed. Both assumptions are, however, not warranted. In this regard, imogolite would serve as a reference surface system. Electron microscopy indicates that imogolite consists of a tubular structure with the inner and outside diameters of about 7-1 0 A and 17-2 1 A, respectively (Wada et al., 1970). Using the measured density 2.65 g/cm3 (Wada and Yoshinaga, 1969) and neglecting the edge surface area, the specific surface would then be calculated to be 1400 to Top: A general appearance of halloysite spherules and flakes. Scalemarker: 1000 A. Bottom: Lattice imaged which appear on halloysite spherules and flakes due to (001) with a separation of 7 A ( a) and due to (02, 11) with a separation of 4.4 A (b). Scalemarker: 100 A.

224

KOJI WADA AND M. E. HARWARD

1500 m*/g. Egashira and Aomine (1974) obtained values for surface area in the order of 1000-1100 m*/g for freeze-dried imogolite specimens by measuring retention of EGME and applying the cross-sectional area of 52 A for the EGME molecule in the monolayer. On the other hand, an area of about 700 m*/g is calculated from the adsorption of water vapor (Yoshinaga, 1968; Wada and Yoshinaga, 1969; Wada and Henmi, 1972). Water retention at p / p o = 0.2 and the recommended cross-sectional area of 10.8 AZfor the water molecule are used in the calculations. The data indicate that the cross-sectional area occupied by the EGME and water molecules on imogolite should be higher than that of the expanding 2: 1 clays by as much as 40 and loo%, respectively. Wada and Henmi (1972) attempted to correlate the maximum retention of quaternary ammonium chlorides and water by imogolite with its micropore space. It was concluded from the data that both the inter- and intrastructure-unit pores can accommodate water or quaternary ammonium chloride preferably with deliquescent nature. The intra-structure-unit pores exclude the quaternary ammonium ion larger than 9-1 1 A in an equivalent spherical diameter. The porosity in air-dry imogolite accounts for 55-60% of the total volume. Roughly 50 and 25% of the total porosity belong to the inter- and intra-structure-unit pores, respectively.

4 . Structure

a. X-ray and Electron DifJraction. X-ray and electron diffraction analyses are very usefuI in studying clay materials which have long-range order even in only one dimension. Imogolite gives a unique electron diffraction pattern depending on the orientation of its threads (Yoshinaga et al., 1968; Russell et al., 1969; Wada and Yoshinaga, 1969; Cradwick et al., 1972). The electron diffraction pattern from randomly oriented threads of imogolite shows a series of rings at 1.4, 2.1, 2.3 (broad), 3.3 (broad), 3.7, 4.1, 5.7 (broad), 7.8 (broad), 11.8 (broad), and 21-23 A. A parallel alignment of the threads results in strong arcing of the reflections at 1.4 and 4.15 A on the meridian and at 5.7, 7.8, 11.8, and 21-23 A on the equator. The meridian reflections were interpreted as indicating a repeat unit of 8.4 A along the fiber axis, and the equatorial reflections were related to the lateral arrangement of the structure unit. All the corresponding reflections except those at 21-23 A and 11.8 A appear on the X-ray diffraction pattern for the randomly oriented specimen of imogolite. The reflections with maxima at 13-14, 7.8, and 5.6 A were enhanced for the “parallel” orientation specimen. A striking feature in the X-ray diffraction of imogolite is a remarkable intensity increase at 18-19 upon heating at 100-200°C. This was attributed to rearrangement of the structure units upon dehydration and partial dehydroxylation (Wada and

AMORPHOUS CLAY CONSTITUENTS OF SOILS

225

Yoshinaga, 1969). This reaction was prevented either by the presence of dithionite-citrate extractable oxides (Wada and Tokashiki, 1972) or by presence of KCH,COO, alkylammonium chlorides, and humified material (Inoue and Wada, 1963, 1971b; Wada and Henmi, 1972). Imogolites from certain volcanic ash soils have shown only a weak diffraction effect at 18-19 A or a mere collapse of the broad band in the region upon heating (Aomine and Miyauchi, 1965; Greenland et al., 1969; Tokashiki and Wada, 1972b). Destruction of the imogolite structure occurs upon heating at 350-400"C (Yoshinaga and Aomine, 1962b). Allophane gives two broad reflections with intensity maxima at 3.3 and 2.2 A on both the electron and X-ray diffraction patterns (Yoshinaga and Aomine, 1962a; Wada and Yoshinaga, 1969). b. Infrared Spectroscopy. The usefulness and potentialities of infrared spectroscopy for the study of amorphous clay materials have been pointed out by Farmer and Russell (1967) and White (1971 ) in their reviews on application of this technique to clay studies. Mitchell et al. (1964) and Leonard et af. (1964) showed that there are correlations between the inlrared spectra and the composition of the synthesized silica, aluminosilica, and alumina gels. The principal differences appear in the absorption bands in the region 650 to 1200 cm-I. The differences between allophanes with different SiO,/Al,O, ratios and imogolite were shown by Wada (1966) on the spectra recorded by use of D,O as a mulling reagent. Allophanes with the SiO2/Al20, ratios from 2 to 1.5 show a broad absorption band with maxima at 1010 and 945 cm-l, whereas imogolite shows an absorption band with maxima at 990, 955, and 925 cm-I. Fieldes and Furkert (1966) observed for the fine clay fractions from New Zealand volcanic ash soils that the distribution of absorption intensity in the region from 800 to 1300 cm-' varies in a systematic way with increasing maturity of the soil. They considered that these differences probably reflect changes in the Si02/AI,0, ratio of the aluminosilicates and in the amounts of associated oxides. Kanno et al. (1968) showed that the Si-0 absorption maximum of the clay and silt separates from weathered volcanic ash and pumice shifts from 1060 to 970 cm-' with increasing Al,O, content. Lai and Swindale (1969) found a similar relationship for fine clay fractions of three Hawaiian and one Japanese volcanic ash soils. The spectral feature of imogolite was discussed by Wada (1966) and by Russell et al. (1969). Wada (1966, 1967) interpreted a strong band at 925 cm-1 as resulting from a high proportion of Si-O-A1 linkages relative to Si-0-53, while Russell et al. (1969) interpreted it as indicating the presence of isolated Si,Oi or SiO, groups. As described previously in Section 111, C, 1, a chemical procedure for differentiating silicate anions of

226

KOJI WADA AND M. E. HARWARD

low degrees of polymerization showed that isolated SiO, groups form an overwhelming majority of silicate groups in imogolite (Cradwick et aZ., 1972). Russell et al. (1969) also pointed out that the absorption bands near 1000,700, and 600 cm-l are sensitive to sample orientation. For example, when oriented, the “Si-0” stretching absorption maxima of imogolite were resolved at 990-1010 and 925-935 cm-I. They then interpreted the higher frequency band as resulting from the presence of the Si-0 bonds whose vibrations are perpendicular to the fiber axis of imogolite. The ease of complete OH-OD exchange with D,O at room temperature for allophane and imogolite (Wada, 1966; Russell et al., 1969) suggests that all the OH groups and adsorbed water are accessible to the ambient solution. Physically adsorbed water, either H,O or D,O, seems to be almost completely removed by heating at or above llO°C or by evacuation at room temperature. Iimura ( 1971) questioned this interpretation on the basis that small absorption band around 1630 cm-l remains on the spectra obtained by Russell et al. (1969) for samples which had been heated and/or evacuated. Iimura maintained that this suggests the presence of difficultly removable water in allophane. However, he overlooked that this absorption band appeared equally for allophane and imogolite which were completely deuterated. c. Diflerentiat Thermal Analysis. The presence of large endothermic peaks between 100-200°C and no endothermic effects between 500 and 7OOOC upon differential thermal analysis of soil clays indicates large amounts of hygroscopic moisture and no sharply defined dehydroxylation. These features are regarded as evidence of substantial amounts of amorphous silicates in the clay (Mitchell et al., 1964). In addition to these features, the absence of an exothermic peak at 900-1000°C in DTA was suggested by Fieldes (1955) to be evidence for discrete hydrous silica and alumina in amorphous fine clays, for which he proposed the name allophane B. It is necessary to keep in mind that, subsequently, participants in the United States-Japan seminar on amorphous clay materials recommended that the term allophane not be used for discrete silica and alumina. In a study on clays separated from young volcanic ash soils in Japan, Miyauchi and Aomine (1964) showed that the <0.2 pm fractions give a strong exothermic peak near 9OO0C, whereas the 0.2-2 pm fractions bear only slight or no evidence of an exothermic reaction. They correlated the absence of the exothermic reaction with the predominating presence of cristobalite, feldspar, and quartz, and pointed out that the absence of the exothermic peak does not signify the presence of discrete silica or alumina. Fieldes and Furkert (1966) then showed that the content of discrete silica exceeds that of crystalline silica, such as quartz and cristobalite, in the coarse clay fractions of some New Zealand volcanic

AMORPHOUS CLAY CONSTITUENTS OF SOILS

227

ash soils. They maintained their interpretation on the presence of allophane B as coarse-clay size, discrete amorphous hydrous silica, some of which were assumed to derive from disordered volcanic glass and feldspar. Uchiyama et al. (1968a) also presented data which indicate that the absence of the 9OOOC exothermic peak relates to the chemical composition of amorphous constituents in the clay fractions of volcanic ash soils. No exothermic reaction was found when the Si02/A1,0, molar ratios of the 0.5 N NaOH soluble fractions of these clays are in the range from 4 to 25. This observation led Shoji and Masui (1969a,b, 1971) to identify opaline silica by electron microscopy in these siliceous clays (see Section 111, A). Fieldes (1955) indicated that the temperature and intensity of the high temperature exothermic peak increased with increasing order of structure from allophane B through allophane AB and A to kaolinite. There are indications that the exothermic reaction varies with the exchangeable cation (Miyauchi and Aomine, 1966a) and with the pretreatment of the material (Miyazawa, 1966; Campbell et al., 1968). The exothermic reaction was observed for synthetic aluminosilica gels with the Si02/A1203ratios of about 0.33 to 2 (Ossaka, 1962; van Reeuwijk, 1967; Wada and Kubo, 1972), but not for those with the higher SiO,/Al,O, ratio. Ossaka (1962) noted that the intensity of the reaction correlated with the amount of mullite formed upon heating at 1000°C, whereas Wada and Kubo (1973) found that this simple relationship does not hold for the gels prepared by different methods. The amount of mullite formed, however, correlated with the Si02/A1203ratio of the gels, and is at its maximum with the gels of the SiO,/Al,O, ratio of 0.47 (Ossaka, 1962) and of 1.0 (Wada and Kubo, 1972). Thus, the presence of the high temperature exothermic peak may be taken as an indication of Si-0-A1 bond formation in amorphous aluminosilicates. Pretreatment with slightly acid salt solution (e.g., at pH 5) to keep the content of exchangeable “bases” as low as possible would assist differentiation of allophane from amorphous oxides and hydroxides of silicon, aluminum, and iron, and from their mixture. Imogolite shows a DTA curve similar to that of allophane, but gives a small yet eminent endothermic peak due to dehydroxylation at 420°C in macro to semimicro DTA (Yoshinaga and Aomine, 1962b) and at 39OOC in micro DTA (Wada et al., 1972). d . Structure Models for Allophaize and Zmogolite. Predictions of behavior of clay and its application are severely limited by absence of knowledge of structure. Thus, attempts have been made to propound structure models for allophane and imogolite. Many of them have had value only as working hypotheses; they are briefly reviewed to follow development in research and to test their validity in current and future studies.

228

KOJI WADA AND M. E. HARWARD

Wada (1967) reviewed data available for allophane and imogolite and suggested that they have structures composed of similar silica tetrahedral and alumina octahedral chains in which a corner of the tetrahedra and octahedra are shared. Udagawa et al. (1969) compared the thermal transformation of allophane with that of kaolinite and determined the coordination number of aluminum by X-ray fluorescence analysis. Formation of spinel as a thermal transition phase to mullite both for allophane and kaolinite was considered to indicate a sheet structure of allophane in which aluminum in 4-fold coordination was included. Brindley and Fancher (1970) then found various kaolinite defect structure models to yield compositions of allophanes which generally contain less silica and more water with respect to alumina than kaolinite. Iimura (1969) calculated an average structural formula of allophanes by equating the Ba( OH) ,titer with the aluminum, and 0’and 0” represent oxygen in Si-0-Sj and Si-0-A1 bonds, respectively.

(OH)1~.as011.alSil.330111.~3A1~(OH)114.77 where (OH)’ and (OH)“ represent OH groups bonded to silicon and aluminum, and OI and 011represent oxygen in Si-0-9 and Si-0-A1 bonds, respectively. By assuming a parallel between synthetic silica-alumina gels and allophane, Cloos et al. (1969) presented a structure model with composition Si) < 1. In this model, a central core varying between 0 < Al/(Al made from a tetrahedral network in which silicon is partially substituted by aluminum carries a net negative charge. This negative charge is balanced by more or less polymerized hydroxy-aluminum cations forming a coating around the core. A similar model, “permutite” core plus hydroxy-aluminum cations, was previously presented by de Villiers and Jackson (1967) by analogy between chloritized 2 : l layer silicates and allophanes with respect to CEC variation with pH. More recently, de Villiers (1971) again proposed this model based on the analytical data for synthetic aluminosilica gels. He expressed the opinion that the aluminosilicate core rather than the core plus hydroxy-aluminum coating should be equated with allophane and the core with a Si0,/A1,O3 molar ratio of 6.0 may serve as an analog for allophane. The choice of a SiO,/Al,O, molar ratio of 6 was based on maximum CEC as a function of composition. It was proposed that aluminum substituted for Si with associated increase in exchange sites up to a certain amount; above this point the aluminum was proposed to exist only as hydroxy polymers with a decrease in CEC due to a dilution effect. However, there are alternative explanations for the phenomen. If some of the aluminum were incorporated in octahedral sites analogous to the kaolinite structure, CEC per gram would also decrease since the for-

+

AMORPHOUS CLAY CONSTITUENTS OF SOILS

229

mula weight would increase. In addition, Dingus (1973) observed that CEC of synthetic gels was highest at molar Si0,/A1,03 ratio of 4. In derivations of the model, de Villiers did not consider available analytical data on allophane, imogolite, and opaline silica present in volcanic ash soils. The state of our present knowledge on chemical structure of allophane is not much different from that summarized at the 1969 United States-Japan seminar on amorphous clay materials (see introduction of Section 111, C) . The chemical composition of allophane as expressed in the SiO,/Al,O, ratio has, however, been narrowed down to some extent. There are indications that allophane with the SiO,/A1,0, ratio 2.0 may occupy a central position in the series which coincides with that defined by Brown (1955). The upper limit of the SiOr/Al,O, ratio has not been well defined, but is expected to be much lower than the value of 6.0 proposed by de Villiers (1971). Whether or not a continuous transition of allophane to imogolite and to opaline silica occurs is left to further study. That part of aluminum atoms in allophane is in 4-fold coordination, but not in imogolite, is important in considering the differences in chemical structure and in negative charge characteristics (see Section VI, A, 1) . An understanding of the structure of imogolite is important not only in itself, but also because of its relationship to allophane. Earlier electronoptical studies (Russell et al., 1969; Wada and Yoshinaga, 1969) indicated that imogolite has a chain structure with a repeat distance of 8.4 A along the chain axis. Two proposed structures assign this 8.4 A repeat distance to a gibbsite-like chain or ribbon, but they differ in the state of the silicate anion attached to it. Also the two proposals do not account for the tubular feature of the structure which is evident from later electron microscopic studies (Wada ef al., 1970). The most recent version of the imogolite structure was advanced by taking the tubular morphology and the presence of isolated orthosilicate anions into consideration (Cradwick et al., 1972). The proposed cylindrical structure units have circumferences of 10, 11, or 12 gibbsite unit cells, with b of the unit cell being along the circumference and a parallel to the cylinder axis. Each orthosilicate anion displaces hydrogen from the three hydroxyl groups surrounding a vacant octahedral site in a gibbsite sheet. The fourth Si-0 bond points away from the sheet, forming a Si-OH group. This structure requires a considerable shortening of the 0-0 distance around the vacant octahedral site, and this contraction may account for the shortening of the repeat distance from 8.6 A in gibbsite to 8.4 A in imogolite and also for the curling of the gibbsite sheet to form the cylinder. The outside diameters of the cylinders range from 18.3 to 20.2 A, consistent with those observed in the electron microscope (17 to 21 A) or

KOJI WADA AND M. E. HARWARD

230

with the estimated interaxial separations of 18 to 23 A for an aligned array of cylinders. The resulting structure has a composition (OH),A1,03 SiOH or Si0,-A1,03*2H,0. This compares with 1.1 Si0,.AI,03.2.3-2.8 ~ 1 . 2H,O (coordinated) given by Wada and Yoshinaga (1969). H,O( The smaller content of both OH and coordinated H,O groups in the proposed structure should be taken into consideration in further refinement. Fourier transforms of the three proposed structures projected perpendicular to the cylinder axis showed fairly good agreement with the most detailed electron diffraction pattern of imogolite. Infrared spectra (Wada, 1966; Russell et al., 1969), retention of quaternary ammonium chlorides (Wada and Henmi, 1972), and negative charge characteristics and aluminum-coordination data (T, Henmi and K. Wada, unpublished) for imogolite can also be accounted for by the proposed structure.

+)

IV.

Identification and Quantitative Estimation

Most of the information on identification of amorphous clay constituents by application of X-ray analysis, electron microscopy, infrared spectroscopy, and differential thermal analysis was reviewed in the previous section since these methods were used to elucidate proposed structures. Consequently, it will not be repeated here. Jackson (1965) gave an account of the analysis of free oxides, hydroxides, and amorphous aluminosilicates in soils and pointed out that sensitive chemical methods are important for independent determination of these constituents. Difficulty has been experienced with many physical techniques in estimating or even recognizing amorphous clay constituents. Mitchell et al. (1964) reviewed the use of selective dissolution treatments which had been utilized at that time and emphasized the importance of this technique as one of the most promising approaches to soil clay studies. This section will be primarily devoted to recent progress with selective dissolution methods. Before proceeding, other chemical methods will be briefly reviewed. It is well known that sorption phenomena for anions are particularly significant for clays rich in allophane, sesquioxides, and hydroxides. Interest has been revived in the large OH release which takes place when NaF is added to allophanic soils (Kawaguchi et al., 1954). Fieldes and Perrott (1966) proposed to use this reaction as a rapid test for allophane. Bracewell et al. (1970) described a procedure for monitoring the fluoride uptake. The release of OH by treating clays with 1 N NaF at pH 7 was in the range from 600 to 1000 me per 100 g for allophanes and 10 to 120 me per 100 g for crystalline layer silicates. It is important to keep in mind that the reagent will react with any available source of aluminum,

AMORPHOUS CLAY CONSTITUENTS OF SOILS

23 1

and it is not specific for allophane. Brydon and Day (1970) have shown that soil materials having reactive A1(OH) groups such as the Bf horizon of Podzols, give a positive test. A combined use of dissolution treatment with an analysis of kinetics (Langston and Jenne, 1964; Follett et al., 1965a,b; Segalen, 1968) or with difference infrared spectroscopy (Wada and Greenland, 1970; Wada and Tokashiki, 1972) have been used for characterization and estimation of amorphous aluminosilicates. In the former procedure, the steady state portion of the dissolution-time curve was taken to indicate a limited attack on crystalline materials. The difference in the rate constants for various reagents was interpreted in terms of soil clays, being a continuum from completely disordered, through poorly ordered, to well crystallized material (Follett et al., 1965a). In the latter procedure, the difference spectra representing the infrared absorption of the materials removed by the dissolution treatments were obtained. These spectra and the spectra of the residue were used for identification and characterization of the major constituents and for allocation of the weight loss data. Tweneboah et al. (1967) proposed a 12-hour extraction using 0.5 M CaCl, at pH 1.5 for “active” aluminum oxides. This proposal is based on the observed kinetics for the removal of iron, aluminum, and silicon from soils and clays, and on the resulting reduction of the positive charges developed at low pH. As a measure of the amorphous portion of the iron oxides, a modified Tamm’s oxalate method, involving a 2-hour extraction with 0.2 M ammonium oxalate-oxalic acid at pH 3.5, has been used widely (Schwertmann, 1964; Blume and Schwertmann, 1969). The validity of the method is based on (1) a significant correlation between goethite content (X-ray, DTA) and the residue, (2) the absence of a correlation between dithionite-soluble and oxalate-soluble (Fe) in numerous soil samples, ( 3 ) nature of the solution-time curve, and (4) the observation that the hematite-goethite ratio in mixtures of crystalline and amorphous oxides does not change consistently upon oxalate treatment. However, the oxalate solution also extracts aluminum from amorphous aluminum oxides and hydroxides (McKeague, 1967), and iron and aluminum from organic matter complexes (McKeague, 1967; McKeague and Day, 1966; Blume and Schwertmann, 1969) as well as from allophane (Miyazawa, 1966; Hattori and Morita, 1966; Dudas and Harward, 1971; Higashi and Ikeda, 1974). It was shown that 0.1 M sodium pyrophosphate more specifically extracts iron and aluminum in sesquioxide-organic complexes which have accumulated in soils such as Spodosols. Both the dithionite-citrate (Mehra and Jackson, 1960) and the pyrophosphate-dithionite (Franzmeier et al., 1965) procedures remove some iron from crystal-

232

KOJI WADA AND M. E. HARWARD

line oxides as well as from amorphous and crystalline clays (McKeague, 1967; BIume and Schwertmann, 1969; Dudas and Harward, 1971). The association of amorphous silicates with “free” iron oxides was inferred from dissolution of silicon and aluminum in addition to iron during treatment for removal of iron oxides from soil clays (Mitchell and MacKenzie, 1954; Mehra and Jackson, 1960; Follett et al., 1965b; Yoshinaga, 1966; Weaver et al., 1968). For volcanic ash soil clays in which allophane predominates, the difference spectra have indicated that allophane-like aluminosilicates, with Si02/A1,03 ratios ranging from 0.4 to 0.8, are dissolved by the dithionite-citrate treatment (Wada and Greenland, 1970; Wada and Tokashiki, 1972; Tokashiki and Wada, 1972a). The spectra are characterized by the “Si-0” stretching absorption maximum in a range from 940 to 980 cm-’, and by the fairly strong absorption in the region from 600 to 800 cm-l. Relatively small amounts of unidentified constituents which show weak absorption as well as absorption maxima different from those described above, were dissolved from soil clays in which gibbsite and/or crystalline layer silicates were the principal constituents. The spectra of the constituents dissolved in 2% Na,C03 solution were also related to the major mineral composition of the clays. The allophanelike constituents dissolved give the “Si-0’ stretching absorption maximum in a very narrow range from 950 to 960 cm-I, and absorb relatively less in the region from 600 to 800 cm-’ compared with the dithionite-citrate soluble fraction. Dissolution of allophane, imogolite, and gibbsite by the 0.5 N NaOH treatment were shown on the difference spectra (Wada and Greenland, 1970; Wada and Tokashiki, 1972). No crystalline layer silicate except for halloysite was dissolved in any significant amount unless the clay was nearly free from allophane and/or allophanelike constituents. The allophanes dissolved in 0.5 N NaOH gave the “Si-0” stretching absorption maximum in the widest frequency range from 940 to 1050 cm-I. This variation correlates with the largest variation in the Si0,/A1,03 ratio of the soluble fraction (from 1.1 to 2.5). The smaller the SiO,/Al,O, ratio, the lower the absorption frequency (Tokashiki and Wada, 1971, 1972a). Tokashiki and Wada (1972a) compared the weight loss of the clays and the sum of extracted Si02, A1203,and Fe,03 by the successive dithionite-citrate, 2% Na,C03, and 0.5 N NaOH treatments, They found that the combined weight loss for the former two treatments and the weight loss for the last treatment can approximately be equated with the amounts of the constituents dissolved by these reagents. Then, the weight losses and the weight of the residue remaining after the 0.5 N NaOH treatment can be used to estimate the amounts of amorphous and crystalline constituents by following the indications from infrared and other instrumental analyses.

AMORPHOUS CLAY CONSTITUENTS OF SOILS

233

- O.S. -

A': Humus AI(Fe) : Humus

AI(Fe) : Humus

'

- Mt - Vt a

~

Ht

/

A'BA

-

Gb

Irn-

-

Mt : Ch

-V t : C h

2

- Gb

6 0 X 1000 years

4

10

12

14

FIG. 4. Formation and transformation of clay minerals important in volcanic ash soils developed in humid, temperate zones (Wada and Aomine, 1973; modified on the basis of more recent data). Abbreviations: A, allophane; A', allophane-like; AI(Fe), sesequioxides; Ch, chlorite; Gb, gibbsite; Ht, halloysite; Im, imogolite; Mt, montmorillonite; Mt :Ch, montmorillonite-chlorite intergrades; O.S., opaline silica; Vt, vermiculite; Vt :Ch, vermiculite-chlorite intergrades. Horizontal bars indicate approximate duration of the respective constituents.

V.

Formation a n d Transformation

Formation and transformation of amorphous clay materials in soils proceeds along with those of crystalline layer silicates and accumulation of soil organic matter, and are in response to environmental changes during soil development. This recognition is important in understanding the nature of the processes in which amorphous clay materials are involved. Figure 4 illustrates the processes for soils developed from volcanic ash in a humid, temperate climatic zone. Similar situations, though different in the trend and magnitude, would also occur in any other soils.

OPALINESILICA The occurrence of the laminar opaline silica was studied in detail by Shoji and Masui (1971). Opaline silica particles are abundant in recent volcanic ash soils, particularly in Hokkaido soils less than 500 years old. Although the contents are low, soils of this area which are more than 7000 years old still contain some opaline silica particles, whereas they are rare in Kanto soils more than 6000 years old and in Kyushu soils more than 4000 years old. These differences were attributed to differences in climatic conditions which result in slower weathering of opaline silica in the Hokkaido soils. The relative abundance of opaline silica in the A-horizon, compared with that in the B- or C-horizons, was interpreted as formation A.

234

KOJI WADA AND M. E. HARWARD

by supersaturation of silica and by concentration of bases due to surface evaporation of soil water. The possibility of biogenesis should also be considered. There is an inverse relationship between opaline silica and allophane in their occurrence in volcanic ash soils (Fig. 4). Shoji and Masui (1972) noted the presence of opaline silica in the fine clay fractions of A horizons, but it was absent in B and C horizons. The reverse was true for allophane. “Siliceous allophane,” though it does not seem to be well defined, was claimed to be present in both the horizons, The presence of both opaline silica and allophanelike constituents, but not of allophane, at relatively early stage of weathering of volcanic ash was also indicated by selective dissolution and infrared spectroscopy for some surface soils (Tokashiki and Wada, 1972b). In these soils, organic matter accumulated in amounts equivalent to as much as 15-17% carbon. These observations suggest that aluminum released by weathering of the parent material is retained by organic matter which is stabilized against biotic degradation and leaching. The resulting suppression of aluminum activities in soil solution would then favor the formation of opaline silica and prevent the formation of allophane. The occurrence of precious opal in Australia was reviewed by Darragh et al. (1966) and has implications to the genesis of opal in soils. They estimated that opal has been deposited under a cover of 5-40 m of rock during the Tertiary period with climatic conditions not greatly different from those at present. Field evidence suggested that opal was formed by slow evaporation of very limited and highly localized supplies of groundwater, a process similar in some respects to the formation of surface duricrust but occurring under a protective cover of rock sufficient to isolate the system from seasonal drying and flooding. Opal host cavities have characteristically been those situated in positions permitting steady-state concentration of groundwater by evaporation. Soluble salts have been removed by diffusion to zones above the cavities, where these salts crystallized as capillary flow ceased.

B. ALUMINUM AND IRON OXIDES,HYDROXIDES, AND ORGANIC COMPLEXES Reports of the occurrence of discrete, amorphous aluminum oxides and hydroxides in soils are still rare, though some of them have been referred to in Section 111, B. This may be due in part to the difficulties in establishing their presence, and in part to the reactivity of these compounds with silica in solution or with organic matter in soils. The interlayer space of expandable 2: 1 layer silicates, however, provides sites for precipitation

AMORPHOUS CLAY CONSTITUENTS OF SOILS

235

of aluminum hydrous oxides in isolation from interfering agents. Rich ( 1968) considered that the most favorable soil conditions for the hydroxyinterlayer formation is moderate pH (4.6-5.8), frequent wetting and drying cycles, and low organic matter content. In volcanic ash soils, Uchiyama et al. (1968b,c) found that the aluminum interlayering in the expandable 2: 1 layer silicates takes place in soils more than 500 years old, possibly paralleling the formation of allophane (Fig. 4). It has been recognized that there are large amounts of nonsilicate forms of aluminum together with iron in the illuvial horizons of Podzols. Recognition that the aluminum and iron is primarily in complexes with organic matter has been obtained by extracting the soils with either pyrophosphate solution (Franzmeier et af., 1965; McKeague, 1967; Bascomb, 1968) or acid ammonium oxalate solution (McKeague, 1967; Blume and Schwertmann, 1969). That the crystallization of amorphous ferric hydroxide is significantly retarded or even inhibited by organic compounds was confirmed either by boiling amorphous ferric hydroxide from a bog soil in a strong alkaline solution with and without pretreatment with H,O, (Schwertmann, 1966) or in vitro using citrate as a model organic compound (Schwertmann et al., 1968; Schwertmann, 1970). The presence of sesquioxides in combination with organic matter in volcanic ash soils has also been indicated by extracting the soils with dithionite-citrate solution (Oba and Okano, 1967; Oba and Hayashi, 1971). The content of extractable aluminum was higher in the surface compared with the subsurface soils and the SiO,/Al,O, ratio of the extracted material was generally lower than 0.5. Kato (1970a) also emphasized the importance of dithionite-citrate soluble sesquioxides, particularly alumina, in the accumulation of humus in these soils. By applying selective dissolution and difference infrared spectroscopy Tokashiki and Wada ( 1972b) found allophane to be essentially absent in a deeply buried volcanic ash soils which was about 10,000 years old and contained organic matter equivalent to 14.4% carbon.

C. ALLOPHANE, IMOGOLITE, AND RELATED SILICATES 1. Soils Developed on Volcanic Ash It has been established that the formation of allophane from volcanic ash constitutes the central feature in the development of soils called Andosols, Andepts, or Humic Allophane soils. These soils occur throughout a wide range of climatic conditions from the cold subhumid regions to the humid equatorial tropics except for those in desert and semiarid regions (Wright, 1964; Flach, 1964). Under subhumid to humid, temperate cli-

236

KOJI WADA AND M. E. HARWARD

mate, nearly all kinds of volcanic ash, either basaltic (e.g., Kanno, 1961; Siefferman and Millot, 1969: Hamblin and Greenland, 1972), andesitic (e.g., Fieldes, 1955; Kanno, 1961), dacitic (e.g., Chichester et al., 1969; Wada and Tokashiki, 1972) or rhyolitic ash (e.g., Fieldes, 1955) produce allophane and allophanelike constituents, though they may be different in nature, stability, and amount. Typical Andosols are not usually formed from consolidated tuffs (Wright, 1964), but there are reports that a considerable amount of allophane forms from these parent materials and affects soil properties (Saly and Mihalik, 1970; Sakr and Meyer, 1970). Allophane does not appear at a very early stage of weathering of volcanic ash (see Section 2V, A and Fig. 4). Also, not all soils with well developed Andosol characteristics contain allophane (Tokashiki and Wada, 1972b). An analysis by selective dissolution-difference infrared spectroscopy of a series of deposits in the same profile in the Kuju district has shown that allophane is absent or nearly absent, both in the present day and in the oldest buried, surface soils. All other buried surface soils and subsoils contain allophane. Allophanelike constituents were present in all the soils examined except for the oldest, buried surface soil. A balance between the release rate of aluminum from volcanic ash by weathering and the supply rate of organic matter seems to control the formation of allophane in these soils. Hisingerite is normally found as a product of hydrothermal alteration of pyroxene and olivine (Whelan and Goldich, 1961). Its occurrence in association with ironstone (a,y-Fe,O,.H,O) in a deep soil overlying limestone in Victoria was reported by Ingles and Willoughby (1967). Silica adsorption by the hydrated surface of the ironstone and replacement of aluminum in allophane adjacent to the ironstone by Fe was suggested as the probable genetic processes. Imogolite as well as allophane imparts typical Andosol characteristics to the soils. The number of reports on occurrence of imogolite has steadily been increasing by use of high-resolution electron microscopy. Imogolite has been found in weathered volcanic ash in Japan (Yoshinaga and Aomine, 1962b; Aomine and Miyauchi, 1965; Kawasaki and Aomine, 1966; Kanno et al., 1968; Wada and Tokashiki, 1972; Shoji and Masui, 1972; Aomine and Mizota, 1973), Chile (Besoain, 1968/1969; Aomine et al., 1972), Oregon, U S A . (Dingus et al., 1973), Papua (Greenland et al., 1969; Parfitt, 1972), and Africa (Sieffermann and Millot, 1969) and in pumice tuff soils in West Germany (Jaritz, 1967). Occurrence of relatively pure imogolite as gels in several pumice beds in Japan (Miyauchi and Aomine, 1966b; Yoshinaga, 1968; Wada and Matsubara, 1968; Yoshinaga and Yamaguchi, 1970b, Tazaki, 1971) has also been reported. The youngest soil containing imogolite has been found in volcanic ash with the 14C

AMORPHOUS CLAY CONSTITUENTS OF SOILS

237

age of 1610 years (Shoji and Masui, 1972) and the oldest one in the pumice bed with the 14C age of 30,200 years (Tazaki, 1972). Association of allophane and imogolite is usually found in volcanic ash soils. A gradation from allophane to imogolite was observed in a series of soils developed from the glassy volcanic ash of Kyushu “Imogo” (Aomine and Miyauchi, 1965). In the pumice beds, allophane with the higher Si02/Al,0, ratio was found within the pumice grains, whereas imogolite occurred exclusively as macroscopic gel films (Wada and Matsubara, 1968; Aomine and Mizota, 1973). Formation of imogolite in a particular ash or pumice bed in the northern Kanto district usually occurs with a relatively thin depositional overburden which serves as a silica source (Aomine and Mizota, 1973). The association of irnogolite and gibbsite, but not of imogolite and halloysite, has also been noted (Wada and Matsubara, 1968; Yoshinaga and Yamaguchi, 1970b; Tazaki, 1971; Aomine and Mizota, 1973). Development of structural order in imogolite has often been taken as indicating that it constitutes an intermediate phase from allophane to crystalline layer silicates, particularly to halloysite, but its SiQ2/Al,0s ratio close to 1.0 and unique tubular structure unit do not support this view. It is more likely that imogolite represents an intermediate phase in transformation from allophane to gibbsite in a desilication process (Fig. 4). As Fieldes (1955) and a number of subsequent investigators have observed, transformation of allophane to halloysite and sometimes to metahalloysite takes place in old and buried volcanic ash soils (Fig. 4). Depositional overburden generally favors this transformation (Mejia et al., 1968; Aomine and Mizota, 1973) through effects of either the weathering time since deposition and/or the silica supply for resilication. The effects of overburden may be indirect. There are observations that stagnant moisture regime favors formation of halloysite (Kanno, 1959; Aomine and Wada, 1962; Dudas, 1972). Authigenic halloysite occurs just above the paleosol contact in well drained sites of Mazama ash (Dudas and Harward, 1974). This localized formation was related to increased moisture regime above the less permeable paleosol. Halloysite has formed from ashes and pumices of various compositions, from basalt (Kurabayashi and Tsuchiya, 1960; Sieffermann and Millot, 1969) through andesite (Birrell et al., 1955; Aomine and Wada, 1962; Calhoun et al., 1972) to dacite (Dingus et al., 1973; Dudas and Harward, 1974). It usually appeared as unique spherules, but not tubes. These spherules have diameters of 0.1-0.5 pm and consist of curled short flakes (Birrell et al., 1955; Kurabayashi and Tsuchiya, 1960; Aomine and Wada, 1962; Sieffermann and Millot, 1969). Development of crystallinity is seen in the lattice images which appear in these flakes (Fig. 3 ) . Substantial dissolution of these spherules takes place

238

KOJI WADA AND M. E. HARWARD

in hot 0.5 N NaOH, but allophane is absent in them as was shown by difference infrared spectroscopy (Wada and Tokashiki, 1972). The age of the ash deposits containing halloysite has been found to range from 8650 (Aomine and Miyauchi, 1963) to 10,000 years old or more in the temperate zone, while it was found in a 4000-year-old ash soil in tropical St. Vincent (Hay, 1960). Transformation of allophane and imogolite to gibbsite takes place in the final stages of weathering of volcanic ash in an environment which favors dedication (Fig. 4 ) . In an old soil (<33,000 years old) derived from dacitic ash, no allophane was detected and gibbsite was found to amount to 30-45% of the clay fraction (Wada and Aomine, 1966; Wada and Greenland, 1970). The association of allophane and gibbsite as well as imogolite and gibbsite was found in many volcanic ash soils and pumice beds. There are, however, some differences in the occurrence of gibbsite; it occurs often as white concretions in the pumice, scoria, and volcanic sand and gravel beds, but not in the fine-textured volcanic ash soils. These associations do not mean that gibbsite always forms by transformation either from allophone or from imogolite. Field and microscopic observations made on pumice and scoria beds (Wada and Matsubara, 1968; Aomine and Mizota, 1973) indicate that gibbsite forms separately, possibly because of the heterogeneity of parent materials or changes in environment from time to time. The transformation of allophane to expandable 2: 1 layer silicates was suggested by Masui et al. (1966) as an important process in weathering of volcanic ash. This inference was made on the basis of their finding that the contents of the 2:l layer silicates increase and those of material soluble in dithionite-citrate and 0.5 N NaOH decrease with weathering, where the degree of weathering was measured by the content of clay in the soils. They overlooked, however, the effect of different kinds of volcanic ash on weathering. A comparison of the mineralogical analyses between the soils derived from andesitic and dacitic ashes in Kyushu (Tokashiki and Wada, 1972b; Wada and Aomine, 1973) indicated that the 2: 1 and 2 : 1: 1 layer silicate contents are much higher in soils derived from dacitic ashes than in those from andesitic ashes, where the amount of quartz in the clay fraction served in distinguishing the two ashes. The infrared spectra and petrographic data obtained by Masui et al. (1966) also support this interpretation and show that the contents of the 2: 1 layer silicates in the soil is high when the quartz content in the clay fraction is high and the anorthite content in feldspar is low. Higher contents of mica and hornblende in dacitic ash may suggest that these 2: 1 layer silicates are inherited, but their origin is beyond the scope of the present article. It is desirable to inject a word of caution at this point. Before interpreting a particular clay mineral as a weathering product of volcanic ash, it

AMORPHOUS CLAY CONSTITUENTS OF SOILS

239

is necessary to establish that it was not of detrital origin and not an initial constituent of the parent material. These are often assumed and not verified. In cases where quartz was not present as phenocrysts in the ash, it may be used as an indicator of detrital components, including clay minerals (Dudas and Harward, 1974). Other primary minerals may be used in a similar manner to check the validity of interpretations of clay mineral formation. In summary, it seems evident that the formation and transformation of allophane and allophanelike constituents in weathering of volcanic ash are primarily controlled by availability of silica and aluminum in the soil solution. At an early stage of soil formation, the organic matter added to the soil would retain aluminum released. Formation of allophane would thereby be inhibited, formation of opaline silica favored, and the organic matter itself stabilized against biotic degradation and leaching. The presence of allophane in buried surface soils and that of allophane and/or imogolite in subsoils would take place as a result of desilication process under limited supply of organic matter. Transformation of allophane to halloysite would then take place in relatively silica rich environments resulting from either limited or stagnant moisture regime or by coupling with the desilication process in the overburden deposit. On the other hand, transformation of allophane and imogolite to gibbsite would take place as a result of desilication in an open and strongly leaching environment. Although the forms of aluminum retained by the organic matter need further clarification, it is likely that they are subject to transformation in weathering just as allophane and allophanelike constituents are, as illustrated in Fig. 4. 2 . Soils Developed on Parent Materials Other Than Volcanic Ash

The presence of allophane in clay fractions of soils derived from parent material other than volcanic ash has been suggested by a number of investigators. DeMumbrum and Chesters (1964) reported that allophane was present in several Wisconsin soils including Podzols. The X-ray patterns of the acid-dispersed clay fractions apparently gave no indication of crystalline layer silicates, but in contradiction to their interpretation, the infrared spectra indicated predominating presence of crystalline layer silicates rather than allophane. Fieldes (1966) took dissolution of soil clay by hot 0.5 N NaOH as evidence of the presence of allophanic material in New ZeaLand soils derived from sedimentary rock as well as volcanic ash. Raman and Mortland (1969/1970) used extraction at pH 3.7 to separate part of the clay fraction from Onaway sandy loam, a Spodosol, and thereby concentrated amorphous material which was soluble in 0.5 N NaOH. The fractions separated from Ap, B2ir, B,t, and C horizons contained 50, 64,

240

KOJI WADA AND M. E. HARWARD

58, and 54% amorphous material, respectively. Their SiO,/Al,O, ratio varied from 1.25 to 1.71, and their OH water content varied from 15 to 22%. Coen and Arnold ( 1972) also found 0.5 N NaOH soluble amorphous silicates in some New York Spodosols, and suggested that chlorite in the albic horizon is weathered to amorphous aluminosilicates and translocated to the spodic horizon. On the other hand, De Kimpe (1970) concluded from his study of a Quebec Podzol that the content of amorphous material was less than 2% of the clay fraction on a quartz-free basis, and that this material consisted almost exclusively of Fe,O,. The presence of poorly ordered aluminosilicates in a Podzol with thin iron pan and noncalcareous humic gley in considerable amounts was indicated by successive extraction of soil clays with Na,C03 solution (Follett et al., 1965a,b). The materials differ from normal finely particulate allophane and are associated closely with the crystalline layer silicates as a film or coating. The small granules were soluble in dithionite and were suggested to be ferruginous complexes containing considerable quantities of silica and alumina. Disordered silica and aluminosilicates as crust on clay particles were also observed in a Red-Brown Earth by Greenland and Wilkinson ( 1969) by electron microscopy of carbon replicas and selective dissolution. The presence of allophane in some red, basaltic soils in Australia, referred to as Krasnozems or Red Loams, was suggested from an X-ray study of their clay fractions. Briner and Jackson (1969, 1970) reported that the clay fractions of Victorian soils contain 1 5 2 0 % allophane having SiO,/A1,0, ratios from 2 to 4, and which dissolve in hot 0.5 N KOH. Sargeant and Skene (1970) had thought that the higher CEC of clay fractions from the soils derived from the newer basalts compared with the older basalts might be due to higher contents of allophane, but their determination showed no consistent difference in the allophane content between the two groups of soils. Prior to these studies, Simonett and Bauleke (1963) also determined the amounts of allophane by boiling in NaOH. Dissolution amounted to as much as 30% of the clay, but the SiO,/Al,O, ratios of the dissolved material were found to be very close to that of kaolinite. They interpreted the results in terms of the dissolution of somewhat poorly crystalline kaolinite and halloysite of small particle size. A selective dissolution-infrared spectroscopic analysis of a Glencoe soil, a Krasnozem from Pleistocene basalt ( Wada and Greenland, 1970) supported this interpretation. Dissolution in 0.5 N NaOH after the dithionite-citrate and 2 % Na,C03 treatments amounted to 13.5% of the clay fraction, but patterns of poorly crystalline layer silicates predominated in the difference infrared spectrum. In Hawaii, the occurrence of allophane in weathering of basaltic and

AMORPHOUS CLAY CONSTITUENTS OF SOILS

24 1

andesitic rocks has been reported. Bates (1962) studied alteration of Honolua andesite and found that an amorphous transition state, probably ranging in composition from allophane to aluminum hydroxide gels, exists as part of the change from halloysite to gibbsite. Evidence was obtained by electron microscopy and diffraction work on pseudomorphs after halloysite tubes which were found in certain samples that had been studied in detail with the light microscope. Patterson (1964) found fairly pure allophane and alumina-silica gel in basalt saprolite at one locality on Maui. As illustrated in Fig. 1 (top), the main constituent of the gel was later identified to be imogolite (Wada et al., 1972). Imogolite occurs as veins and coatings on vesicles and cracks in the saprolite at a depth of 2-6 feet under tropical rainforest. The allophane is white, and most of it is covered by the clear or light-colored gel. Both the allophane and dried imogolite gel contain about 50% alumina, 22 to 26% silica and 25% hydroxyl water. A minor amount of gibbsite is present in the imogolite gel, but no more than a trace occurs in the allophane. Allophane with an approximate composition S O , . 2A1,0, 4.5H20 was found in granodiorite in California. Its occurrence as cores to plagioclase crystals suggest a secondary origin, probably representing the first stage of weathering of the rock (Snetsinger, 1967). Formation of optically isotropic material in plagioclase at early stages of granite and gneiss weathering was also observed by Grant (1963, 1964). The appearance of an amorphous alkali-soluble component, judged to be allophane, was also reported in weathered andesite saprolite from the Cascade Range of Northern California (Hendricks et al., 1967). The allophane comprises 30-50% of the clay fractions, the amounts decreasing with weathering. Halloysite, found to be present in all saprolites, is highest in concentration in the more strongly weathered members. The SiOr/Al,O, ratios of allophane and halloysite dissolved by 0.5 N NaOH treatment after heating at llO°C and then after heating at 55OoC, respectively, are, however, not much different, and both are in the range from 1.3 to 1.9. Yuan (1968) analyzed the composition of the 0.5 N NaOH and dithionite-citrate soluble fraction in Regosols in Florida. This fraction comprises 20-60% of the clay fraction of these very sandy soils and has the Si0,/A1,03 ratios less than 1.7. The presence of allophane in tropical Latosols has also been inferred from selective dissolution analysis. Townsend and Reed (1971) found considerable amounts of alkali-soluble silica and alumina in a Panamanian Latosol which they thought to be possible components of allophane. They attributed the measured high CEC values of 11 me per 100 g of silt and 25 me per 100 g of fine clay to this amorphous constituent rather than kaolin minerals present. Le Roux (1973) reported that an alkali-extractable fraction constitutes up to one-third of the total clay fraction in some

242

KOJI WADA AND M. E. HARWARD

Natal Oxisols, but the major constituent of this fraction is gibbsite. Parfitt (1972) found that both A and B horizons of a Latosol in Papua contains significant amounts of 0.5 N NaOH soluble fraction after treatment of the clay fractions with dithionite-citrate and 2 % Na,CO,. Difference infrared spectroscopy indicated that this fraction consists of disordered crystalline layer silicates. There are some difficulties with interpretation and evaluation of the data referred to in this section. First of all, it is not always clear whether the authors are using the term allophane as synonymous with amorphous or as a particular aluminosilicate. Second, and more importantly, although the boiling alkali treatment gives a useful approximation to the amounts of amorphous components, it is not specific to allophane; some crystalline clays as well as other amorphous components may be dissolved. It can be seen from the above that the presence of amorphous aluminosilicate, specifically of allophane, in the soils derived from parent material other than volcanic ash is largely inconclusive and awaits further study. A probable exception is the rapid and intense weathering under tropical, humid climate which results in formation of allophane even from massive rock of basalt. VI.

Relationship to Soil Properties

Interest in amorphous constituents is predominantly motivated by their properties and those they impart to soil systems. They are distinctly different from the crystalline clays. Presence of significant amounts of amorphous materials creates special problems in chemical and physical analyses and in management of the soil systems. Often it is the unusual properties of materials which suggest the presence of amorphous components. A.

CHEMICAL PROPERTIES

1 . Cation and Anion Exchange

A marked influence of amorphous clay material on the surface charge characteristics of soils has been emphasized in the literature. It was pointed out that either allophane (van Olphen, 1971) or oxides and hydrous oxides of iron and aluminum (van Raij and Peech, 1972; Mekaru and Uehara, 1972) can be treated as a system with a constant surface potential. This implies that the ion exchange capacity is not a constant for these minerals but depends on the environmental conditions. Soils containing allophane have values for cation exchange capacities which are strongly dependent on concentrations of leaching solution, the cation in solution

AMORPHOUS CLAY CONSTITUENTS OF SOILS

243

and the volume and nature of the washing media; adsorption is strongly influenced by pH (Birrell and Gradwell, 1956; Birrell, 1961a) (Table I ) . Conventional methods for determination of the exchange capacity therefore require some revision. Perhaps the least ambiguous method is to determine the cation and anion exchange capacity (CEC and AEC) by equilibrating the soil sample with a dilute salt solution containing the index ions at an appropriate pH and then measuring the retention of these ions without removing the excess salt by washing (Schofield, 1949; Wada and Ataka, 1958; Wada and Harada, 1969; Chichester et al., 1970; van Raij and Peech, 1972). Most of the values described here have been obtained by such procedures unless otherwise specified. Van Raij and Peech (1972) determined the charge characteristics of two highly weathered Oxisols and one Alfisol from Brazil containing gibbsite, iron oxides, and kaolinite as major clay constituents. These soils develop both pH-dependent negative and positive charges in an amount less than 10 and 5 me per 100 g soil, respectively, for the pH range 4 to 7 and at electrolyte concentration of 0.01 N ; they exhibit a point of zero charge at pH 3-6. Barber and Rowel1 (1972) obtained similar data for an iron-rich kaolinitic soil. They explained the variations in charge characteristics in terms of the presence of a small permanent negative charge, pH dependent positive and negative charge influenced by the indifferent electrolyte concentration, and the overlap of diffuse double layers causing a neutralization of charge. In addition to this, Atkinson et al. (1967) presented evidence for ion-pair formation on iron oxides, which would inhibit the development of diffuse double layers. Shifts of the isoelectric point in more than 2 pH units toward an alkaline side were found for sesquioxide minerals with the increase both in the degree of their hydration and disordering in their structural organization (Parks, 1965). The pH-dependent CEC also occurs for chloritized 2 : 1 layer silicates, where permanent negative charge is blocked by positively charged hydroxyaluminum groups. Accessibility to exchange sites is restored by deprotonation of the latter on addition of base (de Villiers and Jackson, 1967; Chichester et al., 1970). Sawhney et al. (1970) thus assigned the irreversible portion of the pH-dependent CEC to the interlayers and coatings, and the reversible portion to organic matter, and found the pH-dependent CEC of several acid soils to be primarily due to weakly dissociated organic matter groups. The negative charge development for imogolite, allophane, and synthetic sesquioxides was found to be reversible (Sawhney and Norrish, 1971; Harada and Wada, 1973). Exchange capacity data for allophane and imogolite of established purity has been very scarce. The CEC values of 135 for allophane (Si0,/A1,03 ratio = 2.0) and 30 me per 100 g for imogolite were obtained by equili-

244

KOJI WADA AND M. E. HARWARD

TABLE I Effects of Changing the Conditions for the Measurement and Predrying of Samples on Cation Exchange Capacity (CEC) and Anion Exchange Capacity (AEC) Values of Imogolite-Allophane, Allophane, Halloysite, and Montmorillonitea Cation and anion exchange material

Changes in conditions for measurement and predrying of samples Salt concentration 0.2 N 0.005 N pH: 6.0 Cation species: N H 4 Temperature: 55'C Predrying of sample Evacuated over P205 Heated a t 150°C

Measured item

CEC CEC AEC CEC AEC CEC CEC CEC AEC CEC AEC

ImogoliteMontallophane Allophane (Si02/A1203; (Si02/A1203; Halloy- morillonite 1) 2) site

++ -

++ ++ + * + -

f

* ND -

++ * ++

** +++ *

i

If:

f

k

ND

ND

ND &

ND

I

If:

Referencesb

(1) (1) (1) (8, 3) (2,3) (1)

+

i f

i k

ND

15)

ND ND ND

(5) (5)

-

f

(4)

(5)

+,

a - and f respectively stand for about a 20% increase and decrease, and less than a 20% change in CEC and AEC, in comparison with their reference values, when conditions for the measurement and predrying of samples were changed as specified in the first column from those for the reference measurement. The reference CEC and AEC values were measured for air-dried samples by determining the cation and anion retention from 0.05 N Ca(CH&00)2 or CaClz of p H 7.0 a t 10-25°C. N D means not determined. References (1) Wada and Harada (1969); (2) Wada and Ataka (1958); (3) Harada and Wada (1973); (4) Wada and Harada (1971); (5) Harada and Wacla, (1972).

brating with 0.05 N NaCH,COO at pH 7.0 (T. Henmi and K. Wada, unpublished). The CEC increases with the increasing SiO,/A1,0:+ ratio for soil clays containing allophane and imogolite ( Wada, 1967). The point of zero net charge was found at about pH 7 for the soils containing imogolite and allophane, while the CEC was much higher than the AEC at the same pH for weathered pumice containing allophane with a Si0,/A120, ratio 2.0 (Wada and Harada, 1969). The large affinity and capacity of allophanic clays for anion retention were illustrated in studies by Singh and Kanehiro (1969) and Kinjo and Pratt (1971a,b) on nitrate adsorption by subsoils of Andepts in comparison with other soils. Table I illustrates the effects of changing the conditions on the CEC and AEC values measured for materials of different clay mineral compositions. The results demonstrate the greater effect of conditions on assessing

AMORPHOUS CLAY CONSTITUENTS OF SOILS

245

electric charges of soils containing allophane and imogolite compared with crystalline layer silicates. The data have implications to methods employed. The observed effect of pH and electrolyte concentration on both the CEC and AEC of imogolite and allophane (Table I) was accounted for by their effects on the proton dissociation from and the uptake by the functional OH groups bonded to silicon and aluminum atoms, respectively (Iimura, 1966; Wada, 1967). Wada and Harada (1969) ascribed a gradual decrease in CEC with the decreasing concentration below 0.05 N to hydrolysis of exchangeable cation, while the relatively rapid increase in CEC at concentrations above 0.1 N were attributed to a non-coulombic adsorption of cation-anion pairs. By analogy to coprecipitated silica and alumina, Birre11 (1961a) suggested physical adsorption of salts by allophane. There is a clear-cut difference in the hydrolysis of adsorbed cations between the soils containing imogolite-allophane and those of weathered pumice containing allophane with a Si0,/A1,03 ratio of 2.0. This may suggest a difference in the mechanism of development of negative charge between imogolite and allophane, and may be correlated with the absence of aluminum in 4-fold coordination in imogolite and its presence in allophane (T. Henmi and K. Wada, unpublished). Another possible interpretation involves the presence or absence of overlap of diffuse double layer at lower electrolyte concentrations as proposed by Schofield (1949) and others. Different values of the CEC of clay materials depending on the cation species have often been reported. The effect is more manifested for imogolite-allophane and least for montmorillonite (Table I). Since the proton is involved in the development of its negative charge, the observed phenomenon may be interpreted broadly in terms of relative bonding energies (strength of adsorption) between the proton and the index cation being added. Therefore, this will vary with valence and effective radius of the cation. Approximately, the CEC decreased in the order: Ba > Ca > K, Mg > NH,, for imogolite-allophane, and the ratio of the lowest to the highest CEC was 0.5 to 0.6 (Wada and Harada, 1969). Different values of the AEC depending on the anion species were also found for imogolite and allophane (Wada and Tsuji, 1973). The AEC increased in the order: NO, < C1 < CH3CO0<< SO,. The much higher value for sulfate anion was interpreted in terms of specific adsorption (Section VI, A, 2). Quite remarkably, the CEC of imogolite-allophane and allophane was found to increase when the temperature of the solution was raised, whereas a slight decrease was found in the AEC (Table I). The increased CEC at higher temperature in a neutral 1 N acetate solution, was only partly reduced by lowering the temperature again (Wada and Harada, 1971). It was proposed that this large CEC increase includes reactions by which some “bound” functional groups are set free for ionization and/or by which

246

KOJI WADA AND M. E. HARWARD

an AI-OH-AI bonding is formed between two terminal AI-H,O groups (Harada and Wada, 1973). The effect of previous drying or heating of the samples containing imogolite and/or allophane shows another interesting feature (Table I). Egawa et al. (1959) found that the CEC decreased remarkably for an allophane on air-drying or heating at 105OC when the CEC was measured using concentrated, e.g., 2.5 N , NH,CH,COO and 80% C,H,OH for washing. A significant increase was, however, found after the same treatment when the CEC was determined with 0.3 N NH,CH,COO and 80% C,H,OH. The data shown in Table I indicate that the negative charge could develop on dehydration, possibly including changes in the coordination of surface aluminum atoms. The changes of surface functional groups resulting in increase of net negative charge upon drying may also be inferred from the changes in electrophoretic mobility of allophanic clays or more simply from the changes in pH of allophanic soils suspended in salt solutions. Watanabe (1961, 1962) observed that the isoelectric point of allophane with a SiO2/A1,O, ratio of 1.11 was lowered from pH 6.8 to 4.1 by air drying. Sadzawka et al. (1 972) observed that the pH of Chilean volcanic ash soils suspended in 0.01 A4 Na,SO, is lowered by about 0.5 pH unit or more when the soils were dried at 110°C. Imogolite and allophane have weak acid properties in soils as expected from the pH-dependence of the negative charge developed on them. Yoshida (1970, 1971 ) confirmed this by treating allophane, imogolite, and crystalline layer silicate clays with 1 N AlCI, and determining exchangeable aluminum and hydrogen on these clays. All the exchange sites on allophane and imogolite were occupied by hydrogen but not aluminum, whereas more than 60% of exchange sites in crystalline layer silicates were occupied by aluminum. The cation selectivity of soils in cation-exchange reactions is markedly influenced by the nature of the cation-exchange material. Yoshida (1961) determined the ratio of exchangeable NH, to Ca after treating the soils with 1 : 1 mixture of 1 N ammonium and calcium acetates. The soils in which humus and/or allophane predominated gave ratios in the range of 0.2 to 0.4, whereas those predominated by crystalline layer silicates gave ratios in the range of 1.4 to 4. 2. Sorption of Cation and Anion

A strong sorption of certain anions such as phosphate in soils is usually associated with amorphous ahminosilicates and hydrous iron and aluminum oxides. The reaction is primarily attributed to aluminum and iron atoms present on the clay surface, and the high reactivity of soil amorphous constituents is correlated with their large aluminum and iron specific surface area.

AMORPHOUS CLAY CONSTITUENTS OF SOILS

247

Recent studies of anion adsorption by hydrous iron and aluminum oxides have shown that “specific” adsorption in addition to “nonspecific” adsorption of anions normally occurs in soils (Hingston et al., 1967, 1968). The “nonspecific” adsorption refers to adsorption of anions by simple coulombic interaction with positive charges on Al-OH,+ or Fe-OH,+ groups, while the specific adsorption refers to incorporation of anion in the coordination shell of an iron or aluminum atom as a ligand. The anion thus bound cannot be displaced from the soil simply by leaching with a solution containing a nonspecifically adsorbed anion, such as chloride. The process is strongly pH dependent, and a maximum or inflection occurs in the adsorption maximum-pH curve at pH values corresponding to the pK values of the acid species formed by the anion. Silicate, arsenate, fluoride, ortho-, pyro-, and tripolyphosphates, selenite, and molybdate are specifically adsorbed. Specific adsorption of molybdate by some crystalline and amorphous soil clays was studied by Theng (1971 ). The high value of phosphate absorption obtained by a conventional method (more than 1500 mg as P,O, by 100 g of air-dry soil from 200 ml of 2.5% ammonium phosphate solution at pH 7 ) has often been used in Japan for roughly distinguishing soils derived from volcanic ash and other parent materials. The greater role in this phosphate sorption has recently been assigned to the dithionite-citrate soluble sesquioxidic components rather than allophane (Kato, 1970a,b; Miyauchi and Nakano, 19711. Difficulty in obtaining and defining an adsorption maximum at a pH was also reported for adsorption of phosphate by volcanic ash soils (Tsukada et al., 1967; Miyauchi and Nakano, 1971). This is probably due to overlapping of the reaction with a phosphate-induced decomposition of clay minerals and with precipitation of complex or discrete new aluminum and iron phosphates. The rapidness of the latter reaction for allophane was demonstrated by its transformation to taranakite-like minerals in weak acid, ammonium, or potassium phosphate solutions (Wada, 1959; Birrell, 1961b). 3 . Interaction with Organic Conipounds

There is ample evidence which indicates that the level of organic matter in soils is affected by the presence of amorphous inorganic material (see Section V) . Recent advances in the study of possible mechanisms of interaction between humic and fulvic acids and soil clays have been reviewed by Greenland ( 1971) . The high ratio of the pyrophosphate or oxalate soluble aluminum and iron to dithionite-citrate soluble iron and aluminum was obtained for Podzol B horizons (Bascomb, 1968; McKeague, 1967; Blume and Schwertmann, 1969). The high solubility of aluminum and iron in the pyrophos-

248

KOJI WADA AND M. E. HARWARD

phate solution was interpreted as indicating that aluminum and iron or hydroxy-aluminum and iron ions form organic complexes (McKeague et al., 1971). Schnitzer (1969) explained the dissolution and precipitation of organic matter in the Podzol A and B horizons by change in the solubility of Fe(II1)- and aluminum-fulvic acid complexes. While 1 :1 molar Fe(II1)-fulvic acid complexes were completely water soluble, 6 : 1 molar Fe( 111)- and aluminum-fulvic acid complexes were water insolubIe. By differential thermogravimetric analysis, the ironpan in the Humic Podzol from Newfoundland who also identified as being essentially a 6 : 1 molar Fe (111)-fulvic acid complex. In these complexes, the formation of electrovalent bonds between negatively charged carboxyl groups and positively charged aluminum and iron ions and or hydroxy-aluminum and iron ions was inferred from infrared spectroscopy. Accumulation of humus constitutes one of the striking features of Andosols. It is not uncommon for the carbon content of the surface soil to be 20% within 5000 years. This occurs under a warm, humid climate and grass vegetation, and at well-drained sites. The role of allophane and related materials in accumulation of organic matter in Andosols has, therefore, been studied by a number of investigators. Recent investigations may be classified under the following four groups. The first is concerned with possible effects of allophane on the action of enzymes such as protease and amylases (Aomine and Kobayashi, 1964, 1966; Kobayashi and Aomine, 1967). These authors pointed out the importance of protective action of allophane against biotic degradation of organic materials which become incorporated in the soils. The second is concerned with a catalytic effect of allophane in an oxidative polycondensation of phenolic units, which results in formation of stable skeletons in soil humic materials. The greater catalytic effect of allophane and sesquioxides compared with crystalline layer silicates was demonstrated for a chestnut tannin-containing substance (Kyuma and Kawaguchi 1964) and pyrogallol (Kumada and Kato, 1970). The third possibility is that allophane acts as a source of aluminum and/or iron which form insoluble humates. The fourth is concerned with the adsorption of humic materials. Kobo and Fujisawa (1964) studied adsorption of humic acids extracted from soils by various clays. Adsorption studies by Inoue and Wada (1968, 1971a,b) with an extract of humified clover showed that allophane and imogolite have a much greater sorption capacity than crystalline layer silicates. The importance of this adsorption in the overall accumulation of humus was inferred from a soil incubation experiment (Wada and Inoue, 1967). In all these studies, the preformation of allophane in the soil has implicitly been assumed. As described in Section V, C, the importance of sesquioxidic constituents, rather than allophane, in accumulation of organic mat-

AMORPHOUS CLAY CONSTITUENTS OF SOILS

249

ter in some volcanic ash soils has been increasingly recognized. In fact, the ratio of pyrophosphate soluble aluminum to dithionite-citrate soluble aluminum as high as 1.0 was found for such soils in which little or no allophane was detected by selective dissolution and infrared spectroscopy (K. Wada and T. Highashi, unpublished).

B.

PHYSICAL AND

ENGINEERING PROPERTIES

The importance of “free iron oxide” in affecting physical properties such as aggregate formation has often been mentioned. Deshpande et al. (1968) subjected “red” soils to different dissolution treatments and observed changes in the physical properties. They suggested that the acid-soluble “active” aluminum oxide rather than iron oxide is important in aggregate formation in red soils. It has been established that the physical properties of soils having a high content of amorphous clay material, specifically allophane, are unique. Birrell (1 964) and Forsythe et al. (1969) summarized features of physical properties of Ando soils, and Sherman et al. (1964) summarized those of Hydrol-Humic Latosols, which were thought to contain amorphous aluminum, iron, and titanium hydrous oxides as well as allophane. Soils containing a high content of allophane and imogolite have low bulk densities. Values as low as 0.25 and 0.3 were found even for such volcanic ash soils, which contain little organic matter and develop under a temperate, humid climate (Wada and Aomine, 1973). In Hawaii, Sherman et al. (1964) reported that the ferruginous aluminous soils have bulk densities as low as 0.09 with an average range between 0.3 and 0.5, while the ferruginous soils have a bulk density of 0.8 to 1.0. Soils developed on the Mazama ash in Oregon commonly have bulk densities in the range of 0.7 to 0.8. A very high natural moisture content is a common feature of soils described above, and is associated with a very low bulk density value. The moisture contents on an oven-dried basis for volcanic ash soils in the Kanto district range from 80 to 180% and are mostly from 100 to 140% (Suyama and Oya, 1965). The subsoils compared with the surface soils have high amounts of both available or free water ( p F 2.5-4.2 or 1/3 to 15 bar) and nonavailable or nonfree water. Water retention, liquid limit, and plasticity index show a marked decrease when the samples have been previously air-dried (Birrell, 1952; Suyama and Oya, 1965; Tada, 1969). There seems to be a critical moisture content below which this effect of predrying of the sample appears both for the liquid limit and plasticity limit values.

250

KOJI WADA AND M. E. HARWARD

The change is, however, not parallel, and is much more pronounced for the liquid limit. Wells and Furkert (1972) inferred that an irreversible change takes place in water status of natural allophane by a simple mechanical disturbance. The tendency toward irreversible drying of soils containing allophane is manifested in formation of stable aggregates. Kubota (1972) found that there is a critical pF value, which may vary in a range from pF 3 to 4, depending on the natural moisture status of the soils. Sherman et al. (1964) reported that the Akaka soils undergo a tremendous loss of volume on drying and will not rehydrate. The original samples, which showed little or no crystallinity, underwent a change of state to form crystalline gibbsite, poorly crystallized iron oxide, and allophane after dehydration. They attributed the change of soil properties to crystallization of hydrous oxides from the corresponding gel of cryptocrystalline materials. The difficulty in determination of particle size distribution for volcanic ash soils has often been mentioned. The high content of organic matter combined with amorphous inorganic constituents in volcanic ash soils poses a problem. Undesirable effects of H,O, treatment resulting in a partial destruction of the clay constituents were discussed by Mitchell et al. ( 1964). Lavkulich and Wiens (1970) compared effects of H,O, and NaOCl treatments for removal of organic matter on selected mineral constituents. It was found that NaOCl extracted more organic matter with less destruction of the oxides than procedures employing H,O,. Harris (1973) reported that the clay yield from Parkdale soils containing considerable amounts of allophane was 14 % less in average after NaOCl treatment than after H,O, treatment. Treatment of these soils at pH 10 (NaOH) or at pH 4 (HCl) after peroxidation and washing with water and with the aid of ultrasonic or sonic oscillations seems to have been partially accepted as a means for dispersion of the clay fraction. The alkaline medium is used for soils in which crystalline layer silicates predominate in the clay fraction, while the acid medium is used for those containing allophane and imogolite in substantial amounts. Matsuo (1964) claimed that sodium metaphosphate can be used as efficiently for dispersion of volcanic ash soils as other soils in Japan, but neither his nor other subsequently published data on highly allophanic soils support his conclusion (Kobo and Oba, 1965; Sherman et al., 1964; Miyazawa, 1966). Adsorption of metaphosphate in an appreciable amount by amorphous clay constituents (Yoshinaga and Yamaguchi, 1970a) also is a disadvantage in the use of metaphosphate at recommended concentrations such as 0.032 to 0.02 M . Kobo and Oba (1965) and Kanno and Arimura (1967) showed that when sonic oscillation is used, the removal of iron oxide as a pretreatment is not necessary for dispersion of volcanic ash soils. The use of sonic or ultrasonic oscillation for mechanical analysis has, however, some disadvantages as pointed out by Watson (1971).

AMORPHOUS CLAY CONSTITUENTS OF SOILS

25 1

Birrell (1966) proposed to indirectly determine clay contents in soils containing allophane by measuring adsorption of nitrogen, acetic acid, and water vapor. Aomine and Egashira ( 1968 ) tested effectiveness of various electrolytes as flocculants for allophanic clays in comparison with montmorillonite. The flocculation of allophane is primarily determined by the valence of anions and that of montmorillonite by the valence of cations. This probably reflects their predominating positive and negative charges. From this observation, they proposed to use the ratio of flocculation value of CaCI, to that of Na,SO, for differentiation of predominating surface charge of soil colloids. The values 20-30 were obtained for soil clays in which allophane and imogolite predominated while the values 0.04-0.12 were found for those with crystalline layer silicates or humus. The void ratios of fine-textured volcanic ash soils are in the range from 2 to 5, and mostly from 3 to 4. These values are higher than 0.8 to 1.0 for sandy alluvial soils and 1.5 to 2.5 for clayey alluvial soils. A peculiarity common to volcanic ash soils subjected to the standard consolidation test is that preconsolidation load (e.g., 20-40 tons/m?) far exceeds the present overburden pressure or even the bearing capacity (10-15 tons/m2), Since there is no geologic evidence for this preconsolidation, one might imagine the same effect being produced by aggregation of soil particles with allophane and related colloids, possibly through drying during weathering of volcanic ash (Birrell, 1964; Suyama and Oya, 1965). An anomalous compaction behavior has also been noted for volcanic ash soils. The soil normally gives a single compaction curve with a welldefined maximum, which is primarily determined by the density and the size distribution of soil particles. The compaction curve shows the dry density values attained at the respective moisture contents when the soil is subjected to the same compaction procedure. However, a volcanic ash soil gives a variety of compaction curves, and hence, the maximum dry density and the optimum moisture at which the former is attained varies depending upon the initial moisture content of the sample. Kuno (quoted by Suyama and Oya, 1965) for example, reported a variation of the maximum dry density from 1.02 to 0.58 tons/m3 in association with the corresponding variation of the optimum moisture content from 60 to 135%, when the initial moisture content of a volcanic ash soil from the Kanto district has been changed from 40 to 130%. Another anomaly which was noted for the allophanic soil is that the clear maximum in the dry density curve was attained only on the moistening cycle but not on the drying cycle. This behavior was again attributed to the fact that allophane can contain a relatively high amount of nonfree water and that there seems to be no sharp boundary between the nonfree and free water. It is clear that building structures that involve soils with allophanic materials, particularly those disturbed and compacted, make special care necessary.

252

KOJI WADA AND M. E. HARWARD

c .

CURRENT STATUS AND

MANAGEMENT OF SOIL

SYSTEMS

Historically, problems have been encountered in the use and management of soils high in amorphous constituents. This is true for the United States as well as other Pacific r i d countries such as Japan and New Zealand. The problems involve failure to recognize and evaluate the importance of the amorphous components. Laboratory analyses either have not been oriented to these materials or the wide range in sensitivity of the amorphous materials to the index tests has not been recognized. To a large degree the problems encountered involve a high degree of porosity in an undisturbed state, large water-holding capacities with the tendency for the clays to be saturated or nearly so in nature, and rearrangement of the gel structure or matrix upon manipulation. Although not always understood, some of the relationships to engineering properties have been at least recognized. In view of this it is somewhat surprising that soil scientists have not done more to relate the nature and occurrence of amorphous materials to problems of soil management which involve dispersion, flocculation, aggregation, infiltration, erosion, and landscape stability. Perhaps this is partially due to our overzealous efforts to “clean-up” the sample prior to the analysis and obtain data more like those of standard or reference specimens which are easier to interpret. In doing so, we may have removed or modified components so that the samples no longer reflect the properties of the systems in situ. A significant accomplishment toward the recognition and evaluation of amorphous components has been made by Jones and Uehara (1973). They have perfected techniques for transmission electron microscopy which show the presence of gel-hulls that enclose the bridge between crystalline particles. Their success is due to a combination of factors including absence of chemicals and use only of mild sonification for dispersion, use of holey substrate for mounting of specimens, operation of the equipment at high voltage and low current to minimize effects of the beam, and exposure of an area only long enough to focus and obtain a film record. Subsequent interpretations were then made from photo micrographs. These procedures permitted them to detect gellike materials in aluminosilicate systems, high aluminum soils and on quartz surfaces. In view of previous experience with relationships of amorphous constituents to engineering properties, relationships to landscape stability are to be expected. In a study of a watershed in the Cascade Range, Youngberg et al. (1973) found amorphous inorganic colloids to be dominant components of readily dispersible soils which were also major contributors to stream turbidity. Amorphous components tended to remain in suspension in the reservoir longer than smectite clay minerals. Similar studies of a

AMORPHOUS CLAY CONSTITUENTS OF SOILS

253

watershed on older geologic formations and under a different climate indicate that samples which contain a combination of smectite and amorphous clays tend to remain in suspension and contribute to continued turbidity of the reservoirs (Silvernale et al., 1973). In view of current interest in preservation of the environment and our natural resources, efforts to develop similar relationships should be increased. VII.

Summary

There are different classes of amorphous clay constituents in soils. They have a common characteristic due to structural randomness, but react differently in physical and chemical reactions in soils because of differences in chemical composition and structure. The amorphous clay constituents in soils derived from volcanic ash have been studied most extensively and intensively, simply because they are predominating in the clay fraction and have predominating influence on the physical and chemical properties of the soils. Considerable progress has been made in characterization of these amorphous clay constituents, specifically allophane, imogolite, and opaline silica, by application of selective dissolution techniques combined with physical methods as represented by infrared spectroscopy and electron microscopy. Amorphous constituents have a significant effect on the properties of soils. Unique physical properties associated with these components include high water holding capacity, slippery but nonsticky consistence, high Atterberg limits, greater values for liquid and plastic limits on undried than dried samples, low bulk densities, high void ratios, and large preconsolidation loads, and anomalous compaction behaviors. Some of the more important chemical properties include ability to retain large amounts of organic matter, pH-dependent cation and anion exchange capacities, large phosphate fixation capacities, weak strength of ion adsorption, and high pH-low base saturation relationships. Knowledge about structure, morphology, and charge characteristics of allophane and imogolite has provided a basis for interpreting their effect on the soil properties which pose a number of problems in practical management and use of soil systems. A better understanding of the processes of soil development from volcanic ash has been obtained by applying advanced methods of mineralogical analysis, and by relating the formation and transformation of the amorphous clay constituents to crystalline clay constituents and to accumulation of organic matter. One of the problems left to future studies is characterization of amorphous sesquioxidic constituents, which are probably associated either with allophane or with organic matter. There are indications that these relatively

254

KOJI WADA AND M. E. HARWARD

poorly organized constituents have a great role in many reactions in the soil. Another problem in need of elucidation involves the chemical composition and confirmed arrangement of atoms in specific amorphous compounds. The studies on the amorphous clay constituents in the soils derived from parent material other than volcanic ash have shown that in many cases they are different from those in volcanic ash soils. Their characterization has been hampered by the relatively small content, and possibly by the very nature of these amorphous constituents. Exploring the methods for their characterizaton and obtaining the data for verifying their effects on soil properties are essential, since the soils containing those amorphous constituents compose a great portion of the soils on earth. Many difficulties will arise, but recent developments in the study of allophane and related constituents in volcanic ash soils hold out hope in future studies.

REFERENCES Ahlrichs, J. L. 1968. Clays Clay Miner. 16, 63-71. Aomine, S., and Egashira, K. 1968. Soil Sci. Plant Nutr. (Tokyo) 14, 94-98. Aomine, S., and Kobayashi, Y. 1964. Soil Sci. Plant Nutr. (Tokyo) 10, 28-32. Aomine, S., and Kobayashi, Y. 1966. Soil Sci. Plant Nutr. (Tokyo) 12, 7-12. Aomine, S., and Miyauchi, N. 1963. Nature (London) 199, 1311-1312. Aomine, S., and Miyauchi, N. 1965. Soil Sci. Plant Nutr. (Tokyo) 11, 212-219. Aomine, S., and Mizota, C. 1973. Int. Clay Conf., Proc., 1972. pp. 207-213. Aomine, S., and Otsuka, H. 1968. Trans. In?. Congr. Soil. Sci. 9th, 1968 Vol. 1, pp. 731-737. Aomine, S., and Wada, K. 1962. Amer. Mineral. 47, 1024-1048. Aomine, S., Inoue, A., and Mizota, C. 1972. Clay Sci. 4, 95-104. Atkinson, R. J., Posner, A. M., and Quirk, J. P. 1967. J . Phys. Chem. 71, 550-558. Barber, R. G., and Rowell, D. L. 1972. J . Soil Sci. 23, 135-146. Bascomb, C. L. 1968. J . Soil Sci. 19, 251-268. Bates, T. F. 1962. Clays Clay Miner, 9, 315-328. Besoain, E. 1968/1969. Geoderma 2, 151-169. Birrell, K. S. 1952. Proc. Aust. New Zealand Conf. Soil Mech. Found. Eng., lst, 1952. pp. 30-34. Birrell, K. S . 1961a. J . Soil Sci. 12, 307-316. Birrell, K. S. 1961b. N.Z. J . Sci. 4, 393-414. Birrell, K. S. 1964. F A 0 World Soil Resour. Rep. 14, 74-81. Birrell, K. S. 1966. N . Z . J . Agr. Res. 9, 554-564. Birrell, K. S., and Gradwell, M. 1956. J . Soil Sci. 7 , 130-147. Birrell, K. S., Fieldes, M., and Williamson, K. I. 1955. Amer. Mineral. 40, 122-124. Blume, H. P., and Schwertmann, U. 1969. Soil Sci. SOC. Amer., Proc. 33,438-444. Bower, C . A., and Truog, E. 1940. Soil Sci. SOC.Amer., Proc. 5, 86-89. Bracewell, J. M., Campbell, A. S., and Mitchell, B. D. 1970. Clay Miner. 8, 325-335. Brindley, G. W. 1969. US.-Jap. Semin. Amorphous Clay Muter. Abstract, p. 11.

AMORPHOUS CLAY CONSTITUENTS OF SOILS

255

Brindley, G. W., and Fancher, D. 1970. Int. Clay Conf.,Proc., 1969 Vol. 2, pp. 29-34. Brindley, G. W., and Pedro, G. 1970. AZPEA Newslett. No. 4, p p . 3-4. Brindley, G. W., and Pedro, G. 1972. A I P E A Newslett. No. 7, pp. 8-13. Briner, G. P., and Jackson, M. L. 1969. Aust. J . SoilRes. 7, 163-169. Briner, G. P., and Jackson, M. L. 1970. Isr. J . Chem. 8, 487-500. Brown, G. 1955. Clay Miner. Bull. 2, 294-302. Brydon, J. E., and Day, I. H. 1970. Can. J . Soil Sci. 50, 35-41. Brydon, J. E., and Kodama, H. 1966. Amer. Mineral. 51, 875-889. Calhown, F. G., Carlisle, V. W., and Luna, Z. C. 1972. Soil Sci. SOC. Amer., Proc. 36, 480-485. Campbell, A. S., Mitchell, B. D., and Bracewell, J. M. 1968. Clay Miner. 7, 451-454. Chichester, F. W., Youngberg, C. T., and Harward, M. E. 1969. Soil Sci. SOC. Amer., Proc. 33, 115-120. Chichester, F. W., Harward, M. E., and Youngberg, C. T. 1970. Clays Clay Miner. 18, 81-90. Clark, J. S. 1964. Soil Sci. 98, 302-306. Cloos, P., Herbillon, A., and Echeverria, J. 1968. Trans. Znt. Congr. Soil Sci., 9th, 1968 Vol. 2, pp. 733-743. Cloos, P., Leonard, A. J., Morean, J. P., Herbillon, A., and Fripiat, J. J. 1969. Clays Clay Miner. 17, 279-287. Coen, G. M., and Arnold, R. W. 1972. Soil Sci. Soc. Amer., Proc. 36, 342-350. Coleman, N. T. and Thomas, G. W. 1964. Soil Sci. SOC. Amer., Proc. 38, 187-190. Cradwick, P. D. G., Farmer, V. C., Russell, J. D., Masson, C. R., Wada, K., and Yoshinaga, N. 1972. Nature (London) Phys. Sci. 240, 187-189. Darragh, P. J., Gaskin, A. J., Terrell, B. C., and Sanders, J. V. 1966. Nature (London) 209, 13-16. D e Kimpe, C. 1970. Can. J . Soil Sci. 50, 317-330. DeMumbrum, L. E., and Chesters, G. 1964. Soil Sci. SOC.Amer., Proc. 28, 355-359. Deshpande, T. L., Greenland, D. J., and Quirk, J. P. 1968. J . Soil Sci. 19, 108-122. de Villiers, J. M. 1969. Soil Sci. 107, 454-461. de Villiers, J. M. 1971. Soil Sci. 112, 2-7. de Villiers, J. M., and Jackson, M. L. 1967. Soil Sci. SOC. Amer., Proc. 31, 473-476. Dingus, D. D. 1973. Dim. Abstr. 34, 3037B. Dingus, D. D., Jones, R. C., and Harward, M. E. 1973. Agron. Abstr. p. 152. Dudas, M. J. 1972. Diss. Abstr. 33, 4418B. Dudas, M. J., and Harward, M. E. 1971. Soil Sci. Soc. Amer., Proc. 35, 134-140. Dudas, M. J., and Harward, M. E., 1974. Soil Sci. SOC. Amer. Proc. 38, (in press). Egashira, K., and Aomine, S. 1974. Clay Sci. 4, 231-242. Egawa, T. 1964. Clay Sci. 2, 1-7. Egawa, T., Watanabe, Y., and Sato, A, 1959. Advan. Clay Sci. Tokyo 1, 260-270. Eswaran, H. 1972. Clay Min. 9, 281-285. Farmer, V. C., and Russell, J. D. 1967. Clays Clay Min. 15, 121-142. Fieldes, M. 1955. N . Z. J . Sci. Techno/., Sect., B 37, 336-350. Fieldes, M. 1966. N. 2. J. Sci. 9, 599-607. Fieldes, M., and Furkert, R. 1. 1966. N . Z. J . Sci. 9, 608-622. Fieldes, M., and Furkert, R. J. 1966. N. Z. J . Sci. 9, 623-629. Flach, K. 1964. F A 0 World Soil Resour. Rep. 14, 61-70. Follett, E. A. C., McHardy, W. J., and Mitchell, B. D. 1965a. C[ay Miner. 6, 23-34.

256

KOJI WADA AND M. E. HARWARD

Follett, E. A. C., McHardy, W. J., Mitchell, B. D., and Smith, B. F. L. 1965b. Clay Miner. 6, 35-43. Forsythe, W. M., Gavande, S . A., and Gonzalez, M. A. 1969. Panel Volcanic Ash Soils La!in Amer. 1969 Paper B3, pp. 1-6. Franzmeier, D. P., Hajek, B. F., and Simonson, C. H. 1965. Soil Sci. SOC. Amer., Proc. 29, 737-743. Furkert, R. J., and Fieldes, M. 1968. Trans. In!. Congr. Soil. Sci., 9!h, 1968 Vol. 3, pp. 133-141. Gotz, J., and Masson, C. R. 1970. J. Chem. Soc., A. pp. 2683-2686. Gotz, J., and Masson, C. R. 1971. J . Chem. Soc., A . pp. 686-688. Grant, W. H. 1963. Clays Clay Miner. 11,65-73. Grant, W. H. 1964. Clays Clay Miner. 12, 455-463. Greenland, D.J. 1971. Soil Sci. 111, 34-41. Greenland, D. J., and Wilkinson, G. K. 1969. Int. Clay Conf., Proc., 1969 Vol. 1, pp. 1861-1870. Greenland, D. J., Wada, K., and Hamblin, A. 1969. Am!. J. Sci. 32, 56-58. Grim, R. E. 1953. “Clay Mineralogy,” McGraw-Hill, New York. Gruner, J. W. 1935. Amer. Mineral. 20, 475-483. Hamblin, A. P., and Greenland, D. J. 1972. Aust. J . Soil Res. 10,61-79. Harada, Y., and Wada, K. 1972. Abstr. Pap., SOC. Sci. Soil Manure Jap. 18, 22. Harada, Y.,and Wada, K. 1973. Soil Sci. Plan! Nutr. ( T o k y o ) 19, 73-82. Harris, B. L. 1973. Diss. Abstr. 34, 951B. Hashimoto, I., and Jackson, M. L. 1960. Clays Clay Miner. 7, 102-113. Hattori, T.,and Morita, S. 1966. J. Sci. Soil Manure, Jap. 37, 189-193. Hay, R. C. 1960. Amer. J . Sci. 258, 354-368. Hendricks, D. M., Whittig, L. D., and Jackson, M. L. 1967. Clays Clay Miner. 15, 395-407. Higashi, T., and Ikeda, H. 1974. Clay Sci. 4, 205-211. Hingston, F. J., Atkinson, R. J., Posner, A. M., and Quirk, J. P. 1967. Nature (London) 215, 1459-1461. Hingston, F. J., Atkinson, R. I., and Posner, A. M. 1968. Trans. In!. Congr. Soil Sci., 9!h, 1968 Vol. 1, pp. 669-678. Hsu, P. H. 1968. Clays Clay Miner. 16,303-3 11. Iimura, K. 1966. Bull. Nu!. Inst. Agr. Sci. (Japan) B17, 101-157. Iimura, K. 1969. Proc. lnt. Clay Conf., 1969 Vol. 1, pp. 161-172. Iimura, K. 1971. In “Dojo Hiryo no Kenkyu (Studies on Soils and Soil Fertility),” Vol. 2, pp. 56-61. Yokendo, Tokyo. Ingles, 0. G., and Willoughby, D. R. 1967. Soil Sci. 104, 383-385. Ionue, T.,and Wada, K. 1968. Trans. Int. Congr. Soil Sci., 9th, 1968, Vol. 3, pp. 289-298. Ionue, T., and Wada, K. 1971a. Clay Sci. 4, 61-70. Inoue, T., and Wada, K. 1971b. Clay Sci. 4, 7 1-80. Jackson, M. L. 1956. “Soil Chemical Analysis-Advanced Course.” Published by the Author, Madison, Wisconsin. Jackson, M. L. 1962. Clays Clay Miner. 11, 29-46. Jackson, M. L. 1965. In “Methods of Soil Analysis” (C. A. Black, ed.) Part 1, pp. 578-603. Amer. SOC.Agron., Madison, Wisconsin. Jaritz, von G. 1967. Z. Pflanzenernaehr., Dueng., Bodenk. 117, 65-77. Jones, J. B., and Segnit, E. R. 1971. 1. Geol. SOC.Am!. 18, 57-68. Jones, J. B., Sanders, J. V., and Segnit, E. R. 1964. Nature (London) 204, 990-991.

AMORPHOUS CLAY CONSTITUENTS OF SOILS

257

Jones, R. C., and Uehara, G. 1973. Soil Sci. SOC.Amer., Proc. 37, 792-798. Kanno, I. 1959. Advan. Clay Sci. Tokyo 1, 213-223. Kanno, I. 1961. Bull. Kyushu Agr. Exp. Stn. 7, 1-185. Kanno, I., and Arimura, S. 1967. Soil Sci. Plant Nutr. ( T o k y o ) 13, 165-170. Kanno, I., Onikura, Y . , and Higashi, T. 1968. Trans. Int. Congr. Soil Sci., 9 t h 1968 Vol. 3, pp. 111-122. Kato, Y. 1970a. Pedologist 14, 16-21. Kato, Y. 1970b. J. Sci. Soil Manure, Jap. 41, 218-224. Kawaguchi, K., Fukutani, H., Murakami, H., and Hattori, T. 1954. Bull. Res. Inst. Food. Sci., Kyoto Univ. 14, 82-91. Kawasaki, H., and Aomine, S. 1966. Soil Sci. Plant Nutr. ( T o k y o ) 12, 144-150. Keller, W. D. 1964. In “Soil Clay Mineralogy” (C. I. Rich and G. W. Kunze, eds.), pp. 3-76. Univ. of North Carolina Press, Chapel Hill. Kinjo, T., and Pratt, P. F. 1971a. Soil Sci. SOC.Amer., Proc. 35, 722-725. Kinjo, T., and Pratt, P. F. 1971b. Soil Sci. SOC.Amer., Proc. 35, 725-728. Kitagawa, Y. 1971. Amer. Mineral. 56, 465-475. Kitagawa, Y. 1973. Clay Sci. 4, 151-154. Kobayashi, Y., and Aomine, S. 1967. Soil Sci. Plant Nutr. ( T o k y o ) 13, 189-194. Kobo, K., and Fujisawa, T. 1964. J . Sci. Soil Manure, Jap. 35, 40-46. Kobo, K., and Oba Y. 1965. J . Sci. Soil Manure, Jap. 36, 207-210. Kubota, T. 1972. Soil Sci. Plant Nutr. ( T o k y o ) 13, 189-194. Kumada, K., and Kato, H. 1970. Soil Sci. Plant Nutr. ( T o k y o ) 16, 195-200. Kurabayashi, S., and Tsuchiya, T. 1960. Advan. Clay Sci. Tokyo 2, 178-196. Kyuma, K., and Kawaguchi, K. 1964. Soil Sci. SOC.Amer., Proc. 28, 371-374. Lai, S. H., and Swindale, L. D. 1969. Soil Sci. SOC.Amer., Proc. 33, 804-808. Langston, R. B., and Jenne, E. A. 1964. Clays Clay Miner. 12, 633-647. Lavkulich, L. M., and Wiens, J. H. 1970. Soil Sci. SOC. Amer., Proc. 34, 755-758. Leonard, A,, Suzuki, S., Fripiat, J. J., and De Kimpe, C. 1964. J. Phys. Chem. 68, 2607-2617. Le Roux, J. 1973. Soil Sci. 115, 137-144. Lindqvist, B., and Jannson, S. 1962. Amer. Mineral. 47, 1356-1362. Lynn, W. C., and Whittig, L. D. 1966. Clays Clay Miner. 14, 241-248. MacEwan, D. M. C. 1961. In “The X-ray Identification and Crystal Structures of Clay Minerals” (G. Brown, ed.), pp. 143-207. Mineral. SOC.,London. McKeague, J. A. 1967. Can. J . Soil Sci. 47, 95-99. McKeague, J. A., and Day, J. H. 1966. Can. J. Soil Sci. 46, 13-22. McKeague, J. A., Brydon, J. E., and Miles, N. M. 1971. Soil Sci. SOC. Amer., Proc. 35, 33-38. Masui, J., Shoji, S., and Uchiyama, N. 1966. Tohoku J . Agr. Res. 17, 17-36. Matsuo, K. 1964. Bull. Nut. Inst. Agr. Sci., Ser. B 14, 285-356. Matsusaka, Y., and Sherman, G. D. 1961. Soil Sci. 91, 239-244. Mehra, 0. P., and Jackson, M. L. 1960. Clays Clay Miner. 7, 317-327. Mejia, G., Kohnke, H., and White, J. L. 1968. Soil Sci. SOC. Amer., Proc. 32, 665-670. Mekaru, T., and Uehara, G. 1972. Soil Sci. SOC.Amer., Proc. 36,296-300. Mitchell, B. D., and MacKenzie, R. C. 1954. Soil Sci. 77, 173-184. Mitchell, B. D., Farmer, V. C., and McHardy, W. J. 1964. Advan. Agron. 16, 327-3 83. Miyauchi, N., and Aomine, S. 1964. Soil Sci. Plant Nutr. ( T o k y o ) 10, 199-203. Miyauchi, N., and Aomine, S. 1966a. Soil Sci. Plant Nutr. ( T o k y o ) 12, 13-17.

258

KOJI WADA AND M. E. HARWARD

Miyauchi, N., and Aomine, S. 1966b. Soil Sci. Plant Nutr. ( T o k y o ) 12, 187-190. Miyauchi, N.,and Nakano, A. 1971. Bull. Fac. Agr., Kagoshima Univ. 21, 143-152. Miyazawa, K. 1966. Bull. Nut. Inst. Agr. Sci., Ser. B 17, 1-100. Oba, Y.,and Hayashi, M. 1971. Abstr. Pap., SOC.Sci. Soil Manure Jap. 17, 43. Oba, Y.,and Okano, K. 1967. Abstr. Pap., SOC.Sci. Soil Manure Jap. 14,23. Ossaka, J. 1962. Advan. Clay Sci. Tokyo 4, 33-47. Parfitt, R. L. 1972. Soil Sci. SOC.Amer., Proc. 36, 683-686. Parks, G.A. 1965. Chem. Rev. 65, 177-198. Patterson, S.H. 1964. Clays Clay Miner. 12, 153-172. Raman, K. V., and Mortland, M. M. 1969/1970. Geoderma 3,37-43. Rich, C. I. 1968. Clays Clay Miner. 16, 15-30. Rich, C.I., and Thomas, G. W. 1960. Advan. Agron. 12, 1-39. Rooksby, H. P. 1961. In ‘The X-ray Identification and Crystal Structures of Clay Minerals” (G. Brown, ed.), pp. 354-392. Mineral. SOC.,London. Russell, J. D., McHardy, W. J., and Fraser, A. R. 1969. Clay Miner. 8, 87-99. Sadzawka, R. M. A,, Melendez, A. E., and Aomine, S. 1972. Soil Sci. Plant Nutr. (Tokyo) 18, 191-197. Sakr, R., and Meyer, B. 1970. Goetfinger Bodenk. Ber. 4, 1-47. (Soifs Fert. 34, 292.) Saly, R., and Mihalik, A. 1970. Z. Pflanzenernaehr., Dueng., Bodenk. 127, 200-210. Sargeant, I. J., and Skene, J. K. M. 1970. Aust. J. Soil Res. 8, 281-295. Sawhney, B. L., and Norrish, K. 1971. Soil Sci. 112, 213-215. Sawhney, B. L., Frink, C. R., and Hill, D. E. 1970. Soil Sci. 109,272-278. Schnitzer, M. 1968. Trans. Int. Congr. Soil Sci., 9th, 1968 Vol. 1, pp. 635-644. Schofield, R. K. 1949. J. Soil Sci. 1, 1-8. Schwertmann, U. 1964. 2. Pflanzenernaehr., Dueng., Bodenk. 105, 194-202. Schwertmann, U. 1966. Nature (London) 212, 645-646. Schwertmann, U. 1970. Geoderma 3, 207-214. Schwertmann, U., Fisher, W. R., and Papendorg, H. 1968. Trans. Int. Congr. Soil Sci., 9, I968 Vol. 1, pp. 645-655. Segalen, P. 1968. Cah. ORSTOM (Off. Rech. Sci. Tech. Outre-Mer) Ser. Pedol. 6, 105-126. Sherman, G. D., Matsusaka, Y., Ikawa, H., and Uehara, G. 1964. Agroclzimica 8, 146-153. Shoji, S., and Masui, J. 1969a. Soil Sci. Plant Nutr. ( T o k y o ) 15, 161-168. Shoji, S., and Masui, J. 1969b. Soil Sci. Plant Nutr. ( T o k y o ) 15, 191-201. Shoji, S.,and Masui, J. 1971. J. Soil Sci. 22, 101-108. Shoji, S.,and Masui, J. 1972. J. Sci. Soil Manure, Jap. 43, 187-193. Sieffermann, G.,and Millot, G. 1969. Proc. Int. Clay Conf.,I969 Vol. 1, pp. 417-430. Silvernale, C. E., Simonson, G. H., and Dingus, D. D. 1973. Agron. Abstr. p. 153. Simonett, D. S.,and Bauleke, M. P. 1963. Soil Sci. SOC.Amer., Proc. 27, 205-212. Singh, B. R., and Kanehiro, Y. 1969. Soil Sci. SOC. Amer., Proc. 33, 681-683. Singleton, P. C., and Harward, M. E. 1971. Soil Sci. SOC.Amer., Proc. 35, 838-842. Snetsinger, K. G. 1967. Amer. Mineral. 52, 254-262. Suyama, K., and Oya, S. 1965. In “Kanto Loam (Volcanic Ash Deposits in Kanto District),” pp. 335-357. Tsukiji Shokan, Tokyo. Tada, A. 1969. In “Dojyo Butsuri (Soil Physics),” pp. 323-332. Yokendo, Tokyo. Tazaki, K. 1971. Geol. Soc. Jap. J . 77, 407-414. Tazaki, K. 1972. Chikyu Kagaku 26, 1-11. Theng, B. K. G. 1971. N . 2.I . Sci. 14, 1040-1056.

AMORPHOUS CLAY CONSTITUENTS OF SOILS

259

Tokashiki, Y., and Wada, K. 1971. Abstr. Pap. SOC.Sci. Soil Manure Jap. 17, 39. Tokashiki, Y., and Wada, K. 1972a. Clay Sci. 4, 105-114. Tokashiki, Y., and Wada, K. 1972b. Abstr. Pap., SOC. Sci. Soil Manure Jap. 18, 23. Townsend, F. C., and Reed, L. W. 1971. Clays Clay Miner. 19, 303-310. Tsukada, T., Nakano, T., and Deguchi, M. 1967. J. Sci. Soil Manure, Jap. 38, 23 2-23 8. Turner, R. C. 1965. Can. J . Soil Sci. 45,331-336. Tweneboah, C. K., Greenland, D. J., and Oades, J. M. 1967. Aust. J . Soil Res. 5, 247-26 1. Uchiyama, N., Masui, J., and Shoji, S. 1968a. J . Sci. Soil Manure, Jap. 39, 149-153. Uchiyama, N., Masui, J., and Shoji, S. 1968b. Soil Sci. Plant Nutr. ( T o k y o ) 14, 125-132. Uchiyama, N., Masui, J., and Shoji, S. 1968c. Soil Sci. Plant Nutr. (Tokyo) 14, 133-1 40. Udagawa, S., Nakada, T., and Nakahira, M. 1969. Znt. Clay Conf.,Proc., 1969 Vol. 1, pp. 151-159. van Olphen, H. 1971. Science 171, 90-91. van Raij, B., and Peech, M. 1972. Soil Sci. SOC.Amer., Proc. 36, 587-593. van Reeuwijk, L. P. 1967. Master’s Thesis, University of Natal. Wada, K. 1959. Soil Sci. 87, 325-330. Wada, K. 1966. Soil Sci. Plant Nutr. ( T o k y o ) 12, 176-182. Wada, K. 1967. Amer. Mineral. 52, 690-708. Wada, K. and Aomine, S . 1966. Soil Sci. Plant Nutr. ( T o k y o ) 12, 151-157. Wada, K., and Aomine, S. 1973. Soil Sci. 116, 170-177. Wada, K., and Ataka, H. 1958. Soil Plant Food (Tokyo) 4, 12-18. Wada, K., and Greenland, D. J. 1970. Clay Miner. 8, 241-254. Wada, K., and Harada, Y. 1969. Znt. Clay Conf., Proc., 1969 Vol. I , pp. 561-571. Wada, K., and Harada, Y. 1971. J . Soil Sci. 22, 109-117. Wada, K., and Henmi, T. 1972. C/ay Sci. 4, 127-136. Wada, K., and Inoue, T. 1967. Soil Sci. Plant Nutr. ( T o k y o ) 13, 9-16. Wada, K.,and Kubo, H. 1972. Abstr. Pap., SOC.Sci. Soil Manure Jap. 18, 22. Wada, K., and Kubo, H. 1973. Abstr. Pap., SOC. Sci. Soil Manure Jap. 19, 37. Wada, K., and Matsubara, I. 1968. Trans. Int. Congr. Soil Sci., 9th 1968. Vol. 3, pp. 123-131. Wada, K., and Tokashiki, Y. 1972. Geoderma 7, 199-213. Wada, K., and Tsuji, S. 1973. Abstr. Pap., SOC. Sci. Soil Manure Jap. 19, 34. Wada, K.,and Yoshinaga, N. 1969. Amer. Mineral. 54, 50-71. Wada, K., Yoshinaga, N., Yotsumoto, H., Ibe, K., and Aida, S . 1970. Clay Miner. 8, 487-489. Wada, K., Henmi, T., Yoshinaga, N., and Patterson, S. H. 1972. Clays Clay Miner. 20, 375-380. Watanabe, Y. 1961. Advan. Clay Sci. Tokyo 3, 83-89. Watanabe, Y. 1962. Advan. Clay Sci. Tokyo 4, 249-257. Watson, J. R. 1971. Soils Fert. 34, 127-134. Weaver, R. M., Syers, J. K., and Jackson, M. L. 1968. Soil. Sci. SOC. Amer., Proc. 32, 497-501. Weismiller, R. A., Ahlrichs, J. L., and White, J. L. 1967. Soil Sci. SOC. Amer., Proc. 31, 459-463. Wells, N., and Furkert, R. J. 1972. Soil Sci. 113, 110-115.

260

KOJI WADA AND M. E. HARWARD

Whelan, J. A., and Goldich, S. S. 1961. Amer. Mineral. 46, 1412-1423. White, J. L. 1971. Soil Sci. 112, 22-31. Wright, A. C . S. 1964. F A 0 World Soil Resour. Rep. 14, 9-22. Yoshida, M. 1961. Kagaku ( T o k y o ) 31, 310-313. Yoshida, M. 1970. J . Sci. Soil Manure, Jap. 41, 483-486. Yoshida, M. 1971. J . Sci. Soil Manure, lap. 42, 329-332. Yoshinaga, N. 1966. Soil Sci. Plant Nutr. ( T o k y o ) 12, 47-54. Yoshinaga, N. 1968. Soil Sci. Plant Nutr. ( T o k y o ) 14, 238-246. Yoshinaga, N., and Aomine, S. 1962a. Soil Sci. Plant Nutr. ( T o k y o ) 8(2), 6-13. Yoshinaga, N., and Aomine, S. 196213. Soil Sci. Plant Nutr. ( T o k y o ) 8 ( 3 ) , 22-29. Yoshinaga, N., and Yamaguchi, M. 1970a. Soil Sci. Plant Nutr. ( T o k y o ) 16, 121-127. Yoshinaga, N., and Yarnaguchi, M. 1970b. Soil Sci. Plant Nurr. ( T o k y o ) 16, 215-223. Yoshinaga, N., Yotsumoto, H., and Ibe, K. 1968. Amer. Mineral. 53, 319-323. Youngberg, C. T., Harward, M. E., Sirnonson, G. H., and Rai, D. 1973. In “Proceedings of the Fourth North American Forest Soils Conference” (B. Bernier and C. H. Winget, eds.). Lava1 University, Quebec (in press). Yuan, T. L. 1968. Soil Sci. 107,242-248.

THE CALIBRATION AND USE OF NET RADIOMETERS’ Sherwood B. ldso US.

Department of Agriculture, Agricultural Research Service, Conservation laboratory, Phoenix, Arizona

US.

Water

Introduction .................................................... Instruments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Calibration Methods ............................................. Utilizing the Basic Net Radiometer ................................ V. Modifications for Different Applications ............................. VI. Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References ...................................................... I. 11. 111. IV.

1.

261 262 263 268 269 272 272

introduction

Net radiation is the algebraic sum of all radiative energy fluxes incident upon and leaving a given object. Without further specification, it generally refers to the radiation balance of a particular land or water surface. In this context, net radiation is the difference between the total incoming flux of short-wave radiation of solar origin plus long-wave radiation of atmospheric origin and the total outgoing flux of short-wave radiation reflected from the ground or water surface plus the long-wave radiation emitted therefrom. It also includes a reflected component of long-wave radiation; but since the reflectance of ground, water, and vegetative surfaces for longwave radiation is universally a rather small quantity (Anderson, 1954; Idso and Jackson, 1969; Idso et al., 1969a; Lorenz, 1966; Sparrow and Cess 1966), this component is often not explicitly included in descriptive formulations of the net radiation flux, although it is automatically included in any net radiation measurement. Net radiation is the most important factor in the energy balance of earth’s surface (Budyko, 1956; Geiger, 1965; Kondrat’yev, 1970) ; it is a necessary requisite of most usuable schemes for predicting evaporation (Ferguson, 1952; Penman, 1948; McIlroy and Angus, 1964; Rosenberg et al., 1968; Sellers, 1964; Slatyer and McIlroy, 1961; Tanner, 1960, 1967; Tanner and Fuchs, 1968; Tanner and Pelton, 1960; Van Bavel, 1966) Contribution from the U.S. Department of Agriculture, Agricultural Research Service.

26 1

262

SHERWOOD B. IDS0

and for scheduling irrigations (Franzoy and Tankersley, 1970; Heermann and Jensen, 1970; Jensen, 1969, 1972; Jensen and Heermann, 1970; Jensen et ai., 1970, 1971). Thus, the study of net radiation has become a subject of considerable importance for many soil scientists and agronomists. Over the years, several attempts have been made to develop techniques €or estimating net radiation from other more easily measured parameters (Davies, 1967; Fritschen, 1967; Gay, 1971; Idso et al., 1969b; Linacre, 1968, 1969; Monteith and Szeicz, 1961; Pasquill, 1949; Penman, 1948 Richardson, 1931; Scholte-Ubing, 1961; Stanhill et al., 1966). Owing to effects of clouds, however, such procedures have not achieved a degree of accuracy that is satisfactory for all-weather utilization. Indeed, even when these procedures are applied only to clear sky conditions, Idso (1968, 1971a) and Nkemdirim (1972, 1973) have shown that they may not be adequately generalized. Direct measurement thus remains the only reliable method for obtaining precise information on net radiation under all the varied conditions existing in the natural environment. Consequently, this paper presents a summary of some of the basic aspects of net radiometers and techniques for using them. It.

Instruments

The primary component of all net radiometers is the thermopile. This device usually takes the form of a flat black plate of some insulating material with a series of thermocouple junctions located on each side. The temperature difference across the plate is proportional to the net radiation to which it is exposed; and by means of a calibration constant, the electrical output of the thermopile is converted to yield this flux. The basic design of the more common thermopiles in use today was described by Funk (1959, 1962a) and Fritschen (1963, 1965). Other variants were reported by Baumgartner (1952), Campbell et al. (1964), Frttschen and Van Wijk (1959), and Hofmann (1952). Gates (1964), Monteith (1959), and Monteith and Szeicz (1962) outlined the manufacture of simple versions. In addition, there is the economical net radiometer transducer which utilizes a pair of thermometers. It has been described by Suomi and Kuhn (1958) and Tanner et al. (1960, 1969). There are two major types of net radiometers-ventilated and shielded. In the first type, a forced stream of air is directed over the thermal transducer, or thermopile, to equalize as much as possible the effects of convection from its upper and lower surfaces. Examples of such instruments are those developed by Gier and Dunkle (1951) and Suomi et al. (1954). In the second type, transparent wind shields are used to reduce unequal convection effects. Some of the materials that have been used to construct

CALIBRATION AND USE OF NET RADIOMETERS

263

FIG. 1. The thermopile of a Fritschen net radiometer showing the epoxy bobbin around which are wound the several thermocouple junctions. The bobbin is encased in a flat disk of similar epoxy and is now ready to be painted black.

these wind shields are rock salt coated with selenium (Aleksandrov and Kurtener, 1941), sylvite coated with selenium (Khvoles, 1952), thallium bromoiodide (Houghton and Brewer, 1954; Kreitz, 1953; Stern and Schwartzmann, 1954), filters of germanium and silicon (Mills and Crawford, 1955), sheets of mica (Fritschen and Van Wijk, 1959), Saran plastic film (Fritschen, 1960), and polystyrene (Fritschen, 1963). The material used by most manufacturers today is polyethylene (Monteith, 1972). Since many good reviews dealing with the various commercially available net radiometers and their operating characteristics already exist (Gates, 1962, 1965a; Kondrat’yev, 1965; Monteith, 1972; Robinson, 1962; Sellers, 1965; Tanner, 1963), this information is not repeated, nor is the detailed theory underlying their operation. Instead, this review is devoted to practical discussions of calibration techniques, basic considerations of data acquisition, and methods of modifying net radiometers for a variety of different applications. Ill.

Calibration Methods

The most common method for calibrating net radiometers is the occulting technique (Collins and Rabich, 1971) or, more simply, the so-called “shading technique.” Calibration is accomplished by positioning a net radiometer alongside a pyranometer or solarimeter and alternately shading

264

SHERWOOD B. IDS0

and unshading both instruments simultaneously from the direct rays of the sun (or some other light source) with identical small, opaque shields. In the laboratory, a high-pressure xenon arc lamp is generally used as the light source (Collins, 1970), although tungsten filament lamps have also been used (Drummond, 1956; Fritschen, 1963; Hill et al., 1966). If the net radiometer is positioned high enough above the ground so that the shadow from the shield used to shade it from the direct beam of the light source is insignificant in altering the radiation it receives from below, the electromotive force (emf) change of the net radiometer’s output

FIG. 2. Calibrating a net radiometer by the common shading technique. The lower left picture shows obscuration of the sun by the shield over the net radiometer; and the lower right picture depicts the shading of the Eppley pyranometer used as the standard.

CALIBRATION AND USE OF NET RADIOMETERS

265

between shaded and unshaded conditions can be equated to the known energy flux change recorded by the standard pyranometer to yield the Calibration factor of the net radiometer. Important points to remember when using this technique include the following: (a) the net radiometer must not be positioned too close to the ground when calibrating; (b) the accuracy of the resulting calibration factor will be no better than that of the solarimeter used to obtain it; and (c) this technique gives the calibration factor for response to shortwave radiation (0 5 A _< 3 p m ) , which may or may not be the same as the net radiometer’s response to long-wave radiation (A > 3 pm). Chamber techniques are generally used to obtain long-wave radiation calibration factors. Two major variations are presently in vogue, one due primarily to Fritschen (1963) and the other to Funk (1959, 1962a,b).

FIG. 3. Back and front views of the calibration chamber constructed by L. J. Fritschen for the calibration of net radiometers at the US. Water Conservation Laboratory. It can be used for both short- and long-wave calibrations. In the former instance, an artifical light source is used to alternately illuminate a net radiometer and a pyranometer located at the central chamber; in the latter instance, the central chamber divider is removed, and the net radiometer is positioned there and calibrated by calculation of the net long-wave radiation exchange between the upper and lower chamber halves, which are kept at different temperatures by circulating water.

266

SHERWOOD B. IDS0

Fritschen’s chamber, developed from a prototype of Johnson (U.S. Weather Bureau, unpublished manuscript, 1956), is composed of two cubical halves with their common side removed. The inner copper walls are painted black and have copper tubing attached to their backsides. Two water streams of different temperature flow through the tubes attached to the upper and lower sets of five walls to create a net flux of thermal radiation. The net radiometer is positioned in the center of the chamber; and its output is equated with the calculated net long-wave radiative flux; this flux is derived from thermocouple-obtained wall temperatures and a knowledge of the wall emittance, which may be measured by a variety of techniques (Buettner and Kern, 1965; Conaway and Van Bavel, 1966; Fuchs and Tanner, 1966, 1968; Idso and Jackson, 1969). The formula given by Fritschen for calculating the net radiation in his chamber is not exact, however, and the correct expression derived by Idso (1970a) should be used when employing this technique. The long-wave calibration technique of Funk was developed from an earlier version of MacDowall ( 1955 ) which makes use of a more complex arrangement of two chambers. The net radiometer is positioned in the center of one of the insulated chambers and allowed to view the second (a blackbody cavity of different temperature) through a small aperture €or only a short period of time. The blackbody cavity most used today is an electrically heated cylindrical cavity of cast aluminum described by Collins (1968), the emissivity of which is known to be better than 0.999. Both Funk‘s and Collin’s analyses of the net radiation to which the net radiometer is exposed, however, are slightly in error. Idso (1970a) provides a more exact evaluation of this flux. Chamber techniques of the Funk and Fritschen type require no other standard radiometer or solarimeter for comparison; and they can yield very accurate long-wave calibration factors when properly utilized. The chambers are somewhat complex, however, and hardly worth the effort to construct for the calibration of just a few radiometers. Thus, they generally are used only in major laboratories, such as the CSIRO Division of Atmospheric Physics, which calibrates some 400 instruments annually (Collins and Rabich, 1971), and the U.S. Water Conservation Laboratory, which at its peak production of Fritschen-type net radiometers calibrated about 200 annually. A simpler technique for obtaining long-wave calibration factors of net radiometers was developed by Idso ( 1970a,b). In this method a net radiometer is positioned anywhere from 5 to 20 cm above the floor of a constant-temperature room. Then, a blackened metallic plate about 40 cm on a side (which was previously cooled about 10°C below the temperature of the room) is brought into the room and set beneath the radiometer.

CALIBRATION AND U S E OF N E T RADIOMETERS

267

FIG.4. The setup of the flat-plate technique for obtaining long-wave calibration factors of net radiometers.

The output of the radiometer is subsequently monitored, along with the temperature of the flat plate (by means of 4 or 5 embedded thermocouples), as the cool plate gradually warms to ambient room temperature. Since all radiative fluxes in the room are constant, except for the flux emitted by the warming plate, the millivolt output of the net radiometer can be plotted against the radiation received from the plate and its calibration factor determined as the slope of this line. The radiation from the plate is specified by the plate’s emittance, temperature, and view factor with respect to the radiometer, which may be determined from a variety of standard heat transfer texts (Eckert and Drake, 1959; Jakob, 1957; Sparrow and Cess, 1966). This technique has also been shown to be suitable for calibrating soil heat flux plates (Idso, 1972b). Both Fritschen and Funk concluded from their studies of short- and longwave calibration factors of polyethylene-shielded net radiometers that these instruments were about 5% more sensitive to short-wave radiation than to longwave radiation. A similar conclusion was also reached by Collins and Kyle ( 1966). Thus, for years, practically all polyethylene-shielded net radiometers have had a small strip of white paint applied to their thermopiles to reflect about 5 % of the incident solar radiation and thereby equate their short- and long-wave calibration factors. Idso ( 1970a), however, showed that the inaccuracies in the original derivations of the equations representing the net long-wave radiation fluxes in the chambers of Fritschen and Funk were such as to completely, and rather exactly, negate the need for this paint. Experimentation with the flat plate technique also confirmed this analysis. Thus, for the most exacting requirements of net radiation measurement, polyethylene-shielded instruments without any white paint on their transducers should be employed.

SHERWOOD B. IDS0

268

IV.

Utilizing the Basic N e t Radiometer

Most of the primary considerations involved in net radiometry are discussed by Fritschen (1963, 1965) and Funk (1959, 1962a). In addition, there is an instruction manual by Fritschen and Mullins (1965) that is whollly devoted to this subject. The methodology covers the physical installation of net radiometers, leveling of net radiometers, preventing of internal and external condensation of water on radiometer domes, cleaning of domes, and replacement of damaged domes. Only a few additional comments are required. First, it has been my experience that radiometer calibration factors are not adversely affected unless the domes are extremely dust laden. Thus, cleaning the domes should be a rather conservative practice, since they are so easily scratched. Second, although a standard placement height of 1 meter is often suggested for net radiometers, this topic is a subject in itself. For instance, routine experimentation has been conducted with net radiometers at heights of 0.1 m (Van Bavel and Fritschen, 1965), 0.3 m (Van Bavel, 1967), 0.5 m (Rosenberg, 1966; Stanhill et al., 1966), 0.75 m (Idso et al., 1969b), 1.0 m (Davies and Buttimor, 1969; Fritschen, 1967; Stanhill and Fuchs, 1968; Thompson and Boyce, 1967), 1.5 m (Ekern, 1965; Fitzpatrick and Stern, 1966; Monteith and Szeicz, 1961), 2.0 m (Begg et al., 1964; Denmead, 1969; Morgan et al., 1970; Rosenberg, 1969), and 2.5 m (Stanhill et al., 1966). Of this group of investigators, only the ones using the lowest (0.1 m) and highest (2.5 m) heights gave any reasons for their choices-and these were only qualitative at that. Idso and Cooley (1971, 1972) conducted a comprehensive study of net radiometer height placement. They showed that if net radiation measurements are required over a surface such as a dry, bare soil that becomes significantly warmer than ambient air during clear days, net radiometers should be located at a height of only 20 or 25 cm. The reason for choosing this level is that the intervening air between the net radiometer and the ground interacts with the radiation from the soil and modifies it significantly over even smaller distances than this. Thus, true net radiation is theoretically measured only at the surface, if surface and air temperatures are different. However, as the surface is approached, the shadow of the net radiometer begins to alter significantly the radiation emitted and reflected by the surface. The level of compromise between these two effects was found to be between 20 and 25 cm for radiometers of the Funk and Fritschen type (about 6 cm in diameter). If for some reason a different level is selected, Kondo (1972) has developed procedures for estimating the required corrections to surface conditions.

CALIBRATION AND USE OF NET RADIOMETERS

269

When surface and air temperatures do not differ greatly, such as over an irrigated field or crop, considerations of homogeneity of field of view tend to predominate in selecting the height of net radiometer placement. View factors must then be calculated for various heights to determine the best level for representative sampling of the total system or a specific portion of it, i.e., the bare soil between crop rows. Standard heat transfer texts may be consulted for this exercise or the two more agriculturally oriented papers of Reifsnyder and Lull (1965) and Reifsnyder (1967), which contain some useful graphs that often obviate the need for an investigator to make the actual calculations. Other examples of view factor utilization in net radiometry have been published by Gates (1962, 1965a) and Waggoner and Reifsnyder (1961).

V.

Modifications for Different Applications

The basic Fritschen and Funk net radiometers are extremely versatile instruments. Solar Radiation Instruments' (SRI) of Australia markets a Funk type that is claimed to be suitable for use in nine different forms. These are: (a) a regular net radiometer, (b) a short-wave balance meter, (c) a solarimeter, (d) an albedometer, (e) a total hemisphrical radiometer, ( f ) a total albedo and long-wave emission radiometer, (g) an underwater short-wave balance meter, ( h ) an underwater solimeter, and (i) an underwater albedometer. Most of the procedures required to transform the basic net radiometer into these other forms are rather simple and straightforward. By merely removing the polyethylene domes and replacing them with glass, for instance, the net radiometer is transformed into a short-wave balance meter. The SRI instrument may be purchased with interchangeable glass domes specifically manufactured for this purpose. Working with the Fritschen net radiometer, I have also made this transformation with the glass domes that are normally used for the Kipp solarimeter, as they are of the proper size to be held in place by the net radiometer's clamping rings. By painting either the inner or outer surface of one of the glass domes black to restrict light entry to only one side of the thermal transducer, the shortwave balance meter may be transformed into either a solarimeter or an albedometer, depending upon whether it is operated in an upright or an inverted position. In both of those applications, however, additional steps must be taken to ensure that the painted dome does not absorb an excessive *Trade names and company names are included for the benefit of the reader and imply no endorsement or preferential treatment of the product listed by the U.S.Department of Agriculture.

270

SHERWOOD B. IDS0

FIG. 5 . Some modifications of net radiometers. Upper left: The basic Fritschen net radiometer operated at optimum height (25 cm) above a surface of dry, bare soil. Note the presence of the strip of white paint on the thermal transducer, which recent research has proven to be unnecessary. Upper right: An underwater solarimeter formed by replacing polyethylene domes with glass, painting the lower dome black, and waterproofing the lead connecter. Lower right: A solarimeter for terrestrial use made by applying white paint over the blackened lower glass dome of the upper right instrument and attaching the double-layered shield to reduce lower dome heating. Lower left: A hemispherical all-wave radiometer made by applying black, and then white, paint to the outside of the lower dome of a polyethylene-shielded net radiometer, taping a thermocouple to it, and attaching the double-layered shield.

amount of either short- or long-wave radiation and thereby become warmer than the clear dome, resulting in a net nonzero exchange of long-wave radiation between the two domes and the thermal transducer. White paint applied upon the black that reflects considerable short-wave radiation and an opaque circular shade that shields the painted dome from the direct rays of the sun or the thermal emission and reflected short-wave radiation from the ground can effectively eliminate this source of error (Idso, 1971b). This technique thus eliminates the need for black and white-hot and cold junctions of standard solarimeter thermopiles, such as the original KimbaII and Hobbs (1923) instrument and later models of Blackwell-Anderson (Anderson, 1967), Dirmhirn (Gates, 1962; Robinson, 1966), Ianishevsky (Monteith, 1959), and Monteith (1959), which cannot be transformed

CALIBRATION AND USE OF NET RADIOMETERS

27 1

into total all-wave radiometers because of the differing spectral properties of black paint and of white paint in the short- and long-wave regions. It also allows for a more accurate measurement of solar radiation at low intensities, where solarimeters based upon the black and block-hot and cold junction design of the Moll thermopile (Moll, 1923; Gorczynski, 1924) often exhibit large zero offset errors (Maxwell, 1969). One of the polyethylene domes of the original net radiometer can also be similarly painted and shielded, resulting in either a total hemispherical radiometer or a total albedo and long-wave emission radiometer. In this case the temperature of the painted dome must also be known. Originally, Idso (1971b, 1972a) obtained this temperature from three or four fine thermocouples sandwiched between a double-layered polyethylene dome. Later research, however, demonstrated that equally good results could be obtained with a single thermocouple merely taped to a normal singlelayer painted dome (Idso, 1 9 7 2 ~ ) . The three underwater applications of the net radiometer derive from merely waterproofing the lead connectors of some of the previously mentioned modifications. The short-wave balance meter thereby becomes an underwater short-wave balance meter; and the solarimeter and albedometer convert to their underwater counterparts. For shallow applications, even polyethylene versions can be used in this way (Idso, 1972d), since no long-wave radiation moves in water and it makes no difference whether polyethylene or glass envelops the transducer. In this case, increased air pressure must be supplied to keep the domes from collapsing; whereas for the glass-domed instruments, air need not be supplied for most applications. Idso and Gilbert (1974) and Idso and Foster (1974) have kept glass versions underwater for as long as 6 months without any leakage or other deleterious effects. In addition to these nine variations of the basic net radiometer, there is also a tenth use to which it may be put, although this modification is somewhat more difficult to make and use. It is the transformation of a net all-wave radiometer into a net long-wave radiometer described by Paltridge ( 1969). The procedure involves constructing two hemispheres of black polyethylene and joining them at their circumference so that they completeIy encase a regular net radiometer. Both the domes of the net radiometer and the outer black polyethylene sphere are then inflated by dry air; and the outer shell is spun continuously about the stationary net radiometer. In this way the asymmetric heating of the black polyethylene due to solar radiation absorption is neutralized; and the output of the net radiometer becomes proportional to the net long-wave radiation transmitted by the black polyethylene and absorbed at the radiometer’s transducer.

272

SHERWOOD B. IDSO

VI.

Summary

Knowledge of net radiation is essential to many agricultural research endeavors and to most practical schemes of irrigation scheduling. For both of these applications, estimation techniques are not suitable for all-weather utilization; and direct measurement must be relied upon to obtain sufficiently accurate net radiation data. The primary component of a net radiometer is a thermal transducer or thermopile. Two different approaches are used to minimize or equalize convection effects from its upper and lower surfaces; these are to ventilate both surfaces equally or to shield them with transparent domes. The material most used for this latter purpose today is polyethylene. Net radiometers are calibrated for sensitivity to short-wave radiation (0 5 A 5 3pm) by simple shading techniques employing standard solarimeters for comparison. Long-wave calibrations ( A > 3pm) may be obtained from either special calibration chambers or from a simple flat-plate technique, neither of which approaches requires any other radiometer. The most recent intensive work in this area has indicated that popular polyethylene-shielded net radiometers have comparable sensitivities for both short- and long-wave radiation and that they therefore do not need the small strip of white paint that has routinely been applied to their black transducers by most manufacturers in an attempt to correct for an erroneously assumed inequality in these two calibration factors. Height placement of net radiometers is dependent upon the homogeneity of the underlying surface and its temperature relative to that of ambient air. When these two temperatures differ, common polyethylene-shielded net radiometers ( 6 cm diameter) are preferably located at a height of 20-25 cm. Polythylene-shielded net radiometers are extremely versatile instruments, being easily transformed for utilization in nine additional applications. As such, they enjoy a wide range of use in many fields of environmental research. REFERENCES:

Aleksandrov, B. P., and Kurtener, A. V. 1941. Meteorol. Gidrol. No. 3. Anderson, E. R. 1954. US.,Geol. Surv., Prof. Pap 269, 71-119. Anderson, M. C. 1967. J . Appl. Meteorol. 6 , 941-947. Baumgartner, A. 1952. Forstwiss. Centralbl. 71, 337-349. Begg, J. E., Bierhuizen, J. F., Lemon, E. R., Misra, D. K., Slatyer, R. O., and Stern, W. R. 1964. Agr. Meteorol. 1, 294-312. Budyko, M. I. 1956. "Teplovoi balans zemnoi poverkhnosti." Gidrometeo-logicheskoe Izdatel'stvo, Leningrad. Buettner, K. J. K., and Kern, C. D. 1965. J . Geophys. Res. 70, 1329-1337.

CALIBRATION AND USE OF N E T RADIOMETERS

273

Campbell, G. S., Ashcroft, G. L., and Taylor, S, A. 1964. J . Appl. Meteorol. 3, 640-642. Collins, B. G. 1968. J. Sci. Instrum. [2] 1, 62-64. Collins, B. G. 1970. Pure Appl. Geophys. 80, 371-377. Collins, B. G., and Kyle, T. G. 1966. Pure Appl. Geophys. 63, 231-236. Collins, B. G., and Rabich, H. A. H. 1971. Aust. Meterol. Mag. 19, 83-90. Conaway, J., and Van Bavel, C. H. M. 1966. Tech. Rep. ECOM 2-67P-1. Davies, J. A. 1967. Quart. J . Roy. Meteorol. SOC.93, 109-115. Davies, J. A., and Buttimore, P. H. 1969. Agr. Meteorol. 6, 373-386. Denmead, 0.T. 1969. Agr. Meteorol. 6, 357-371. Drummond, A. J. 1956. Arch. Meteorol., Geophys. Bioklimatol., Ser. B 7, 413. Eckert, E. R. G., and Drake, R. M., Jr. 1959. “Heat and Mass Transfer.” McGrawHill, New York. Ekern, P. C. 1965. J . Geophys. Res. 70,785-793. Ferguson, J . 1952. Aust. J . Sci. Res. 5, 315-330. Fitzpatrick, E. A., and Stern, W. R. 1966. Agr. Meteorol. 3, 225-239. Franzoy, C. E., and Tankersley, E. L. 1970. Trans. Amer. SOC.Agr. Eng. 13, 814-816. Fritschen, L.J. 1960. Bull. Amer. Meteorol. SOC. 41, 180-183. Fritschen, L. J. 1963. J . Appl. Meteorol. 2, 165-172. Fritschen, L. J. 1965. I . Appl. Meteorol. 4, 528-532. Fritschen, L. J. 1967. Agr. Meteorol. 4, 55-62. Fritschen, L. J., and Mullins, K. 1965. U.S. Water Conserv. Lab. Rep. No. 5. Fritschen, L. J., and Van Wijk, W. R. 1959. Bull. Amer. Meteorol. SOC.40, 291-294. Fuchs, M., and Tanner, C. B. 1966. Agron. .I. 58, 597-601. Fuchs, M., and Tanner, C. B. 1968. J. Appl. Meteorol. 7, 303-305. Funk, J. P. 1959. J . Sci. Instrum. [2] 36, 267-270. Funk, J. P. 1962a. J . Geophys. Res. 67, 2753-2760. Funk, J. P. 1962b. Arch. Meteorol., Geophys. Bioklimatol., Ser. B 11, 70-74. Gates, D. M. 1962. “Energy Exchange in the Biosphere.” Harper, New York. Gates, D. M. 1964. Sci. Teacher 31,No. 4. Gates, D. M. 1965a. Agr. Meteorol., Meteorol. Monogr. 6, No. 28, 1-26. Gates, D. M. 1965b. Proc. Symp. Remote Sensing Environ., 3rd, 1960 pp. 573-600. Gay, L. W. 1971. Arch. Meteorol., Geophys. Bioklirnatol., Ser. B 19, 1-14. Geiger, R. 1965. “The Climate Near the Ground.” Harvard Univ. Press, Cambridge, Massachusetts. Gier, J. T., and Dunkle, R. V. 1951. Trans. Amer. Inst. Elec. Eng. 70, 339. Gorczynski, L. 1924. Mon. Weather Rev. 52, 299-301. Heermann, D. F., and Jensen, M. E. 1970. Proc. ASAE Nut. Irrig. Symp., 1970 00-1 to 00-10. Hill, A. N., Latimer, J. R., Drummond, A. J., and Greer, H. W. 1966. Sol. Energy 10, 1-11. Hofmann, G. 1952. Forstwiss. Centralbl. 71, 330-337. Houghton, J. T., and Brewer, A. W. 1954. J . Sci. Insfrum. [2] 31,NO. 5. Idso, S. B. 1968. J . Appl. Meteorol. 7, 716-717. Idso, S. B. 1970a, Rev. Sci. Instrum. 41, 939-943. Idso, S. B. 1970b. Agr. Meteorol. 8, 235-243. Idso, S. B. 1971a. J. Meteorol. SOC.Jap. 49, 1-12. Idso, S. B. 1971b. Agr. Meteorol. 9, 109-121. Idso, S. B. 1972a. Rev. Sci. Instrum. 43, 506-508. Idso, S. B. 1972b. Agr. Meteorol. 10, 467-471.

274

SHERWOOD B. IDS0

Idso, S. B. 1972c. Agr. Meteorol. 10, 473-476. Idso, S. B. 1972d. Limnol. Oceanogr. 17, 462-466. Idso, S. B., and Cooley, K. R. 1971. I . Meteorol. SOC.lap. 49, 343-349. Idso, S. B., and Cooley, K. R. 1972. J. Meteorol. SOC.lap. 50, 49-58. Idso, S. B., and Foster, J. M. 1974. Water Resour. Res. 10, 129-132. Idso, S . B., and Gilbert, R. G. 1974. 1. Appl. Ecol. 11, 399-401. Idso, S. B., and Jackson, R. D. 1969. I . Appl. Meteorol. 8, 168-169. Idso, S. B., Jackson, R. D., Ehrler, W. L., and Mitchell, S. T. 1969a. Ecology 50, 899-902. Idso, S. B., Baker, D. G., and Blad, B. L. 1969b. Quart. I. Roy Meteorol. SOC. 95, 244-257. Jakob, M. 1957. “Heat Transfer,” Vol. 11. Wiley, New York. Jensen, M. E. 1969.1. Soil Water Conserv. 24, 193-195. Jensen, M. E. 1972. In “Optimizing the Soil Environment Toward Greater Crop Yields” (D. Hillel, ed.), pp. 133-161. Academic Press, New York. Jensen, M. E., and Heermann, D. F. 1970. Proc. A S A E Nut. Irrig. Symp., 1970 NN-1 to ”-10. Jensen, M. E., Robb, D. C.,and Franzoy, C. E. 1970. ASCE I . Irrig. Drain. Div. 96, 25-38. Jensen, M. E., Wright, J. L., and Pratt, B. J. 1971. Trans. A S A E (Amer. SOC. Agr. Eng.) 14, 954-959. Khvoles, S. B. 1952. Agrofiz. Inst. Sb. Tr. Po Agronom. Fiz. No. 5. Kimball, H. H., and Hobbs, H. E. 1923. Mon. Weather Rev. 51, 239-242. Kondo, J. 1972. I . Meteorol. SOC.lap. 50, 381-383. Kondrat’yev, K. Ya. 1965. “Luchisty: Teplobmen atmosfere.” Gidrometeozidat, Leningrad. Kondrat’yev, K. Ya. 1970. In “Radiation Including Satellite Techniques,” WMO-No. 248, TP. 136. World Meteorological Organization, Geneva. Kreitz, E. 1953. Geofis. Pura Appl., 26. Linacre, E. T. 1968. Agr. Meterof. 5, 49-63. Linacre, E. T. 1969. I . Appl. Ecol. 6, 61-75. Lorenz, D. 1966. I . Appl. Meteorol. 5, 421-430. MacDowall, J. 1955. Meteorol. Mag. 84, 65. McIlroy, I. C., and Angus, D. E. 1964. Agr. Meteorol. 1, 201-224. Maxwell, R. K. 1969. Ph.D. Thesis, University of Minnesota, Minneapolis. Mills, I. M., and Crawford, B., Jr. 1955. I . Opt. SOC.Amer. 45, No. 6. Moll, W. J. H. 1923. Proc. Pkys. SOC.,London Sect. A 35, 257-260. Monteith, J. L. 1959. I . Sci. Instrum. [2] 36, 341-346. Monteith, J . L., ed. 1972. “Survey of Instruments for Micrometeorology.” Blackwell, London. Monteith, J. L., and Szeicz, G. 1961. Quart. J . Roy. Meteorol. SOC.87, 159-170. Monteith, J. L., and Szeicz, G. 1962. Arch. Meteorol., Geophys. Bioklimatol., Ser. B 11, 491-500. Morgan, D. L., Pruitt, W. O., and Lourence, F. J. 1970. R & D Tech. Rep. ECOM 68-618-2. Nkemdirim, L. C. 1972. Arch. Meteorol., Geophys. Bioklimatol., Ser. B 20, 23-40. Nkemdirim, L. C. 1973. Agr. Meteorol. 11, 229-242. Paltridge, G. W. 1969. Quart. 1. Roy. Meteorof. SOC.95, 635-638. Pasquill, F. 1949. Proc. Roy. SOC. Ser. A 198, 116-140. Penman, H. L. 1948. Proc. Roy. Soc., Ser. A 193, 120-145.

CALIBRATION AND USE OF NET RADIOMETERS

275

Reifsnyder, W. E. 1967. Agr. Meteorol. 4, 255-265. Reifsynder, W. E., and Lull, H. W. 1965. U S . Dep. Agr., Tech. Bull. 1344. Richardson, B. 1931. Trans. Amer. SOC. Civil. Eng. 95, 966-1019. Robinson, G. D. 1962. In “Catalogue of IGY/IGC Meterological Data,” WMO/OMM No. 135--IGY/AGI, 4, Vol. B. World Meteorological Organization, Geneva. Robinson, N., ed 1966. “Solar Radiation.” Elsevier, Amsterdam. Rosenberg, N. J. 1966. Agr. Meteorol. 3, 197-224. Rosenberg, N. J. 1969. Agr. Meteorol. 6, 179-184. Rosenberg, N. J., Hart, H. E., and Brown, K. W. 1968. “Evapotranspiration Review of Research.” University of Nebraska, Lincoln. Scholte-Ubing, D. W. 1961. Neth. J. Agr. Sci. 9, 81-93. Sellers, W. D. 1964. J . Appl. Meteorol. 3, 98-104. Sellers, W. D. 1965. “Physical Climatology.” Univ. of Chicago Press, Chicago, Illinois. Slatyer, R. O., and McIlroy, I. C. 1961. “Practical Microclimatology.” Plant Ind. Div., CSIRO, Canberra, Australia. Sparrow, E. M., and Cess, R. D. 1966. “Radiation Heat Transfer.” Books, Cole, Belmont, California. Stanhill, G., and Fuchs, M. 1968. Agr. Meteorol. 5, 183-202. Stanhill, G., Hofstede, G. J., and Kalma, J. D. 1966. Quart. J . Roy. Meteorol. SOC. 92, 128-140. Stern, S. C., and Schwartzmann, F. 1954. J . Meteorol. 11, No. 2. Suomi, V. E., and Kuhn, P. M. 1958. Tellus 10, 160-163. Suomi, V. E., Franssila, M., and Islitzer, N. F. 1954. J. Meteorol. 11, 276-282. Tanner, C. B. 1960. Soil Sci. SOC., Amer., Proc. 24, 1-9. Tanner, C. B. 1963. “Basic Instrumentation and Measurement for Plant Environment and Micrometerology.” Univ. of Wisconsin Press, Madison. Tanner, C. B. 1967. I n “Irrigation of Agricultural Lands,” pp. 534-574. Amer. SOC. Agron., Madison, Wisconsin. Tanner, C. B., and Fuchs, M. 1968. J. Geophys. Res. 73, 1299-1304. Tanner, C. B., and Pelton, W. L. 1960. J. Geophys. Res. 65, 3391-3413. Tanner, C . B., Businger, J. A., and Kuhn, P. M. 1960. J . Geopltys. Res. 65, 3647-3 667. Tanner, C. B., Federer, C. A., Black, T. A., and Swan, J. B. 1969. Wis., Univ., Coll. Agr. Life Sci., Res. Rep. 40. Thompson, G. D., and Boyce, J. P. 1967. Agr. Meteorol. 4,267-279. Van Bavel, C. H. M. 1966. Water Resour. Res. 2, 455-467. Van Bavel, C. H. M. 1967. Tech. Rep. ECOM 2-67P-2. Van Bavel, C. H. M., and Fritschen, L. J. 1965. I n “Methodology of Plant EcoPhysiology,” Proc. Montpellier Symp., pp. 99-107. UNESCO. Waggoner, P. E., and Reifsnyder, W. E. 1961. Soil Sci. 91,246-250.

This Page Intentionally Left Blank

QUANTITATIVE GENETICS-EMPIRICAL RESULTS RELEVANT TO PLANT BREEDING' R. H. Moll* and C. W. 'Department

Stuber*t

of Genetics, North Carolina Stote University, and t U . 5 Department of

Agriculture, Agricultural Research Service, Raleigh, North Carolina

I. Introduction

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

11. Genetic Variability

111.

IV. V.

VI.

A. Basic Concepts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B. Experimental Estimates in Crop Species .......................... Inbreeding Depression and Heterosis ............................... A. Inbreeding Depression . . ................................... B. Heterosis .................................................... Genotype-Environmental Interactions ............................... A. Measurement of Genotype-Environmental Interactions . . . . . . . . . . . . . . B. Evaluation of Stability ......................................... Response to Selection ............................................. A. Genetic Variances and Expected Response ........................ B. Experimental Evaluation of Selection Procedures . . . . . . . . . . . . . . . . . . C. Correlated Responses and Selection Indexes ...................... Implications of Quantitative Genetics to Breeding Methodology . . . . . . . . A. Breeding Objectives ........................................... B. Development of Genetic Material with Breeding Potential . . . . . . . . . . C. Testing and Evaluation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References ......................................................

I.

277 278 278 281 284 284 285 287 289 290 295 296 296 302 305 305 306 308 310

Introduction

Quantitative genetics deals with the inheritance of those differences among individuals that are expressed in terms of degree rather than kind. In contrast with qualitative traits, in which variation is characterized by discrete classes, variation in quantitative traits forms a continuous array of values from one extreme to the other. Nearly every organ or function, including most economically important traits of crop species, show differences of a quantitative nature. The relevance of quantitative genetics to plant breeding lies in the fact that manipulation of genetic variability of quantitative traits through in-

' Paper No. 4236 of the Journal Series of the North Carolina State University Agricultural Experiment Station, Raleigh, North Carolina. 277

278

R. H. MOLL AND C. W. STUBER

breeding, crossbreeding, and selection are essential features of any plant breeding program. A primary objective of quantitative genetic research is an understanding of the genetic consequences of such manipulations. A basic premise of quantitative genetics is that the genes that affect quantitative traits follow the same laws of transmission as genes that affect qualitative traits. Usually many loci with small individual effects are involved; therefore, it is necessary to study these traits through statistics appropriate for continuous variables, such as means, variances, and covariances. Fisher ( I 918) provided the initial framework for the study of quantitative inheritance. Since that time, his developments have been clarified, elaborated, and extended by numerous geneticists and statisticians. Unfortunately, the experimental aspects of quantitative genetics have lagged behind theory. Because it is difficult to design quantitative genetic experiments with definitive alternative hypotheses, many of the experimental conclusions have been reached from the experience of numerous individual empirical investigations that have shown similar results. Most of our emphasis will be concentrated on reviewing and interrelating recent research results in areas of most significance to plant breeders, such as ( 1 ) kinds of genetic variability found, (2) effects of inbreeding and crossbreeding, ( 3 ) genotype-environmental interaction, and (4) selection methodology and response. It is not our purpose to present a detailed description of quantitative genetic theory. Cursory summaries (using nomenclature from Falconer, 1960) are included to provide background for readers untrained in quantitative genetics and to aid in understanding results from experimental research. 11.

Genetic Variability

A. BASICCONCEPTS Evaluations of inheritance mechanisms in quantitative genetics research depend on valid assessments of genotypic values. However, the genotypic value of an individual must be ascertained from measurements made on its phenotype. Phenotypic value then, is defined as the performance of a particular genotype in the environment in which it is grown. The two components of the phenotypic value (P)-genotypic value ( G ) and environmental deviation (E)-are usually represented in the equation for phenotypic value as: P = G E. A genotype is considered as the particular assemblage of genes possessed by an individual, and genotypic value for a given genotype is defined as the average of all possible phenotypic values, expressed as a deviation from the population mean. In other words, it is the average phenotypic value

+

279

QUANTITATIVE GENETICS

when genotypes are grown over all possible environments (the mean environmental deviation is zero) . Diagrammatically, the relationship between genotypes and genotypic values for a single locus may be represented as follows:

Genotypic value

-a

0

a

f a

In this representation, the origin, or zero point, is midway between the value of the two homozygotes, and B , represents the allele that increases the genotypic value. The value, d, of the heterozygote depends on the degree of dominance. In order to determine the contribution of this locus to the population mean, values of the genotypes must be weighted by their respective frequencies. Contributions of each individual locus must be combined in the computation of the population mean. Estimation of genetic effects and variances requires some type of family structure. To analyze the properties of a population composed of various family structures, it.is necessary to deal with concepts concerning transmission of value from parent to offspring. This cannot be done with only genotypic values, because parents pass on their genes, not their genotypes, to the next generation. Genotypes are created anew by uniting gametes in each generation. Therefore, a measure is required that allows the assignment of values associated with the genes carried by an individual and transmitted to its offspring, i.e., a measure is needed that reflects the average effects of genes. The average effect of a gene at a locus is defined as the average difference resulting from the substitution of one allele for the other. For example, if B , genes could be changed at random in a population to B , genes (see the diagram above), the resulting change in value is the average effect of the gene substitution. Because the average effect of a gene substitution depends on the gene frequency, it is a property of the population as well as of the gene. Although the average effects of genes cannot be measured directly, breeding values, which are weighted functions of average effects of genes, can be measured experimentally. If an individual is mated with a number of random individuals from the population, the breeding value is estimated as twice the mean deviation of its progeny from the population. This deviation must be doubled because only one-half the genes in its progeny are contributed by the individual being evaluated. In addition to the breeding value, another component of the genotypic value that must be considered is referred to as the dominance deviation. For a single locus, it is defined as the difference between the genotypic value and the breeding value. Although dominance deviations are within-

280

R. H. MOLL AND C . W. STUBER

locus interactions, their importance depends on gene frequencies in the population, so they are not simply measures of the degree of dominance. Most quantitative traits are conditioned by genes at many loci. However, the aggregate genotypic value may or may not be simply an additive combination of the genotypic values for individual loci. When this combination is not additive, genes are said to interact or to show epistasis. The phenotypic value has been expressed as P = G E. Therefore, the phenotypic variance, up2,may be expressed as:

+

up2 =

UG'

+ + U E ~

~ U G E

where uG2= genotypic variance, uK2= environmental variance, covariance of genotypic and environmental effects. The genotypic partition may be further divided as follows:

uGE

=

where uA2= additive genetic variance, uD2= dominance variance, u12= epistatic variance. The additive variance, which is the variance of breeding values, is the primary measure of the resemblance between relatives and is relevant to the effectiveness of selection. The objective of selection is to increase the frequency of favorable alleles in a population by substituting favorable genes for unfavorable genes. The effectiveness of gene substitution in changing the population mean is directly related to the average effects of genes. Average effects, in turn, are reflected in breeding values. The average effect of a gene substitution is actually a weighted regression coefficient (weighted by gene frequency). Therefore, additive genetic variance at a single locus may be considered as variance caused by the weighted linear regression of genotypic values on number of favorable alleles. Dominance variance, then, is variance attributed to deviations from regression. If the total variation over loci is larger than the summation of additive and dominance variances for individual loci, the differences will be variance caused by epistasis. Resemblance between relatives is reflected in similarities of expression of quantitative traits. The degree of resemblance expected provides the basis for estimating genetic variance components. Estimation procedures require systematic mating schemes that result in different types of relatives. Using appropriate experimental designs and statistical analyses, design variance components can then be calculated. Genetic interpretations of these design components are facilitated by translating them into covariances among relatives. Theoretical considerations of the genetic causes for resemblances between relatives permit the translation of these covariances into functions of genetic variance components. For example, with the assump-

28 1

QUANTITATIVE GENETICS

tions of no epistasis and an inbreeding level of zero in the population, the covariance between a parent and its offspring produced from mating at random in the population is 1/2 uA2,the covariance among half-sibs is % uA2,and the covariance among full-sibs is '/2 uA2 1/4 uBZ,These relationships change with different levels of inbreeding. A comprehensive summary of methods for estimating genetic variances was presented by Cockerham (1963). He also discussed many of the problems and limitations inherent in the definition and estimation of genetic variances when inbreeding is present. For most cross-pollinated species, genetic variances can be estimated with mating designs that do not depend on inbred relatives. However, in many self-pollinators, production of sufficient seed for replicated evaluation trials is nearly impossible without the use of inbred generations. The development by Stuber (1970) of methods using inbred relatives has facilitated the estimation of genetic variances in self-pollinated species. Diallels and modified diallels are often used to estimate genetic variances. In this type of design, general and specific combining ability components of variance are routinely estimated. The general combining ability component is primarily a function of additive genetic variance. However. if epistasis is present, it may include functions of additive types of epistasis. The specific combining ability component is primarily a function of dominance variance, but it may include all types of epistatic components. The relative proportions of genetic variances in the two combining ability components depends on the inbreeding level of the parents. Although diallels can be generated with parents chosen at random from some random mating reference population, this type of mating design generally has been used for specific sets of parents. With specific sets, the variance estimates must be interpreted as characteristic of only the set of parents involved and should not be used to characterize more broadly defined reference populations.

+

B. EXPERIMENTAL ESTIMATES IN

CROP SPECIES

Excellent reviews of genetic variance estimates available before 1962 for important crop species are given by Gardner (1963) and Matzinger (1963). Since 1962, a large number of experiments have been reported which cover essentially all major crop species and many different kinds of traits. Most of the data reported points to one general conclusion: genetic variability of important agronomic traits is predominantly additive genetic variance. Nonadditive variance also exists in nearly all species and for many important traits, but it is generally smaller than additive genetic variance.

282

R. H. MOLL AND

C . W. STUBER

Quantitative inheritance of various traits of maize has been studied most extensively, and studies reported in the literature deal with a wide range of maize populations. Estimates of genetic variability summarized by Gardner (1963) and Moll and Robinson (1967), as well as evidence in several more recent reports, indicate that additive genetic variance exceeds dominance variance in many different kinds of populations, including open-pollinated varieties, synthetics, variety hybrids, and variety composites. Furthermore, when sampling errors of variance estimates are considered, it appears that locally adapted open-pollinated varieties have genetic variances of essentially the same order of magnitude. Important differences in genetic variability seem to occur only between populations of distinctly different kinds. For example, composite populations formed by intermating a number of varieties tend to have greater variability than the parental varieties themselves. Composites of more genetically diverse populations have greater variances than composites of less diverse populations. Even so, estimates reported for such composites show additive variance to be larger than nonadditive variance for most traits. Although a number of extensive variability studies have been reported, epistatic variability has not been shown to be an important component of genetic variances of maize populations (Chi et al., 1969; Eberhart el al., 1966; Stuber et al., 1966). Studies of inbred line hybrids of various kinds, however, frequently reveal significant epistatic effects (Wright et al., 1971; Russell and Eberhart, 1970; Stuber and Moll, 1971; and others). Joint consideration of these two kinds of evidence leads to the conclusion that epistatic interactions must occur in maize populations, but they contribute very little variability beyond that accounted for by additive and dominance variances. The patterns seen in variance estimates in many other cross-pollinators is similar to that reported for maize; however, the data are much less extensive. Although sugarcane presents difficulties for quantitative genetic studies because of irregular meiosis and mating incompatibilities, estimates reported by Hogarth (1 97 1 ) indicate that additive genetic variance exceeds dominance variance, Variance estimates in alfalfa also provide evidence that additive genetic variance exceeds nonadditive variance, but there is some evidence for variance caused by trigenic, quadragenic, or epistatic effects (Dudley et al., 1969). Hill et al. (1972) also report a preponderance of variance caused by general combining ability (which wouId be largely additive genetic variance) and, although variance caused by specific combining ability was detected, it was relatively small in magnitude. Significant general combining ability has also been found in several forage grasses, including Bromus inermis Leyss. (Mishra and Drolsom, 1972; Dunn and Wright, 1970), Dactylis glumerata L. (Kalton and Leffel, 1955), and Lolium perenne L. (Hayward and Lawrence, 1972).

QUANTITATIVE GENETJCS

283

A series of studies of genetic variability in flue-cured tobacco (a selfpollinator) have also shown that additive genetic variance is predominant for a number of traits in several different populations. There appears to be some epistatic variability in certain populations for certain traits, especially plant height and leaf measurements. Dominance variance tends to be small and usually nonsignificant (Matzinger, 1968; Matzinger et al., 1960, 1966, 1971). Changes in means after several generations of random mating were interpreted as evidence for epistasis. Segregation of epistatic gene complexes may have caused a breakdown of internal balance by forced intercrossing in a naturally self-pollinated species (Humphrey et al., 1969). Genetic variability for a number of traits of soybeans has also been shown to be predominantly additive, but nonadditive variability is significant for many of the traits (Brim and Cockerham, 1961; Hanson et al., 1967; Weber et al., 1970). A series of diallel studies involving a number of pulse crops, such as mungbean, cowpeas, blackgram, and lentil, have shown significant general combining ability for a number of traits. Specific combining ability appears to be important for certain traits, but it is usually less important than general combining ability (Singh and Jain, 1971, 1972; K. B. Singh and Singh, 1971; T. P. Singh and Singh, 1972). Diallel studies in small grains, particularly in wheat and oats, and in sorghum have also indicated that general combining ability for yield and related traits is more important than specific combining ability, even though specific combining ability is statistically significant in some instances (Collins and Pickett, 1972; Lee and Kaltsikes, 1972; Ohm and Patterson, 1973a,b; Gyawali et al., 1968; Walton, 1972; Widner and Lebsock, 1973; and others). A genetic study of pearl millet by Bains (1971), which used the analysis proposed by Kearsey and Jinks (1968), found that additive genetic variance was important for three agronomic traits, but epistasis also appeared to be prevalent. In diallel studies, however, Gupta and Singh (1971) found the additive component to be nonsignificant for grain yield and ear number in a study of eight diverse lines of pearl millet. General combining ability was either smaller than specific combining ability or nearly the same size in four traits studied by Ahluwalia et al. ( 1962). Although many diallel studies in cotton indicate that additive variance is more important than nonadditive types, there are several examples of deviation from this pattern. For example, Gupta and Singh (1970) reported dominance variance to be larger than additive variance in several seed and fiber traits. Their study involved eight diverse strains of upland cotton. Baker and Verhalen (1973) reported similar results in a study of 10 selected upland cotton lines. They also presented an extensive literature review in which they documented several instances in which additive variances predominated and several in which dominance predominated.

R. H. M O L L AND C. W. STUBER

284

111.

Inbreeding Depression and Heterosis

Inbreeding results from matings between related individuals. The degree of inbreeding is measured by the inbreeding coefficient, which is the probability that two genes at a locus (in a diploid) are identical by descent. The effect of inbreeding upon the population mean can be shown to be a function of the gene frequency, dominance effects, and the coefficient of inbreeding. If dominance is directional, i.e., the majority of loci show dominance for the favorable allele, inbreeding will result in a decrease in the mean proportional to the inbreeding coefficient. Heterosis, which results from crossing unrelated strains, is the reverse of inbreeding depression. It also depends upon directional dominance for its expression. Therefore, inbreeding depression and heterosis both refer to differences in mean performance directly related to differences in heterozygosity, and in diploids the level of heterozygosity is directly related to the coefficient of inbreeding. Inbreeding depression is the decline in trait expression with decreased heterozygosity, and heterosis is the enhancement of trait expression with increased heterozygosity. A.

INBREEDING DEPRESSION

The relationship between the mean expression and the coefficient of inbreeding tends to be linear for most traits of maize, and yield of grain shows a steeper rate of depression than other agronomic traits (Sing et al., 1967). Papers by Levings (1964) and Busbice (1969) show that in autotetraploids the loss of heterozygosity is not directly related to the inbreeding coefficient, and is much slower than in diploids. Levings et al. ( 1967) reported that the relationship between heterozygosity and performance in autotetraploid maize was linear for three quantitative traits. The decrease in ear weight relative to the inbreeding coefficient was slightly less than half as rapid in the autotetraploid as it was for yield of ear corn in diploid maize studied by Sing et al. ( 1967). Comparison of these results agrees qualitatively, at least, with theoretical expectations. A direct comparison of inbreeding effects in crested wheatgrass [Agropyron cristatum (L. ) Gaertn.] suggests that inbreeding depression is greater than expected at higher levels of ploidy (Dewey, 1966). Forage yields of self-pollinated progeny for diploids, tetraploids,, and hexaploids were 35.9, 50.4, and 67.4%, respectively, less than yields of noninbred progenies. Also contrary to theoretical expectations, diploid and tetraploid sugar beets (Beta vulgaris L.) were reported to show the same rate of inbreeding depression (Hecker, 1972). In a study involving diploid and

QUANTITATIVE GENETICS

285

autotetraploid rye (Secale cereale L.), Lundqvist (1969) found less inbreeding depression in the autotetraploids than in the diploids, but not as much less as expected from genetic theory. Inbreeding in alfalfa is accompanied by impaired reproductive fertility, as well as by loss of vigor and productivity. The relationship between inbreeding coefficient and inbreeding depression was linear for several traits, and most drastic for yield and spring vigor. The inbreeding rates for yield and spring vigor approach a theoretical curve for tetragenic inheritance, whereas plant height approximates the curve for duplex inheritance (Aycock and Wilsie, 1968).

B. HETEROSIS 1 . Genetic Diversify and Heterosis

Heterosis, in quantitative genetic terminology, is usually measured as the superiority of a hybrid over the average of its parents, and has been reported for a wide range of crop species, which include both self- and cross-pollinators.The expression of heterosis is greatly influenced by the magnitude of genetic differences for some traits, but not for others. For example, several recent reports of diallel crosses among strains of wheat show greater heterosis associated with crosses of more distantly related parents (Fonesca and Patterson, 1968; Widner and Lebsock, 1973; Sun et d., 1972). On the other hand, Gyawali et al. (1968) found no evidence for an increase in heterosis associated with interclass differences between soft red and hard red parents. A study of crosses of nine strains of tall fescue (Festuca arundinaceae Schreb) suggests that heterosis increased with genetic divergence with respect to morphological traits and flowering time, and also with respect to geographical origin of the parents (Moutray and Frakes, 1973). Comparisons of inter- and intraspecific hybrids of alfalfa and cotton show greater heterosis associated with greater diversity (Sriwatanapongse and Wilsie, 1968; Marani, 1963, 1968). Heterosis in lint yield of cotton tended to be associated with a greater number of bolls rather than boll size, especially in interspecific hybrids, in which boll size was often less than the average of the parents. A relationship of heterosis to diversity is also reported for several traits of cotton relating to plant growth, such as plant height, leaf area index, and dry matter accumulation (Marani and Avieli, 1973). Studies involving interracial crosses of maize and interspecific crosses of tobacco indicate that the relationship between diversity and heterosis may not be linear over very wide ranges of diversity. There is considerable evidence that increased genetic differences between inbred lines of maize

286

R. H. MOLL AND C. W. STUBER

result in greater heterosis in their hybrids, More recent studies have involved variety crosses of parents of different geographical origin, and show that the association of increased heterosis with increased diversity extends over a considerable range of maize types (Paterniani and Lonnquist, 1963; Moll et al., 1962). However, as the range of diversity was expanded further, crosses of the most distantly related populations showed less heterosis than crosses of populations assumed to be less distantly related. This would suggest that maximum heterosis occurs at an optimal or intermediate level of genetic diversity (Moll et al., 1965). A similar pattern of heterosis and diversity has been observed in tobacco. Hybrids between flue-cured varieties (Nicotiana tabacum L. ) and primitive strains of Central and South America (the assumed center of origin of N . tabucum) gave heterosis values similar to those observed in crosses of flue-cured and Oriental varieties (Vandenberg and Matzinger, 1970; Matzinger and Wernsman, 1968). The greatest heterosis was found in crosses of flue-cured varieties to progenitor species, N . otophora and N . tornentosiforrnis Goodsp. Crosses of flue-cured varieties with more distantly related species resulted in less heterosis, which also suggested an optimum degree of diversity for maximum heterosis (Matzinger and Wernsman, 1967). 2. Genetic Causes of Heferosis There are three possible genetic causes of heterosis: partial to complete dominance, overdominance, and epistasis. From the point of view of a plant breeder, a basic issue is whether the best genotypes are homozygotes or heterozygotes. If overdominance is important, the best genotype is a heterozygote. With partial to complete dominance, the best genotype would be a homozygote, and, rather than capitalize on heterosis directly, it might be desirable to isolate transgressive segregates. Evidence for heterozygote superiority in barley was reported by Jain and Allard (1960). A bulk population, which had originated from intercrosses of 3 1 barley varieties and was heterozygous for several recognizable mutants, was studied at intervals that encompassed 18 generations. The proportion of heterozygotes at the marked loci did not decrease at the rate expected, apparently because of the selective advantage of heterozygotes. Overdominance was suggested as a possible explanation. However, no evidence for overdominance was found for four lethal chlorophyll mutants of barley (Rasmusson and Byrne, 1972). The rate of elimination of the recessive lethals for three of the loci was compatible with the rate expected if there were complete dominance for fitness. The fourth allele showed deviations from both complete dominance and overdominance models, and behaved somewhat like a frequency-dependent phenomenon. Apparent heterozygote superiority for forage yield of cocksfoot (Dac-

QUANTITATIVE GENETICS

287

tylis glornerata) was shown to result from genotype x environmental interactions (Breese, 1969; Knight, 1971). Through the application of the regression method of Finlay and Wilkinson (1963), the linear adjustments for environmental differences resulted in estimates of dominance in the partial dominance range. The relative importance of dominant versus overdominant gene action has been studied most intensively in maize. Extensive data reviewed by Gardner ( 1963), Moll et al. ( 1964), and Moll and Robinson ( 1967) provide evidence that if overdominance occurs in maize, it is either infrequent in occurrence or small in magnitude. The evidence clearly shows, however, that linkage between loci with partial to complete dominance does result in heterozygous effects that mimic effects of overdominance for several generations after a cross. The issue of whether overdominance occurs to some extent in maize is not entirely resolved, and the possibility of the existence of effects like overdominance must be recognized. For example, comparisons involving changes in heterosis after recurrent selection for hybrid performance are similar to changes expected if overdominant gene action were increased in importance by selection (Moll and Stuber, 197 1 ) . Epistasis, particularly, epistasis that involves dominance effects, may also contribute to heterosis, as was shown for certain traits in interspecific crosses of cotton (Marani, 1968). Epistasis has also been shown to occur in crosses of certain inbred lines of maize (Sprague and Thomas, 1967; Eberhart and Hallauer, 1968; Stuber and Moll, 1971; Stuber et al., 1973). On the other hand, epistasis does not appear to be a major component of genetic variability in varieties or variety hybrids (Castro et al., 1968; Eberhart and Gardner, 1966; Eberhart et al., 1966; Chi et al., 1969; Stuber et al., 1966). The curvilinear relationship between heterosis and genetic diversity noted previously might be a result of epistasis. However, Cress (1966) has pointed out that multiple alleles would result in negative dominance effects among some of the combinations, and could account for the observed results in the absence of epistasis.

IV.

Genotype-Environmental Interactions

Valid interpretations of mechanisms of inheritance as well as predictions of performance in breeding programs depend on accurate assessments of genotypic values (Section 11, A ) . These assessments must be made from data on phenotypes that reflect both nongenetic and genetic influences on plant development. Unfortunately for the geneticist and plant breeder, the genetic effects are not independent of the nongenetic environmental effects. For example, the relative rankings of genotypes often differ in different

288

R. H. MOLL AND C. W. STUBER

environments. This interplay of genetic and nongenetic effects, genotypeenvironmental interaction, reduces the correlation between genotype and phenotype, which in turn reduces confidence in inferences from experimental data relevant to both plant improvement and inheritance mechanisms. As Allard and Bradshaw (1964) indicated, the nature of genotype-environmental interactions is extremely complex. In their attempt to classify types of interactions, they showed that for only 10 genotypes and 10 environments, there are possible types of interactions. This number is larger than the total number of plants that have ever existed on the earth. Therefore, consideration of genotypes and environments separately may provide the only reasonable means of gaining an insight into the nature and significance of the interactions. Environmental variations can be classified into two types, predictable and unpredictable (Allard and Bradshaw, 1964; Allard and Hansche, 1964). Predictable variations include the more permanent features of environments, such as climate and soil type, as well as cyclic fluctuations such as day length. In addition, factors that can be fixed at will (e.g., planting date, plant density, fertility levels, and harvest methods) are considered in this category. Unpredictable variations include fluctuations in weather such as distribution and amount of rainfall, temperature changes, and insect or disease infestations. Although distinctions between the two categories of variation may not always be clear, they have distinctly different impacts on breeding programs, both on the operational procedures of selection and on the testing phases. Performances of genotypes (varieties) may or may not change with environmental fluctuations, even when there are large differences in environmental factors. Although workers disagree on their concepts of stability (to be discussed later), it is generally agreed that the more stable genotypes can somehow adjust their phenotypic responses to provide some measure of uniformity in spite of environmental fluctuations, Allard and Bradshaw ( 1964) and Allard and Hansche (1964) equated stability with the term “well-buffered.” They defined two types of buffering, individual bdff ering and populational buffering. A homogeneous variety must depend largely on individual buffering to achieve stability over a range of environments, whereas a heterogeneous variety may use both individual and populational buffering for this purpose. The significance of genotype-environmental interactions to the plant breeder depends on his objectives. If he desires varieties that perform well over a broad spectrum of environments, then his program is favored by small genotype-environmental interactions and/or well-buffered varieties. However, if he desires varieties that are adapted to very specific environ-

289

QUANTITATIVE GENETICS

ments that can be predicted or specified in advance, then his program may benefit by large interactions, and buffering may be of little consequence. A major objective of the quantitative geneticist is the ascertainment of the magnitudes of genetic variances as the basis for predicting genetic improvement in selection programs. In this situation, genotype-environmental interactions may significantly affect the reliability of the variance estimates. Such interactions are the source of part of the random errors of variance estimates and often introduce an upward bias. Discrepancies between realized and expected response to selection wiII undoubtedly occur if expectations of progress are calculated from biased estimates of genetic variances. A. MEASUREMENT OF GENOTYPE-ENVIRONMENTAL INTERACTIONS Several procedures are available for characterizing behavior of varieties, lines, or genotypes in varying environmental conditions. One of the most commonly used methods is the use of replicated performance tests over a series of environments. These tests are often analyzed as follows:

Source of variation

d.f.

Environments Genotypes Genotype x environment Error

e-1 g--1 (g - l ) ( e - 1) ge(r - 1)

Mean square

Expectation of mean square

MS, MSz

u2

MSI

u2

u2

+ ruse2+ reup2

+ rugc*

The above analysis provides the mean squares for a simple F test (MSJMS, ) for evaluating the overall significant of genotype-environmental interactions. Also, the component of variation attributable to these interactions, uge2,can be estimated as (MS2-MS,)/r. This may be useful when comparisons of vse2 with genotypic variances, us2, are desired. If the experiment is designed so that the genotypic variance can be partitioned into separate components (such as additive, dominance, and possibly epistatic variances), it is often desirable to partition the genotypic-environmental source of variation so that specific tests of each type of variance can be made. Another measure of genotype-environmental interactions that has some appeal to empirical plant breeders involves the correlation of performances of an array of genotypes in one environment with their performances in other environments (Stuber et al., 1973). Large positive values for this

290

R. H. MOLL AND C. W. STUBER

type of correlation coefficient indicate little effect of genotype-environmental interactions, whereas the converse is true when evaluating the magnitudes of variance components attributed to such interactions. Although genotype-environmental interaction effects appear to vary somewhat in different species, no general pattern has been noted that could be associated with specific types of mating systems or with specific types of traits. Results from numerous experiments with maize indicate that estimates of genotype-environmental interaction variance components are significant for most traits evaluated and are relatively large when compared with estimates of genetic variances (Gardner, 1963; Moll and Robinson, 1967; Stuber and Moll, 1971). Both additive and nonadditive effects show significant interactions with environments. Second-order interactions (i.e., genotype-year-location) tend to be much greater than either genotypeyear or genotype-location effects. Both genotype-environmental interaction effects and epistatic genetic effects may contribute to biases when standard methods are used for predicting three-way and double-cross hybrids from single-cross performances. Studies with maize by Otsuka et al. (1972) and Stuber et al. (1973) indicated that genotype-environmental interaction effects produce greater biases than epistatic effects when predictions are made from data obtained in a single environment. When several environments are used for prediction values, biases from the two types of effects are nearly equal. Data from yield traits in three autogamous species-cotton, soybeans, and tobacco-showed highly significant variances attributable to secondorder interactions (Matzinger, 1963). Except for tobacco, these variance components tend to be large in comparison with total genetic variances. More recent studies in cotton (Lee et al., 1967; Bridge et al., 1969; Baker and Verhalen, 1973) produced similar results, with the additive component of variance for lint yield showing particularly large interactions with environments. Cotton quality traits, in general, show relatively little genotype-environrnental interaction. Baker ( 1969) found highly significant genotype-environmental effects for grain yield in hard spring wheat, with the second-order interaction component of variance nearly as large as the genotypic component of variance. In a durum wheat study, Widner and Lebsock (1973) reported that genotypic-environmental interaction mean squares were significant for Fl’s and parents, but not for Fz‘s in their grain yield data.

B. EVALUATION OF STABILITY Although, in general, genetic effects are not independent of environmental effects, a number of authors (Yates and Cochran, 1938; Finlay and

291

QUANTITATIVE GENETICS

Wilkinson, 1963; Rowe and Andrew, 1964; Eberhart and Russell, 1966; Perkins and Jinks, 1968; Johnson et d.,1968; Breese, 1969; Baker, 1969) have observed that the relation between the performance of different genotypes in various environments and some measure of these environments is often linear, or nearly so. From these observations, Freeman and Perkins (1971) concluded that there is strong evidence indicating a genuine underlying linear relation between performances of specific genotypes and environmental conditions, even though this relation does not always account for all of the interaction observed. Because of this linear relationship, several authors have used regression techniques to characterize responses of genotypes in varying environmental conditions. In particular, regression analyses have been used to provide measures of phenotypic stability. Many of the regression analyses used for this purpose do not entirely satisfy rigorous statistical requirements. Even so, the regressions computed have been shown to be useful predictors of stability, and would appear to be particularly meaningful in practical plant improvement work. One of the basic statistical objections to many of the papers cited above is the improper choice of sums of squares and degrees of freedom from which to subtract the regression components. In fact, one author, Baker (1969), divided the genotype-environmental sum of squares, with (g - 1 ) (e - 1) degrees of freedom, into a separate partition associated with each genotype. The total degrees of freedom for these partitions is g(e - 1 ) . This, of course, is statistically invalid, because any sum of squares has a unique number of degrees of freedom irrespective of the partitioning scheme used. A method for partitioning sums of squares, as presented by Freeman and Perkins (1971 ), avoids the difficulties cited above. Their analysis, in which all the terms are orthogonal, and in which comparisons are possible with F tests, is as follows:

Source of variation

d.f.

Genotypes (C) Environments ( E ) Combined regression Residual Interaction (C X E ) Heterogeneity of regressions Residual Error (between replicates)

g-1

1 e-2 g-1

(9

- I)(e - 2) ge(r - 1)

292

R. H. MOLL AND

C . W. STUBER

In this analysis, the “combined regression” term partitions the variation associated with fitting a single line over all genotypes. The term “heterogeneity of regressions” partitions the sum of squares associated with the differences of the individual regression lines for the different genotypes from the combined regression line. Tai ( 1971 ) partitioned the ‘‘interaction” term in a similar manner. Eberhart and Russell (1966) used an analysis which (although it was developed earlier) might be considered as a modification of the FreemanPerkins analysis. It is as follows: Source of variation

d.f.

Genotypes (G) Environments ( E ) Interaction (C X E ) Environments (linear) G X E (linear) Pooled deviations Genotype 1 Genotype 2

9 - 1

Genotype g Pooled error

s(e - 1)

1 g-1 g(e - 2) e - 2

e-e

e-52

49 -

- 1)

The above analysis differs from the Freeman-Perkins analysis by considering the total variation among genotypes as a single component by combiing “environments” and “interaction” sums of squares. The two “residual” sums of squares from the Freeman-Perkins analysis form the “pooled deviations” partition in this analysis, which can then be further subdivided to provide individual genotype (or variety) partitions. Another basic objection to many of the regression analyses cited is the choice of measurement of environmental effects on which the regression is made. Some type of environmental measurement independent of the experimental organism would be highly desirable. It would be even better if environmental values could be measured without error. These requirements cannot be met, and the best measure of the combined environmental effects is probably provided by the organism itself., Several of the authors cited earlier have used the mean of all the genotypes (varieties) grown in a specific environment as the value assigned to the environment (i.e., variety performance is regressed on the mean of all varieties grown at a specific environmental site). This does not pro-

QUANTITATIVE GENETICS

293

vide an independent measure of environmental effects and, therefore, does not satisfy the requirements of a regression analysis as rigorously as one would like. However, it appears to provide sufficiently reliable estimates to be useful. Some independent measures that are possible include: (1) division of the replicates of each genotype into two groups, using one group to measure the environment and the other to evaluate interactions, or (2) use of one or more genotypes as standards to assess the environment. Although these measures provide the desired independence between environmental and genetic effects, they require additional experimental costs or the discarding of some data from the interaction analyses, and are inefficient with regard to minimizing sampling errors. Perkins and Jinks (1973) compared the use of the usual dependent assessment of environments with three independent measures in a study of 82 lines of Nicotiuna rusticu L. In their analyses of significance of heterogeneity of regressions and the ranking of inbred lines based on their linear regression coefficients, they found that it made little difference whether they used dependent or independent environmental measures. In fact, the increased size of the sampling variances, because fewer experimental units were available for the independent environmental assessments, was probably more serious than the lack of independence that resulted when all the experimental material was used for the environmental assessment. Although the regression analyses appear to be very useful in providing stability measures, there is not complete agreement on the best definition for the term stability. In a study of 277 barley varieties, Finlay and Wilkinson ( 1 943) (basing their analyses on methods developed by Yates and Cochran, 1938) regressed variety mean yield on site mean yield, using a logarithmic scale. They indicated that a regression coefficient of: (1) unity indicates average stability, (2) greater than unity indicates below average stability, and (3) less than unity indicates above average stability. Absolute phenotypic stability would be expressed by a coefficient of zero. Although their definition implied that a stable cultivar performs relatively well in poor environments and relatively poorly in favorable environments, they defined an ideal variety as one with maximum yield potential in the most favorable environment and with maximum phenotypic stability. Using these definitions, their barley data showed that the varieties with high phenotypic stability all had low mean yields and were unable to exploit highly favorable environments. They then concluded that the breeder must compromise in his search for an ideal variety. Eberhart and Russell (1966) improved the regression technique for evaluating stability by considering two empirical parameters, the slope of the regression line and the deviations from the regression line. In their pij, Yij is the ith variety mean at the jth enmodel, Y i j = pi piZi

+

+

294

R. H. MOLL AND C. W. STUBER

vironment, pi is the ith variety mean over all environments, pi is the regression coefficient that measures the response of the ithvariety to varying environments, &! is the deviation from regression, and If is the environmental index. They then defined a stable variety as one with a regression coefficient of unity ( b = 1.0) and with a minimum deviation from the regression line ( ~ d = , ~ 0). Using their definitions, a breeder would usually desire a variety with a high mean and one that meets the above requirements for stability. Data on corn inbreds and hybrids presented by Eberhart and Russell ( 1966) showed considerable variation for stability parameters. However, the data suggested that both parameters, the regression coefficient and the parameter measuring deviations from regression, are of prime importance for varietal evaluations. Regression techniques were used by Breese (1969) in a study of five populations of Dactylis glornerata. He also concluded that a stability concept must include a measure of deviations from regression as proposed by Eberhart and Russell (1966). Comparisons of performances of 12 varieties of hard red winter wheat (Triticum aestivum L.) grown in regional performance nurseries produced regression coefficients that differed little from unity (Johnson et al., 1968). Although they did not compute the deviations from regression, comparisons with a standard wheat variety, KHARKoF, also indicated that the deviations from regression should be considered when making varietal comparisons relating to stability. In a study of sorghum yield stability parameters, Jowett (1972) compared evaluation methods of Eberhart and Russell (1966), Finlay and Wilkinson (1963), and Wricke (1960), using empirical data. Wricke’s method, based on a single parameter called ecovalence, was determined to be the least informative of the three methods. Jowett concluded that the Eberhart-Russell method, which uses an arithmetic scale, was probably preferable because it is more explicit than the Finlay-Wilkinson procecdure, which uses a logarithmic scale. However, the logarithmic scale may be preferable if varieties differ markedly. In the empirical analyses, Jowett (1972) found that single cross and three-way cross sorghum hybrids were more stable over environments than the inbred varieties. In an analysis of regional potato trials, Tai ( 1971) considered stability in a slightly different context and outlined a procedure that removes the error deviate from the stability parameters defined by Eberhart and Russell ( 1966). This provided genotypic stability statistics which Tai compared with the phenotypic statistics of Eberhart and Russell. He concluded that with only a small number of varieties tested over a small number of environments, the phenotypic estimates may differ substantially from the genotypic estimates. However, in a potato experiment with eight entries and

QUANTITATIVE GENETICS

29 5

six environments, the estimates of genotypic statistics differed very little from phenotypic estimates. Tai also found that the highest yielding experimental lines were unstable, and those with average stability yielded about the same as the check varieties. Until recently, hopes that genotype-environmental interactions could be successfully accommodated in the study and manipulation of quantitative traits have not been very optimistic. Sprague (1966) showed considerable pessimism when he indicated that the possibility for reducing such interactions under field conditions seemed questionable. As Breese (1969) suggested, perhaps the attitude that we should seek to minimize genotype-environmental interactions has inhibited our approach to this problem. From the studies discussed above, it now seems apparent that the combined effects of genotype and environment do not behave in a disorderly fashion, but might be envisioned as reasonably predictable responses to some type of regulatory system. In the past, workers also may have been deterred in the study of genotype-environmental concepts because they have been too concerned with the identification of individual components of the environment. However, Breese (1 969) reminded us that the phenotype is the product of the genotype and its environment. Therefore, it is just as appropriate to qualify an environment by its mean expression over a range of genotypes as it is to measure a genotype by its mean expression over a range of environments. We certainly cannot specify all the underlying biochemical and physiological processes that characterize a genotype. Therefore, the fact that we are unable to specify all the factors that contribute to an environmental expression should not be a deterrent to our study of quantitative traits and their relationships to environmental conditions. In fact, the recent evidence, strongly indicating a genuine underlying linear relationship between performances of specific genotypes and environmental conditions, should provide new opportunities for quantitative genetic research, as well as provide better tools for the plant breeder to predict response over varying environments. V.

Response to Selection

The genetic potential of a plant population can be considered in two different ways: (1) the mean performance of the population itself or (2) the performance of the population in hybrid combination with other germplasm. Selection procedures have been devised for the direct improvement of each kind of genetic potential. When the goal of the breeding program is a superior variety or pure line, it would be logical to choose among selection procedures designed to improve population performance itself.

R. H. MOLL AND C. W. STUBER

296

However, improvement of the popuIation itself may also result in improvement of its performance in hybrid combinations. Therefore, when the goal is a superior hybrid, the choice between the two kinds of procedures is not incisive. A.

GENETIC VARIANCES

AND

EXPECTED RESPONSE

A major role of quantitative genetics is to provide a genetic basis for the development of effective and efficient selection schemes for particular objectives. Heritability, which is the ratio of additive genetic variance to total phenotypic variance, measures the degree of association between phenotypic values and breeding values (Section I1,A). Expected gain from selection among individuals can be shown to be A = (2, -

Z)UA~/U~~,

in which E8 is the mean of the individuals selected, and 2 is the sample mean of the population. The selection differential, (X, - R), is often represented in standard measure as k, so that ku, = (X, - 2). By substitution, the expression for expected selection response becomes A =

kuA'/up.

Selection among families rather than among individuals requires modification of the expression. For example, the degree of association between full-sib family means and the average breeding values of members of each family would be (1/2)uA2/~p,Z,in which m~plzis the variance among family means. Therefore, the expected response to selection among full-sib families is AF9 = kmA2/2up,. Expressions for expected response to various kinds of selection have been tabulated by Sprague (1966), Empig ei al. (1972), and Brim and Stuber ( 1973). Recent reports of experimental selection programs provide empirical evidence that bears on the effectiveness of various selection procedures and comparisons between rate of response observed with the rate predicted from theory. In addition, there are a few reports that provide direct experimental comparisons between alternative selection procedures. B.

EXPERIMENTAL EVALUATION OF SELECTION PROCEDURES 1 . lntrapopulation Selection Response

Mass selection, which simply involves the propagation of the most desirable individuals, based on their phenotype, is the least complex and least expensive procedure for improving populations. A disadvantage of this ap-

QUANTITATIVE GENETICS

297

proach is that the effect of environmental variation upon phenotypic expression may overwhelm genotypic differences. Earlier in this century, mass selection techniques were considered to be relatively ineffective for traits of low heritability such as grain yield of maize, and their use was restricted largely to less complex traits with higher heritabilities. Gardner (1961) introduced an effective means of reducing the effect of environmental variation by selection among plants within small subdivisions of the experimental field. In this way phenotypes to be compared for selection are grown in close proximity and are affected by only a small portion of the variation in soil and microclimate over the entire field. These refinements have consistently improved corn yields for as many as 13 generations in the variety HAYS GOLDEN (Harris er al., 1972). The rate of progress has been essentially linear, at a rate slightly less than 3% per generation, and is consistent with expectations based on quantitative genetic theory and estimates of variance components. Several other workers who have used similar techniques have also reported satisfactory progress. Johnson ( 1963) reported spectacular gains in two Mexican corn varieties. Matzinger and Wernsman (1968 demonstrated the effectiveness of mass selection for yield in a self-pollinator, Nicotiana tabacum, when maximum genetic recombination was assured by controlled random mating between selection cycles. The responses they observed were linear over four selection cycles and were in excellent agreement with theoretical expectations. Mass selection techniques have also been used to improve grain yields indirectly by selecting more highly heritable traits that are associated with yield (Chandhanamutta and Frey, 1973; Lonnquist, 1967). Hallauer and Sears (1969) have reported two instances in which mass selection for yield of corn failed to give significant improvement. However, they eliminated lodged plants from consideration for selection, which markedly reduced selection intensity for yield, Darrah et d. (1972) reported preliminary findings of an extensive investigation, which will provide comparisons among several alternative procedures, including mass selection. The authors suspect that the high plant densities used in their studies may have contributed to the disappointing progress observed so far under mass selection. Other intrapopulation selection procedures involve family structures, with selection based on average phenotype of replicated progeny. Although these methods result in more reliable measures of genotypic differences, the controlled matings that are required and the replication feature add considerable complexity and expense in comparison with mass selection. Comparisons of rates of selection response between mass selection and family selection techniques should consider the inherent differences in selection

298

R. H. MOLL AND C . W. STUBER

intensity and the number of generations required per selection cycle. Because more individuals than families can be evaluated in a limited space, a greater selection intensity is possible with mass selection than with more complex methods. Selection can occur in each generation of mass selection, but the generations of matings involved in family selection methods usually are not subject to selection for the primary trait, and selection can occur only in alternate generations, Theoretical comparisons that take these factors into account suggest that response to mass selection may be greater than response to family selection in terms of gain per generation, even for a trait of low heritability, such as yield of maize (Moll and Robinson, 1967). However, family selection is expected to show greater response in terms of gain per cycle. Relative rates of gain per year then would depend on the feasibility of making controlled matings in an off-season nursery. Full-sib family selection, half-sib progeny test, and S1 progeny test are intrapopulation selection procedures that have been investigated experimentally, and data are available that bear on their relative effectiveness in plant improvement, Full-sib family selection has been found effective in four populations of corn (Moll and Robinson, 1967; Moll and Stuber, 1971). Selection responses over as many as seven cycles were essentially linear and generally in good agreement with theoretical expectations. In a comparison between full-sib family selection and half-sib progeny test, da Silva and Lonnquist (1968) reported good agreement between observed and expected response. There was a small difference in response in favor of full-sib family selection, which was attributed to a difference in selection intensity. Half-sib progeny result from a testcross involving an individual mated with a sample of tester plants, The kind of population from which the testers are drawn affects the kind of selection response expected (Empig et al., 1972). Allison and Curnow (1966) concluded from theoretical considerations that the best choice of tester would be the parental population itself. Lonnquist ( 1968) found that selection based on testcross progeny, using the population itself as tester, was superior to selections based on S, progeny tests and testcrosses using an unrelated tester. Selection studies reported by Horner et al. (1969, 1973) involved comparisons of testcrosses with two kinds of testers, population per se and an inbred line, and S, progeny tests. Selection based on the two kinds of testcrosses led to similar quadratic changes, but caused no overall increase in the yield of the population itself. Selection based on S, progeny tests also resulted in a quadratic response and an overall decrease in the yield of population. After adjustment for inbreeding effects, responses were essentially linear and positive, and showed a small advantage in favor of the inbred tester. Other workers who have compared S, progeny selection with half-sib

QUANTITATIVE GENETICS

299

testcrosses in maize have used unrelated double crosses as testers. In these cases, the S, progeny test has been found to be superior to, or at least equal to, half-sib testcrosses for the improvement of the population per se (Burton et al., 1971; Carangal et al., 1971; Duclos and Crane, 1968). Slprogeny test has also been shown to be effective in reducing inbreeding depression in the S, generation (Genter and Alexander, 1966; Genter, 1971). 2. Interpopulation Selection Response

Recurrent selection for specific combining ability (Hull, 1945) and reciprocal recurrent selection (Comstock et al., 1949) are designed to apply direct selection pressure for crossbred performance. However, crossbred performance is also expected to improve indirectly with intrapopulation selection. Although crossbred response might be somewhat less with intrapopulation selection than with interpopulation selection, there are advantages of the former, such as greater flexibility and economy. Therefore, comparisons of rates of progress between intra- and interpopulation selection methods are particularly pertinent when the breeding objective is a superior hybrid. Recurrent selection for specific combining ability is similar to half-sib testcross selection, except that the tester is an inbred line suitable as a parent of a potential hybrid. A study reported by Russell et al. (1973) compared the response of two corn belt populations over five cycles of selection for combining ability with the same inbred tester (B14). The rate of gain of the testcross was much greater and in better agreement with expectations for the variety ALPH than for (WF9 X B7)F2. Rates of improvement estimated for the populations per se, and in test crosses with an unrelated synthetic, were not significantly different from rates observed in testcrosses with B14. The crossbred of the two populations, ALPH X (WF9 x B7)F2, showed the greatest improvement, which was nearly equal to the sum of the gains in the two testcrosses. Horner et al. (1973) has also demonstrated significant response to selection involving an inbred tester. They compared results of three selection methods over five selection cycles for yield of corn. The methods involved testcrosses with an inbred tester, testcrosses with the parental population as tester, and S2 progeny tests. The responses of all three methods evaluated in both kinds of testcrosses were linear, and the greatest response was observed for the inbred-tester method. The lowest rate of improvement resulted from S2 progeny selection. Reciprocal recurrent selection as originally proposed (Comstock et al., 1949) involves half-sib testcrosses, in which one population serves as the tester for evaluating parents in another population. The two populations

300

R. H. MOLL AND C. W. STUBER

are interchanged as testers, so that two sets of test crosses are evaluated, and selection is applied to each population for combining ability with the other. Comparisons between reciprocal recurrent selection for interpopulation improvement and full-sib family selection for intrapopulation improvement were reported by Moll and Stuber (1971). The study included two openpollinated varieties of corn, the variety hybrid, and a composite population derived by random mating in the variety hybrid. Full-sib family selection in the composite population showed much less response to selection for yield of grain than that expected from genetic theory, and was no better than the response in the other kinds of populations. There was very little difference in response of the variety hybrid to the two methods of selection. Much greater differences in response were found in the varietal populations themselves; full-sib family selection was more effective in increasing population yields than reciprocal recurrent selection. As a result, the variety hybrid showed a marked difference in heterosis after selection by the two methods. Heterosis (F,-midparent) after full-sib family selection was slightly less than in the original variety hybrid, whereas after reciprocal recurrent selection, heterosis had increased by more than 50%. Since differences in heterosis reflect differences in genetic effects, it appears that the two methods resulted in improvements (in the variety hybrid) of similar magnitude, but through different kinds of genetic changes in the two populations. Eberhart et al. (1973) also reported a marked increase in heterosis in a hybrid between two synthetic corn varieties after reciprocal recurrent selection. Changes in each of the two parental populations contributed about equally to the improvement observed in the hybrid between them, even though the populations themselves showed little or no change in mean performance. One of the parental synthetics was also subjected to selection by half-sib testcross procedures involving a double cross tester. The population that resulted from half-sib testcross selection was crossed to the other population after reciprocal recurrent selection, and the resulting hybrid showed almost the same rate of improvement as the hybrid resulting directly from reciprocal recurrent selection. Preliminary data of comparative selection studies in Kenya show that reciprocal selection was much more effective for crossbred improvement than the intrapopulation method (Darrah et al., 1972). Response of the population per se was about the same for the two methods, but the response of the variety hybrid to reciprocal selection was almost triple the response to ear-to-row selection. Heterosis increased slightly after reciprocal recurrent selection, and decreased markedly after ear-to-row selection. Agreement of observed response with response predicted from genetic

QUANTITATIVE GENETICS

30 1

theory was satisfactory in the studies reported from Kenya (Darrah et al., 1972) and from North Carolina (Moll and Stuber, 1971) for intra- and selection interpopulation selection, with one exception-intrapopulation in the varietal composite reported from North Carolina. In the studies reported from Iowa (Eberhart et al., 1973), the responses predicted from genetic theory appear to be higher than the observed response, but the discrepancies may be within sampling errors. Full-sib reciprocal recurrent selection has been proposed by Hallauer and Eberhart (1970) and examined theoretically by Jones et al. (1971). In this procedure, selection is based on the performance of crossbred fullsib families, rather than crossbred half-sib families. The primary advantage is that twice as many parents can be evaluated in a given space, which in turn permits more intense selection with an equivalent effective population size. However, the advantage tends to be offset by the difference in phenotypic variances, which are larger for full-sib family means than for half-sib family means. Experimental data for full-sib reciprocal recurrent selection are available for only one selection cycle (Hallauer, 1973). Although corn yields improved significantly in each of the populations themselves, no improvement was detected in the population hybrid. There are indications, however, that further evaluation, particularly evaluation of the second cycle, will reveal a significant response in the hybrid. 3. Effectsof Selection on Genetic Variance

Improvement in mean performance with selection is a result of changes in gene frequencies, which also affect the magnitude of genetic variances. It is possible that genetic variances may increase in early selection cycles if the initial frequencies of favorable alleles are low. However, it is more likely that genetic variances will decrease and thus limit rate of response in later selection cycles, It is anticipated that for complex traits, such as those that are influenced by many loci with small effects, genetic variances would change very slowly with selection. This is supported by experimental evidence from a number of selection experiments. Estimates of genetic variances in later cycles of selection have been found to be within sampling errors of estimates in the original populations in studies involving corn and tobacco (Horner et al., 1973; Lonnquist et al., 1966; Matzinger et al., 1972; Moll and Robinson, 1966). The continued linearity of selection response and the agreement of observed response with expectations based upon original estimates is further evidence that variances have not changed materially (Moll and Stuber, 1971; Matzinger and Wernsman, 1968). However, instances have been reported in which variances decreased after selection (Hallauer, 1970; da Silva and Lonnquist, 196)s; Harris et

302

R. H. MOLL AND C. W. STUBER

al., 1972). In the mass selectioq studies for yield of corn at the Nebraska Agricultural Experiment Station, no changes in genetic variances were detected through the first six generations of selection (Lonnquist et al., 1966), but after nine generations it appeared that genetic variability had decreased (Harris et al., 1972). The selection experiments of longest duration involve divergent selection for oil and protein percentages in maize. After 65 generations of selection, Dudley and Lambert ( 1969) reported significant genetic variability remaining in all four strains. Heritabilities are still high enough so that predicted progress deviates only slightly from the average changes so far. It appears that drastic changes in variances with selection are unlikely, at least for traits of low heritability, and variance estimates in original populations may serve for predictions over several selection cycles. C.

CORRELATED RESPONSES AND SELECTION INDEXES

In many selection experiments that were designed to evaluate different kinds of selection, the emphasis has been on a single primary trait in order to facilitate interpretations. However, correlated changes in other traits have been reported in most selection studies and are frequently compared with expectations based on estimates of genetic correlations. Comparisons between expectation and observation for correlated responses have been less consistent than for the direct responses discussed previously. For example, after four selection cycles for leaf weight in tobacco, Matzinger and Wernsman (1968) reported fairly good average agreement between observed and expected correlated responses, but the correlated responses were not linear over selection cycles. In another tobacco experiment, which involved a different primary trait, increased alkaloids, observed and expected correlated responses were in generally poor agreement, although yield did increase as expected (Matzinger et al., 1972). In selection experiments for yield of maize, agreement between observed and expected increases was good for number of ears per plant, but poor for ear height and days to tassel (Moll and Robinson, 1966). Other reports of selection in maize also show a consistent positive association between yield of grain and number of ears per plant, but correlated responses reported for ear height and yield of grain are somewhat variable (Acosta and Crane, 1972; Horner et al., 1973; Harris et al., 1972; Russell et al., 1973; and others). Correlations among traits may be utilized to enhance the rate of selection response in the primary trait. For example, if the primary trait, say trait Y , has low heritability, and it is highly correlated with trait X,which has high heritability, there might be an advantage in improving trait Y indi-

QUANTITATIVE GENETICS

303

rectly by selection for trait X . Searle (1965) has tabulated the minimum combinations of heritabilities and correlations necessary for indirect selection to be more efficient than direct selection. From a practical point of view, the discrepancies between theoretical expectations and correlated responses observed raise doubts about the reliability of indirect selection. Nevertheless, certain traits with obvious physical relationships to each other, such as components of yield and yield itself, have been used with considerable success. Indirect selection for increased yield of oats by selection for certain yield components has been shown to be effective. Frey (1967) reported 9% increase in grain yield by selection for seed width, and Chandhanamutta and Frey (1973) showed an average increase of 5.6% per cycle in grain yield when selection was for increased panicle weight. The grain yield of a population of maize was improved markedly through indirect selection for the number of ears per plant (Lonnquist, 1967). Five cycles of mass selection for prolificacy resulted in an average increase of 6.28% per cycle in yield of the variety HAYS GOLDEN. Gardner (1961) showed only 3.8% gain per cycle with direct selection for yield in the same population. Although Lonnquist applied more intense selection than Gardner, the difference between the selection differentials was not sufficient to account for the difference in response. The greater yield response through indirect selection was partly because of the high heritability of prolificacy and the high correlation between prolificacy and yield. Although indirect selection may be particularly useful in certain instances, it has been proposed that a better way to capitalize upon genetic correlations with more heritable traits is to construct an index that combines information on a11 traits (Searle, 1965; Smith, 1936). Brim et aE. (1959) constructed a series of indexes to improve soybeans according to the method proposed by Smith (1936). They discussed the uncertainty that arises because of errors in estimation of the statistics upon which the index weights are based. In an attempt to circumvent some of the problems of estimation, Hanson and Johnson (1957) proposed a modification of the scheme proposed by Smith (1936), in which the weights would be based on averages of statistics for several populations. An experimental comparison of the two kinds of indexes for soybean yields shows that only the specific index as proposed by Smith was more effective than selection for yield alone (Caldwell and Weber, 1965). Furthermore, their results agree with a more recent study of indexes involving yield components of soybeans in showing indexes slightly more effective than yield alone, but not sufficiently better to justify the extra expense (Pritchard et al., 1973), Practical plant breeding objectives frequently require modification of several traits of a crop species. Methods devised to manipulate multiple

304

R. H. MOLL AND

C. W. STUBER

traits must take correlations between the traits into account. Three general procedures to alter several traits are tandem selection, independent culling levels, and selection indexes. Tandem selection involves selection for each trait separately. Independent culling levels requires setting of minimum levels of acceptability for certain traits, with selection for other traits restricted to those selection units that exceed these minimum levels. Index selection involves a selection criterion based on a combination of measurements of the various traits. The best known indexes involve discriminant functions hased on the relative economic importance of the traits and their genotypic and phenotypic variances and covariances (Smith, 1936; Hazel, 1943). It can be shown that a selection index is expected to be at least as efficient as independent culling, and more efficient than tandem selection (Hazel and Lush, 1942; Young, 1961). Independent culling might be particularly useful when traits are manifest at different stages. For example, one of the advantages cited for full-sib reciprocal selection in maize is that it automatically culls nonprolific parents, Tandem selection is expected to be inferior to selection index methods or independent culling procedures. One of the problems of the conventional selection index approach is in the choice of appropriate economic weights. Pesek and Baker (1969) have proposed that the index be computed on the basis of desired gains for each trait rather than economic weights. An experimental evaluation of two indexes based upon desired gains shows poor agreement between observed and expected gains for three of the four traits involved, probably because of errors of estimation of variances and covariances (Pesek and Baker, 1970). Elston (1963) proposed a multiplicative index that is independent of weights. On the basis of estimates of variances and covariances in two variety hybrids in maize, Subandi et al. (1973) show that the multiplicative index is expected to be as efficient as the weighted index as proposed by Smith ( 1936), and has considerable advantage in simplicity of application. Two kinds of selection indexes were compared with tandem selection and independent culling for five traits of alfalfa (Elgin et al., 1970). The indexes included one in which the weights were computed from estimates of variances, covariances, and economic weights as proposed by Smith (1936), and a base index in which the traits were weighted directly by the economic weights as proposed by Williams (1 962). After five selection cycles, the greatest total improvement resulted from the base index, although the estimated index was a close second. Response to the base index was fairly consistent from cycle to cycle, whereas the response to the estimated index tended to fluctuate considerably. Tandem selection was the least effective method, and independent culling was intermediate in overall effectiveness.

QUANTITATIVE GENETICS

305

Selection indexes do not appear to be in widespread use in practical breeding programs, and there is limited optimism regarding their potential utility. Those that require a minimum number of parameters to estimate have obvious advantages and appear to be at least as effective as the more complex indexes. Independent culling offers advantages in plant improvement, because culling for certain traits can be accomplished in seed nurseries when testcross matings are rhade. Although independent culling is frequently used in this way in plant improvement, no reports could be found of experimental evaluations of the refinements proposed by Young ( 1964), which are expected to lead to greater efficiency by establishment of more realistic culling levels.

VI.

Implications of Quantitative Genetics to Breeding Methodology

A.

BREEDING OBJECTIVES

Plant breeders are primarily concerned with improvement of traits that are directly or indirectly related to economic worth. For example, traits with obvious economic significance would include yield, chemical composition, and various attributes of quality. Traits such as maturity, height, and standability primarily affect efficiency of production. Disease and insect resistance may fall into either category, but they are certain to be of concern in any plant improvement program. Many of these traits are inherited strictly as quantitative traits, but there are some, in certain species, that show segregation for major genes. Examples would include dwarf genes in many crops; the opaque gene and the dull gene of maize, which affect chemical composition; and many instances of disease resistance. However, for most of these instances, the expression of the trait is modified quantitatively by genes with small individual effects. Therefore, the primary objectives of the plant breeder will necessarily involve manipulation of quantitative variation, and the decisions he faces will be influenced by his knowledge of quantitative genetics. Plant breeders must choose the kinds of cultural conditions in which to evaluate their material. The issue involves questions of genotype-environmental interaction, and the choices vary between conditions of deliberate stress to the most optimum growing conditions available. The basic questions have not been satisfactorily resolved, and most breeding programs involve stress conditions and optimum conditions at different stages and for different traits. However, final evaluations are usually conducted under cultural conditions practiced by the most progressive farmers. Decisions on the kind of testing program also must consider the geo-

R. H. MOLL AND C. W. STUBER

306

graphic range desired in the improved strain. The relevant issues here involve phenotypic stability (Section IV, B). As the range of diversity of environments is expanded for which a superior strain is sought, genotype-environmental interactions become more serious, and selection of strains with satisfactory stability becomes more difficult. Procedures most appropriate for a given breeding problem depend to a large extent on whether the breeder is seeking an improved variety, a pure line, or an F, hybrid. Questions concerning the desirability of F, hybrids for commercial use depend to some extent upon the amount of heterosis relative to improvements possible by capitalizing upon transgressive segregation. However, the ultimate decision of whether or not the breeding objective is an F, hybrid often rests on many other factors, such as the availability of sterility mechanisms to facilitate cross-pollination, the desirability of the genotypic control afforded by F, hybrids, and various economic considerations. Once the decision is made to develop F, hybrids, the breeder must choose the most appropriate selection procedures. Comparisons between different kinds of selection procedures in Section V are most important and suggest that, in the initial stages at least, the more simple methods of intrapopulation selection may be quite satisfactory. More complex methods of interpopulation selection may become advantageous in later stages, after significant improvements have been achieved. B.

OF GENETICMATERIAL WITH DEVELOPMENT BREEDINGPOTENTIAL

A broad range of genetic diversity is available in all major crop species. However, breeding procedures for plant improvement have severely limited germplasm diversity in materials available for commercial production. In fact, some crops are being grown almost in monocultures over large areas. This creates a serious hazard, in that the crop becomes extremely vulnerable to disasters, as was evidenced by the 1970 southern corn leaf blight epidemic in the United States (Committee on Genetic Vulnerability of Major Crops, 1972). Because of the narrowing germplasm base in breeding populations, new emphasis is being directed toward the use of broad-based genetic populations, in which recurrent selection is being initiated. The first step in the development of such populations is the selection of superior genetic material. For many species, this will involve an evaluation of exotic as well as local or native materials. As Eberhart et al. (1967) suggest, two or three years spent introducing and evaluating exotic germplasm in addition to

QUANTITATIVE GENETICS

307

the local sources may often produce greater results than a ten-year program recycling the local materials. Eberhart ( 1971) and Goodman ( 1965 ) have demonstrated the benefits from incorporating exotic germplasm into United States maize breeding materials. Large numbers of collections of germplasm are available in most economically important species; therefore, some type of systematic plan for evaluation is required. For preliminary screening, general combining ability is the most important consideration. Therefore, the screening might be accomplished by making testcrosses of collections to a locally adapted line or variety. Agronomic notes can be recorded in the testcross nursery, with yield comparisons among the testcrosses and other performance evaluations conducted in following seasons. As Comstock and Moll (1963) suggest, adequate information for selection of the best entries should be possible with results from a single season, if evaluations are made at several locations. Whether the final goal is to develop one or two breeding populations, some technique should be used to ensure thorough recombination as the selected entries are being composited. Eberhart et al. (1967) proposed a method specifically designed for corn, which might be modified for use in other plant species. In their method, individual entries are planted in a replicated manner and detasseled. Rows of bulked seed of all entries are planted between ranges of the individual entries to provide pollen. At harvest, ears of each entry are saved and bulked over replications to represent that entry in the following season. If desired, selection might be imposed, and only ears from the best plants of each entry would be saved. Individual entires are handled in a similar manner for a minimum of four generations with the variation among entries decreasing as recombination progresses. The total number of ears saved in each generation should be reasonably large (possibly 800-1 000) to minimize loss of favorable genes. Compositing is more difficult in most self-pollinating species. However, the presence of genetic male-sterility mechanisms in many of these species affords an opportunity for recombination and adaptation of recurrent selection procedures to improvement problems (Brim and Stuber, 1973; Doggett, 1972; Doggett and Eberhart, 1968; Gilmore, 1964). To minimize the contribution of the cytoplasm of the male-sterile source in the composite, heterozygous fertiles from a maintainer line can be used as male parents in the synthesis of the initial population. (This normally will require hand pollinations,) The genetic contribution of the male-sterile genotype can be minimized by subsequent backcrossing. This should be followed by several cycles of intermating, in which only male steriles are saved to plant the succeeding cycles. If pollen is transmitted primarily by insects, mating may not be random. Therefore, a sampling scheme, such as a grid

308

R. H. MOLL AND C. W. STUBER

system, to divide the intermating nursery into subblocks should be imposed before harvesting male-sterile plants (Brim and Stuber, 1973). After the intermating cycles are completed, recurrent selection schemes (discussed in Section V) can be initiated. Genetic male-sterility systems can be used effectively to provide recombination between cycles of selection, particularly in self-pollinating species.

C. TESTINGAND EVALUATION Before initiation of a recurrent selection program, it is desirable to have some measure of the response that can be expected per unit of time. Decisions concerning the selection scheme to be used and the selection intensity to be imposed are influenced by the magnitude of genetic variances. However, it may not be necessary to devote time and resources to produce precise estimates of variance components for crops that have already been investigated thoroughly. As pointed out previously, the magnitude of additive variance for particular traits appears to be similar among populations of the same kind. Therefore, the breeder may rely on information from investigations in related populations to aid in practical decision making. Quantitative genetic studies of a wide range of crop species have indicated that the additive genetic (or general combining ability) component is usually more important than the nonadditive (or specific combining ability) component, and that epistatic variance components can be ignored in predictions of selection response in many cases. Therefore, the assumption of predominantly additive genetic variance in a breeding population should be reasonably safe. If the breeder feels a definite requirement for variance component estimates before initiating a selection program, then he should be certain that the estimation experiments are adequate to provide reliable estimates of the kind required. Estimation of genetic variances requires the use of appropriate mating and environmental designs. Dudley and Moll ( 1969) compared various designs and suggested that the most preferable design is the simplest one that will provide the required information. Results from numerous studies in corn (Moll and Robinson, 1967) indicate that 256 progenies (each with two common ancestors, e.g., full-sib families) would be a minimum to estimate additive and dominance variance components, and these progenies need to be grown in at least two environments. Adequate seeds for this number of progenies may be difficult to produce in many self-pollinating species, and experimental procedures using inbred relatives (Stuber, 1970) may be more appropriate to provide the estimates desired.

QUANTITATIVE GENETICS

309

Genetic variances can be estimated during the evaluation phases of the selection program, if the selection scheme includes some type of family structure. After two or three cycles of selection have been completed, reasonably precise estimates of these variances should be available. These estimates can then be used to predict further selection response. As indicated previously (see Section V), genetic variances in corn and tobacco have not changed significantly over several selection cycles for traits with low heritabilities such as yield. Therefore, predictions made from early cycle variance estimates should be reasonably reliable over several cycles for such traits. Although prior estimates of genetic variances are desirable for a population improvement program, the choice of selection scheme will be dictated primarily by the breeder’s specific objectives, the mode of reproduction of the species, and resources available. In a recurrent selection program, decisions concerning the number of parents selected for each cycle of intermating and the selection intensity have far reaching effects as they relate to long- and short-term gains. Progress over the short term may be the primary aim of the plant breeder; however, conservation of the genetic potential of a population over the long term should maintain a high priority in a breeding program. Most breeding programs involve both short-term and long-term goals, and replicated selection similar to that proposed by Baker and Curnow (1969) may provide useful flexibility, Short-term objectives can be realized best with very intense selection, but unless population sizes are large, the rate of inbreeding would prohibit long-term goals. Thus, the replicates that showed greatest improvement could provide the sources for short-term objectives, and replicates could be intermated to minimize inbreeding in more advanced selection cycles. Theoretical developments are available to provide the breeder some guidelines in making these decisions regarding population size. Robertson (1960, 1963) showed that expected total advance and “half-life” of recurrent selection processes are proportional to effective population size ( N ) , an that in long-term selection programs, N should be as large as possible. Rawlings (1970) proposed an effective population size of 30-45, with a selection intensity of 0.10 as a reasonable compromise to satisfy both longand short-term objectives. Similar conclusions were reported by Baker and Curnow ( 1969 ) . Although maintenance of a control (or check) population may be difficult to justify for the empirical plant breeder, it is difficult to assess the progress achieved without appropriate points of reference. An inbred line or single cross will provide a constant genotype as a control population; however, a genetically homogeneous population normally shows more interaction with environments than a genetically heterogeneous population

310

R. H. MOLL AND C. W. STUBER

(Sprague and Federer, 1951). Therefore, the original random mating population in which selection was initiated will probably provide the best source. With cold storage facilities, this population can be maintained over long periods of time with little chance for significant changes caused by natural selection. However, even heterogeneous populations interact differently with changing environments. This is evidenced by the observation that many open-pollinated varieties of corn produce relatively poorly when subjected to high plant densities and high fertility regimes. Therefore, an element of caution must be exercised with the use of any control population. REFERENCES Acosta, A. E., and Crane, P. L. 1972. Crop. Sci. 12, 165-167. Ahluwalia, M., Shanker, K., Jain, S. K., and Joshi, A. B. 1962. Indian J . Genet. Plant Breed. 22, 45-53. Allard, R. W., and Bradshaw, A. D. 1964. Crop Sci. 4, 503-508. Allard, R. W., and Hansche, P. E. 1964. Advan. Agron. 16, 281-325. Allison, J. C. S., and Curnow, R. N. 1966. Crop Sci. 6, 541-544. Aycock, M. K., Jr., and Wilsie, C. P. 1968. Crop. Sci. 8, 481485. Bains, K. S. 1971. Theor. Appl. Genet. 41, 302-305. Baker, J. L., and Verhalen, L. M. 1973. Crop. Sci. 13, 444450. Baker, L. H., and Curnow, R. N. 1969. Crop. Sci. 9,555-560. Baker, R. J. 1969. Can. J . Plant Sci. 49,743-751. Breese, E. L. 1969. Heredity 24, 27-44. Bridge, R. R., Meredith, W. R., Jr., and Chism, J. F. 1969. Crop Sci. 9, 837-838. Brim, C. A., and Cockerham, C. C. 1961. Crop. Sci. 1, 187-190. Brim, C.A., and Stuber, C. W. 1973. Crop Sci. 13, 528-530. Brim, C. A., Johnson, H. W., and Cockerham, C. C. 1959. Agron. J . 51, 42-46. Burton, J. W., Penney, L.. H., Hallauer, A. R., and Eberhart, S. A. 1971. Crop Sci. 11, 361-365. Busbice, T. H. 1969. Crop Sci. 9, 601-604. Caldwell, B. E., and Weber, C. R. 1965. Crop Sci. 5, 223-226. Carangal, V. R., Ali, S. M., Koble, A. F., Rinke, E. H., and Sentz, J. C. 1971. Crop. Sci. 11, 658-661. Castro, G. M., Gardner, C. O., and Lonnquist, J. H. 1968. Crop Sci. 8, 97-101. Chandhanamutta, P., and Frey, K. J. 1973. Crop. Sci. 13, 470-473. Chi, R. K., Eberhart, S. A., and Penny, L. H. 1969. Genetics 63,511-520. Cockerham, C. C. 1963. Nat. Acad. Sci.-Nat. Res. Counc., Publ. 982, 53-94. Collins, F. C., and Pickett, R. C. 1972. Crop. Sci. 12, 5-6. Committee on Genetic Vulnerability of Major Crops. 1972. “Genetic Vulnerability of Major Crops.” Nat. Acad. Sci., Washington, D.C. Comstock, R. E., and Moll, R. H. 1963. Nat. Acad. Sci.-Nat. Res. Counc., Publ. 982, 169-196. Comstock, R. E., Robinson, H. F., and Harvey, P. H. 1949. Agron. J. 41, 360-367. Cress, C. E. 1966. Genetics 53, 269-274. Darrah, L.L., Eberhart, S. A., and Penny, L. H. 1972. Crop Sci. 12, 605-608.

QUANTITATIVE GENETICS

31 1

daSilva W. J., and Lonnquist, J. H. 1968. Crop Sci. 8, 201-204. Dewey, D. R. 1966. Crop Sci. 6, 144-147. Doggett, H. 1972. Heredity 28, 9-29. Doggett, H.. and Eberhart, S . A. 1968. Crop. Sci. 8, 119-121. Duclos, L. A., and Crane, P. L. 1968. Crop. Sci. 8, 191-194. Dudley, J. W., and Lambert, R. J. 1969. Crop. Sci. 9, 179-181. Dudley, J. W., and Moll, R. H. 1969. Crop. Sci. 9, 257-262. Dudley, J. W., Busbice, T. H., and Levings, C. S., 111. 1969. Crop Sci. 9, 228-231. Dunn, G. M., and Wright, J. A. 1970. Crop Sci. 10, 56-58. Eberhart, S. A. 1971. Crop Sci. 11, 911-914. Eberhart, S. A., and Gardner, C. 0. 1966. Biornetrics 22, 864-881. Eberhart, S. A., and Hallauer, A. R. 1968. Crop Sci. 8, 377-379. Eberhart, S. A,, and Russell, W. A. 1966. Crop Sci. 6, 36-40. Eberhart, S. A., Moll, R. H., Robinson, H. F., and Cockerham, C. C. 1966. Crop Sci. 6, 275-280. Eberhart, S. A., Harrison, M. N., and Ogada, F. 1967. Zuechter 37, 169-174. Eberhart, S. A., Debela, S., and Hallauer, A. R. 1973. Crop Sci. 13,451-456. Elgin, J. H., Jr., Hill, R. R., Jr., and Zeiders, K. E. 1970. Crop Sci. 10, 190-193. Elston, R. C. 1963. Biornetrics 19, 85-97. Empig, L. T., Gardner, C. O., and Compton, W. A. 1972. Nebr., Agr. Exp., Sta., Bull. MP26 (revised). Falconer, D. S. 1960. “Introduction to Quantitative Genetics.” Ronald Press, New York. Finlay, K. W., and Wilkinson, G. N. 1963. Aust. J . Agr. Res. 14, 742-754. Fisher, R. A. 1918. Trans. Roy. SOC.Edinburgh 52, 399-433. Fonseca, S., and Patterson, F. L. 1968. Crop. Sci. 8, 85-88. Freeman, G. H., and Perkins, J. M. 1971. Heredity 27, 15-23. Frey, K. J. 1967. Euphyrica 16, 341-349. Gardner, C. 0. 1961. Crop. Sci. 1, 241-245. Gardner, C. 0. 1963. Nut. Acad. Sci.-Nut. Res. Corinc., Publ. 982, 225-252. Genter, C. F. 1971. Crop. Sci. 11, 821-824. Genter, C. W., and Alexander, M. W. 1966. Crop Sci. 6,429-431. Gilmore, E. C., Jr. 1964. Crop. Sci. 4, 323-325. Goodman, M. M. 1965. Crop. Sci. 5,87-90. Gupta, M. P., and Singh, R. B. 1970. Indian J . Genet. Plant Breed. 30, 590-598. Gupta, S. P., and Singh, T. H. 1971. Zndian J. Agr. Sci. 41, 324-328. Gyawali, K. K., Qualset, C. O., and Yamazaki, W. T, 1968. Crop. Sci. 8, 322-324. Hallauer, A. R. 1970. Crop. Sci. 10, 482-485. Hallauer, A. R. 1973. Eygpr. J . Genet. Cytol. 2, 84-101. Hallauer, A. R., and Eberhart, S. A. 1970. Crop. Sci. 10, 3 15-3 16. Hallauer, A. R., and Sears, J . H. 1969. Crop. Sci. 9, 47-50. Hanson, W. D., and Johnson, H. W. 1957. Genetics 42,421-432. Hanson, W. D., Probst, A. H., and Caldwell, B. E. 1967. Crop Sci. 7, 99-103. Harris, R. E., Gardner, C. O., and Compton, W. A. 1972. Crop. Sci. 12, 594-598. Hayward, M. D., and Lawrence, T. 1972. Can. J . Genet. C y t d . 14, 601-607. Hazel, L. N. 1943. Genetics 28, 476-490. Hazel, L. N., and Lush, J. L. 1942. J . Hered. 33, 393-399. Hecker, R. J. 1972. Euphytica 21, 106-111. Hill, R. R., Jr., Leath, K. T., and Zeiders, K. E. 1972. Crop Sci. 12, 627-630. Hogarth, D. M. 1971. Aust. J . A g r . Res. 22,93-102.

312

R. H. MOLL AND C. W. STUBER

Horner, E. S., Chapman, W. H., Lutrick, M. C., and Lundy, H. W. 1969. Crop Sci. 9, 539-543. Horner, E. S., Lundy, H. W., Lutrick, M. C., and Chapman, W. H. 1973. Crop Sci. 13, 485-489. Hull, F. H. 1945. J . Amer. SOC.Agron. 37, 134-145. Humphrey, A. B., Matzinger, D. F., and Cockerham, C. C. 1969. Crop Sci. 9, 495-497. Jain, S. K., and Allard, R. W. 1960. Proc. Nat. Acad. Sci. US.46, 1371-1377. Johnson, E. C. 1963. Inform. Reun. Annu. PCCMM, 9th pp. 56-57. Johnson, V. A., Shafer, L., and Schmidt, J. W. 1968. Crop. Sci. 8, 187-191. Jones, L. P., Compton, W. A., and Gardner, C. 0. 1971. Theor. Appl. Genet. 41, 36-39. Jowett, D. 1972. Crop Sci. 12,3 14-3 17. Kalton, R. R., and Leffel, R. C. 1955. Agron. J. 47, 370-373. Kearsey, M. J., and Jinks, J. L. 1968. Heredity 23, 403-409. Knight, R. 1971. Theor. Appl. Genet. 41, 306-311. Lee, J., and Kaltsikes, P. J. 1972. Crop. Sci. 12, 770-772. Lee, J. A., Miller, P. A., and Rawlings, J. 0. 1967. Crop. Sci. 7, 477-481. Levings, C. S., 111. 1964. J . Hered. 55, 262-266. Levings, C. S., 111, Dudley, J. W., and Alexander, D. E. 1967. Crop Sci. 7, 72-73. Lonnquist, J. H. 1967. Zuechfer 37, 185-188. Lonnquist, J. H. 1968. Crop Sci. 8, 50-53. Lonnquist, J. H., Cota A., .O., and Gardner, C. 0. 1966. Crop Sci. 6, 330-332. Lundqvist, A. 1969. Hereditas 61,361-399. Marani, A. 1963. Crop Sci. 3, 552-555. Marani, A. 1968. Crop Sci. 8, 299-303. Marani, A,, and Avieli, E. 1973. Crop Sci. 13, 15-18. Matzinger, D. F. 1963. Nat. Acad. Sci.-Nut. Res. Counc., Publ. 982, 253-279. Matzinger, D.F. 1968. Crop. Sci. 8, 732-735. Matzinger, D. F., and Wernsman, E. A. 1967. Zuechter 37, 188-191. Matzinger, D.F., and Wernsman, E. A. 1968. Crop Sci. 8, 239-243. Matzinger, D.F., Mann, T. J., and Robinson, H. F. 1960. Agron. J. 52, 8-11. Matzinger, D. F., Mann, T. J., and Cockerham, C. C. 1966. Crop Sci. 6, 476-478. Matzinger, D. F., Wernsman, E. A., and Ross, H. F. 1971. Crop Sci. 11, 275-279. Matzinger, D. F., Wernsman, E. A., and Cockerham, C. C. 1972. Crop Sci. 12, 40-43. Mishra, S. N., and Drolson, P. N. 1972. Crop. Sci. 12, 497-499. Moll, R. H., and Robinson, H. F. 1966. Crop Sci. 6, 319-324. Moll, R. H., and Robinson, H. F. 1967. Zuechter 37, 192-199. Moll, R. H., and Stuber, C. W. 1971. Crop Sci. 11, 706-711. Moll, R. H., Salhuana, W. S., and Robinson, H. F. 1962. Crop Sci. 2, 197-198. Moll, R. H., Lindsey, M. F., and Robinson, H. F. 1964. Genetics 49, 411-423. Moll, R. H., Lonnquist, J. H., Velez Fortuno, J., and Johnson, E. C. 1965. Genetics 52, 139-144. Moutray, J. B., Jr., and Frakes, R. V. 1973. Crop. Sci. 13, 1-4. Ohm, H.W., and Patterson, F. L. 1973a. Crop. Sci. 13, 27-30. Ohm, H.W., and Patterson, F. L. 1973b. Crop Sci. 13, 55-58. Otsuka, Y.,Eberhart, S. A., and Russell, W. A. 1972. Crop Sci. 12, 325-331. Paterniani, E., and Lonnquist, J. H. 1963. Crop. Sci. 3, 504-507. Perkins, J. M., and Jinks, J. L. 1968. Heredity 23, 339-356.

QUANTITATIVE GENETICS

313

Perkins, J. M., and Jinks, J. L. 1973. Heredity 30, 111-126. Pesek, J., and Baker, R. I. 1969. Can. .I. Plant. Sci. 49, 803-804. Pesek, J., and Baker, R. J. 1970. Can. J . Plant Sci. 50, 267-276. Pritchard, A. J., Byth, D. E., and Bray, R. A. 1973. Aust. J . Agr. Res. 24, 81-89. Rasmusson, D. C., and Bryne, I. 1972. Crop. Sci. 12, 640-643. Rawlings, J. 0. 1970. In “Papers Presented at the Second Meeting of the Working Group on Quantitative Genetics” (G. Narnkoong and K. Stern, eds.), Sect. 22, pp. 1-15. Int. Union Forest. Res. Organ., Raleigh, North Carolina. Robertson, A. 1960. Proc. Roy. Sac., Ser. B 153, 234-249. Robertson, A. 1963. Nat. Acad. Sci.-Nut. Res. Counc., Publ. 982, 108-115. Rowe, P. R., and Andrews, R. A., 1964. Crop. Sci. 4,563-564. Russell, W. A., and Eberhart, S. A. 1970. Crop Sci. 10, 165-169. Russell, W. A., Eberhart, S. A., and Vega, O., U. A. 1973. Crop. Sci. 13, 257-261. Searle, S. R. 1965. Biometrics 21, 682-707. Sing, C. F., Moll, R. H., and Hanson, W. D. 1967. Crop. Sci. 7,631-636. Singh, K. B., and Jain, R. P. 1971. Theor. Appl. Genet. 41, 279-281. Singh, K. B., and Jain, R. P. 1972. Indian J. Genet. Plant Breed. 31, 62-66. Singh, K. B., and Singh, J. K. 1971. Zndian J . Genet. Plant Breed. 31, 491-498. Singh, T. P., and Singh, K. B. 1972. Indian J Genet. Plant Breed. 31, 67-72. Smith, H. F. 1936. Ann. Eugen. London 7, 240-250. Sprague, G. F. 1966. In “Plant Breeding” (K. J. Frey, ed.), pp. 315-354. Iowa State Univ. Press, Ames. Sprague, G. F., and Federer, W. T. 1951. Agron. J. 43, 535-541. Sprague, G. F., and Thomas, W. I. 1967. Crop Sci. 7, 355-356. Sriwatanapongse, S., and Wilsie, C. P. 1968. Crop. Sci. 8, 465-466. Stuber, C. W. 1970. CropSci. 10, 129-135. Stuber, C. W., and Moll, R. H. 1971. Genetics 67, 137-149. Stuber, C. W., Moll, R. H., and Hanson, W. D. 1966. Crop. Sci. 6, 455-458. Stuber, C. W., Williams, W. P., and Moll, R. H. 1973. Crop Sci. 13, 195-200. Subandi, Compton, W. A., and Empig, L. T. 1973. Crop Sci. 13, 184-186. Sun, P. L. F., Shands, H. L., and Forsberg, R. A. 1972. Crop Sci. 12, 1-5. Tai, G. C. C. 1971. CropSci. 11, 184-190. Vandenberg, P., and Matzinger, D. F. 1970. Crop. Sci. 10,437-440. Walton, P. D. 1972. Euphytica 21, 553-556. Weber, C. R., Empig, L. T., andThrone, I . C. 1970. Crop. Sci. 10, 159-160. Widner, J. N., and Lebsock, K. L. 1973. Crop. Sci. 13, 164-167. Williams, J. S. 1962. Biometrics 18, 375-393. Wricke, G . 1960. Rundschr 2 Arb-Gem. Biometric DLG-Pflunzenzuechtabt. 1, 1-5. Wright, 3. A., Hallauer, A. R., Penny, L. H., and Eherhart, S. A. 1971. Crop. Sci. 11, 690-695. Yates, F., and Cochran, W. G., 1938. J . Agr. Sci. 28, 556-580. Young, S. S. Y. 1961. Genet. Res. 2, 106-121. Young, S. S. Y. 1964. Heredity 19, 131-145.

This Page Intentionally Left Blank

THE DEVELOPMENT OF TRlTlCALE F . J . Zillinsky International Maize and Wheat Improvement Center (CIMMYT). Mexico City. Mexico

I. Historical Review ................................................ A . The Development of Octoploid Triticale .......................... B. The Development of Hexaploid Triticale .......................... I1. Breeding and Research in Eastern Europe ........................... A . Hungary .................................................... B. Russia . . . . . . . . . . . . . . ..................................... 111. Breeding and Research in tern Europe .......................... A . Sweden ..................................................... B. Spain . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . C . Germany . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . I V. Breeding and Research in North America ............................ A . United States ........... ................................... B. Canada . . . . . . . . . . . . . . . ................................... V. Triticale Improvement at CIMMYT ................................ A . The Establishment of an International Base ...................... B. Breeding Program ............................................ VI . Recent International Developments ................................ A . Adaptation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B . Industrial and Nutritional Quality ............................... C . Recent Cytological Research . . . . ............................. D . Nomenclature ................................................ E . General Comments ............................................ References ......................................................

315 315 318 318 318 321 322 322 323 324 324 324 325 326 326 327 338 338 340 342 344 345 346

I . Historical Review

A . THE DEVELOPMENT OF OCTOPLOID TRITICALE

The history of triticale extends back almost a century . It is highlighted by a series of contributions from many scientists in several countries across three continents (Fig. 1 ) . Historical reviews have been prepared by numer315

316

F. J. ZILLINSKY

FIG.1. Scientists engaged in research on triticale, taken at the International Maize and Wheat Improvement Center (CIMMYT) Headquarters, El Batan. Reading left to right: N. E. Borlaug, CIMMYT; L. H. Shebeski, University of Manitoba; A. Kiss, Hungary; A. Muntzing, Lund, Sweden; E. SAnchez-Monge, Madrid, Spain; K. D. Krolow, West Berlin, Germany; E. Larter, University of Manitoba; F. J. Zillinsky, CIMMYT.

ous authors, but those Muntzing (1973a) and Briggle (1969) have been used freely in the preparation of this manuscript. Triticale is an artificially created derivative of a cross between wheat and rye and possesses the chromosome complements of both parental species. There are two main groups of triticale: the octoploid triticales, which are amphiploids of hybrids between hexaploid wheats and rye; and hexaploid triticales, which are amphiploids of hybrids between tetraploid wheat and rye. Recently tetraploid forms have been reported (Krolow, 1973). The first report of hybrids between wheat and rye was published by Wilson in 1875. The hybrids were highly sterile and did not reproduce. Rimpau, a German scientist, obtained a fertile, true-breeding strain from a cross between bread wheat and rye in 1891. It was not until 1935 that this strain was proved to be an amphiploid with 2n = 56 chromosomes (Lindschau and Oehler, 1935; Muntzing, 1936). According to Muntzing ( 1973a) an unusual outcrossing phenomenon was observed in 1918 by Meister at the Saratov Experiment Station in Russia. Thousands of natural wheat-rye hybrids occurred in wheat plots which had been adjacent to rye plots the previous year. He reproduced plants from these hybrids for several generations and eventually obtained true-breeding, more or less fertile derivatives. In 1930 Meister gave a botanical description of the new species and named it Triticum seculotricum surutoviense Meister. Lewistsky and Benetzkaja (193 1 ) produced cytological evidence that the new forms produced by Meister from the bread

THE DEVELOPMENT OF TRITICALE

317

wheat X rye crosses amphiploids with 2n = 56 chromosomes. They also observed univalents and other meiotic irregularities. They assumed that incompatibility existed among chromosomes between parental genomes. Since disturbed pairing could not be due to a lack of chromosome homology, they believed that the amphiploids arose as a result of an apogamous development of F, ovules having a somatic chromosome number and that this was doubled during the first division of the egg cell. In his review on early research on triticales, Muntzing (1973a) pointed out that Lebedeff, working in the Ukraine, produced some rather advanced cytological work in the early 1930’s. Lebedeff made investigations on amphiploids that he had created as well as on those of his co-workers. He suggested that the poor fertility found in the amphiploids was due to the detrimental effects of inbreeding on the rye genome. This influence might be avoided by using vigorous self-fertile rye parents. Lebedeff also demonstrated the formation of unreduced ovules with 28 chromosomes which produced a 35-chromosome plant when fertilized with rye pollen. Lebedeffs work on triticale appears to have stopped in the mid 19303, as did the work at the Saratov Experiment Station under Meister. It was during this period that Lysenko was gaining a powerful influence in agricultural research in Russia. Up to this time, triticale was more or less a biological curiosity, as it appeared to have little or no potential as a commercial crop. Arne Muntzing, at the University of Lund in Sweden, began research on triticale in 1934 and has continued this work to the present time. His contributions in cytology, genetics, and plant improvement in triticale have been outstanding. His work did much to encourage other scientists to undertake triticale research. He not only produced new primary octoploid triticales of his own and secondary forms from crosses between different octoploids, but also developed a new form in 1933 which was produced by crossing a tetraploid wheat, Triticum turgidum, with rye and pollinating the F, hybrid with hexaploid wheat, A triple hybrid with 42 chromosomes resulted, possessing the complete genome of rye and a combination of chromosomes from bread and durum wheat (Muntzing, 1935). Nakajima (1942) produced triple hybrids in the same way. Shulyndin at Kharkov produced what he identified as 3-species hybrids, by crossing the F, of common wheat X rye with durum wheat (Lukyanenko, 1972; Shulyndin, 1972). Muntzing (1936) observed that occasionally F, plants of wheat-rye hybrids produce anthers having viable pollen. The viable pollen among these anthers ranged from 20 to 60%. A plant having 56 chromosomes was produced from controlled self-pollination with this pollen, thus providing evidence that new amphiploids, which occur occasionally, arise from the

318

F. J. ZILLINSKY

spontaneous formation of small somatic sectors with a doubled chromosome number (Fig. 2 ) . This probably represents the mechanism by which previous octoploids originated from sterile, haploid hybrids. Muntzing (1939), reporting on triticale research between 1934 and 1939, observed that crosses between octoploid triticales from different sources were not as cross-compatible as expected and that fertility of the hybrids was quite low although the hybrids were vigorous. In later generations more-fertile recombinants could be obtained; these were more productive than the parental strains. He suggested the use of self-fertile ryes as parents in the production of primary amphiploids to overcome sterility problems. B. THE DEVELOPMENT OF HEXAPLOID TRITICALE

Two important developments occurred during the late 1930’s that dramatically affected triticale research. First was the discovery that colchicine could be used to induce chromosome doubling so that new amphiploids could be produced routinely (Kostoff, 1938). Second, during the same period improvements in embryo culturing had developed so that hybrids could be obtained from normally cross-incompatible parental combinations. These developments paved the way for the production of hexaploid triticales from hybrids between tetraploid wheat and rye (Fig. 3 ) . The first hexaploid tritical was reported by Derzhavin (1938) from the cross durum wheat x Secale montanum. A hexaploid triticale from a durum wheat X cultivated rye, S. cereale, by O’Mara (1948) was to play an important role in the development of triticale in North America and Europe. Soon numerous new hexaploid triticales were being produced from combinations of different tetraploid wheats and diploid ryes (Nakajima, 1952, 1958, 1963; SBnchez-Monge et al., 1956, 1959; Pissarev, 1963; Kiss, 1966; Larter, 1968; Jenkins, 1969; etc.). Muntzing (1972) indicated that the first hexaploids to be produced had such poor seed development that researchers were not encouraged to work on a form which appeared to have so little economic potential. However, those produced by O’Mara and Sgnchez-Monge were more promising. It.

Breeding and Research in Eastern Europe

A.

HUNGARY

Some outstanding work on the development of triticale as a commercial crop has been done by Arpad Kiss, a Hungarian plant breeder. Kiss started his investigations on triticale in 1949. Using the species T . turgidum as

THE DEVELOPMENT OF TRITICALE

FIG.2A. Development of octoploid triticale.

Fro. 2B. Octoploid triticale and its parental forms.

319

HEXAPLOID TRlTlCALE

FIG. 3A. Development of primary hexaploid triticale.

FIG.3B. Hexaploid triticale and its parental forms.

THE DEVELOPMENT OF TRITICALE

321

female parent in crosses to Hungarian rye varieties, he obtained his first primary hexaploid triticale in 1951 (Kiss, 1965). In 1952 he produced a primary octoploid, using local varieties of bread wheat and rye. He started crossing octoploid and hexaploid triticale in 1954 and obtained a secondary hexaploid that was more productive than either of the parental forms in 1960. It was too susceptible to lodging to be suitable as a commercial crop. In 1958 he reported “that most of the results obtained by researchers in triticale from the standpoint of production, fertility, seed type and cytological stability would lead one to accept that the idea of producing a highly fertile amphiploid competitive with other cereals as being completely illusory.” At that time it was very difficult to justify a practical breeding program in triticale. However, Kiss succeeded in developing secondary hexaploids with improved fertility from octoploid x hexaploid triticale crosses. This encouraged him to continue breeding work on this crop. In 1965, selections from crosses between secondary octoploids and secondary hexaploids were increased and tested in national yield trials. From this source, two secondary hexaploids, Triticale No. 57 and No. 64, were released for commercial production in Hungary. It is believed that these were the first triticale varieties to be grown commercially. These varieties could be grown on sandy soil, where they were competitive with rye in grain production but were not as resistant as rye to frost damage. Top yield of about 7 tons per hectare have been obtained from commercial plantings under favorable conditions. The success of Kiss in the improvement of triticale has stimulated workers in all Central European countries to include triticale in their cereal breeding program. Kiss (1973) reported that, as a feed grain, triticale is equivalent to other cereals for poultry and hogs provided that the grain is free of ergot. In spite of laboratory results indicating that satisfactory bread can be made from triticale grain and that the protein content is higher than that of wheat, triticafe bread has not been produced commercially in Hungary. Triticale is classified as a feed grain and is competitive with rye, but not with bread wheat. B.

RUSSIA

V. Pissarev working in Eastern Siberia first attempted to produce amphiploids of wheat X rye cross about 25 years before colchicine was introduced as a chromosome doubling agent. All attempts ended in failure due to sterility in F,. In 1940, he renewed his efforts to produce rye-wheat amphiploids and after one year had obtained 23 primary octoploids (Pissarev, 1963). In 1959 work on the development of hexaploid triticale was

F. J. ZILLINSKY

322

initiated. He soon found the fertility and plant type to be considerably more advantageous than that encountered in octoploids. In the early 1960's secondary hexaploids were being produced from crosses between octoploid and hexaploid triticales; these secondary forms were found to be more promising than either of the primary forms. A program to introduce winter hardiness in triticale was started in 1962 because winter cereals have a competitive advantage over spring types in Western Russia. Considerable breeding and research work is being carried on in the U.S.S.R., as evidenced by the contributions to the Triticale Conference at Leningrad, sponsored by EUCARPIA, in July 1973. Unfortunately, the proceedings from this conference were not available to the author at the time of writing. A. F. Shulyndin conducts an active triticale improvement program at Kharkov. Most of the material comprises hexaploid winter types. He reported that strains were now being increased for farm-scale production, although he estimated that only a few thousand hectares were being grown commercially in Russia in 1973.

111.

Breeding and Research in Western Europe

Triticale research has been conducted in Sweden, Spain, Germany, and Switzerland for several decades. More recently, institutions in Britain, France, Denmark, and other countries have also become involved in triticale research and improvement.

A. SWEDEN Research on triticale was started at the University of Lund, Sweden, by Dr. A. Muntzing in 1931, and he has directed a triticale research program to the present time. In the early years his major interests were in the areas of cytology and genetics. Breeding work was maintained on a limited basis during most of this period. A summary of his early work on triticale improvement was published in 1939. Numerous research publications provide a history of research achievements under his guidance. A general summary of triticale improvement work was prepared and dedicated to his colleague Dr. H. Stubbe on the occasion of his 70th birthday in 1972. Muntzing found that the rye genome had a very definite influence in octoploid triticale: the seeds and spikes had some resemblence to rye; winter hardiness of rye was evident in winter triticale; its adaptation to sandy soil is similar to that of rye. It was observed that fertility and yield

THE DEVELOPMENT OF TRITICALE

323

were reduced in the primary types but could be improved by selection in the progenies of secondary forms. Muntzing found that some triticale strains of the same ploidy leveI were usually difficult to hybridize. The same phenomenon was later observed in other programs. Research on industrial and biochemical properties was maintained with the cooperation of the Svalov Experiment Station. Agronomic improvement of octoploid winter triticales was advanced to a level almost equaling that of the best bread wheats in grain yield. The most promising material was derived from crosses of his octoploids with similar types obtained from Dr. M. Ingold of Switzerland. The best selections were found to be well adapted to the lighter soils, but they tended to lodge readily on heavier soils. The grain quality was not generally accepted by industry for bread making. With increased interest in hexaploid triticale, Muntzing has increased his efforts to produce new secondary forms from crosses between octoploid and hexaploid triticale. In an effort to broaden the genetic base of triticale, crosses between octoploid winter and hexaploid spring triticales have been made. Populations from early generations of these crosses have been sent to the CIMMYT triticale program in Mexico. Selections for all four combinations of growth habit and ploidy level are being made at both locations. In an effort to improve the grain quality and test weight of triticale, seeds from the Swedish and CIMMYT programs were treated with different mutagens by Dr. Ake Gustaffson. Selection work from this program is still continuing on a cooperative basis. B.

SPAIN

SAnchez-Monge introduced triticale research to Spain in 1950 and has been actively engaged in triticale improvement ever since. New primary hexaploids were developed using different species of tetraploid wheat as parents (1956). He soon found that the hexaploids performed better than the octoploid forms and suggested that the chromosome number is closer to the optimum in the hexaploids (1959). Improvement in fertility was obtained among the progenies of amphiploids involving self-fertile strains of rye as male parents. He was not able to demonstrate an improvement in cross-compatibility from wheat embryos grafted onto rye endosperm, as was suggested by Pissarev and Vinogradova in 1944. Mutation research has been undertaken more recently. SBnchez-Monge ( 1973) reported that seed with smoother endosperm was obtained among the progenies of plants irradiated in an gamma field with 1500-3000 r. He reported (1973) on the use of cytoplasm for Aegilops species and Triticum timopheevi to develop cytoplasmic male sterile triticales for producing commercial hybrids. A hexaploid triticale cultivar CACHIRULO was released for commercial pro-

F. J. ZILLINSKY

324

duction in 1969. It has a higher protein and lysine content than bread wheat. The major problems are susceptibility to lodging, difficulty of threshing, and a lack of quality as compared to bread wheat (SinchezMonge, 1973 ) .

C. GERMANY Some important contributions in triticale improvement have come from research at the Technical University of West Berlin by K. D. Krolow. Krolow has studied the relationships between cytological behavior and fertility among different types of amphiploids. He observed that aneuploid frequency could be reduced by successively selecting the most fertile plants, and that a further reduction resulted from reselections within lines (Krolow, 1962, 1963). His most recent contribution has been the production of tetraploid triticales possessing 7 pairs of rye chromosomes and 7 pairs derived from the A and B genomes of wheat (Krolow, 1973). Numerous investigators have attempted to produce tetraploid triticales from crosses between Triticum monococcum x rye crosses. So far these attempts have failed. Krolow obtained fertile tetraploids by selfing the F, hybrids between hexaploid triticale X rye. He is attempting to develop tetraploid strains having the genomic constitutions AARR and BBRR (Krolow, 1973).

IV.

Breeding and Research in North America

A.

UNITEDSTATES

Several investigators reported making crosses between wheat and rye in the first quarter of the 20th century in the United States, but very few if any of these produced useful amphiploids. Florell (1936) reported that amphiploids of hexaploid wheat x rye crosses had been produced by J. W. Taylor at the Arlington Experiment Station. O’Mara (1940) reported cytological research using octoploid triticale produced by E. R. Sears between “Chinese” spring wheat X rye. One of the first primary hexaploid triticales was produced from a cross between durum wheat and rye by O’Mara ( 1948). This provided the initial material for several breeding programs in North America and Europe. Jenkins ( 1969) initiated triticale breeding in California with material introduced from the University of Manitoba in 1966. F. C. Elliott, Michigan State University, concentrated on the development of winter triticale using material from the same source and has produced some of the most winter hardy triticales. This work stimulated many universities and breeding institutions to initiate research breeding and testing work. By 1973 research

THE DEVELOPMENT OF TRITICALE

325

on triticale was carried on by either public or private institutions in almost every state of the United States mainland. Yield data from experiment stations and universities collected in 1970 by K. Lebsock, USDA, were disappointing. More encouraging results are now being obtained particularly in the western and southern states where breeding and research programs are being conducted. B.

CANADA

The first concerted effort to develop triticale as a commercial crop in North America was initiated by Professor L. H. Shebeski at the University of Manitoba, Winnepeg. A private endowment to the University of Manitoba by the Bronfman Family Foundation was used to establish a Research Chair in 1954. A program of breeding and cytogenetic research was conducted under the leadership of B. C . Jenkins and L. Evans until 1966, when E. Larter was appointed to head the triticale research. During the early years, emphasis was placed on collecting and evaluating numerous primary triticale from scientists in many countries. Crossing between the most promising primary types was started in 1958. The University of Manitoba program included the production of both winter and spring habit triticales. However, the winter types were generally insufficiently hardy to withstand the low temperatures during the winter months. Progressively more effort was directed toward the spring triticales. Synthesis of new primary triticales of both octoploid and hexaploid forms was maintained as a part of the improvement program. Secondary hexaploids derived from crosses between 8 X by 6X triticale are produced regularly in the breeding program following observations by European scientists that hexaploid plants recovered from these crosses are superior in agronomic characteristics to primary hexaploids or hexaploids derived by intercrossing hexaploids. Research on reciprocal crosses between 6 X and 8>< triticale revealed that among the beneficial effects contributed to the progeny by the octoploid parents were better meiotic stability, improved fertility, plumper seed, lower amylase activity, and higher lysine content. Larter (1973) suggested that the beneficial effects were due to the hexaploid cytoplasm which had already been modified to coexist with a foreign genome ( D from Aegilops sp.). During the winter of 1963-1964 a cooperative arrangement between the University of Manitoba and the Wheat Improvement Program in Mexico, now the International Maize and Wheat Improvement Center (CIMMYT), was established to grow a triticale nursery in the Yaqui Valley of Sonora during the winter months. Daylength-insensitive triticale strains were selected in 1965 from progenies of crosses among hexaploid triticales and among Mexican bread wheat x triticale crosses (Quifiones, 1967). By 1968

326

F. J. ZILLINSKY

the performance of triticale had improved to such an extent that grain yields were approaching yields of the bread wheats in Manitoba, but were considerably less than that of barley. ROSNER,the first variety of the triticale developed in North America, was released for commercial production in 1970. Strains of triticale have now been developed which outyield ROSNER (Larter, 1973). These strains are a product of multiparent crosses utilizing ROSNER and highly fertile Armadillo types from the CIMMYT program. The triticale program at the University of Manitoba provided the initial germplasm for practically all subsequent programs established in North America and many countries around the world.

V.

A.

Triticale Improvement a t CIMMYT

THE ESTABLISHMENT OF AN INTERNATIONAL BASE

Dr. N. E. Borlaug first observed triticale in the breeding program at the University of Manitoba in 1958. At that time, the crop left much to be desired, but the vigor of the plant and its apparent potential for improvement fired his imagination. If problems of sterility, seed shriveling, disease resistance, and cytological instability could be overcome, it would have great potential as a cereal grain to improve the quality of nutrition in food-deficient areas. Borlaug imagined that progress in overcoming the deficiencies on triticale could be achieved much more rapidly if work was located in an area where two crop cycles per year could be obtained instead of one, and where the program had access to material from a diversified wheat and durum breeding programs which were already underway in Mexico. Triticale research in Mexico was initiated by Dr. Borlaug in 1964 as a cooperative project between CIMMYT and the University of Manitoba. Funds were provided by the Rockefeller Foundation. The main objective was to develop a grain crop that would be competitive with other cereals, particularly in improving human nutrition in developing countries. During the first two years, the program was directed by Borlaug with the assistance of R. Rodriguez and M. Quiiiones. In 1966 Dr. Charles Krull assumed the responsibility until January 1968, when the leadership of the program was taken over by the author with the assistance of Alfonso L6pez. In 1971, the Government of Canada, through its agencies C.I.D.A. and I.D.R.C. undertook the funding of an expanded triticale program with the University of Manitoba cooperating. The establishment of a cooperative triticale improvement program between CIMMYT and the University of Manitoba greatly expanded research and improvement opportunities in triticale. A new dimension was added

THE DEVELOPMENT OF TRITICALE

327

by utilizing three entirely different climatic zones at which the populations were grown and selected. In the winter, the triticale nursery is grown at Centro de Investigaciones Agricolas del Noroeste (CIANO), State of Sonora, at 28O N latitude, 35 meters elevation. The summer nursery in Mexico is grown in the Toluca Valley at 1 8 S 0 N latitude and 2600 meters elevation. The University of Manitoba summer nursery is grown in the Winnipeg area at 50° N and 230 meters elevation. Thus not only did the program expand in area and number of workers involved, but the number of generations grown per year doubled, the number of environments tripled, and for the first time triticale breeding was introduced to tropical regions requiring daylength insensitivity similar to that of the wheats which are adapted to regions between 30° N and 30° S latitudes. The alternation of generations between CIANO and Toluca permitted screening for strains capable of performing well at two widely different environmental conditions. The disease infestations differed greatly at the two locations. This has greatIy enhanced the possibilities of obtaining selections having wider adaptation. A further advantage was gained by having access to large, diversified material in aggressive durum and bread wheat programs to which triticale crosses could be made and from which new primary triticales could be produced. B.

BREEDING PROGRAM

Preliminary work on triticale was started in 1963, when some triticale strains were included among wheat populations obtained from Dr. J. A. Rupert in Chile. Ingenieros Ricardo Rodriguez and Marco Quiiiones because of scientific curiosity made a number of crosses between these triticales and several Mexican dwarf wheats (Quiiiones, 1967). The triticale lines originated from the University of Manitoba. The hybrids and selected plants were brought to the Toluca nursery (State of Mexico) for observation. The need to overcome daylength sensitivity, reduce plant height, and improve resistance to stripe rust became apparent immediately. Some degree of daylength insensitivity was recovered from crosses between triticales, but this was insufficient for crop production in Mexico. Improved daylength insensitivity and disease resistance, were recovered in later generations from triticale X bread wheat crosses made at this time. By 1967, a number of strains with enough resistance to disease, and insensitivity to daylength had been developed to be included in replicated yield tests. Results from tests at CIANO and Toluca indicated that the triticale strains produced about one half as much grain as the best wheat cultivars under similar conditions. The triticale strains were tall, late maturing, and at least as vigorous in the production of total plant material as the best wheat cultivars. The depressed grain yields were attributed to the high incidence of sterility and severe endosperm shriveling.

328

F. J. ZILLINSKY

1. Improving Fertility In 1968 a very intensive selection effort was devoted to finding plants having better fertility. A few plants with improved fertility were found in an F, population of a cross between two hexaploid triticales. The average percentage of seed set of two of the original lines was about 6 % below that of adapted bread wheat strains, and 15% above the best original hexaploid triticales. These few plants eventually provided a major contribution to triticale improvement. Among the characters associated with these selections, which were later identified as Armadillo strains (Zillinsky and Borlaug, 1971), were high fertility, improved test weight, better grain yield, insensitivity to daylength, one gene for dwarfness, early maturity, and good nutritional quality. Each of the factors were found to be heritable and could be easily transmitted to its progeny. Furthermore the Armadillo strains were generally more cross compatible with bread wheat, durum wheat, and rye than were the normal hexaploid strains in the program. This improved compatibility might be attributed to a higher proportion of viable pollen when used as the pollen parent in crosses. A similar improvement in the production of F, hybrids was observed when the strains were used as the female parent. Investigations on the origin of the unusual characteristics of the Armadillo strains (Fig. 4) revealed that the majority of the characteristics, such as dwarfing, disease resistance, earliness, erect juvenile growth habit, short spike, and smaller plumper kernels, must have been introduced from a Mexican bread wheat having a NORIN 10 dwarfing gene. The bread wheat is believed to have been introduced via spontaneous outcrossing on the F, hybrid of cross X308, since the original cross, X308, from which the Armadillo strains were selected, combined secondary hexaploid triticale parents having no bread wheat in the progenitors. The outcrossed hybrid was subsequently pollinated with hexaploid triticale pollen from neighboring plants in the triticale nursery. A verification that a bread wheat progenitor was involved in the origin of Armadillo was obtained in 1973 when a D chromosome was found to be substituted for one of the rye chromosomes (Gustafson and Zillinsky, 1973; Gregory, 1973; Merker, 1973b). Although the rye genome in Armadillo appears to have lost one pair of chromosomes by a substitution, the total genotype was considerably improved by the modification. Whether this was due to the deletion of that particular rye chromosome or to the favorable effects of the D genome chromosome has not yet been determined. The Armadillo strains were used frequently as parents in crosses to other hexaploid triticales both primary and secondary forms, to bread and durum wheats, and to primary octoploid triticales during the following generations.

THE DEVELOPMENT OF TRITICALE

329

FIG.4. The plant type of the fertile selection Armadillo.

Selections with fertility approaching that of the Armadillo parent were obtained among the segregating populations. By 1970 practically all the material in the CIMMYT triticale program originated from crosses having Armadillo as a progenitor. 2. Lodging

Susceptibility to lodging was a common problem encountered by many of the early investigators, including Muntzing, Kiss, and Sfinchez-Monge. The problem was intensified under Mexican conditions owing to the tendency of long day-sensitive material to grow taller under short-day conditions. Even the single factor for dwarfing possessed by the Armadillo strains was not sufficient to prevent lodging since the increase in fertility and grain density increased the weight of the mature spike (Fig. 5 ) . Attempts to improve lodging resistance included increasing straw thickness and incorporating more dwarfing genes from wheat. New primary amphiploids, both hexaploid and octoploid, were produced using dwarf durum and bread wheats. These were used as parents in crosses to

330

F. J. ZILLINSKY

FIG.5. Dwarfing in triticale: differences in plant height.

hexaploid triticale. Early attempts to incorporate more dwarfing genes from bread and durum wheats having NORIN 10 dwarfing were discouraging. It was very difficult to maintain fertility among the dwarf selections. The grain quality tended to deteriorate conspicuously. Similar problems were encountered in the early stage of the wheat breeding program in Mexico when NORIN 10 was used as a dwarfing source. Borlaug 1968; Zillinsky and Borlaug, 1971 ) pointed out that only semidwarfs (singlegene dwarfs) with good fertility and acceptable grain type could be isolated from crosses between NORIN 10 X tall wheats. All double dwarf segregates were highly sterile and possessed very shriveled grain. Subsequent recrossing and selection for fertility and grain plumpness resulted in the development of excellent double dwarfs, such as SONOM 64, INIA 66, possessing complete fertility and excellent grain type. Kiss (1968) reported similar difficulty with sterility and grain shriveling when using the NORIN wheat dwarfing sources. He subsequently used TOM THUMB with more success. Crosses between Armadillo strains and stiff-strawed, normal-height triticales resulted in only moderate improvements in lodging resistance. It became obvious that if grain yields competitive with the Mexican dwarf wheat were to be achieved, the straw length of triticale had to be reduced. Since all the triticale germplasm possessed genomes of tall ryes, a major obstacle to expression of the dwarfing characteristic in triticale was the

THE DEVELOPMENT OF TRITICALE

33 1

tall genotype of rye. An obvious solution was to replace these with genes from dwarf ryes. A search for dwarf ryes among collections of spring ryes resulted in the discovery of a single heterozygous dwarf plant in a rye population received from Dr. Darrell Morey of The Coastal Plains Experiment Station, Tifton, Georgia (Zillinsky and Borlaug 1971 ) . The dwarf segregates among the progeny of this plant were identified as “Snoopy” selections. Unfortunately the original plant was susceptible to several diseases (stripe rust, bacterial stripe, and scab) and had some other unfavorable agronomic characteristics. It was necessary to improve the phenotype by crossing to selected tall strains before crossing to triticale. Dwarf segregates from crosses between Armadillo x Snoopy rye had sterility and seed shriveling problems similar to those from the Armadillo X dwarf wheat. Two-gene dwarf hexaploids were eventually obtained in 1972 which were equal in fertility to Armadillo (Zillinsky and Lopez, 1973). These originated from two sources: (a) hexaploid triticale x bread wheat, and (b) octoploid triticale x hexaploid triticale. The F, hybrids from both sources were equal to the triticale parent in height. The hybrids having a bread wheat parent were much more sterile than those from octoploid x hexaploid crosses. It was necessary to overcome the sterility by growing the F, plants in rows alternating with normal fertile hexaploid triticale as a source of viable pollen for two generations. 3. Diseases

Triticale was first released for commercial production in Hungary in 1968. Even today only a few countries are growing limited acreages commercially. Information on diseases is rather scarce. Wherever the crop is grown, disease symptoms appear, apparently caused by plant pathogens which parasitize wheat and rye species. They have not been reported as a serious limiting factor in triticale development. Fuentes ( 1973) summarized the literature on diseases of triticale. Larter el al. (1968) reported that in higher latitudes ergot caused by Claviceps purpurea is a serious problem. Grain contaminated with sclerotia of ergot causes toxicity problems in animal feed. There is very little genetic resistance to the disease, although considerable protection from infection can be obtained among highly selffertile strains. This form of protection is present among cultivars of wheat and other cereals. European investigators have reported that triticale generally is more resistant to diseases than wheat (Pissarev, 1963; Shulyndin, 1972; Kiss, 1973). Leaf rust (Puccinia recondita) and stem rust ( P . graminis) attack triticale and are considered the most serious diseases at many of the international triticale yield nurseries. Leaf rust and stripe rust ( P . glumarum) are serious pathogens of triticale

332

F. J. ZILLINSKY

in Mexico, and natural infestations occur regularly in the summer nurseries in the State of Mexico. Rajaram et ul. (1972) observed that the many of the triticale strains in the CIMMYT program were susceptible in the seedling stage to 4 races of leaf rust which attack INIA, and SIETE CERROS and other cultivars or bread wheat. However, some of these (19 out of 75) were resistant in the adult plant stage. During investigation of patterns of leaf rust development in triticale, it was observed that some strains which are susceptible to leaf rust in the seedling and early adult stage abruptly produce the telial stage prior to maturation (Zillinsky, 1973). This would tend to restrict the production of inoculum and thus provide some degree of protection. Quiiiones et ul. (1972) reported that each of the strains 6A-190, ROSNER, Armadillo, BRONCO, and TOLUCA 16a have a single dominant gene for resistance to leaf rust which was derived from the wheat parent, and that resistance carried by the rye parent was not expressed in the amphiploid. This is probably true for seedling resistance, but adult plant resistance is carried by octoploid triticale strains derived from cross between INIA wheat and several ryes to races that attack INIA in both seedling and adult stages (Rajaram et ul., 1972). Resistance to stripe rust was essential to maintain a nursery in Toluca, where stripe rust infestations can be devastating. Quiiiones and Rodriguez (1973) observed almost 100% of the triticale strains were destroyed by stripe rust in the first season the triticale nursery was grown at Toluca. Resistance was obtained from intercrosses among resistant plants and backcrosses to resistant wheats. The continued use of resistant strains as parents and heavy selection pressure for resistance has resulted in a degree of resistance superior to that found in most durum and bread wheats to races currently prevalent in Mexico. Bacterial diseases attack triticale strains in the Mexico nurseries and other areas of North America. Dr. Bradbury of the Commonwealth Mycological Institute, Kew, isolated Pseudomonus striufuciens from a bacterial leaf stripe lesions on triticale from the Toluca nursery in 1972. During the next growth cycle at CIANO in the Yaqui Valley, he isolated the bacterium Xanthornonus trunslucens from bacterial lesions on infected leaves (J. M. Waller, private communication W-1549 and W-1557). A very serious outbreak of bacterial stripe, probably due to Xanthornonus trunslucens, occurred on triticale in the nursery at Navojoa, Sonora, in February and March 1970. Many of the strains were susceptible, and they were almost completely defoliated. Resistant plants were selected and used in crosses. The spread of the disease is highly dependent upon a favorable environment, which occurs occasionally in the nursery areas, and continuous dependable screening for resistance to the disease has not been possible. In Mexico, rye strains have generally been more severely damaged by bacterial diseases than the wheats.

THE DEVELOPMENT OF TRITICALE

333

Leaf blight caused by Fusarium nivale occurs regularly on triticale and wheat in the Toluca nursery throughout the growing season. This fungus is of little importance to spring crops in other regions of the world (Richardson et al., 1972), but it is devastating on susceptible strains of triticale and wheat in the Toluca Valley. There does not appear to be clear cut resistance among the triticale strains, although some strains are killed and others damaged only slightly. The more tolerant strains are infected much later in the growth cycle. This disease has not been observed at elevations below 7000 feet in Mexico. It is possible that the disease is indigenous on grass species in regions at high elevations in Mexico and other countries. Fuentes ( 1973) has investigated leaf blotch on Septoria tritici wheat and triticales in the CIMMYT breeding program. He has found that triticales are generally more resistant than wheats to the strains of this pathogen found in Mexico. He assumes that other areas, such as North Africa, the Middle East, and South America, may have strains that are more virulent, since some reports on the reaction of triticales to Septoria tritici have indicated high susceptibility. Lesions of infected triticale leaves from the nurseries in Mexico have been examined regularly since 1971. The pycnidia of Septoria tritici have been isolated only rarely although pycnidia resembling Septoria nordorum and Septoria avenue f. sp. triticea occur regularly. More intensive investigations need to be carried out on diseases causing leaf blotching, particularly in the cooler and more humid areas of the tropics. It may be possible to replace wheat with triticale, which is more resistant to these diseases. Triticale appears to be more resistant to powdery mildew (Erisiphe graminis) and the smuts (Ustilago spp.) than wheat. However D. D. Morey observed powdery mildew on triticale in the winter nursery at Tifton, Georgia. A few spikes infected with loose smut have been found in the Toluca nursery. Occasional plants infected with downy mildew (Sclerophthora macrospora) are found in the CIANO nursery each year. Head blights, foot rots, and seed infections occur regularly in the summer nurseries. M. J. Richardson of East Craigs, Scotland, and J. M. Waller of the C.M.I. Kew investigated diseases of triticale in the Mexico nurseries during the fall of 1973. They observed fruiting structures of Ophiobolus graminis, Cochliobolus sativus, and Fusarium graminearum on triticale plants. They also isolated several seed-borne pathogens on seed produced in Toluca and El Batan (Richardson and Waller, 1973). Several virus diseases have appeared in triticale nurseries. The aphidtransmitted barley yellow dwarf virus infects triticale strains in the Mexico nurseries. The proportion of plants infected is generally higher than among wheats in the same area, but much less than either oats or barley. Symptoms on infected triticale plants are similar to those developed on infected

334

F. J. ZILLINSKY

bread wheats. Triticale plants infected with African cereal streak virus were observed in nurseries at Njoro, Kenya, and Debre Zeit, Ethiopia, during February and March 1973. Viruslike symptoms on the leaves of juvenile triticale plants were observed in the nursery at Ankara, Turkey, in 1972, but positive identification was not made. Diseases have not generally appeared as a serious problem in the CIMMYT triticale breeding program. However, as commercial production increases, diseases that parasitize triticale will increase. A close watch on disease development will have to be maintained as production spreads. Genetic resistance appears to be available for most of the diseases observed to date. It is extremely important that breeding programs maintain a broad germplasm base to provide protection against present and future disease infestations. Dr. Alejandro Ortega, an entomologist in the CIMMYT corn program, observed that triticale are generally attacked by the same insects as other cereal crops. Infestations of corn leaf, English grain, and cereal root aphids are common. Occasionally heavy infestations of shoot fly, frit fly, and stink bugs have been observed on triticale in localized areas in Mexico. Triticale plants infected with Hessian fly and root knot nematodes have been found in North Africa. Care must be exercised in using insecticides on triticale. Some strains of triticale are quite sensitive to pesticides applied as foliar sprays. 4. Yield

The introduction of the Armadillo strains into replicated tests resulted in a significant increase in grain yield. The degree of improvement was influenced by soil fertility, diseases, and other environmental factors. The Armadillo strains approached the Mexican bread wheats in grain yields at low to moderate levels of nitrogen, but dropped off sharply with increases in levels of nitrogen. The Armadillo strains were less responsive to nitrogen fertilization and more susceptible to lodging than the Mexican dwarf bread wheats. An estimate of the rate of improvement in grain yield of triticales compared to that of bread wheats in Mexico during the past 6 years can be obtained from yield data from replicated tests at CIANO Experiment Station, Sonora (Fig. 6 ) . The rapid increase in yield improvement of triticale between 1967 and 1969 was due to the introduction of Armadillo strains. Improvement in yield during the next 2 years (1969-1970, 1971-1972) occurred at more or less the same rate in both wheat and triticale. The introduction of fertile two-gene dwarf triticales into the 1972-1 973 yield tests resulted in a second significant increase in the rate of yield improvement. It is expected that as more dwarf triticale strains from the breeding

THE DEVELOPMENT OF TRITICALE

335

FIG.6. Yields of triticale at the Sonora, Mexico, winter nurseries. Comparison of yields of top wheat vs the average of the top triticale strains.

program are advanced to the replicated trials, triticale yields will equal or surpass those of the best bread wheats in the Yaqui Valley of Sonora. Grain yields of triticale are already competitive with wheat in some of the high mountain valleys and on some sandy soils areas of Mexico. Further increments in grain yield in triticales are expected with the introduction of more dwarfing genes and improvements in tillering capacity, grain density, plant structure. An immediate increase of 10-1 5 % could be achieved if triticales could produce grain of equal density to wheat. Increasing spike length may also result in yield increases perhaps compensating for the present deficiencies in tillering capacity. 5. Grain Quality The most important unsolved problem in triticale breeding is abnormal endosperm formation resulting in seed shriveling, low test weight, and low germination rate. As the spikes approach maturity, abnormalities appear and seed development becomes progressively more abnormal as it ripens. The ripe seeds have a wrinkled seed coat, lack luster, and have a deep crease. The endosperm is chalky in contrast to the hard vitreous seed of durum and bread wheat. The test weights range from 58 to 72 kg/ha, while the best bread wheats have test weights in excess of 80 kg/ha.

336

F. J. ZILLINSKY

Lebedeff (1934) and others have suggested that abnormal seed development may be due to the deleterious effect of inbreeding on the rye genome. Sinchez-Monge (1969) showed that improvement in fertility and grain quality could be achieved by using self-fertile ryes as parents in triticale breeding. Inbred ryes are being used as parents in the CIMMYT program although no naturally self-pollinating ryes are available. More intensive research has recently been undertaken at the University of Manitoba in an effort to identify the causes of seed shriveling and the means to overcome the problem. Larter (1973) reported that research was being conducted on cytogenetics and cytology (by Darvey and Kaltsikes) , histology (by Shealy) , and biochemistry (by Hill). This research has provided information on the physical and developmental aspects of seed shriveling. Bennett (1973) suggested that differences in the duration of the meiotic cycle might influence endosperm development. Distinctive chromatin formation in the chromosomes of the rye genome may require more time for replication than for the wheat chromosomes. Thus, disturbances in cell reproduction in endosperm tissue result from segments of late-replicating heterochromatin at the telomeres of rye chromosomes. Improvements in grain quality have been achieved by breeding, although progress is slow. Triticale is more sensitive to environmental influence than wheat. A higher protein content tends to be associated with increased endosperm shriveling. Thus, the higher protein content of the grain observed in the earlier triticale strains is deceiving and may be the result of abnormal development at the expense of other nutrients. Villegas (1973) has shown that a marked decrease in protein content has occurred with improvement in yield capacity and kernel plumpness. The increase in yield has more than compensated for the loss in protein so that the production of protein per unit area has increased (Figs. 7 and 8). Visual screening for plumper seed had to be applied with considerable caution. There is a strong tendency to eliminate all selections from wide crosses possessing dwarfing genes, since these forms produce shriveled seeds. Improvement in seed type is obtained at the expense of desired plant types or those having wide genetic diversity unless care is exercised to avoid discarding those selected for characters other than plump seed. Dr. Ake Gustaffson of Lund, Sweden, initiated mutation research to improve seed quality in triticale in 1969, using mutagenic chemicals and radiation at several concentrations. Some improvement in seed type was obtained from this material among selections made in Mexico on the third and fourth generations after treatment.

337

THE DEVELOPMENT OF TRITICALE

FIG.7. Comparison of triticale seed produced at CIANO, Sonora, in 1967, 1970, and 1973.

0-

- 20

9-

- 18

8-

- 16

7-

-

6-

- I 2

-

c

- 10

-&

-

a

0

I

..

5-

VI f=

0

-

-

14

-

L

4-

- 8

3-

- 6

-

+

a,

8

-

2-

4

-

l-

-

-

2

FIG.8. Grain yield (0-0) vs percent protein ( O - - - O )in triticales at CIMMYT, 1967-1973.

F. J. ZILLINSKY

338

The most encouraging progress in seed improvement resulted from screening segregating generations of octoploid X hexaploid triticale crosses for high fertility. The progeny of the most fertile selections were then screened visually for grain plumpness. It has been possible to obtain in this way highly productive strains with grain test weights of up to 76 kg/ha. Mechanical separation for higher seed density is possible with several commercial separators. Bulk seed lots of early generation material is being screened routinely in the CIMMYT program. The material treated in this way has gone through only a single cycle, and it is not yet possible to conclude that the screened seed produces progeny having a higher seed density, It appears that improvement in seed plumpness and density will be achieved through breeding and selection. The rate of improvement will depend upon the degree of selection pressure placed on the segregating populations, the number of cross-combinations made, and the obtaining of a fortuitous combination of compatible rye and wheat chromosomes. VI.

Recent International Developments

A. ADAPTATION CIMMYT initiated an international triticale testing program in 1969. Scientists at universities and public plant breeding institutions in numerous countries around the world have cooperated. This program has made it possible to estimate the productivity of triticale strains relative to wheat, from year to year over a wide range of environments. Results from the first international triticale yield nurseries (MacKenzie, 1972) indicated that the strains were more narrowly adapted than the Mexican wheats. The daylength-insensitive strains tend to produce few tillers and short spikes under long-day conditions. Development under high temperatures during the early growth period is equally unproductive. There has been considerable improvement in the overall average performance of the triticale strains during the three years of testing from which results have been obtained (Fig. 9 ) . Areas of high elevation and cool growth temperatures appear to provide a suitable environment for triticale. Pinto (1973) reported that triticale significantly outyielded durum and bread wheats in the national trials in Ethiopia. Wabwoto (1973) indicated that resistance to stem and stripe rust was more easily obtained in triticale than in wheats, thereby giving this crop a competitive advantage in the higher elevations of Kenya where rusts are a serious problem. Chauhan and Srivastava (1973) observed

THE DEVELOPMENT OF TRITICALE

339

FIG.9. Relative yields of top five triticale strains compared to the top wheat check in the international triticale yield nurseries at all locations, 1969-1972.

greater resistance to frost damage and better growth response of triticale over bread wheat during the winter season in the mountains of Northern India. Triticale is more competitive with wheat in the Toluca nursery (2600 meters elevation) than in the Yaqui Valley of Sonora (35 meters elevation). Triticale appears to have a competitive advantage under cool growth conditions at intermediate to low elevations as well. Most of the commercial triticales grown in the United States are produced during the winter season in the Central Gulf coast states. Vigorous growth is obtained even when night temperatures approach freezing, making it a suitable forage and pasture crop. In 1972, A. Kiss observed vigorous growth and resistance to winter injury among some day length-insensitive, spring-type triticale strains when grown during the winter season in Hungary. Generally, the winter triticale strains so far developed are not as resistant to severe winter frosts as the best winter ryes. Shulyndin (1973) claims to have strains of hexaploid triticale as hardy as winter ryes. Triticale could serve as a substitute crop for rye on sandy soils. SBnchezMonge (1973) indicated that the triticale variety CACHIRULU performs well on sandy soils in Spain. Muntzing (1973a) observed that the yields of the octoploid triticales equaled those of wheat on sandy soils or under lower fertility conditions but not under conditions of high fertility. Kiss (1965) reported that much of commercial production of triticale in Hungary is located in lighter sandy soils.

340

F. J. ZILLINSKY

Experience gathered from the CIMMYT triticale program indicates that the only requirements are modifications in agronomic practices used for wheat production. These will differ according to cultivars and environment. Compared to bread wheat, triticale is generally less responsive to nitrogen, more susceptible to lodging, and preharvest sprouting, more sensitive to stress due to drought and high temperatures, and more sensitive to injury from pesticides. The seed is more susceptible to damage by seed-borne fungi and insects. Triticale appears to be more resistant to light frosts during the early growth period and to many leaf diseases. Shallow seeding is important, since emergence at seeding depths below 9 cm is very poor.

B.

INDUSTRIAL AND

NUTRITIONAL QUALITY

Muntzing (1956, 1963), in collaborative work between the University of Lund and the Department of Cereal Chemistry at Svalof, reported that octoploid triticales had good baking characteristics and high protein content. One of the octoploid triticale strains produced by Meister gave loaf volumes equal to those of good bread wheat varieties. The early hexaploid triticale strains, on the other hand, produced loaf volumes and texture of inferior quality. More recent studies at Lund have demonstrated that improved baking quality can be obtained by modifying the technique and by selection. Scales and standards suited for evaluating wheat quality are not suitable for the evaluation of triticale (Muntzing, 1973a). He indicated that several features of triticale, such as predisposition to preharvest germination, low gluten strength, and amylase activity, detract from good breadmaking quality. Eastern European workers generally have observed that although satisfactory bread can be made from some strains of triticale, there appears to be little interest in its use over conventional bread wheat (Kiss, 1965; Pissarev, 1963; Shulyndin, 1973), and they have considered triticale to be more suitable as a feed grain. Lorenz et al. (1972) compared the mixing and baking properties of some triticale strains with properties of spring and winter wheats. They found that, although triticale grain had a higher protein content, the flour made from triticale grain had less protein than did wheat flour. However, good bread could be produced by appropriate changes in absorption and mixing time and mixing procedures. Tsen et al. ( 197 1 ) and Amaya ( 1973) have reported that bread of acceptable taste, texture, and loaf volume can be produced from triticale flour if modified bread-making procedures are used. Villegas et al. (1973) have indicated the potential use of triticale in the production of chapatis, tortillas, bread, and pancakes. They indicated that flour yields of triticales were generally much lower than those of bread wheat and that the mixing properties of triticale flour were very weak.

THE DEVELOPMENT OF TRITICALE

341

Several food products are manufactured from triticale grain in the United States. Briggs (1973) reported that pancakes made from wholegrain triticale flour, are nutritive and have appealing flavor. He indicated that industry was hesitant in developing commercial food products until adequate supplies of grain of uniform and acceptable quality were commercially available. Commercial grade standards have not yet been established in North America, although studies are now in progress to develop such standards (Strand, 1973). In the past more emphasis was placed on the physical properties of triticale, more recently nutritional quality has received considerable attention. Some nutritional quality characteristics can be determined chemically such as total protein and the amino acid patterns. An inadequate balance of essential amino acids depresses nutritional quality of the protein. Villegas’ group reported in 1968 a range in protein content among triticale strains of 10-20%. The range in percentage of lysine was greater than among the bread wheats and ryes. Villegas (1973) noted that strains with the highest protein content usually had severely shriveled seed. Selecting for plumper kernels and higher grain yields has tended to reduce protein content. The average protein content of strains included in yield tests dropped from 17.5% in 1968 to 13.4% in 1972. Nutritional research on triticale grain indicated that lysine was a limiting amino acid €or humans (Kies and Fox, 1970), poultry (McGinnis, 1973; Avila et al., 1971) and rats (MacDonald and Ahmend, 1973). Bragg and Sharby (1970) observed an increase in growth of broiler chicks by supplementation of the triticale diet with DL-methionine, but not with lysine. Studies on small-animal feeding trials at the University of Manitoba indicate that growth responses of animals fed pure triticale were consistently better than those of animals fed wheat. Ergot contamination was suggested as a factor in poor nutritional results with poultry and hogs in earlier studies (Larter, 1973 ) . Villegas et al. (1973) found that the proportion of lysine in triticale protein changed only slightly with a decrease in protein content. The average percentage of lysine in the protein rose from 3.2 to 3.4 from 1968 to 1972. For some strains, a lysine content of 3.9% was reported. Growth responses of small animals and chicks frequently differ from calculated values based on chemical analysis of the feed. This may be due to palatability, food preference, growth inhibitors, or differences in genetic requirements of animals. MacAuliffe and McGinnis (1971) observed that high levels of rye in chick diets depressed growth and feed efficiency. A greatly improved growth response was obtained when low levels of antibiotics were added to the rye diet. Similar responses were obtained with some strains of triticale used in chick diets. However, McGinnis (1973) observed that triticale generally can be used to supply a high proportion

342

F. J. ZILLINSKY

of the total protein for chicks, broilers, and laying hens and also for young turkeys. Elliott ( 1973) proposed that meadow voles (Microtus pennsylvanicus) could be used effectively as a screening technique for better nutritional quality in the early generations of breeding programs. An evaluation from five animals can be obtained in 1 week with about 100 g of grain. Bauer ( 1973) observed that the heterozygosity of the population produced a wide range in feed efficiency values. There was also a strong preference for triticale and wheat over corn by most animals in the colony. Weiringa ( 1967) reported that growth-inhibiting substances occurring in rye grain caused growth depression when the rye content exceeded 50% of the diet fed to rats and swine. The growth-inhibiting substances were found to be soluble in petroleum ether and acetone and was identified as a mixture of alkyl resorcinols. Villegas et al. (1973) and Larter (1973) found that, among the triticale strains analyzed, the levels of resorcinol compounds were much too low to cause growth depression. Munck ( 1964), McGinnis (1973) and Elliott (1974) have indicated that protein content, and growth efficiencies are influenced by variety, location, environment, and agronomic practices. Disease organisms, such as ergot and scab (Gibberella fujikori) , produce toxins that influence growth responses. Nutritional evaluations should be based on disease-free seed of a specific cultivar. Since triticale strains have a considerable range in values that influence nutritional quality, it is quite possible to improve the nutritional quality by breeding, provided adequate screening techniques are available. Pomeranz et al. (1970) studied several strains of triticale grain from four different locations in North America for suitability in malting and brewing. Some lines produced malts with high extract values, high amylase activity, and satisfactory brew yields and wort runoff times. The wort and beer colors were slightly dark. Some beers had excellent gas-stability and clarity-stability indexes along with acceptable taste.

C. RECENTCYTOLOGICAL RESEARCH Most of the early cytological investigations on triticale dealt with the nature and frequency of cytological disturbances, particularly on octoploid triticale. OMara (1940) reported that triticale could be used to develop chromosome addition lines of wheat. He was able to produce wheat lines with three different rye chromosome additions. More recent cytological research indicates that some earlier concepts need revision. Merker (1971, 1973b) and Larter and Hsam (1973) have reported fertility and meiotic instability are independent. However, the most fertile selections in terms

THE DEVELOPMENT OF TRITICALE

343

of number of viable seeds produced per floret, are also among the most stable meiotically (Muntzing, 1972). Larter (1973) and Merker (1973b) observed that aneuploidy in hexaploid triticale involved both wheat and rye chromosomes more or less at random, indicating imperfect genetic control in chromosome pairing during meiosis. Bennett et al. (1971 ), Kaltsikes (1971 ), and Bennett and Kaltsikes (1973) studied the duration of meiosis in triticale and the parental species to determine whether this was a factor causing meiotic instability. It was observed that meiosis in hexaploid bread wheats (24 hours) was shorter than in the tetraploid wheats (30 hours), hexaploid triticale (34 to 37 hours), or diploid rye (51 hours). Bennett ( 1973) suggested that the difference between the rate of meiotic development of rye chromosomes in triticale might be a major cause of meiotic instability. He pointed out that genome incompatibility and subsequent aberrant endosperm formation might also be due to the presence of segments of late-replicating heterochromatin at the telomeres of the rye chromosomes, but not of the smaller chromosomes of wheat. Kaltsikes (1973) suggested that since rye chromosomes are larger, and carry 1.5 times the amount of DNA compared to wheat chromosomes, it seems likely that differences in rates of meiotic development could result in meiotic abnormalities. The presence of chromosome substitutions between chromosomes of the R genome of rye and the D genome of bread wheat opens up a new field of triticale research. Gustafson and Zillinsky (1973) reported that a single pair of rye chromosomes of the hexaploid triticale Armadillo had been substituted by a pair of D genome chromosomes (2D) The substitution appears to be highly beneficial agronomically. Further substitutions are possible. A hexaploid derivative from a cross between octoploid triticale X Armadillo “S” identified as “Camel” was found to possess two substituted chromosomes (Merker, 1973a,b) ; Bennett, 1973). The chromosome substitutions are usually accompanied by distinct morphological changes in plant development. The “Camel” strain also has a distinctly shaped kernel with more vitreous textured endosperm. Chromosome substitutions which provide a competitive advantage are useful in plant improvement although they tend to create difficulties in gene transfer. How far this chromosome substitution can continue, and still maintain characteristics that are of agronomic advantage, is not known. Antagonism between the cytoplasms and nuclear constitution has been implicated in abnormal cytological and endosperm development (Muntzing, 1935, 1939). Recently Larter (1968), Sisodia and McGinnis (1970), Lacadena and Perez (1973), Kies and Fox (1973), and Brandes et al. (1973) have observed cytoplasmic influences in triticales. It appears that cytoplasm for hexaploid wheat is more compatible with the hexaploid

344

F. J. ZILLINSKY

and octoploid triticale nuclei than cytoplasms from either tetraploid wheat or rye. O’Mara (1940) studied the influence of rye chromosome additions to the genome of hexaploid wheat as a possible means of plant improvement. Darvey ( 1973 ) used chromosome addition lines to determine the influence of individual rye chromosomes on seed shriveling. Rupert et at. (1973) observed that improvement in fertility and seed quality could be obtained by using F, durum hybrids as female parent in the development of primary amphiploids and also by using self-fertile ryes as male parents, as suggested by Sinchez-Monge (1959, 1968). Hybrid necrosis, a physiological disorder resulting from a combination of genes occurring in a single genotype, has created breeding and research problems in wheat (Hermsen, 1963). Gregory (1973) pointed out that similar problems occur in triticale, especially when crossing hexaploid triticale with bread wheats or octoploid triticale. He is now studying methods of overcoming these problems in triticale breeding. D. NOMENCLATURE

Triticale has become widely accepted as a common name to designate all allopolyploids derived from crosses between wheat (genus Triticum), and rye (genus Secale) . It includes octoploid, hexaploid, and tetraploid forms and both primary and secondary strains. The name triticale was reported by Lindschau and Oehler (1935) to have been coined on a suggestion by Tchermak for Triticum and Secale amphiploids, which at that time were all octoploids. They also proposed to add the name of the scientist developing the new form as a means of identifying different strains. According to Baum (1971 ), Wittmack proposed the name Tritico-secale for Rimpau’s stable derivatives from wheat-rye crosses in 1899. The scientific designation Triticum secalotriticum saratoviense Meister was proposed by Meister in 1930 (in Lewitsky and Benetzkaya, 1931). Kiss (1966) proposed that the scientific name be shortened to Triticum triticale. Larter et al. (1970) proposed Triticale hexaploide and Triticale octoploide for hexaploid and octoploid forms. Baum (1971 ) claimed that this nomenclature was unacceptable, but his proposal of Triticale turgidocereale appears to have been no more acceptable. The use of triticale as a generic name assumes the status of genus for the rye-wheat amphiploids. There appears to be little justification for a generic differentiation. Triticales and wheats cross readily, form partially fertile hybrids and possess two (AABB) or three (AABBDD) genomes from the genus Triticum. Furthermore as a result of meiotic irregularities, rye chromosomes are occasionally eliminated and the plants revert to wheat (Muntzing, 1957; Stutz, 1962). Such reverted forms are being used in

THE DEVELOPMENT OF TRITICALE

345

breeding programs at CIMMYT and at Michigan State University by F. Elliott. Furthermore the use of Triticale as a generic name introduces confusion by having the same word as the common and scientific name. Triticales should remain in the genus Triticum as proposed by Meister [Lewitsky and Benetzkaja (1931)l and Kiss (1966). The easiest solution would be to utilize the earliest suggestion by Wittmack (in Baum, 1971) and change it only sufficiently to make it scientifically acceptablefor example, Triticum secalum instead of Tritico-secale. The different polyploid forms could be differentiated by adding ploidy level names as suggested by Larter et al. (1970), that is f.s. octoploide for octoploids, f.s. hexaploide for hexaploids, and f,s. tetraploide for the tetraploid forms. If necessary, credit could be given to the first scientist to report the production of that particular form of triticale. It appears that Rimpau (1891) produced the first octoploid, Derzhavin (1938) the fmt hexaploid, and Krolow (1973) the first stable tetraploid form. E.

GENERAL COMMENTS

Although the first relatively stable triticale was produced in 1890, concerted breeding efforts to develop a commercial crop were not started until the early 1940’s. The development of new techniques in doubling chromosome numbers and culturing excised embryos made possible the development of primary hexaploid triticale in unlimited numbers. Muntzing developed octoploid winter triticales which were as productive as bread wheats on the poorer, sandier soils of southern Sweden. Kiss and Sinchez-Monge in Europe and Larter and Jenkins in North America developed hexaploid triticale strains that were competitive with bread wheats under some conditions. The greatest improvement in fertility was obtained among secondary hexaploids derived from crosses between octopIoid 2n = 56 and hexaploid 2n = 42 triticales or between bread wheat and hexaploid triticale. Triticale does not yet have a competitive advantage over wheat or other cereal grains except in some specific environments, such as high elevations, in areas where cool early growth temperatures prevail, and on sandy or low fertility soils. Considerable breeding work still remains to be done. The most serious agronomic problems are grain shriveling, preharvest germination, the tendency to produce few tillers under stress conditions, a narrow range of adaptation, and susceptibility to ergot. Considerable research is still required to improve its physical properties for the production of commercial food products. Unethical tactics used in the promotion of seed sales and overenthusiastic reporting have created a distorted image of the crop. Most research scientists agree that it has potential but is not yet ready for general produc-

346

F. J. ZILLINSKY

tion in competition with adapted varieties of other cereal crops. Rapid improvements in grain yield usually occur during the early years of triticale breeding programs but tend to fall off as the grain yields approach those of other cereals. However, this is being offset by the increasing number of scientists devoting their efforts to improving the crop. Initially the crop will be used as animal feed, both as grain and forage. Its use as human food will develop more slowly. New techniques are required if triticale flour is used as a substitute for wheat flour. Development research will be needed for the creation of new food products. Its potential for better balance in essential amino acids could be of great benefit in improving the nutrition of people in food-deficit areas of the world. ACKNOWLEDGMENT

I wish to express my sincere appreciation to Dr. Arne Muntzing for permission to use freely from his historical review of triticale in the preparation of this manuscript, and for his encouragement, advice, and cooperation in the CIMMYT triticale program. I wish to thank Mr. Gil Olmos of the CIMMYT photography section for the preparation of photographs and charts. REFERENCES Amaya, A. 1973. Int. Triticale Symp., C I M M Y T , 1973. Avila, E., Cuca, M., and Pro., A. 1971. A L P A M e m . 6, 29-35. Bauer, R. 1973. C l M M Y T Res. Bull. 24, 64-67. Baum, B. R. 1971. Euphytica 20,302-306. Bennett, M. D. 1973. Int. Triticale Symp., C I M M Y T , 1973. Bennett, M. D., and Kalsikes, P. J. 1973. Can. J . Genet. Cytol. 15, 671-679. Bennett, M. D. Chapman, J., and Riley, J. 1971. Proc. R o y . SOC.,Ser. B . 178, 259-275. Borlaug, N. E. 1968. Proc. lnt. Wheat Genet. Symp., 3rd, 1968 1-36. Bragg, D. B., and Sharby, T. F. 1970. Poultry Sci. 44, No. 4, 1022-1027. Brandes, D., Rimpau, J., and Robhelen, G. 1973. Proc. lnt. Wheat Genet. Symp., 1973. Briggle, L. W. 1969. Crop Sci. 9, 197-202. Briggs, C. 1973. Int. Triticale Symp., C I M M Y T , 1973. Chauhan, K. P. S., and Srivastava, J. P. 1973. lnt. Triticale Symp., C I M M Y T , 1973. Darvey, N. L. 1973. Proc. Int. Wheat Genet. Symp. 1973. Derzhavin, A. 1938. Izv. Akad. Nauk SSSR Ser. Biol. No. 3, pp. 663-665. Elliott, F. C . 1973. “Evaluation of Protein in Triticales.” Report to the Triticale Symposium at Lubbock, Texas. Elliott, F. C. 1974. “Triticale, First Man-Made Cereal” (C. C. Tsen, ed.) pp, 212-222. Amer. Assoc. Cereal Chem. Florell, V. H. 1936. J . Agr. Res. 52, 199-204. Fuentes, S. 1973. CIMMYT Res. Bull. 24, 34-38. Gregory, R. S. 1973. Int. Triticale Symp., C I M M Y T , 1973.

THE DEVELOPMENT OF TRITICALE

347

Gustafson, J. R., and Zillinsky, F. J. 1973. Proc. In!. Wheat Genet. Symp. 1973. Hermsen, J. G. 1963. Euphytica 12, 1-16. Ingold, M., Oehler, E., and Nogler, G. A. 1968. Z. Pflanzenzuch. 60, 41-88. Jenkins, B. C. 1969. Wheat Inform. Ser. 28, 18-20. Jenkins, B. C. 1974. “Triticale: First Man-Made Cereal” (C. C. Tsen, ed.), pp. 56-61. Amer. Assoc. Cereal Chem. Kaltsikes, P. J. 1971. Can. J . Genet. Cytol. 13, 656-662. Kaltsikes, P. J. 1973. In!. Triticale Symp. C I M M Y T , 1973. Kaltsikes, P. J. Evans, L. E,, and Larter, E. N. 1969. Can. J . Genet. Cytol. 11, 65-7 1. Kerber, E. R. 1964. Science 143, 253-255. Kies, C. and Fox, H. M. 1970. Cereal Chem. 47, 671. Kies, C., and Fox, H. 1973. Abstr. Annu. Meet. Amer. Assoc. Cereal Chem. Kiss, A. 1958. Dis. KecskemCt. (Sta. Bull.) Kiss, A. 1965. Actu. Agr. (Budapest) 14, 189-201. Kiss, A., 1966. Z . Pflanzenzuecht, 55, 309-329. Kiss, A. 1968. “Triticale.” Mezogazdasagi Kiad6, Budpest. Kiss, A. 1970. Wheat Inform. Serv. (Jap.) No. 31, pp. 24-25. Kiss, A. 1973. Int. Triticale Symp. C I M M Y T , 1973. Kostoff, D. 1938. Nature (London) 142, 573. Krolow, K. D. 1962. Z. Pflanzenzuecht. 48, 177-196. Krolow, K. D. 1963. Z. Pflanzenzuecht. 49, 210-242. Krolow, K. D. 1973. Int. Triticale Symp. C I M M Y T , 1973. Lacadena, J. R., and Perez, M. 1973. Proc. In!. Wheat Genet. Symp. 1973. Larter, E. N. 1968. Agr. Insf. Rev. 33, No. 2, 12-15. Larter, E. N. 1973. In!. Triticale Symp., C I M M Y T , 1973. Larter, E. N.. Hsam, S. L. 1973. Proc. Int. Wheat Genet. Symp. 1973. Larter, E. N., Tsuchiya, T., and Evans, L. 1968. I n t . Wheal Genet Symp., 3rd, 1968, pp. 213-221. Aust. Acad. Sci., Canberra. Larter, E. N., Shebeski, L. H., McGinnis, R. C., Evans, E., and Kaltsikes, P. J., 1970. Can. f. Plant. Sci. 50, 122-124. Lebedeff, V.N. 1934. Pflanzenzrcechtung 509-525. Lebsock, K. 1972. PSR-32-72, USDA, ARS Beltsville, Maryland. Lewitsky, G. A,, and Benetzkaja, F. K. 1931. Bull. Appl. Bot., Genet. Plant Breed. 27, No. 1, 241-264. Lindschau, M. U., and Oehler, E., 1935. Zurchter 7, 228-233. Lorenz, K., Welsh, J., Normann, R., and Maga, J. 1972. Cereal Chem. 49, 187-193. Lukyanenko, P. P. 1972. Proc. In!. Winter Wheat Conf. 1972. pp. 13-21. USDA. MacDonald, C. E. and Ahmend, S. R. 1973. Abstr. Annu. Meet. AACC. McGinnis, J. 1973. Int. Triticale Symp., C I M M Y T , 1973. MacAuliffe, T., and McGinnis, J. 1971. Poultry Sci. 1, No. 4, 1130-1134. MacKenzie, D. R. 1972. C I M M Y T Inform. Bull. No. 1 . Merker, A. 1971. Hereditas 68, 281-290. Merker, A. 1973a. Hereditas (in press). Merker, A. 1973b. Int. Triticale Symp., C I M M Y T , 1973. Munck, L. 1964. Hereditas 52, 151-165. Muntzing, A. 1935. Hereditas 20, 137-160. Muntzing, A. 1936. Zeuchter 8, 188-191. Muntzing, A. 1939. Hereditas 25, 387-430. Muntzing, A. 1956. Conf. Chromosomes Wageningen pp. 161-197.

3 48

F. J. ZILLINSKY

Muntzing, A. 1957. Cytologia, Suppl. 51-56. Muntzing, A. 1963. In “Recent Plant Breed Research at Svalof” (E. Abebert et al., eds.), pp. 167-178. Almquist and Wiksell, Stockholm. Muntzing, A. 1972. Biol. Zentralbl. 91, No. 1, 69-80. Muntzing, A. 1973a. Int. Triticale Symp., C I M M Y T , 1973. Muntzing, A. 1973b. Hereditas 74, 41-56. Nakajima, G. 1942. Proc. Imp. Acad. (Tokyo) 18, No. 2, 100-106. Nakajima, G. 1952. Bot. Mag. 65, 288-294. Nakajima, G. 1958. Kromosomo 34-36, 1194-1206. Nakajima, G. 1963. Kromosomo 55-56, 1829-1850. O’Mara, J. G. 1940. Genetics 25, 401-408. O’Mara, J. G. 1948. Rec. Genet. Sac. Amer. 17, 52. Pinto, F. F. 1973. Int. Triticale Symp., CIMMYT, 1973. Pissarev, V. 1963. Hereditas, Suppl. 2, 279-290. Pissarev, V., and Vinogradova, N. M. 1944. Akad. Nauk URSS 45(3). Pomeranz, Y., Burkhart, B. A., and Moon, L. C. 1970. Proc. Amer. SOC. Brew. Chem., 1970 pp. 40-46. Quiiiones, M . A. 1967. CIMMYT Res. Bull. 6. Quiiiones, M. A., and Rodriguex, R. 1973. Int. Triticale Symp., CIMMYT. 1973. Quiiiones, M. A., Larter, E. N., and Samborski, D. J. 1972. Can. J . Genet. Cytol. 14, 495-505. Rajaram, S., Zillinsky, F. J., Borlaug, N. E. 1972. Indian Phytopathol. 35, No. 3, 442-448. Richardson, M. J., and Walker, J. M. 1973. Int. Triticale Symp., C I M M Y T , 1973. Richardson, M. J., and Zillinsky, F. J. 1972. Plant Dis. Rep. 56, No. 9, 803-804. Rimpau, W. 189 1. “Kreuzungsprodukte landwirtschaftlicher Kulturpflanzen,” pp. 1-39. Parey, Berlin. Rupert, E. A., Rupert, J. A., and Beatty, K. D. 1973. Proc. Int. Wheat Genet. Symp. 1973. SSlnchez-Monge, E. 1956. An. Aula Dei 4, 191-207. SSlnchez-Monge, E. 1959. Proc. Int. Wheat Genet. Symp., 1st 1958 pp. 181-194. SSlnchez-Monge, E. 1969. Proc. Int. Wheat Genet. Symp., 3rd. 1968, pp. 371-372. SSlnchez-Monge, E. 1973. Int. Triticale Symp., CIMMYT 1973. Sisodia, N. S., and McGinnis, R. C. 1970. Crop Sci. 10, 161-164. Shulyndin, A. F. 1972. Ukr. Res. Inst. Plant Breed. Genet. Tech. Bull. Vol. 7, NO. 12, 61-74. Shulyndin, A. F. 1973. EUCARPIA Triticale Conf. Leningrad 1973. Strand, R. D. 1973. “Grade Standards for Triticale.” Report Amer. Ass. Cereal Chem., St. Louis, Missouri. Stutz, H. C. 1962. Genetics 47, 988 (abstr.). Tsen, C . C., Hoover, W. J. and Farrel, E. P. 1971. Abstr. 56th Annu. Meet. AACC. Villegas, E., Amaya, A., and Bauer, R 1973. CIMMYT Res. Bull. 24, 55-62. Wabwoto, N. 1973. Int. Triticale Symp., CIMMYT, 1973. Weiringa, G. W. 1967. Publ. 156. pp. 1-68. Inst. Storage Processing Agric. Produce, Wageningen. Wilson, A. S. 1875. Trans. Proc. Bot. SOC.Edinburgh 12, 286-288. Zillinsky, F. J. and Borlaug, N. E. (1971) CIMMYT Res. Bull. 17, 1-27. Zillinsky, F. J. 1973. Int. Triticale Symp., CIMMYT, 1973. Zillinsky, F. J., and Liipez, B. A. 1973. CIMMYT Res. Bull. 24, 12-32.

Subject Index Barley, 164, 180, 187, 197, 286, 293, 333 Barley yellow dwarf virus, 333 Abscissic acid, 70 Bean, African locust, 100 Acanthomia, 51 African yam, 87, 89-90, 95, 102 Aegilops sp., 323, 325 American yam, 102 Agromyza obtusa, 38 broad, 6 Agropyron cristatum, 284 cluster, 10, 78-79, 96, 99 Albedometer, 269 common, 83 Alfalfa, 204, 285, 304 Congo, 32 Alfisol, 243 dry, 2, 5, 6, 97 Allelopathy, 191-193, 201-202, 204 Egyptian kidney, 81 Allophane, 216, 217, 218-219, 221, four-angled, 88 223-224, 225, 227-230, 231, 232, goa, 88 233, 234, 235-2429 243, 244-246, haricot, 10 248, 249, 250, 251 horse-eye, 102 Allophane A, 227 hyacinth, 6, 7, 10, 76, 81-82, 95, 100 Allophane B, 226, 227 India butter, 81 Ancylostomia stercorea, 38 jack, 2, 83, 86, 87, 92, 93, 96, 101 Andosol, 248 jugo, 79 Anion exchange, amorphous clay, lima, 10, 87-88, 92, 95, 102 242-247 locust, 2, 81, 82-83 Anoplocnemis, 51 Manila, 88 Anthracnose, 53, 88 mat, 78 Aphis craccivora, 21 Mexican yam, 2, 87, 90-91 Arachis, 9 moth, 6, 7, 10, 78, 92, 96, 99 Arachis glabrata, 16, 20 mung, see mung bean Arachis hagenbeckii, 20 rice, 6, 7, 10, 11, 83, 86, 87, 92, 94, Arachis hypogaea, 6, 10, 11, 12, 13, 15, 101 16, 80, 97, 100 sword, 83, 86, 101 Arachis monticola, 16 tepary, 94, 99 Arachis repens, 20 velvet, 2, 10, 11, 91, 94, 96, 102 Arachis villosa, 16, 20 winged, 87, 88-89, 96, 102 Arhar, 32, 35 yam, 10, 11 Armyworm, fall, 23 Bean fly, 50 Ascorbic acid, 26 Bean leaf beetle, 51 Aspergillus gavus, 20, 22, 23 Bean mosaic virus, southern, 116 Aspergillus niger, 20, 23 Beetle, pulse, 69 Atylosia lineata, 33, 37 Belonlaimus gracilis, 53 Atylosia scarabaeoides, 33, 37 Benefin, 30 Aty losia sericea, 33, 37 Benomyl, 53, 54 Azinophosmethyl, 51 N-Benzyl-0-fluorophenoxy-acetamide, 70 Bermudagrass, 168 B Berseem, 76 Beta vulgaris, 284 Bacteria, soil retention of, 141-146 Biochemical oxygen demand (BOD), Bacterial pustule, 53, 84, 116 137-141, 145 A

349

350

SUBJECT INDEX

Biomass, productivity of mixtures, 177-210 soil, 138 Blackgram, 283 Boehmite, 214 Bonavist, 81 Boron, 28, 155-156 Botryodiplodia theobromae, 52 Brachiaria, 30 Breeding methodology, 305-3 10 cowpea, 46-50 5-Bromodeoxyuridine, 70 Bromus inermis, 282 Bruchideae, SO Brughus sp., 38, 52 C

Choenephora sp., 53 Chromium, 157, 158, 161 Cicer, 8 Cicer arietinum, 6 , 97 Claviceps purpurea, 33 1 Clavigralla gibbosa, 38 Clay, amorphous, 21 1-260 cation and anion exchange, 242-247 Cobalt, 73, 157, 162 Cochliobolus sativus, 333 Cocksfoot, 286 Colletotrichum cajanae, 39 Colletotrichum lindemuthianum, 53, 88 Colletotrichum sp., 52 Copper, 157, 158, 159, 160, 161, 162 Corn, 153, 165, 166, 297, 299, 300, 301, 307, 308 see also maize Corn rootworm, southern, 22 Cornstalk borer, lesser, 22 Cotton, 82, 94, 203, 283, 285, 287, 290 coupc, 44 Cowpea, 2, 6, 7, 10, 44-61, 76, 84, 91, 92, 94, 95, 97, 98, 100, 283 description, 45 management, 58-60 physiology, 54-58 pests and diseases, 50-54 plant improvement, 46-50 utilization, 60-61 Cowpea curculia, 51 Cowpea mottle virus, 116 Cowpea yellow mosaic, 116 p-CPA, 75 Cucumber mosaic virus, 116 Cyamopsis, 93 Cyamopsis psoralides, 78 Cyamopsis tetragonolobus, 78, 99 Cyst nematode, 84

Cadmium, 157, 158, 160 Cajanus, 9, 35, 37, 39, 41 Cajanus cajan, 6, 10, 32, 33, 97, 100, 103-107 Calcium, 27, 57, 73, 155 Callosobruchus chinensis, 69 Canarygrass, Reed, 153, 165, 168 Canavalia, 10, 93 Canavalia ensiformis, 86, 101 Canavalia gladiata, 86, 101 Canavalia plagiosperma, 86 Canavalia spp., 101 Carbofuran, 52 Caterpillar, hairy, 69 Cassava, 45 Cation exchange, amorphous clay, 242-247 Ceratoma trifurcata, 117 Cercospora arachidicola, 15, 20, 22 Cercospora canescens, 22, 53 Cercospora cruenta, 53, 116 Cerospora leaf spot, 13, 14, 22, 23 Cercospora personata, 22 D Cercospora sp., 39, 69 Cerotoma ruficornis, 51 Dactylb glomerata, 286-287, 294 Chalsodermus aeneus, 51 DDT, 75 Chemical oxygen demand (COD), Demosan, 54 137-141, 149 Dendorix sp., 51 Chick-pea, 5, 6, 76, 97 Diabrotica sp., 22 Chloramben, 60 Diacrisia obliqua, 69 Chlorite, 233 Diaporthe phaseolorum, 88 Chloroneb, 54 2,2-Dichloropropionic acid, 66

35 1

SUBJECT INDEX

Digitaria, 30 Dimethoate, 51 Dioclea, 10, 93 Dioclea reflexa, 91 Dioscorea spp., 89 Dithane M-45, 54 Dolichos, 9 Dolichos biflorus, 7, 10, 81 Dolichos lablab, 81-82 Dolichos uniflorus, 81, 82, 100 Dormancy, 26 Downy mildew, 88 E

Elasmopalpus lignosellus, 22 Elasmopalpus rubedinellus, 38 Eleusine corocana, 82 Endosulfan, 51 Environment-genotype interactions, 287-29 5 Ergot, 331, 342 Erisphe graminis, 333 Erisiphe polygoni, 69 Ethephon, 70 Ethylene, 25 Exelastis atomosa, 37, 38 F

Fescue, tall, 285 Festuca arundinaceae, 285 Field crops, mixture, productivity of, 177-210 Flax-linseed, 182, 187 Fluorine, 155 0-Fluorophenoxyoc-methylaceticacid, 70 Fly, bean, 69 Frijble, 44 Fungicide, 54 Furadan, 52 Fusarium graminearum, 333 Fusarium nivale, 333 Fusarium oxysporum, 115 Fusarium root rot, 88 Fusarium sp., 52, 203 Fusarium udum, 37, 38 Fusarum wilt, 115

G

Gandul, 32 Gardona, 51, 52 Genetics, cowpea, 67-68 peanut, 15-20 pigeon pea, 36-38, 103-118 quantitative relevance to breeding, 277-3 11 Gibberella fujikori, 342 Gibberillic acid, 70, 71 Gibberellin, 25 Gibbsite, 215, 232, 238, 241, 243, 250 Glycine, 10 Glycine max, 6, 10, 11, 83, 97, 101 Glyricidia, 3 1 Goober, Congo, 79 Gram, Asian, 6, 7, 8, 97, 98 black, 6, 7, 44, 62, 63, 65, 67, 73, 75, 76, 91, 100 golden, 62, 75 green, 7, 62, 63, 65, 67, 75, 76, 91 horse, 7, 10, 81, 92, 94, 96, 100 madras, 81 red, 32 yellow, 7 Grasses, 180, 182, 187 Groundnut, 10 Bambarra, 79-80, 95, 99 Kersting, 80, 99 Guar, 78-79, 94 Guerte, 79 Gusathion, 51 Gyarnopsis tetragoizolobus, 10 H

Halloysite, 222, 223, 233, 237, 241, 243 Halo bright, 69 Helicotylenchus pseudorobructus, 53 Heliothis armicera, 38 Heliothis sp., 38, 50 Heliothis virescens, 38 Helminthosporium sp., 53 Hemiptera spp., 50 Heterodera sp., 84 Heterosis, 285-287 Hisingerite, 217-218, 236

352

SUBJECT INDEX

Horsegram, 10 see also gram, horse Humate, sodium, 73 I

Imogolite, 218, 219-221, 224, 225-226, 227-230, 232, 233, 236, 237, 238, 243, 244-246, 249 Indoleacetic acid, 70 Infrared radiation, 66 Insecticides, 21, 38, 51-52 Iron, 28, 159, 161 K

Kaolin, 217 Kaolinite, 227, 243 Kerstingielia, 9, 93 Kerstingiella geocarpa, 80, 99 Kinetin, 70 1

Lablab, 81 Lablab, 9 Lablab niger, 10, 76, 81, 100 Labtab vulgaris, 82 Lampides sp., 51 Lannate, 52 Laspegresis pychora, 38, 50 Lathyrus, 8 Latosol, 241 Lead, 157,158, 161 Leaf blight, 333 Leaf blotch, 333 Leaf rust, 331 Leaf spot, 53, 69, 116 Leaf stripe, 332 Legume, taxonomy, 8-10 tropical grain, 1-132 Lens, 8 Lens esculenta, 6, 97 Lentil, 6, 97, 283 Leptosphearula sp., 53 Leveillula taurica, 39 Limonite, 214 Lindane, 51 Linuron, 30 Lodging, 329-331

Lolium perenne, 282 Lotus tetragonolobus, 88 Lubia, 44 Lubia, seim, 81 Lysine, 341 M

Macrophomina, 23 Macrophomina phaseoli, 20, 39, 69 Magnesium, 57, 73 Maize, 45, 59, 81, 285, 286, 287, 302, 303 see also corn Manganese, 28, 73, 159, 161 Maruca testulalis, 38, 50 MCPE, 76 Melangromyza phaseoli, 50, 69 Melanagromyza vignalis, 50 Meloidogyne incognita, 53, 84, 117 Mercury, 157, 158 Methomyl, 52 Mildew, downy, 69 Millet, 45, 58 pear, 78 pearl, 283 Molybdenum, 28, 41,73 Monoculture, 179-183, 193-194 Montmorillonite, 233, 244, 251 Mucuna, 9, 93 Mucuna sloanet, 102 Mucuna pruriens, 91 Mucuna pruriens var. utilis, 11, 102 Mucuna sloanei, 11, 91 Mucuna spp., 10 Mucuna urens, 91 Mullite, 227 Mung bean, 62-76, 91, 95, 100, 283 description, 63-65 management, 75-77 physiology, 70-75 plant improvement, 65-68 N

NAA, 75 Nematode, 23, 53 cyst, 69 root knot, 53 Neoscosmopora vasinfecta, 53

SUBJECT INDEX Nickel, 157, 158, 160, 161 Nicotiana otophora, 286 Nicotiana rustica, 293 Nicotiana tabacum, 286 Nicotiana tomenfosiformis, 286 Niebe, 44 Nitrofen, 30 Nitrogen, 27, 135, 146-151, 162, 189, 190 waste water, 146-151, 164-165 Nitrogen fixation, cowpea, 57-58 mung bean, 74-75 NOA. 75 0

Oat, 164, 166, 197, 283, 303, 333 Ootheca mutabilis, 50 Opaline silica, 213-214, 233-234 Ophiobolus graminis, 333 Orthene, 52 Oxisol, 243 Oxygen, soil, 138-141 P

Pachyrrhizus, 9, 93 Pachyrrhizus erosus, 90, 91, 102 Pachyrrhizus spp., 88 Pachyrrhizus tuberosus, 90, 91 Parkia, 93 Parkia biglabosa, 82 Parkia clappertonia, 82 Parkia filicoides, 82 Parkia oliveri, 82 Parkia spp., 100 Pea, asparagus, 88 Australian, 82 blackeye, 44 dry, 6 earth, 79 kaffir, 79 pigeon, 2, 5, 6, 7, 10, 11, 32-44, 92, 93, 95, 97, 98, 100, 103-118 princess, 88 southern, 44 Peanut, 2, 6-7, 11-32, 91, 93, 97, 97, 98, 100 botanical, 12-14 diseases, 22-24 growth process, 24-29

353

insect pests, 21-22 management, 29-3 1 plant improvement, 14-20 seed composition, 3 1 Pesticide, plant uptake, 165-166 Phaseolus, 9, 92, 93, 96 Phaseolus acontifolius, 8, 74, 78 Phaseolus acutifolius var. latifolius, 79, 99 Phaseolus angularis, 8 Phaseolus aureus, 8, 62, 74, 75 Phaseolus calcaretus, 8, 11 Phaseolus lunatus, 87, 93, 102 Phaeolus manihotis, 39 Phaseolus mungo, 8, 74 Phaseolus radiatus, 8 Phaseolus vulgaris, 6, 10, 75, 83, 97, 101 Phosphate, 246 Phosphorus, 27, 41, 57, 58, 59, 60,72-73, 189 wastewater, 151-1 55 Photoperiod, 55, 110 Physalospora cajanae, 39 Phytophthora phaseoli, 88 Piezotrachelus varium, 51 Pigeon pea, 2, 5, 6, 7, 10, 11, 32-44, 92, 93, 95, 97, 98, 100, 103-118 botanical, 33-34 genetics, 103-118 pests and diseases, 38-40 physiology and management, 40-44 plant improvement, 34-38 Pisum, 8 Pisum spp., 6, 97 Plant relative yield (PRY), 185 Pod blight, 88 Polychlorinated biphenyl, 166 Potash, 27, 57, 59 Potassium, 57, 60, 73, 74, 189 Potassium azide, 54 Potato, 294 Powdery mildew, 116, 333 Pratylenchus brachyurus, 23 Pratylenchus sp., 53 Prometryne, 30 Pseudomonas phaseolicola, 69 Pseudomonas solancearum, 20 Pseudomonas striafaciens, 332 Psophocarpus, 10, 94 Psophocarpus tetragonolobus, 88, 102

354 Puccinia Puccinia Puccinia Puccinia Pumice, Pythium Pythium

SUBJECT INDEX arachidis, 20, 22 glumarum, 33 1 graminis, 33 1 recondita, 3 3 1 218, 221, 225, 236, 237 aphanidermatum, 52 sp., 54

R Radiometer, calibration and use, 261-275 Ragi, 82 Relative yield total (RYT), 186, 193, 195, 196, 199, 200-204 Rhizobial symbiosis, 27, 42 Rhizobium, 75 Rhizobium japonicum, 84 Rhizoctonia bataticola, 23, 53, 115 Rhizoctonia solani, 52, 204 Rhizocotnia sp., 54 Rhizopus nigricuns, 23 Rice, 63, 75, 81, 93, 180, 187, 202 Rogor, 40, 51 Root knot nematode, 84, 117 Rosette virus, 21, 23 Rotylenchus reniformis, 39 Rush, 144, 168 Rye, 180, 187, 285, 316, 322, 328, 339 Ryegrass, 181 S

Salt, waste water, 162-163 Scirpus lacustris, 144, 168 Sclerophthora macrospora, 333 Sclerotium bataticola, 20 Sclerotium rolfsii, 20, 22, 54, 69 Secale cereak, 318 Secale montanum, 3 18 Septoria tritici, 333 Sericothrips occipetalis, 50 Short-wave balance meter, 269 Silica, opaline, 213-214, 233-234 Smut, 333 Soil, amorphous clay, 21 1-260 Solarimeter, 269 Sorghum, 45, 59, 78, 82, 92, 283, 294 Soybean, 2, 4, 5, 6, 10, 11, 83-85, 87, 93, 95, 97, 101, 166, 194, 283, 290, 303

Soybean mosaic, 84 Spartina Townsendii, 144 Sphenoptera sp., 51 Sphenostylis, 9, 93 Sphenostylis stenocarpa, 10, 11, 89, 102 Spodoptera frugipcrda, 22 Spodoptera spp., 50 Spodozol, 23 1 Stem blight, 53 Stem rust, 331 Striga gesnerioides, 53 Stripe rust, 331, 332 Subterranean clover, 187 Sudangrass, 59 Sugar beet, 284 Sulfur, 28, 73 Sweet potato, 87 T

Taeniothrips sjostedi, 50 Tassidedes, 5 1 Thiodan, 51, 52 Thrip, 52 2-Thiouracil, 70 Thymidine, 70 Tobacco, 283, 285, 286, 290, 301 Tobacco ringspot virus, 117 Total organic content (TOC), 137, 138, 141 Trifluralin, 30, 60 Triticale, development of, 3 15-348 Triticum aestivum, 294 Triticum monococcum, 324 Triticum secalotricum saratoviense, 3 16, 344 Triticum secalum, 345 Triticum timopheevi, 323 Triticum triticale, 344 Triticum turgidum, 3 17, 3 18 Tropical agricultural, grain legume, 1-132 Tur, 32, 35 Tylenchorhyachus sp., 39 U

Ultraviolet radiation, 66 Uracil, 70 Uromyccs appendiculatus, 53, 69

355

SUBJECT INDEX Uromyces piiaseofi, 88 Uromyces phaseoli var. vignae, 54 Uromyces sp., 39 Ustilago spp., 333 V

Vermiculite, 23 3 Verticillium dahliae, 20 Vicia, 8 Vicia faba, 6, 97 Vigna, 8, 9, 81, 92 Vigna acontifolia, 10, 78, 99 Vigna embellata, 93 Vigna radiata, 62, 100 Vigna radiata var. aureus, 10 Vigna radiata var. mungo, 10 Vigna sesquipedalis, 47 Vigna sp., 7 Vigna trinervius, 62 Vigna umbellata, 10, 11, 101 Vigna unguiculata, 6, 10, 44, 69, 97, 100 Virus, barley yellow dwarf, 333 cowpea mottle, 116 cowpea yellow mosaic, 53, 54, 116 cucumber mosaic, 54, 116 green mottle, 53, 54 rosette, 21, 23 soil retention of, 141-146 southern bean mosaic, 116 soybean mosaic, 84 tobacco ringspot, 117

A 4 6 5 C 6

0 7 € 8 F 9 G O

H 1 1 2 J 3

yellow bean mosaic, 54, 84 yellow mosaic, 53 Voandzeia, 9, 80, 93 Voandzeia subterranea, 10, 79, 99 Voandza, 79 Volcanic ash, 218, 219, 223, 225, 233, 235-239, 247, 249, 250-251 W

Wastewater, land treatment, 133-176 Water, 201 land disposal, 133-176 Weedicide, 76 Wheat, 164, 166, 180, 187, 283, 290, 316, 332, 333, 342 bread, 340 dururn, 318 hard red winter, 294 Wheatgrass, crested, 284 White clover, 190 X

Xanthemonas phaseoli, 84 Xanthomonas translucens, 332 Xanthomonas vignicota, 53, 1I6 L

Zinc, 157, 158, 159, 160, 161, 162 Zonocerus spp., 50

This Page Intentionally Left Blank