Advances in Agronomy, Volume 21

ADVANCES IN AGRONOMY VOLUME 21 CONTRIBUTORS TO THIS VOLUME F. J. CARLISLE D. J. GREENLAND R. B. GROSSMAN S. B. HEA...

Author: N. C. Brady

21 downloads 1091 Views 18MB Size Report

This content was uploaded by our users and we assume good faith they have the permission to share this book. If you own the copyright to this book and it is wrongfully on our website, we offer a simple DMCA procedure to remove your content from our site. Start by pressing the button below!
Report copyright / DMCA form

DOWNLOAD PDF

ADVANCES IN

AGRONOMY VOLUME 21

CONTRIBUTORS TO THIS VOLUME

F. J. CARLISLE D. J. GREENLAND

R. B. GROSSMAN S. B. HEATH CHARLES E. KELLOGG

OLIVER E. NELSON J. M. OADES ARNOLD C . ORVEDAL W. F. RAYMOND

G. D. SWINCER R. W. WILLEY

ADVANCES IN

AGRONOMY Prepared under the Auspices of the

AMERICAN SOCIETY

OF

AGRONOMY

VOLUME 21

Edifed by N. C. BRADY Roberts Hall, Cornell University, ithaca, New York

ADVISORY BOARD

R. R. DAVIS F. A. HASKINS W. D. KEMPER

.I.P. MARTIN

J . W. PENDLETON W. A. RANEY 1969

@

ACADEMIC PRESS 0 N e w York and London

COPYRIGHT ‘C 1969,

BY

ACADEMIC PRESS, INC.

ALL RIGHTS RESERVED. N O PART OF THIS BOOK MAY BE REPRODUCED IN ANY FORM, BY PHOTOSTAT, MICROFILM, RETRIEVAL SYSTEM, OR ANY OTHER MEANS, W I T H O U T WRITTEN PERMISSION FROM T H E PUBLISHERS.

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

United Kingdom Edition published by ACADEMIC PRESS, INC. (LONDON) LTD. Berkeley Square House, London W 1X 6BA

LIBRARY OF

CONGRESS CATALOG C A R D

NUMBER50-5598

PRI N TED IN T H E UNITED STATES O F AMERICA

CONTRIBUTORS TO VOLUME 21 Numbers in parentheses indicate the pages on which the authors’ contributions begin.

F. J . CARLISLE ( 2 3 7 ) , Soil Conservation Service, United States Department of Agriculture, Hyattsville, Maryland D. J . GREENLAND (195), Depurtment of Agricultural Biochemistry and Soil Science, Waite Agricultural Research Institute, University of Adelaide, South Australia R. B. GROSSMAN ( 2 3 7 ) , Soil Conservation Service, United States Department of Agriculture, Lincoln, Nebraska S . B. HEATH(28 11, Department of Agriculture, University of Reading, Reading Berkshire, England CHARLES E. KELLOGG( I09), Soil Survey, Soil Conservation Service, United States Department of Agriculture, Washington, D.C. OLIVER E . NELSON* (17 l ) , Purdue University, Lafayetre, Indiana J . M . OADES(195), Department of Agricultural Biochemistry and Soil Science, Waite Agricultural Research Institute, University of Adelaide, South Australia ARNOLDC. ORVEDAL ( 109), Soil Survey, Soil Conservation Service, United States Department of Agriculture, Washington, D.C. W. F . RAYMOND( I ) , The Grassland Research Institute, Hurley, England G . D. SWINCER (195), Department of Agricultural Biochemistry and Soil Science, Waite Agricultural Research Institute, University of Adelaide, South Australia R. W. WILLEY t (28 I ) , Department ofAgriculture, University of Reading, Reading Berkshire, England

*Present address: University of Wisconsin, Madison, Wisconsin. t Presenr address: Makerere University College, Kampala, Uganda. V

This Page Intentionally Left Blank

PREFACE

This volume marks a significant milestone in the history of Advances in Agronomy. The first 20 volumes were compiled under the very capable editorship of Dr. A. G. Norman, now Vice-president for Research at the University of Michigan. Mounting pressures of other responsibilities have prompted Dr. Norman to ask to be relieved as editor of this serial publication; this is the first volume that does not carry his name. It is fitting that we reflect briefly on the contributions he has made, not only to this work but to his profession as well. As editor of Advances in Agronomy, Dr. Norman has given 20 years of faithful service and leadership to agronomists and soil and crop scientists throughout the world. His extraordinarily good judgment in the selection of authors and of subject matter has been largely responsible for the success of this publication. His guidance to authors has helped both them and the quality of their papers. He has seen Advunces in Agronomy grow from a struggling review journal of concern only to American scientists to a prominent review series with contributors and subscribers in many nations. Dr. Norman has found other ways to benefit his profession. He has contributed directly as an active researcher in soil microbiology and in soil and plant biochemistry. He has served as director of a large, interdisciplinary research unit and has enriched the education and training of many soil and crop scientists as well as biologists. We are also indebted to Dr. Norman for his service in scientific societies. He served as vice-president and later president of the American Society of Agronomy during a very critical period in the Society’s history. In addition, for a period of three years he served as chairman of the Division of Biology and Agriculture of the National Research Council. Even though Dr. Norman has resigned his editorial responsibilities, Advances in Agronomy fortunately will reflect his influence for some time to come. The challenge of maintaining Dr. Norman’s high standards and the broad subject matter coverage he provided is materially aided by the efforts of such men as the eleven who have contributed to this Volume 2 I.

N. C. BRADY Ithaca, New York August, I969

vii


CONTENTS CONTRIBUTORS TO VOLUME 21

.

.

.

.

.

.

.

.

.

.

.

PREFACE .

.

.

.

.

.

.

.

.

.

.

.

vii

2 3

.

.

.

.

V

THE NUTRITIVE VALUE OF FORAGE CROPS W . F . RAYMOND

. . . . . . . . . . . I . Introduction . . . . . . . I 1 . The Components of Nutritive Value . . . . . . 111. The Digestibility of Forage Crops . . . . . IV . The Digestibility of Different Forage Species . . . . . . V . The Voluntary Intake of Forages . . . VI . The Efficiency of Utilization of Digested Nutrients . VII . The Relationship between Forage Quality and Forage Yield . . . . VIII . Forage Breeding for Improved Nutritive Value . IX . The Effects of Processing on the Components of Forage . . . . . . . . . . Nutritive Value . X . The Nutritive Value of Grazed Forage . . . . . . References . . . . . . . . . . . .

.

.

.

.

. . .

. . .

.

.

. .

. .

65

4 16

27 38 68

.

.

.

.

69 80

.

.

97

. . . . .

. . . . .

POTENTIALLY ARABLE SOILS OF THE WORLD A N D CRITICAL MEASURES FOR THEIR USE CHARLES E. KELLOGGA N D ARNOL.D C . ORVEDAL

I . Introduction . . . . . . . . I 1 . The Principle of Interactions in Soil Use . 111 . Higher Production from Existing Arable Soils IV . New Potentially Arable Soils . . . . References . . . . . . . .

. . . . .

. . . . .

. . . . .

. . . . .

109

I12 122 140 169

GENETIC MODIFICATION OF PROTEIN QUALITY I N PLANTS OLIVERE . NELSON

I . lntroduction . . . . . . . . . . . . I I . The Genetic Control of Protein Structure . 111 . The Relative Constancy of Leaf Protein Composition

ix

. .

. .

. .

.

.

.

.

.

.

171 173 177

X

CONTENTS

IV. The Storage Proteins of Seeds . . . . V. Theopaque-2 andJloury-2 Mutations in Maize VI. The Prospects of Improvements in Other Plants VI1. Summary . . . . . . . . References . . . . . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

178 I80 I87 I90 191

THE EXTRACTION, CHARACTERIZATION, A N D SIGNIFICANCE OF SOIL POLYSACCHARIDES G. D. SWINCER, J. M. OADES,A N D D. J. GREENLAND I. Soil Carbohydrates . . . . . . . . . . 11. The Significance of Soil Polysaccharides . . . . . I l l . Studies on Soil Polysaccharides . . . . . . . IV. Methods for the Analysis of Complex Polysaccharide Materials V. Summary and Conclusions. . . . . . . . . References . . . . . . . . . . . .

.

.

. .

. .

.

.

. .

. .

195 196 I99 222 229 230

FRAGIPAN SOILS OF THE EASTERN UNITED STATES R. B. GROSSMAN A N D F. J. CARLISLE 1. 11. 111.

IV. V. VI. VII. VIII. IX.

x.

Introduction . . . . . . Horizons of Fragipan Soils . . Occurrence of Fragipan Soils . . Properties of Fragipans . . . Fragipans and the Soil Water Regime . . . Genesis of Fragipans . Fragipans and Soil Use . . . Classification of Fragipan Soils. . Unresolved Problems. . . . Summary . . . . . . References . . . . . . Appendix . . . . . .

. . .

. . .

. . .

. . .

. . .

. . .

. . .

. . .

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

. .

. .

. .

. .

. .

. .

. .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . . . .

231 240 244 246 254 256 263 265 269 27 I 272 276

THE QUANTITATIVE RELATIONSHIPS BETWEEN PLANT POPULATION A N D CROP YIELD R. W. WILLEYA N D S. B. HEATH

I. 11.

Introduction . . . . . . . . . . Relationships between Plant Density and Crop Yield.

.

.

.

.

.

.

.

.

28 I 283

xi

CONTENTS

I l l . The Relationship between Plant Rectangularity IV. The Variation in Yield of the Individual Plant . . . . . . . V. Conclusions . . . . . . . . References .

AUTHORI N D E X . SUBJECT INDEX .

. .

. .

. .

. .

. .

. .

. .

. .

. . .

. . .

. . .

. . .

. . . .

.

. . .

3 14 317 319 3 20

. .

. .

. .

. .

. .

. .

323

and Crop Yield

338


THE NUTRITIVE VALUE OF FORAGE CROPS W. F. Raymond The Grassland Research Institute, Hurley, England

I. 11. 111.

IV.

V.

VI.

VII. v111. IX.

X.

Page Introduction. ...... ........................................................... 2 The Components of Nutritive Value .................................. .... 3 The Digestibility of Forage Crops ........... .................. 4 A. The Measurement of Digestibility in .................. 4 B. The Prediction of Forage Digestibility from Chemical Composition ....... 6 .... 7 C. Improved Chemical Techniques ..... . . . . . ................... ..... D. Estimation of Forage Digestibility by in Vitro Techniques ................. 10 E. The Relative Utility of Chemical and in Virro Estimations of Forage Digestibility _ . _ _____ . _,_. .._ _ _ _ ___._.__..__, _ _ _ _ . .._....._.___. ._.......... 16 The Digestibility of Different Forage Species ..._...._.... ... . ..... 16 16 A. Basic Patterns of Digestibility ....................... B. The Digestibility of Different Plant Fractions .. _........_____..._. 20 C. The Effect of Environmental and Other Factors on ...... ......... Forage Digestibility , . , . , . . . . . . . _ ._ . _ _ _ _ _._. ._. . _. . ,_..., __............. 23 .............................................. 27 The Voluntary Intake of Forages 27 A. The Factors Controlling Feed Intake ........................... 28 B. Intrinsic Factors Determining Forage Intake ..._.. .........._ 35 C. The Nutritive Value Index ........................................................... D. The Crude Protein Content of Forage and Voluntary Intake .............. 36 E. The Effect of Supplementary Feeds on Forage Intake ..... 37 The Efficiency of Utilization of Digested Nutrients. .................. 38 A. Methods of Expressing Energy Values ....... . , , .. .. .. ................ . .. ........ 38 B. The Role of Volatile Fatty Acids in Ruminant 40 C . The Utilization of the Crude Protein in Forages .................. 48 D. The Use of Nonprotein Nitrogen in Ruminant 51 E. The Mineral Nutrients in Forages ................................................. 54 F. Pharmacologically Active Components in Forage .................. 61 The Relationship between Forage Quality and Forage Yield ...... .............. 65 Forage Breeding for Improved Nutritive Value ....... . . ........... ..... 68 The Effects of Processing on the Components of Forage Nutritive Value , . _.. .._,, . . . _ .. . . . . . , , _ _ _...... . . ..., . . . . . ............. .............. 69 .......................................... 69 B. The Grinding and Pelleting of Dehydrated Forages ........ 71 C . The Ensiling of Forage Crops .......... ................................... 74 The Nutritive Value of Grazed Forage .... ................................... 80 A. Measurement of the Nutrient Intake by Grazing Animals ..... 80

2

W. F. RAYMOND

B. The Botanical Composition of Grazed Forage ................................. C. The Nutrient Intake of Grazing Livestock ...................................... D. The Effect of Management on the Productivity of Grazing Animals ........................................................................ References .......................................................................................

90 91 94 97

1. Introduction

Forages are grown for ruminant feeding, most ruminant animals eat forages. Thus a review of the nutritive value of forages is essentially a review of ruminant nutrition, yet with the difference that the nutritionist can treat the animal and the forage it eats in isolation, whereas the agronomist must also consider the problems that arise when the animal and its feed are brought together in practical systems of forage production and utilization. Ruminant feeding to date has been a nonintensive system of land use, in comparison with crop farming or the feeding of nonruminant livestock. This has been justified on the basis that forages are cheap to grow, and that harvesting by grazing is a cheap method of utilization. However, as Melville ( 1960) emphasized, present extensive pastoral systems produce very low outputs of human food per acre; as world demand for food increases, pastures will either have to become markedly more productive, or be replaced by crops that can be used by nonruminants, or directly by humans. In the latter cases forage-ruminant systems would be confined to noncultivable agricultural areas, and so would contribute only marginally to the nutrition of the world’s population. At that time of apparent food surpluses the replacement of ruminant products seemed remote; today analogue substitutes are already serious competitors to meat and milk in North America, the most sophisticated consumer market in the world. To date this competition has been in terms of cost and convenience; in future it will increasingly be in terms of competition for land, as foreseen by Melville. This means that the efficiency of soil-forage-ruminant systems must be greatly increased if they are to continue as a significant sector of agriculture. Raymond (1968) has considered this problem in terms of (a) the efficiency of use of incident light energy by the growing plant, (b) the proportion of the energy in the plant which is actually eaten by the ruminant animal, and (c) the efficiency with which different animal populations convert the energy they eat into products which can be used by humans. In many cases it appears that stages (b) and (c) are the main factors limiting the output of ruminant products per acre; until we can

THE NUTRITIVE VALUE OF FORAGE CROPS

3

ensure that a high proportion of the forage grown is eaten by efficient animals, there may be little advantage in concentrating effort on growing more forage. Efficiency of feed conversion (c) depends on many factors, including the structure of the animal population (the proportion of adult breeding animals to productive offspring; Spedding, 1965) and the genetic potential of the animals. But a dominant factor is the level of nutrient intake of the animals being fed: the higher the level of nutrient intake, the higher the level of productivity of the animals, and the lower the nutrient requirement for each unit of animal output. Thus, as the daily nutrient intake of the 300-kg. steer increases from 15.3 to 20.3 Mcal. of metabolizable energy, its daily rate of liveweight gain increases from 0.5 to 1.25 kg. per day: the corresponding requirement of metabolizable energy per kilogram gain decreases markedly from 30.6 to 16.2 Mcal. (Raymond, 1968). II. The Components of Nutritive Value

Thus the nutritive value of a forage should be considered not as a single parameter, but as composed of a complex of parameters that determine the nutrient intake of ruminant animals fed on that forage. In this it differs from the classical concept of nutritive value as a feed concentration (starch equivalent, total digestible nutrients, or net energy) by including feed intake as an integral component of nutritive value. With the major economic feeds of earlier feeding systems (cereals and pulses, oilseed residues, and industrial by-products) this was not necessary, as the quantity eaten was controlled by rationing; with forages, on the other hand, there is seldom any formal control of the amount eaten, which therefore depends on factors in the forage and in its method of presentation. This review therefore considers the nutritive value of forages in terms of the factors that determine the level of nutrient intake by ruminant livestock. It has proved useful to treat nutrient intake as the product of three parameters (Raymond, 1969b): Nutrient intake

= intake X

of feed x digestibility of feed efficiency of utilization of digested feed

(1)

each of which can be investigated separately, before their interactions in practical systems of ruminant feeding are considered. The importance of this approach is indicated by the conclusion of lngalls er al. ( 1 965) that 70 percent of the variation in production potential between forages can be accounted for in terms of differences in voluntary intake, compared with 30 percent by differences in digestibility, the nutrient concentration

4

W. F. RAYMOND

measure of earlier systems. Equation (1) is also possibly more informative than the analogous nutritive value index (Crampton et al., 1960), which combines intake and digestibility in one parameter, so that the relative importance of these two factors is not immediately evident. However, nutritive value and nutrient intake can have no real meaning except in relation to the needs of ruminant animals. Different animals have different nutritional requirements, depending on their species, sex, physiological status, and level of production, and this means that the nutritional adequacy of a forage diet can be assessed only in terms of the nutritional needs of the particular animals to be fed. The requirements by different classes of stock for energy, protein, minerals, and vitamins have been tabulated (Morrison, 1957; Agricultural Research Council, 1965; National Academy of Sciences, 1966). The objective must then be to establish relevant parameters to describe the nutritive value of forages which can be equated with these nutrient needs. The components in Eq. (1) provide a framework within which to assess our current knowledge of these nutritional parameters. Of these components, the digestibility of forage is considered first, because of the important influence which digestibility exerts on the other two components, intake and efficiency of utilization. These components are discussed in relation to fresh forages, but particular emphasis is then given to the effects of processing methods, feed interactions, and methods of feeding, all of which can markedly alter the basic nutritional features of forages. The practical aim must be to exploit this new information so as to improve the nutritional potential of forage feeding systems, and the effectiveness with which soil-forage-ruminant systems can compete for the world’s increasingly scarce land resources. Ill. The Digestibility of Forage Crops

A. THEMEASUREMENT OF DIGESTIBILITY in

ViVO

The digestibility of a feed is defined: Digestibility = ‘Ieed

- cfses

X

100

cfeed

Where Cfed is the amount of feed or feed component eaten (organic matter, cellulose, protein), and Cfecesis the corresponding amount of fecal excretion. The measurement of digestibility requires a preliminary feeding period during which the experimental animals adapt to the feed under test, followed by a test period, during which feed eaten and fecal output are measured. For precise measurement preliminary and test


5

periods of at least 10 days are recommended (Raymond et al., 1953); this presents the particular difficulty with fresh forages that the forage must be cut daily, and so may change in digestibility and chemical composition during this experimental period. In many studies this day-to-day variation in feed characteristics has been overcome by cutting at one time sufficient fresh forage for the complete digestibility experiment, and preserving this forage so that it can be fed over an extended period. Storage as hay (J. R. Jones and Hogue, 1963) or after artificial drying (Kivimae, 1959) has been used but, because of the changes in digestibility possible with these methods, cold storage of forages has been adopted by some workers (Raymond et al., 1953; Pigden et al., 196 1 ; Minson, 1966). The technique of storage at 5°F. has been described in detail (Commonwealth Agricultural Bureaux, 1961, pp. 88 and 150); it has been shown to have a negligible effect on the digestibility of the dry matter or organic matter in forage (Raymond et al., 1953) or of the rate of digestion within the rumen (Pigden et al., 1961), but slightly reduces the digestibility of the crude protein fraction (Raymond et al., 1953; Minson, 1966). An alternative technique, the continuous digestion trial with fresh forage, is now being increasingly widely used (Greenhalgh er al., 1960; Commonwealth Agricultural Bureaux, 196 1 ; Ademosum et al., 1968). Herbage is cut and fed daily over an extended period, and the amounts of forage eaten and feces voided are measured daily throughout the experiment. The amounts of forage eaten and feces are summed over 5day subperiods, allowing a 2-day lag for passage of the feces, and digestibility coefficients are calculated on these subperiods, each of which serves as the preliminary (adaption) treatment for the succeeding subperiod. This technique has proved of particular use in association with grazing experiments (see Section X,A,2), but it is less accurate than the cold-storage technique because of the shorter balance periods used. The measurement of the digestibility of forages conserved by natural or artificial dehydration presents no such problem, and most of the reported data on forage digestibility relate to such feeds. Silage is generally removed daily from silos for feeding, but cold storage of silage (Harris and Raymond, 1963) requires less labor, and eliminates any risk of day-to-day variation in silage quality. The many thousands of recorded determinations of forage digestibility have been collated at intervals and provide the broad background to our present understanding of forage nutritive value (Schneider, 1952; Leitch, 1969; tropical forages, Butterworth, 1967). However, such compila-

6

W. F. RAYMOND

tions may be of limited value in indicating the digestibility of an “unknown” forage, because of the difficulty of identifying it with a particular feed class. This problem, long recognized, led to the development of techniques such as the Weende feed analysis for estimating nutritive values; a major advance in the period under review has been in the development of improved laboratory techniques for predicting the nutritive value of forages, to replace wherever possible the laborious and expensive in vivo determination.

B. THE PREDICTION OF FORAGE DIGESTIBILITY FROM CHEMICAL COMPOSITION As Van Soest (1968) has noted, animal nutrition has had a history of inertia and complacency, each further experiment carried out with old techniques and old terminologies making it yet more difficult to adopt new ones. But it is still difficult to create a logical pattern from the torrent of new analytical techniques and new parameters of nutritive value that have recently been put forward to replace these older concepts. The requirement is to establish a relationship between a nutritional parameter (e.g., digestibility) of forages, measured in controlled in vivo experiments, and the chemical composition of the same forages, from which the nutritive value of other forages can be predicted. Digestion of forage by the ruminant is a most complex process; yet for nearly a century the attempt was made to predict the extent of forage digestion in terms of its proximate analysis based on Weende crude fiber, crude protein and nitrogen-free extractives. Sullivan (1 962) and Dijkstra ( 1 966) have both shown that when these analyses are applied to a limited range of forages close relationships between digestibility and chemical composition can be established, but that these relationships become less precise as the range of forages included is increased. As a forage crop matures its fiber content increases and it becomes less digestible; a close negative relationship between fiber content and digestibility is found. But this relationship is likely to differ from that with a different forage species (in particular, tropical forages; Butterworth, 1963) or from that with the same forage species at a different time of year; in each case the forage becomes less digestible as it becomes more fibrous, but at a given fiber content different forages can have markedly different levels of digestibility. To some extent this can be overcome by using tabulations of relationships, each based on a limited feed class (Dijkstra, 1966). But again these pose the problem of allocation to a particular feed class; more seriously, they add little to our basic understanding of the factors that determine forage digestibility.

T H E NUTRITIVE VALUE OF FORAGE CROPS

7

The inadequacy of crude fiber as a determinant of nutritive value was clearly established by Norman ( 1 935). Tentative alternatives to crude fiber were proposed: cellulose (Crampton and Maynard, 1938), holocellulose (Ely and Moore, 1955), modified acid-detergent fiber (Clancy and Wilson, 1966). Each of these aimed to analyze a more precise chemical grouping than crude fiber, but each perhaps reflected the same basic thinking, that the complex process of forage digestion can be quantified by a single chemical analysis. The relationships between these “fiber” components and forage digestibility (reviewed by Miller, 196 1 ; Sullivan, 1962), are often more precise than those based on crude fiber; they are still inadequate for predictive purposes. This conclusion, which had become evident by 1960, stimulated the two main developments discussed below: the study of chemical techniques more relevant to the digestion process, and of biological techniques that attempt to simulate the process of rumen digestion by a laboratory technique. C. IMPROVEDCHEMICAL TECHNIQUES Forage digestibility, Eq. (2), is the summation C% content X % digestibility of all the different chemical components in the forage. Some of these components, such as soluble carbohydrates and organic acids, are completely digested as the forage passes through the ruminant tract; others, in particular the polysaccharides and lignin, are much less completely digested and comprise most of the feed residue excreted as feces. The “classical” approach, discussed above, assumes that the extent to which the fiber fraction is digested is directly related to the proportion of that fraction in the forage. Detailed studies of the digestibility of different fiber fractions, based on in vivo experiments, have clearly shown that this is not so. Thus Jarrige and Minson (1964) found that there was no decrease in the digestibility of the cellulose in S.24 ryegrass as the cellulose content increased from 14.1 to 19.0 percent of the dry matter in early spring, while Gaillard (1962) and others showed that the cellulose in alfalfa is much less digestible than that in grasses with the same content of cellulose. This led to the development of techniques of graded extraction with reagents of increasing concentration (Gaillard, 1958; Jarrige, 196 1 ; Burdick and Sullivan, 1963) and of cellulose solubility in cupriethylenediamine (Dehority and Johnson, 1963) which take some account of the chain length and resistance to digestion of the different polysaccharide fractions. However, no single procedure is likely to give results relevant to the wide range of polysaccharides and lignin that comprise the fiber

8

W. F. RAYMOND

fraction in forages, and Gaillard (1966) has developed a more comprehensive relationship between forage digestibility and composition: Digestibility of organic matter % = 0.37(C-19.19) - 5.51(L-5.58) - 0.51(H-18.10) (3) + 4. I I(U-3.80) 65. I

+

which includes the percent contents of cellulose (C), lignin (L), hemicellulose (H), and anhydrouronic acid (U). More recently Gaillard and Nijkamp (1968) have proposed a less complex analytical system, which replaces the separate determinations of cellulose and hemicellulose with neutral-detergent fiber (N DF, v.i.): Digestibility of organic matter % = 66.7 - 4.64(L-5.19)- 0.14(NDF-48.05) + 2.95(U-3.47)

(4)

An alternative approach, developed by Van Soest (1 967) and Terry and Tilley ( 1964a), emphasizes the contribution to total forage digestibility of the highly digestible cell-contents fraction in forages. These workers have considered forage to contain two main fractions, the cell contents which are almost completely digested, and the cell-wall constituents, which are only partly digested, and they have proposed analytical systems that (a) separate these two fractions and (b) indicate the extent to which the cell-wall fraction would be digested in the ruminant tract. In a series of papers (summarized by Van Soest, 1967) this author has described methods for separating a forage sample into a cell-contents fraction soluble in neutral detergent (S), and an insoluble cell-wall fraction (neutral-detergent fiber, NDF), as well as a fiber fraction insoluble in acid detergent (acid-detergent fiber, ADF) and lignin (L). In a key paper (Van Soest and Moore, 1966), the digestibility of the N D F fraction was shown to be negatively correlated with log X (r = -0.98**) where X , the concentration of lignin in the A D F fraction, effectively measures the extent of lignification of the cellulose in the forage (in that paper X was denoted as L, which was confused with percent lignin). The mechanism by which lignin reduces fiber digestibility probably includes the effects of physical incrustation, of lignin-carbohydrate complexes, and of molecular bonds. Van Soest (1967) also showed that the cell-content fraction in forages is almost completely digested (98 percent) by the ruminant. However, a significant amount of material soluble in neutral detergent occurs in ruminant feces. This is not undigested plant cell contents, but consists of endogenous materials (mucus, salts, bile residues, and undigested bacteria) resulting from the digestion process; digestibility as measured by Eq. (2) is not the “true” digestibility of the forage material, but the “apparent” digestibility, (feed - feces) measuring the amount of feed digested, less this inevitable

THE NUTRITIVE VALUE O F FORAGE CROPS

9

endogenous loss associated with the passage of the feed through the tract. Based on in vivo results with a limited range of feeds, Van Soest (1967) calculated this fecal loss to be 12.9 percent of the dry weight of forage eaten. Van Soest (1967) was then able to compute the apparent digestibility of forage: Apparent digestibility of dry matter % = 0.98s

+ W ( 1 . 4 7 3 - 0.789 log X ) - 12.9

(5a)

comprising the almost completely digested cell-contents (S),plus the cellwall constituents (W=NDF) digested to an extent depending on the degree of lignification of the A D F fraction ( X ) , and less the endogenous excretion. It has not yet been possible to test this relationship on a wider range of forages than those studied by Van Soest. But by taking account of the differing contents and digestibilities of the two main fractions in herbage, the cell contents and the cell-wall material, Eq. (5a) clearly represents an important advance over the more empirical methods summarized by Miller ( 1 96 1) and Sullivan (1 962). In the course of the development of the detergent-fiber methods, Van Soest ( 1 965b) examined the effect of the method of drying herbage samples before analysis on the measured levels of acid-detergent fiber and lignin. Drying temperatures above 50”C., particularly over an extended period, significantly increased the levels of both these fractions; this artifact “fiber” was shown to result from a nonenzymatic browning reaction, in which protein polymerizes with products of carbohydrate breakdown, so that the “lignin” fraction in particular contains an abnormally high percentage of nitrogen. In earlier work this had been corrected by subtracting %N X 6.25 from the apparent lignin analysis. However, Van Soest recognized that natural plant lignins may contain some nitrogen, and derived a relationship that would correct only for the nitrogenous matter which might be attributed to heat damage: % corrected lignin (L,)

=

1.208 X % measured lignin (LA)- 10.75 x %N in A D F + 0.42

(6)

The acid-detergent fiber (ADF) content is then corrected: % A D F corrected = % A D F observed - (LA- L,)

(7)

From Eqs. (6) and (7) the factor log X in Eq. (5a), based on corrected values for A D F and lignin, can be calculated. The need for this correction must reduce the utility (and precision) of Eq. (5a) and emphasizes the importance of adequate drying methods for preparing herbage samples for analysis. The method of choice must surely

10

W. F. RAYMOND

be freeze-drying (lyophilization). But the great majority of freeze-driers in current laboratory use are of small capacity (< 1 kg. water/24 hours), and this can introduce a source of error which is seldom recognized- that the sample of forage which is dried by this ideal method may be so small as to be quite unrepresentative of the material sampled. This possible contradiction between the precision of the drying method and the accuracy of sampling has been discussed (Commonwealth Agricultural Bureaux, 1961, p. 135); until much larger freeze-driers become available, the solution in many cases may be to dry forage samples of adequate size as rapidly as possible at 100"C.,so as to minimize the time during which nonenzymatic browning (which occurs only in the presence of water) can take place. The individual investigator can then test the success of his own drying method by the application of Eq. (6) to analyses on representative samples. Recently Van Soest and Jones (1969) suggested a further refinement to the concept summarized in Eq. (5a), by indicating that the silica present in plant material may exert much the same effect as lignin in reducing the digestibility of the neutral-detergent cell-wall fraction. L. H. P. Jones and Handreck (1967) discussed the forms and reactions of silica in the food chain from soil to plant to animal. They pointed out that silica absorbed by plant roots is carried in solution to the actively metabolizing tissues. As the transporting water is transpired, solid silica is deposited on to the cell walls so that as these develop the polysaccharides are intimately associated with encrusting silica as well as lignin. From examination of the digestibility in vitro of forage samples of silica content ranging from 0.5 percent to 5.4 percent, Van Soest and Jones (1969) proposed a modified form of Eq. (5a): Apparent digestibility of dry matter % = 0.98s + W(1.473 - 0.789 log X) - 3.O(SiO2)- 12.9

(5b)

As yet the evidence is restricted to relatively few forages, but further study may clearly indicate the need for refinement of the biological concepts implicit in Eqs. (5a) and (5b).

D. ESTIMATION OF FORAGE DIGESTIBILITY BY in Vitro TECHNIQUES The inclusion of silica as a further component which may influence forage digestibility illustrates the trend toward multicomponent chemical techniques for predicting digestibility. In effect, this accepts that no single component can quantify the complex process of ruminant digestion, and that this must be treated as a series of stages, each described by a logical chemical evaluation, as in the decreasing digestibility of the N D F fraction as the fiber becomes more lignified.

T H E NUTRITIVE VALUE OF FORAGE CROPS

11

The inclusion of silica also illustrates a basic problem with chemical methods of evaluation, that a relationship such as Eq. (5a), which is found to be adequate with one population of forages, may give inaccurate prediction of the digestibility of other forages-in this case, of forages of unusually high silica content. This could arise from two causes: (a) the original relationship did not include all the components that exert a significant effect on forage digestibility and (b) chemical methods measure the content of different components in forage samples; they do not measure the physical distribution and organization of these different components within the plant, which must to some extent determine how far the plant fibers are digested by the microorganisms within the rumen. The chemical approach tends to treat a forage as a homogeneous material, an increase in lignin content, for instance, being visualized as an increase in lignification throughout the whole plant. In practice the forage plant is more realistically considered as made up of morphologically “distinct” fractions, each of which can be changing in chemical composition and digestibility in a way not necessarily related to the other fractions, so that chemical analysis (an average of the whole plant material) may well not describe the summation of the individual plant fractions that make up the digestibility of the whole plant. Thus, parallel to the development of chemical methods of forage evaluation, described in the previous section, has been the development of biological methods of evaluation, the artificial rumen or in vitro digestion methods. Essentially these have attempted to simulate the process of ruminant digestion by methods that can take account both of the overall chemical composition of the forage plant and of the distribution and physical interrelations of the chemical components within the different morphological parts of the plant. With the recognition that the digestibility of the “fiber” fraction of forages would be most affected by these physical characteristics the initial investigations of biological methods were concerned with fiber digestibility, and in particular with the digestibility of the cellulose fraction in forages. Although details of technique differed, all these methods were based on the incubation, under controlled conditions, of a sample of the test forage with a mixed culture of the microflora taken from the rumen of a forage-fed animal; the aim was to standardize the conditions of incubation so that the fiber in the forage sample was digested to the same extent as in the same forage when fed in an in vivo experiment (Quicke et al., 1959; Lefevre and Kamstra, 1960; Karn et al., 1967). These techniques were also used to measure the extent to which the dry matter (Clark and Mott, 1960), organic matter (R. L. Reid el al., 19601, or energy content (R. L. Reid et al., 1960; Baumgardt el al., 1962; Naga and El-Shazly,

12

W. F. RAYMOND

1963) in forage was digested in vitro. In most cases the extent of digestion in vitro was found to be less than that in vivo, and regression equations were developed to allow prediction of in vivo values. In an alternative system, a sample of dried forage is enclosed in a nylon or dacron mesh bag suspended within the rumen in vivo, and digestibility and rate of digestion are measured by the loss of dry matter or of cellulose from the sample (Lusk et al., 1962; Hopson et al., 1963). This technique could have the advantage that a normal microfloral population will be maintained, although this will tend to be that characteristic of the feed eaten by the host animal, rather than of the sample under test. However the technique does permit the comparison of large numbers of feed samples under standard conditions, and it could be of use in ranking forage samples in a breeding selection program. This approach was analogous to that with the earlier chemical methods, in attempting to predict the complex process of forage digestion by a single procedure. As with the chemical methods, the accuracy of prediction was found to decrease as the range of forages examined was widened; in particular marked divergences were found between results for grasses and legumes (Shelton and Reid, 1960). Tilley and Terry (1963) suggested that these discrepancies might be the result of correlating data from a single digestion with rumen organisms with those from digestion within the animal, which involves a mainly bacterial digestion within the rumen followed by a mainly enzymatic digestion in the remainder of the digestive tract. Within the rumen, the “digestible” polysaccharides, carbohydrates, and protein in the feed are broken down by the action of the microorganisms there; some of the products of digestion are absorbed directly through the lumen wall, but a considerable part serves as the substrate for microbial growth, and is resynthesized into protein, polysaccharides, and lipids within the proliferating bacterial and protozoal population. These microorganisms, entrained in the residues of undigested fiber and other feed components, then pass from the rumen to the abomasum and duodenum. In these organs the digesta are acidified and further digested by secreted enzymes that hydrolyze much of the bacterial and residual plant proteins to amino acids. These are then absorbed as the main source of amino acids for the metabolism of the host animal. The undigested residue from the in vitro digestion of forage material with rumen microorganisms is thus seen to contain, in addition to undigested feed, the rumen organisms which, in vivo, would be enzymatically digested in the ruminant hind tract. Tilley and Terry ( 1 963) proposed that this second stage should be simulated by subjecting the residue from the in vitro bacterial digestion to a second enzymatic digestion. They ex-


13

amined several enzymes and concluded that a two-stage procedure comprising digestion by rumen microorganisms followed by digestion by acid-pepsin gave the closest agreement with in vivo digestibility values for the dry matter and organic contents in forages. This method showed a correlation of 0.97 between in vitro and in vivo values when tested on a wide range of forages, including grasses fertilized with different levels of nitrogen, and legumes: Digestibility in vivo = 0.99 X digestibility in vitro - 1.O I (S.E. = f 2.3 I )

(8)

Similar high degrees of correlation have been found by O’Shea and Wilson ( I 965; r = 0.94), Wedin el al. ( 1966; r = 0.996) and Ademosum et af. ( 1 968; r = 0.96); Dent ( 1 963) found close agreement between two-stage in vitro and in vivo digestibility results with brassicas and forage maize. In a number of studies the precision of prediction of digestibijity in vivo by this in vitro technique has been compared with the chemical methods already discussed. Armstrong et al. (1964a) found that the metabolizable energy and net energy contents of a series of dried grass feeds were more accurately correlated with organic matter digestibility in vitro than with cellulose or lignin contents; Bosman (1967) and Ademosum ef al. (1968) have reported the in vitro method to correlate more closely with in vivo digestibility than the chemical methods tested; Engels and Van der Merwe (1967) have found the same result with lowdigestibility hays in South Africa. However, to date no direct comparison has been reported with the improved chemical technique proposed by Van Soest ( 1 967, Eq. 5 ) , and it is possible that these two techniques differ little in the precision with which they allow prediction of forage digestibility in the laboratory. In fact the stage may be approaching at which little further improvement in precision can be expected. In interpreting the error terms of these relationships, it is important to recognize that this error does not arise solely from deficiencies in the laboratory technique used (chemical or in vitro), but that errors are also associated with the actual measurement of the in vivo digestibility of the forages, and with the fact that digestibility in vivo is not a constant parameter of a particular forage. Thus digestibility determined in an animal experiment may depend on the amount of forage fed, digestibility decreasing as the level of feeding increases (Moe et al., 1963, and can be significantly reduced if the animal is parasitized with stomach worms-probably the rule rather than the exception with sheep (Spedding, 1954; Shumard et al., 1957). The standard deviation of an estimate of digestibility is between 1.0 and 1.3% (Raymond et al.,

14

W. F. RAYMOND

1953); as most digestibility determinations are made with only 2 or 3 sheep it is evident, even where the factors noted above are standardized, that much of the error in the relationships noted must result from errors in the in vivo determination, rather than in the concept or precision of the laboratory determination. Probably the main area for improvement lies then in the better standardization of the in vivo digestibility experiments, of the preparation of the forage samples for analysis, and of the conduct of the laboratory procedures. R. L. Reid et al. (1964) and Noller et al. ( 1966) have both reported significantly higher levels of dry-matter digestibility in vitro in forages prepared by freeze-drying than by oven-drying, presumably because of the production of indigestible artifacts during oven-drying, as suggested by Van Soest ( I965b). Tilley and Terry (personal communication), however, found no advantage of freeze drying compared with rapid ovendrying at 1Oo”C., although the same authors (1963) had reported considerable depression of digestibility in vitro in samples dried at temperatures above 1 10°C. The concept that fiber digestibility is limited by physical incrustation with lignin indicated that digestibility should be increased by fine subdivision, and both Dehority et al. (1962) and Tilley and Terry (1 963) reported increases of up to 50 percent in digestibility as a result of grinding forage samples in a ball mill before in vitro digestion. However, within the range of fineness of grinding found in forage samples ground by hammermill, Tilley and Terry (1 963) found no significant effect of particle size. The most serious problem arises, however, in the lack of standardization of in vitro procedures between different laboratories. Barnes (1967) reported the results of a collaborative study in which the in vitro digestibilities of the dry matter and cellulose in three forages were measured at 17 laboratories. The mean values for cellulose digestibility after 24 hours ranged from 40.0 to 63.9 percent, reflecting the use of different techniques in terms of sample size, preparation of the rumen inoculum, pH control, etc. In contrast, Raymond and Terry (1966) have reported close agreement between in vitro results from two laboratories using identical procedures, and have stressed the importance of different laboratories using the same “standard” forage samples as an additional check on the reproducibility of the method. Tilley and Terry ( 1 963) found that rumen liquors taken from donor animals fed on several contrasting forages were of similar digestive efficiency in the two-stage in vitro system, and Troelsen and Hanel(l966)


15

have reported that the potency of liquors taken from different sheep differs less than the in vivo digestive efficiency between sheep. I n general the most important consideration seems to be that the diet of the donor animals should contain at least 10 percent of crude protein (see below), that it should give a sample of rumen contents from which a strained liquor can readily be separated (i.e., the animals should be fed on a coarsely chopped hay rather than on a pelleted ground feed), and that the sample should be kept in an anaerobic condition and be prepared for inoculation of the digestion tubes as rapidly as possible. It must be accepted, however, that this in vitro procedure, developed with temperate grass and legume species, may not be directly applicable with other temperate species, or with tropical forage species, and Drew (1 966) has stressed that the system should wherever possible be checked with relevant samples of known in vivo digestibility. Thus Raymond and Terry (1 966) reported low in vitro digestibility levels when both the test forage (0.7 percent N) and the feed of the donor animal were of low nitrogen content, as can often occur with tropical forage species. The higher level of digestibility in vivo resulted from the animal’s ability to recycle urea via salivary and ruminal secretions, whereas digestibility in vitro was limited by a deficiency of nitrogen in the combined sample and inoculum. Addition of 6 mg. of N , as urea, to the in vitro system increased sample digestibility to the in vivo level. Engels and Van der Merwe (1967) found that the difference between in vitro and in vivo digestibility values of veldt grasses became greater as the nitrogen content of the test forages decreased. Addition of 20 mg. of urea N to each digestion tube gave in vitro values in close agreement with those in vivo. In a modification of the method of Tilley and Terry ( 1963), Alexander and McGowan ( 1966) have included ammonium sulfate in the buffer added to each tube. However in the author’s opinion this is advisable only where depressed levels of in vitro digestibility result from low nitrogen contents. Engels and Van der Merwe (1967) showed a marked depression in digestibility when 60 mg. of urea N was included in the digestion system, and a similar depression might occur when urea is added to an in vitro digestion of a forage sample already of high nitrogen content. Although some modification may be necessary in particular situations, the study reported by Barnes (1967) does emphasize the importance of the general adoption of standardized in vitro digestibility procedures, without which results reported by different laboratories cannot be directly comparable. The two-stage procedure described by Tilley and Terry (1963) is now used by many laboratories, and there seems a strong case

16

W. F. RAYMOND

that this should be adopted as a standard procedure in agronomic studies. But such a technique, based on digestion in vitro with a mixed culture of rumen organisms, must always be sensitive to uncontrolled biological variation, and there is a clear need for the future for a technique that can be more rigidly defined. A promising development, reported by Dehority et al. (1968), is the use of a pure culture of cellulolytic bacteria to replace the mixed inoculum taken from the rumen of a donor animal. Two of the strains of bacteria tested gave cellulose digestibility values in vitro in close agreement with the measured in vivo values; further development of this work could well lead to significant improvement in the standardization of laboratory in vitro techniques.

E. THERELATIVEUTILITY OF CHEMICAL AND in Vitro ESTIMATIONS OF FORAGE DIGESTIBILITY As has been indicated, the two-stage in vitro technique appears to give a better prediction of in vivo forage digestibility than any of the chemical methods yet investigated, and the use of this technique, reported below, has contributed greatly to our knowledge of forage digestibility. However this is an integrative technique, the measured digestibility being the sum of the digestibilities of the many different chemical fractions within the forage; without additional chemical information it can only describe, rather than explain, the differences in digestibility observed among different forage samples. This suggests that in vitro and chemical techniques should be considered as complementary, rather than competitive, methods of forage evaluation, the in vitro techniques being used to establish that forages differ in digestibility, and the chemical techniques to study the probable reasons for these differences. Such an understanding is essential if further improvement of forage digestibility is to be based on nonempirical concepts. IV. The Digestibility of Different Forage Species

A. BASIC PATTERNSOF DIGESTIBILITY The use of these more precise laboratory techniques for estimating the digestibility of forages has led to considerable progress in extending in vivo studies that were started in the 1950’s. It had long been recognized that as a forage becomes more mature it also becomes less digestible; the possibility that this imprecise statement could be quantified was proposed by Homb (1953), who showed a close relationship between the age (maturity) of a timothy-clover mixture and its digestibility, and by J. T. Reid et al. (1959), who suggested that the digestibility of a range of


17

forage species harvested during first growth during the spring could be estimated: % digestibility of dry matter = 85.0 - 0.48X (9) where X = number of days to harvest from April 30. These reports stimulated further investigations that have greatly increased our understanding of forage digestibility. Clearly the “date of harvest” in Eq. (9) can only be relevant in the area close to Cornell (Ithaca, New York), where the forages used in these in vivo digestibility determinations were harvested, but the basic principle, that digestibility is closely related to forage maturity, with “date of cutting” being used to describe stage of maturity in a particular location, was soon confirmed by other workers (Kane and Moore, 1959, and others; reviewed by Blaser, 1964). However, several divergences from the original concept, that all forages are of similar digestibility at a given date, became evident. Thus the digestibility of timothy (Phfeumpratense) (Mellin et a f . , 1962; Minson et a f . , 1964), of lucerne (Medicago sativa) (Demarquilly, 1966a), and in particular of white clover (Trifofium repens) (Harkess, 1963) were shown to decline less rapidly with advancing maturity than the 0.5 unit per day indicated in Eq. (9). Small differences in the digestibility of a given forage variety cut on the same date in different years may be due to the delayed onset of active spring growth in a “late” season (Minson et af., 1960), or to differences in leaf percentage in the forage grown in different years (R. H. Brown et a f . , 1968). But the most important observation has been that there can be large and consistent differences in digestibility between forage species and forage varieties. In a detailed series of in vivo experiments, Minson et a f . ( I 960, 1964) found that certain species (Lofium spp. and Festuca pratense) were considerably more digestible than others such as cocksfoot (Dactylis gfomerata) and tall fescue ( F . arundinacea); within a species late-maturing varieties (e.g., S.23 ryegrass) maintained a high level of digestibility to a later date than early-maturing varieties (e.g., S.24). These differences, illustrated in Fig. 1 , were considered large enough to be of agronomic and nutritional significance. Figure 1 also shows that the digestibility of each grass studied did not fall at a uniform rate as it matured. There was an initial period of almost constant digestibility before digestibility began to decrease. In the case of Lofium and Dactylis this change to a more rapid fall in digestibility, at a rate very similar to the 0.5 unit per day recorded by Reid et a f . ( 1 959), was closely associated with the first emergence of flowering heads; the digestibility of timothy (S.48), however, decreased well before this stage, and this species also showed the much slower rate of fall in digestibility noted above.

18

W. F. RAYMOND

In establishing these novel digestibility patterns, Minson et al. (1964) had two main advantages compared with their colleagues in North America-in the more general use of the genus Lofium in the United

I

I

10 20 April

30

Ib

20

May

3b

Ib

$0

$0

June

date of first cuttinq

FIG. I . The percent digestibility of the organic matter in grass varieties during first growth in the spring. S.23 and S.24, ryegrass; S.37 and Germinal, cocksfoot; S.215, meadow fescue; S.48, timothy; S. 170, tall fescue; O.M., organic matter. 0 indicates date of first ear emergence. (Data from Minson et al., 1964.)

Kingdom than in North America; and in the availability of widely differing maturity types, within this genus, developed at the Welsh Plant Breeding Station. These results stimulated other in vivo studies, which have in general confirmed the important conclusions implicit in Fig. 1. Thus Castle et a f . (1962) and Harkess (1963) have found consistently higher digestibility of ryegrass than of cocksfoot, and Lowe et af. (1962) reported that late-maturing varieties of grasses were more digestible than early-maturing at a given cutting date in the spring. These in vivo studies, initially on first spring growth, have also been extended to regrowths during the rest of the growing season, and consistent patterns have again emerged. The digestibility of regrowth cocksfoot and tall fescue is always lower than that of the corresponding regrowth of ryegrass, and the rate of fall of digestibility with time is much less in these later, largely vegetative, regrowths than in the first, reproductive, growth (Minson et af., 1960, 1964); a similar slower rate of fall in digestibility has been shown with regrowths of lucerne (Demarquilly, 1966a).


19

However, the digestibility of regrowth forage is likely to be less predictable than that of first growth. The latter comprises largely reproductive tillers which develop from the start of active growth in the spring; regrowths can contain both reproductive and vegetative tillers, the relative proportions of each depending on the management of earlier harvests. Thus the regrowth from a tiller whose reproductive growing point (ear) is harvested will comprise mainly leaf, which will decrease only slowly in digestibility at ca. 0.1 percent per day. In contrast, a tiller which is harvested before the ear has reached the height of cutting or grazing will continue to develop, and its digestibility will decrease at the 0.5 percent per day characteristic of first growth. Harvesting up to the time of first ear emergence will remove only some of the developing ears, and the digestibility of the regrowth will decrease at an intermediate rate, e.g., 0.3 percent per day. A later harvest, which removes most of the potential ears, will give a leafy regrowth which decreases more slowly in digestibility. Under a cutting regime regrowths are mainly leafy, and of relatively predictable digestibility (Minson et al., 1960, 1964). Under a grazing situation the prediction of the digestibility of regrowths is less precise. The grazing animal generally leaves some of the herbage on offer, and this continues to decrease in digestibility, so that the combined regrowth and remaining herbage is of lower digestibility than the corresponding regrowth from a cut sward (Tayler and Deriaz, 1963). As noted in Section IV, C , 2, the digestibility of a regrowth from grazing will also depend considerably on the moisture and nitrogen status of the sward, which determines the proportion of the next harvest that is composed of new growth of high digestibility. Limited information indicates that the digestibility of forage mixtures can be calculated from the proportions and digestibilities of the constituent species at the time of harvest (Harkess, 1963). An exception may be the case where the digestibility of one species is low because it is deficient in protein content (as has been shown with some tropical forages; Smith, 1962). Such a species, grown with another species of higher protein content, could have an enhanced digestibility, so that the digestibility of the “mixture” would be somewhat higher than predicted. The patterns of forage digestibility discussed above were all based on in vivo experiments; they have been considerably extended by the use of laboratory in v i m techniques. The validity of the two-stage in vitro technique was indicated by Terry and Tilley (1964a), who showed that the basic patterns of digestibility shown in Fig. 1, and in particular the nonlinear fall in digestibility with time and the consistently higher diges-

20

W. F. RAYMOND

tibility of ryegrass than of cocksfoot, could be closely reproduced by in vitro measurements. With this method many more forage samples can be examined than would ever be possible by in vivo experiments, but the need to confirm, in vivo, the more important conclusions indicated from in vitro studies must be emphasized. Thus Dent and Aldrich (1968) have measured the digestibility, in both first growth and regrowths, of numbers of varieties within several forage species. They have shown consistent differences in digestibility between species, and between maturity types within species, at different centers and in different years. Their work indicated that different varieties of the same maturity type within a species might differ in digestibility. Thus a small number of varieties of cocksfoot (e.g., Roskilde I1 and Scotia) appeared to be more digestible than the varieties Germinal and S.37, shown in Fig. 1, and Reveille tetraploid ryegrass was more digestible than S.24 ryegrass. These in vitro results, subsequently confirmed in in vivo experiments (Osbourn, unpublished) have particular relevance to the possibility of breeding more digestible forage varieties, discussed in Section VIII.

B. THEDIGESTIBILITY OF DIFFERENT PLANTFRACTIONS The use of in vitro techniques to measure the digestibility of different plant fractions has also provided a more logical understanding of these patterns of forage digestibility. These are in many cases contrary to what would have been predicted in terms of the earlier concepts of forage nutritive value, that leaf was more digestible than stem, and that forage fractions high in protein (N) content would be more digestible than those of lower N content. The results of Minson et al. ( 1 960) showed that these concepts might be incorrect. Thus in first growth, S.24 ryegrass forage cut in 1959 on May 1 , with a nitrogen content of 2.62 percent and 53 percent of leaf lamina, was of exactly the same digestibility as that cut on April 20, which had contained 3.66 percent N and 77 percent leaf, S.37 cocksfoot cut on May 5 , with 2.78 percent N and 58 percent leaf, was 4.5 units less digestible than the ryegrass cut on May 1 . In later experiments Tayler and Rudman (1966) harvested a S.24 ryegrass sward in two horizons, a top fraction cut above 13.5 cm., and a bottom fraction from 6 to 13.5 cm. The digestibilities of the organic matter in the two fractions, in vivo, were 84.0 and 82.6 percent, respectively. This indirect in vivo evidence has been followed up in detail by in vitro digestibility determinations on forage samples separated into fractions of leaf, leaf sheath, stem, inflorescences, and dead material. Terry and Tilley ( 1 964a) analyzed these fractions from the forages fed in vivo by Minson et al. ( 1 960, 1964) (Fig. 1). They showed that in all species the


21

digestibility of the leaf fraction fell only slowly with advancing maturity (0.13 percent per day) whereas that of the leaf sheath (0.4 percent) and stem fractions (0.7 percent) fell much more rapidly. In the immature forages the stem was always more digestible than the other components. On any given date each fraction in S.24 ryegrass was more digestible than the corresponding fraction in S.37 cocksfoot, of equivalent maturity type. A typical set of results is shown in Fig. 2. The digestibility of the

whole plant leaf sheath

----

leaf blade stem

--

-----.

FIG.2. The digestibility in vifro of the dry matter in the whole plant, and in the leaf blade, leaf sheath, and stem fractions of S.37 cocksfoot during first growth in the spring. Figures in parentheses are the percentage of stem in the whole plant. (Data from Terry and Tilley, I 964a. )

whole forage material, calculated from the proportions and digestibilities of the constituent fractions, changed with maturity just as in the in vivo experiments-that is, with a slow fall in digestibility up to the time of ear emergence, followed by a more rapid fall as the stem and leaf sheath fractions, by now less digestible than the leaf, comprised an increasing proportion of the total forage. The slower and more steady fall in the digestibility of S.48 timothy could be explained by the much higher proportion of leaf sheath in this species than in ryegrass or cocksfoot.

22

W. F. RAYMOND

These conclusions have been confirmed in considerable detail by Pritchard et al. (1963), Wedin et al. (1966), Walters et al. (1967), and Dent and Aldrich (1968); Mowat et al. (1965) found similar results with timothy and bromegrass, but were unable to show higher digestibility of cocksfoot stems than of leaves even in immature forage. Similar logical patterns of digestibility have been shown with legume forages (Terry and Tilley, 1964a; Mowat et al., 1965). With lucerne ( M . sativa), red clover ( T . pratense), and sainfoin (,Onobiychis viciifoliu) the older leaves tend to senesce and fall, so that the digestibility of the leaf fraction decreases very little as the plant matures. By separating the stem fraction into 6 inch subfractions, measured from the top of the plant, it was shown that digestibility decreased down the stem, but that the digestibility of any given fraction changed relatively little with maturity. A similar result has been found with the stems of brassicas and forage maize (Zea mays), the stem tip of marrow-stem kale being highly digestible (81.8 percent dry matter) and the stem base of lower digestibility (62.3 percent) (Dent, 1963). Within a particular forage species, Walters et al. (1967) and Dent and Aldrich (1968) have shown, at a similar stage of morphological development, that “late” varieties tend to be less digestible than “early” varieties because both the stem and leaf fractions are somewhat less digestible, and because they contain a higher proportion of senescent and dead material. Small differences in digestibility of a given variety at ear emergence in different years can be attributed to differences in leafistem ratios, stem at this stage being less digestible than leaf (R. H. Brown et al., 1968). Of particular importance is the observation, already noted, of differences in digestibility between varieties of similar maturity type within a species (Dent and Aldrich, 1968). The higher digestibility of REVEILLE ryegrass than S.24 was not accounted for in terms of leafstem ratio, but because both the leaf and stem fractions in REVEILLE were more digestible than the same fractions in S.24. Differences in digestibility have also been shown between the individual plants (genotypes) within a variety (Cooper et al., 1962; Walters et al., 1967; Mowat, 1969), resulting from differences in digestibility of the plant fractions rather than from different leaf stem ratios. However, while the in virro techniques used in these studies can describe the changes in digestibility within and between different forage. species, they cannot explain them; for this the newer chemical techniques (Section 111, C ) are needed, to analyze digestibility measured in vitro into its component parts. In detailed studies by Tilley, Terry, and Outen (unpublished) the forage sample is separated into a cell con-

THE NUTRITIVE VALUE O F FORAGE CROPS

23

tents fraction, soluble in acid pepsin, and a cell wall fraction (analogous to the fraction soluble in neutral detergent and the cell wall residue of Van Soest, 1967). The digestibility of the cell wall fraction is also measured by in vitro (rumen organism) digestion. Consistent differences have been shown between S.24 ryegrass and S.37 cocksfoot, harvested at the same stage of maturity: (a) the content of pepsin-soluble material is higher in the ryegrass than in the cocksfoot; (b) as a result there is a higher content of cell wall fraction in the cocksfoot; and (c) this fraction in the cocksfoot is less digestible than in the ryegrass, so that the content of “digestible cell wall material” in the two species is very similar. As a result the digestibility of the ryegrass (cell contents X 0.98 digestible cell wall fraction) is higher than that of the cocksfoot. Within different plant fractions, the small decrease in digestibility of the leaf fraction as the plant matures (Fig. 2) is accounted for by the consistently high level of cell contents and high digestibility of the cell wall fraction in the leaf. Young stem material contains an even higher proportion of cell contents than the leaf (hence its higher digestibility); but as the stem matures the cell content fraction decreases rapidly and is replaced by cell wall material which becomes less digestible with increasing lignification, so that the digestibility of the stem decreases rapidly with advancing maturity. Extension of these more detailed studies of the components of digestibility to the genotypes within a species which have been found to be of higher digestibility (Section VIII) may provide a more objective basis for selection for improved digestibility than the in vitro techniques that have so far been used.

+

c. THEEFFECTOF ON

ENVIRONMENTAL A N D OTHER FACTORS FORAGEDIGESTIBILITY

1 . Environmental Effects The difference in digestibility between forages cut on the same date at Cornell (J. T. Reid et al., 1959) and in Maryland (Kane and Moore, 1959) appeared to be due to the forages in these two locations being at different stages of physiological maturity. It is of course possible that, even at the same stage of maturity, the digestibility of a forage may differ between locations; thus Aldrich and Dent (1 967) have found an indication of higher digestibility at a northern than at a southern latitude in the United Kingdom in cocksfoot cut 10 days after 50 percent ear emergence. Deinum et al. (1 968) measured the in vivo digestibility of perennial ryegrass grown under high and low light intensities, and with low and high levels of nitrogen manuring. Considerable differences in chemical com-

24

W. F. RAYMOND

position of the grass were found between treatments, but these had no significant effect on dry matter digestibility at any one sampling. These experiments showed lower levels of digestibility on all treatments during the summer when temperatures were higher, confirming earlier results of Deinum, based on chemical analysis of ryegrass grown in controlled environment cabinets. Deinum et al. ( 1 968) postulated that this effect of high temperature might partly account for the generally lower level of digestibility of tropical than of temperate forages. Hiridoglou et al. ( 1966) have also shown that high summer temperatures were associated with low in vitro digestibility levels. Forages growing in summer also tend to contain lower moisture contents than late season forages; however, the only report found on the effect of water intake on forage digestibility (Thornton and Yates, 1968) has indicated a small increase in the digestibility by cattle of the dry matter and fiber in chaffed oat straw-lucerne hay when water intake was restricted. 2 . Fertilizers and Forage Digestibility The effects of fertilizer nitrogen on forage digestibility have been studied in numerous experiments; most of these have reported an insignificant effect from the use of widely differing levels of application (summarized by Blaser, 1964): thus Minson et al. ( 1 960) found no effect on the digestibility of ryegrass or cocksfoot from levels of nitrogen application varying from 0 to 175 pounds/acre. However, Raymond and Spedding (1 965) have indicated several situations in which fertilizer nitrogen is likely to affect forage digestibility: (a) in a mixed grass-clover sward the use of this fertilizer may reduce the contribution, in the forage harvested, of the more digestible clover complement: (b) uneaten herbage left on a sward after stock have grazed will continue to decrease in digestibility, and by diluting the highly digestible new growth will depress the digestibility of herbage available at the next grazing; fertilizer nitrogen will increase the proportion of new growth in this harvest, and so may increase its digestibility; (c) unfertilized herbage may contain an inadequate level of nitrogen for the growth of rumen microorganisms: thus Smith ( 1 962) found an increased level of digestibility after application of nitrogen to such forage; 40 pounds of fertilizer nitrogen per acre increased the crude protein content of late-cut veldt hay from 3.6 to 6.8 percent and the digestibility of the forage dry matter from 5 1.7 to 59.5 percent. These effects of fertilizer nitrogen are consistent with the concepts of forage digestibility already discussed. McIlroy ( 1 967) has summarized results showing that fertilizer nitrogen increases the crude protein content and decreases the soluble carbohydrate content of herbage. But


25

there is little net change in the content of crude protein plus soluble carbohydrate, which largely comprises the cell contents fraction, or in the composition of the cell wall fraction. Thus little change in the digestibility of the forage would be expected except when the digestibility of the cell-wall fraction is limited by the low content of protein in the forage. There is also the possibility that certain forages contain specific components which reduce bacterial activity within the rumen. Hawkins ( 1 959) suggested that the low digestibility of the protein fraction in Sericea lespedeza and vetches might be due to the formation of insoluble protein complexes with the tannin in these forages, and Smart et al. ( 1 96 I ) found a depression of cellulose activity in vitro by an extract from Lespedeza cuneata. Schillinger and Elliott ( 1 966) observed differences of u p to 15 percent between the digestibilities in vitro of different lucerne plants. Low levels of digestibility could be raised by addition of amino acids (glycine, aspartic acid, glutamine) to the in vitro system, and these amino acids also increased the growth rate of voles fed on the lucerne forage. These authors attributed these differences to the presence of water-soluble antimetabolites in the low digestibility plants. Such cases are, however, likely to be exceptional, and the digestibility of most forages appears to be in line with the systems of evaluation proposed by Van Soest ( 1967) and Terry and Tilley ( 1964a).

3. The Effect of Feed Supplements on Forage Digestibility These systems do not, however, predict adequately the digestibility of forages fed in mixed rations. Thus when carbohydrate (starch) supplements are fed with forages, there can be a significant decrease in the digestibility of the fiber (cell wall) fraction of the forage, unaccounted for by any change in the composition of the cell wall (Eq. 5 ) . This has been attributed to a preferential digestion of the starch by the rumen microorganisms, so that the extent of digestion of the plant fibers is reduced, or alternatively that the amylolytic bacteria compete preferentially for ammonia against the cellulolytic bacteria, so reducing cellulose digestion (El-Shazly et al., 1961). More recent work has indicated an alternative explanation. First, it is known that when a starch supplement is fed with a forage there is a reduction in the pH of the rumen contents compared with that when the forage is fed alone (Topps et al., 1965). Second, Tilley et al. ( 1964) have shown a marked reduction in the rate and extent of digestion of dry matter and cellulose in vitro when the pH of the in vitro system is reduced (Table I); a decrease, with decrease in rumen pH, has also recently been

26

W. F. RAYMOND

indicated in the rate of “digestion” of cotton threads suspended within the rumen in vivo (Wilkins, unpublished). Tilley et al. (1964) have therefore suggested that the lower rumen pH when a starch supplement is fed TABLE I The Effect of the pH of the in Vitro Digestion on the Extent of Dry Matter and Cellulose Digestibility of Samples of Cocksfoot, S.37, by the 2-Stage in Vitro Method“

Sample No.

In vivo

Percent of forage dry matter digested In vitro (48 hours) pH 6.8 pH 6.0 (normal)

76 73 72 72 64 63 59 59 5 49 49 “From Tilley et al. ( 1964). *Figures in parentheses are percentages matter.

(15)

(17) (16) (12)

67 (12) 63 ( I 1) 50 (10) 49 (10) 41 (7)

pH 5.5 54 53 37 36 29

(4) (5) (3) (2) (1)

of digestible cellulose in the digestible dry

may provide a less favorable environment for the cellulolytic and other bacteria that are able to digest plant fiber (see also Head, 1961). In anthropomorphic terms, the natural rumen microflora has become adapted to the pH 6.6 to 6.8 characteristic of the rumen contents of the grazing herbivore; any marked divergence from this pH finds a microfloral population increasingly unable to digest fiber. This hypothesis, if substantiated, could lead to the development of feeding regimes, aimed at optimizing rumen pH (and redox potential) to ensure maximum digestion of the cell wall fraction of forages. As already noted, forages may be of low digestibility because they are of very low protein content (< 4% crude protein). The digestion of these forages can be increased by feeding protein supplements, and there has been much interest in the use of urea for this purpose. Thus Campling ef al. (1962) measured an increase in organic matter digestibility from 41 to 50 percent when a urea supplement was fed with oat straw of 3.0 percent crude protein, and other examples have been reported (see M. H. Briggs, 1967). Within the rumen the urea is rapidly deaminated, the ammonia produced then being used by the celluloytic and other bacteria, whose digestive activity would have been limited by deficiency of protein in the unsupplemented forage. The feeding of urea to increase the digestibility of forage is not often used in practice, because a crude protein level in forage of less than 4


27

percent is uncommon. The main use of urea is likely to be as a substitute for protein in productive rations (Section VI, D). V. The Voluntary Intake of Forages

A. THEFACTORS CONTROLLING FEEDINTAKE

It was noted in the Introduction that the quantity of forage that ruminant animals eat is in practice seldom controlled in the same way that most other farm feeds are rationed. Yet the amount of forage that animals eat is in many cases the major factor determining their level of nutrient intake and their output of useful products. Increasing emphasis has been given in the last decade to the study of voluntary intake. Earlier studies of forage intake were undoubtedly hindered by the confusion of “intake” with “palatability” (Blaxter et al., I96 1 ; Campling, 1964), but it is now accepted that forage intake is mainly controlled by largely involuntary physiological reflexes within the animal, rather than by its subjective liking for different feeds. The development of some of the current concepts on voluntary intake have been reviewed by Balch and Campling (1962), Conrad ( 1 966), and L. D. Brown (1966). From these a broad distinction appears between the factors determining intake by ruminants and by nonruminants. Intake by nonruminants is controlled mainly by levels of blood metabolites, the animal ceasing to eat when these reach a threshold level. Intake by ruminants depends much more on the capacity of the digestive tract, particularly the rumen, eating ceasing when a certain degree of “fill” has been reached, and starting again when “fill” has been reduced by digestion and movement of food residues through the digestive tract: only on feeds of high energy concentration does blood metabolite level, rather than gastrointestinal fill, begin to control the amount of food that ruminants will eat (Conrad, 1966). However, as with other aspects of forage nutritive value it is essential to recognize that the amount of forage that animals will eat is likely to be determined by a complex of factors. Some earlier investigations may have oversimplified the problem; the observation by Blaxter et al. ( I96 1) that the newer concepts related to digestibility and rate of passage of foods are “attributes which are hardly consonant with their acceptability to the palate or taste” perhaps dismissed too lightly the possible significance of these other attributes. It is evident also that the amount of forage that animals eat may depend as much on the amount of forage available as on any inherent characteristics of the forage itself. Raymond (1966a) has proposed that the factors determining forage intake can be usefully divided into intrinsic factors (i.e., features inherent in the forage)

28

W . F. RAYMOND

and extrinsic factors, which depend on the method of presentation of the forage, on the effect of processes such as ensilage and dehydration, and on environmental factors. The intrinsic factors determine how much of a forage a ruminant animal could eat under ad libitum conditions; the extrinsic factors determine how much the animal is able to eat under the particular feeding conditions imposed upon it. B. INTRINSIC FACTORS DETERMINING FORAGE INTAKE It is first necessary to emphasize the great importance of rigid standardization of the conditions under which intake measurements are made, if comparisons are to be made between results from different centers (Chalupa and McCullough, 1967). Voluntary intake is generally defined as the amount animals will eat when an excess of 15 percent is offered (Blaxter et al., 1961). But fresh feed should be presented at least twice daily and uneaten residues removed to avoid soiling this fresh feed, so that limitation of intake by extrinsic factors is avoided. Ruminants are also found to vary much more in their capacities for feed intake than, say, in their digestive capacities, and adequate numbers of animals must be used to obtain reliable intake data. Heaney et al. (1968) summarized CV’s* of individual animal intakes ranging from 10.5 percent (Minson et al., 1964) to 16.4 percent in their own experiments. Complicating factors are also the species and size of the experimental animals, and most intake results are now reported in terms of the metabolic weight of the animal, expressed as liveweight to the power 0.75 (although agronomists must have detected that some of their colleagues in ruminant nutrition still prefer 0.73). In this way the voluntary intakes of both sheep and cattle could be expressed on a comparable basis (Blaxter and Wilson, 1962). But it must be accepted that intake cannot be a precise parameter of a forage, because the amount that an animal eats depends on the individuality of the animal (its species, sex, physiological status, health, etc.) as well as on the intrinsic features of the forage. 1 . The Relation between Forage Digestibility and Intake It is now accepted that the major factor limiting the amount of forage eaten by ruminants is the capacity of the rumen and digestive tract. Ruminants are able to eat much more of highly digestible forages than of less digestible forages, because the latter occupy more volume and are within the rumen for a longer time and because from them more indigestible residue has to be passed down the hind tract (Balch and Campling, 1962). A decrease in voluntary intake as forage becomes more mature, *Coefficient of variation.


29

and so less digestible, has been shown in many experiments both with temperate forages (Crampton et a / . , 1960; Minson et al., 1964; Demarquilly, 1966b; Osbourn et al., 1966; Heaney et al., 1966) and with tropical forages (Grieve and Osbourn, 1965; Milford, 1967; da Silva and Gomide, 1967). These results appeared to confirm the concept that rumen fill controls voluntary intake. Unfortunately this led to the more generalized concept that the voluntary intake of a forage could thus be predicted from its digestibility, and that a single forage parameter, related to digestibility, would be adequate to estimate both digestibility and intake, the two main components of nutrient intake. There is now increasing evidence that this is a too simplified concept. While with most forage species intake decreases as the forage becomes less digestible, the relationships between intake and digestibility can differ markedly between different forages. Therefore, different forage species at the same level of digestibility may be eaten in quite different amounts, a fact that is in conflict with the earlier concept. This first became apparent with the observation by several workers of higher intakes from legumes than from grasses of the same digestibility (e.g., Van Soest, 196%; Osbourn et a / . , 1966; Milford, 1967; Weston and Hogan, 1967) and subsequently of different levels of intake between grass species. Thus, low intakes have been reported with timothy (Phleum prarense) (Minson e f al., 1964; R. L. Reid and Jung, 1966; Miles and Walters, 1966), with tall fescue ( F . arundinacea) (Van Soest, 19641, and with Phalaris arundinacea (0. N. Andrews and Hoveland, 1965; O’Donovan et al., 1967) compared with other grass species. With timothy the rate of fall in intake with decreasing digestibility was found to be less than with ryegrass (Minson et al., 1964), and Heaney et a / . ( 1966) found only a negligible change in intake with decreasing digestibility of Phleum nodosum. Within a species, differences in intake characteristics have also been shown. Thus Osbourn et al. ( I 966) found the intake of a diploid variety of Italian ryegrass to be 16 percent greater than that of a tetraploid variety during first growth in the spring. Large differences in intake between different lines of Phalaris arundinacea have been shown by Roe and Mottershead ( 1962) and O’Donovan et al. ( 1967). 2 . Differences in Intake between Forage Species Observations of different intake-digestibility relationships for different forages are of considerable importance. As long as voluntary intake was accepted as being determined mainly by level of digestibility, there appeared to be little prospect of improving the nutritive potential (intake X digestibility) of forages except by an improvement in digesti-

30

W. F. RAYMOND

bility. The evidence that factors in forage other than digestibility can also affect intake now offers a much wider scope for the improvement of forage nutritive value. Two possible lines of development are indicated in the work of Osbourn et al. (1966) and O’Donovan et al. (1967), noted above. Thus Osbourn et al. (1966) and Osbourn (1967) showed marked differences in voluntary intake, in the order lucerne > ryegrass > timothy, at a given level of digestibility. Chemical analysis of these forages showed that the “digestible” fraction in lucerne contained a higher proportion of pepsin-soluble material, and a lower proportion of digestible fiber than the “digestible” fraction in timothy, with the levels in ryegrass intermediate between those for lucerne and timothy (Fig. 3).

PEPSIN SOLUBLE MATERIAL

DIGESTIBLE FIBER

INDIGESTIBLE MATERIAL

LUCERNE Voluntary intake qDM/kqO’~dldoy

82

S 24

5 48

73

bl

FIG. 3. The composition of the digested fraction, and the voluntary intakes of lucerne, S.24 ryegrass, and S.48 timothy forages of the same dry matter (D.M.)digestibility. (From Osbourn, 1967.)

Van Soest ( 1 965c) reported a similar conclusion, that lucerne contains a higher proportion of cell contents (material soluble in neutral detergent) and a. lower proportion of cell wall constituents, than grass of the same level of digestibility. While the grass and lucerne forages are of the same digestibility, it is likely that the lucerne will reach the “digested” stage more rapidly than the ryegrass, and the ryegrass than the timothy. The digestible fraction of the lucerne could thus occupy less volume X time within the rumen; as a result the animal could eat more of it than of the grasses. Osbourn ( 1 967) also reported that, in the experiment described by Osbourn et al. (1966), the diploid ryegrass had a higher content of pepsin-soluble material than the tetraploid ryegrass, which could have been associated with the higher intake of the diploid variety.


31

The observations illustrated in Fig. 3 offer one possible explanation as to why different forages of the same level of digestibility may be digested at different rates, and so have different intake levels. But there is also the possibility that rate of digestion, and in turn the rate of intake, may be affected by conditions within the rumen. As Tilley et al. ( 1 964) have shown (Table 1) the rate and extent of cellulose digestion decreases as the pH of an in vitro system (and by analogy of the rumen in vivo) diverges from the physiologically normal level of about pH 6.8. Some forages, particularly highly buffered, low-sugar, forages such as the legumes, are found to give a characteristically higher rumen pH (6.6 to 6.8) than perennial ryegrass ( hexoses > deoxyhexoses > pentoses > uronic acids. The particular acid used has some bearing on this order: hexosamines are more quickly destroyed in HzS04 than in HCl of equal concentration whereas the reverse applies for the other groups. With amino sugars the exclusion of oxygen may greatly reduce the extent of destruction (e.g., see Walborg and Ward, I963), and the presence of heavy metals may lead to increased destruction (Hartree, 1964). An additional complication is the possibility of undesirable side reactions, such as “acid reversion” [the acid-catalyzed reaction of the reducing group of a liberated monosaccharide with the primary or even secondary hydroxyl groups of another sugar molecule to give disaccharides or oligosaccharides (Pigman, 1957; Whelan, 1960; Overend er al., 1962)], or interaction of the free sugars with amino acids (Francois et al., 1962; Gottschalk, 1966). This last point may be particularly significant in the analysis of soil polysaccharides which have always been found associated with polypeptide materials. Neuberger and Marshall ( 1 966a) consider that it is at present impossible to say whether these possible sources of error affect seriously the analytical results. The reactions being bimolecular, their effect can be minimized by carrying out the hydrolysis at low substrate concentration. Glycosidic bonds in polysaccharides are hydrolyzed at rates that are dependent largely on the nature of the sugar supplying the anomeric carbon atom of the linkage (Wolfrom and Thompson, 1957; Overend er al., 1962; Adams, 1965). Thus furanoside linkages are more labile than pyranoside linkages (Haworth and Hirst, 1930; Shafizadeh, 1958; Reichstein and Weiss, 1962); alpha glycosidic bonds are usually more stable than beta (Wolfrom and Thompson, 1957; Overend et al., 1962); with pyranoside linkages, pentoses and 2-deoxyhexoses allow easier hydrolysis than ordinary aldohexoses; and increased resistance to hydrolysis is conferred by uronic acid groups (Smith and Montgomery, 1959) and also by amino sugars (Gottschalk and Ada, 1956; Johansen et al., 1960). In practice, no perfect conditions for hydrolysis have been devised whereby it is certain that all glycosidic linkages of polysaccharides are cleaved and, at the same time, all the monosaccharides itre still intact at the end of the acid treatment. However, by dividing the sugars into groups, a set of reasonably satisfactory conditions can be derived for all groups except the uronic acids and possibly the pentoses. The optimum hydrolysis conditions for each group vary from one polysaccharide to another and must be ascertained by preliminary experiments in each

SOIL POLYSACCHARIDES

225

particular case. In general it can be said that HCI (4N to 8 N ) has been most satisfactory with hexosamines; H z S 0 4( 1 N or 2 N ) with hexoses; and more dilute acids, such as 0.1 N H2S04or HCI, with deoxyhexoses. The uronic acids and to a lesser extent, the pentoses represent a special problem which at present has no adequate solution. Perry and Hulyalkar ( 1965) state that even in the most favorable cases recoveries of polymerbound uronic acids rarely exceed 70 percent and, where the more acidlabile uronic acids are involved, the yields are considerably below this figure. Little is known about the best conditions for the release of pentoses. Although losses are often observed, it seems that reasonable recoveries are obtainable as long as optimum conditions are established (Saeman el al., 1954; Ivarson and Sowden, 1962; Cheshire and Mundie, 1966). The N-acetylamino sugars also represent a problem, although this is not usually very difficult to solve. Under strong acid conditions the polysaccharides containing these sugars are de-N-acetylated rapidly, and subsequent hydrolysis of glycosidic bonds is inhibited by the resulting free amino groups which, being positively charged in acid solution, confer some protection on the adjacent bonds (Moggridge and Neuberger, 1938; Johansen et at., 1960). However, by diluting the acid, it is possible to reduce the rate of de-N-acetylation relative to that of glycosidic splitting, and thus to recover a high proportion of the original N-acetylamino sugars for the purpose of identification. Once the sugars have been identified, they can be determined quantitatively either (a) by measuring the total N-acetyl content of the polysaccharide material, or (b) (if it is certain that in the polymer all amino sugars are N-acetylated) by determining the free (de-N-acetylated) amino sugars following their complete release by hydrolysis with 4-8 N HCI. A technique with important possibilities for measuring losses of sugars during hydrolysis is the isotope dilution method employed by Frdncois et al. ( 1 962). c . Analysis of Polysaccharide Hydrolyzates. Methods for the separation, identification, and determination of the monosaccharides in hydrolyzates of polysaccharides have been reviewed thoroughly elsewhere (Percival, 1963; Bishop, 1964; Davidson, 1966; Neuberger and Marshall, 1966a; Northcote, 1966; Spiro, 1966). Chromatographic techniques are of primary importance for the separation, detection, and preliminary identification of sugars. However, chromatographic behavior alone does not allow unequivocal identification of an individual sugar; this requires either isolation of the pure sugar in crystalline form or conversion of the sugar to a characteristic crystalline derivative.

226

G . D. SWINCER, J. M. OADES, AND D. J. GREENLAND

Satisfactory removal of the hydrolyzing acid prior to chromatography is often difficult. Volatile acids are removed quite simply by evaporation, although with HCI, particularly in the presence of heavy metals, destruction of some sugars can occur. Strong mineral acids are probably best removed with strong anion exchange resins in the carbonale, bicarbonate, or acetate forms. However, precipitation with barium hydroxide or barium carbonate is still used regularly for the removal of HzS04 in spite of the danger of selective adsorption of sugars by the precipitate. An unusual approach with definite possibilities for the removal of H a S o l is selective extraction of the sulfate with an immiscible organic liquid (Becker and Shefner, 1964). Paper chromatography which has been the most useful routine method for rapid preliminary characterization of sugar mixtures has not proved completely adequate for precise quantitative work. For this reason it is being gradually superseded by such methods as column chromatography with ion exchange resins and gas-liquid chromatography, which offer several distinct advantages. Ion exchange chromatography has been applied successfully not only to the charged sugars (amino sugars and uronic acids), but also, by formation of their borate complexes, to neutral sugars. Fully automated procedures have been developed, and the method can be used as a preparative technique for milligram quantities of sugars. Gasliquid chromatographic analysis of sugars has been developed only very recently, and it is almost certain that further advances will be made. This technique offers the possibility of rapid and precise quantitative analysis of very small samples containing a complex mixture of sugars (Oades, 1967a). Except for gas-liquid chromatography, most of the methods used for quantitative determination of the separated sugars are colorimetric. Because of the relatively low specificity of some of the color reactions, special care is required to ensure that all interfering substances are taken into account. Allowance also has to be made for the variations in color yield from sugar to sugar. Different methods are needed at least for each class of monosaccharide (e.g., hexoses, pentoses, uronic acids, hexosamines). Specific enzyme assays have been worked out for a few monosaccharides. As further assays are developed, this approach is likely to become useful for the analysis of complex hydrolyzates since there is normally no need for prior separation of the sugars. 2 . Physicochemical Analysis. Most of the methods generally used on high polymers for the determination of molecular size, shape and flexibility can be used with polysaccharides. They may be listed as: (a) hydrodynamic methods (e.g.,


227

sedimentation analysis by ultracentrifugation, viscometry, streaming birefrigence measurements, determination of diffusion constants); (b) methods based on the colligative properties of the molecules (e.g., osmometry, isothermal distillation); (c) methods involving measurements which depend directly on the physical size of the molecules (e.g., light scattering); (d) end-group determination by chemical assay; and (e) techniques that are fundamentally separative rather than analytical in nature (e.g., gel filtration, ultrafiltration through membranes of graded pore diameter, electrokinetic ultrafil tration, density gradient centrifugation, free-boundary electrophoresis). The application of these methods to polysaccharides has been reviewed in considerable detail (Greenwood, 1952, 1956; Whistler and Smart, 1953; Whistler and Corbett, 1957; Banks and Greenwood, 1963; Horton and Wolfrom, 1963; Gibbons, 1966). For thorough characterization of a polysaccharide, as many methods as possible should be combined. Many of the methods yield reliable quantitative information only with preparations that are homogeneous (i.e., “consisting of molecules having identical structure but not necessarily the same molecular weight” Banks and Greenwood, 1963), and satisfactory interpretation of the results often demands, in addition, a narrow distribution of molecular weight. For example, the diffuse sedimentation boundary produced during ultracentrifugation of a very polydisperse sample allows computation only of an approximate value for the average molecular weight. These points are of obvious significance with respect to the analysis of soil polysaccharide preparations which have proved particularly difficult to purify and fractionate. Clearly, it is imperative either to ensure that homogeneous polysaccharides have been prepared before attempting physicochemical characterization or to use only the limited range of techniques that can be applied satisfactorily to heterogeneous preparations. Undoubtedly for the most complete and precise information homogeneous polymers must be obtained, but the problems involved both in isolation of the required fraction and in assessment of its homogeneity are extremely difficult. In fact, Banks and Greenwood ( 1 963) pointed out that “it is doubtful if any polysaccharide has been examined by sufficient methods to prove unambiguously that it is homogeneous.” The techniques that can be applied most satisfactorily to the characterization of polysaccharides in heterogeneous mixtures are those based on separation methods.

C. SEPARATION METHODS A wide variety of methods have been used for the isolation of individual polysaccharides from biological materials (Whistler and Smart, 1953;

228

G. D. SWINCER, J. M. OADES, A N D D. J . GREENLAND

Pigman and Platt, 1957; Bouveng and Lindberg, 1960; J . E. Scott, 1960; Banks and Greenwood, 1963; Barker, 1963; Horton and Wolfrom, 1963: Jeanloz, 1963; Kertesz, 1963; Brimacombe and Webber, 1964; Whistler, 1965; Northcote, 1966). Initial extraction is usually the most critical step because it is often at this point that the most drastic treatments are needed. Separation of polysaccharides from cellular material is rarely easy, and the problem is especially formidable with polysaccharides that are associated intimately with an insoluble matrix as in cell walls. There is an obvious relationship here with the problem of separating polysaccharides from the mineral matrix of soils. With almost all the procedures sufficiently powerful to solubilize such polysaccharides, there is a definite risk of degradation (Northcote, 1966). Any modification of the structure of the molecules or the molecular weight distribution, or both, may invalidate many of the subsequent analyses. Once the polysaccharides have been brought into solution, they can be purified and fractionated by a variety of techniques. Fractional precipitation or dissolution of polysaccharides (and polysaccharide acetates or nitrates), either by changing the solvent composition, or pH, or temperature, has been widely used. With complex mixtures this approach is only of limited application, except for the removal of extraneous material, because of the tendency to coprecipitation and occlusion of other polymers. Moreover, in most cases fractional precipitation merely subdivides the polymolecular system into fractions on a molecular weight basis: each individual fraction represents a narrow molecular weight range, but still remains a mixture of polysaccharide types. Gel filtration is assuming a place of primary importance in the study of heterogeneous polydisperse systems. It gives very efficient separation of polydisperse materials into fractions covering a limited range of molecular weight and also provides a simple and effective method for removing low-molecular-weight impurities (Granath and Flodin, 196 I ; Anderson et al., 1965; Anderson and Stoddart, 1966; Granath and Kvist, 1967). Separation techniques based on the ionic properties of the polysaccharides offer the best prospects for isolation of homogeneous materials. These methods are amenable to all soluble polysaccharides as even those polysaccharides with no readily ionizable groups are generally slightly charged, particularly in alkaline solution, due to ionization of the hydroxyl groups, while complex formation with certain ions, notably borate ions, increases the negative charge on a polysaccharide. The techniques that have been applied successfully include selective precipitation with metal ions or quaternary ammonium salts, and ion exchange chromatography, particularly with charged celluloses. These procedures can generally be


229

made very sensitive to small differences in the net charge of the polymers and they are often capable of resolving mixtures of closely related polysaccharide species. The ion exchange procedures in particular have given some excellent separations (e.g., Jermyn, 1962; Antonopoulos et al., 1967) and the technique offers considerable scope for further refinement. In addition, some very good separations of mucopolysaccharides have been achieved recently by a combination of fractional precipitation and column chromatography (e.g., Antonopoulos et at., 1964; Pearce and Mathieson, 1967). Other procedures with distinct possibilities for the fractionation of mixtures of polysaccharides are density gradient centrifugation (e.g., Charlwood, 1966; Franek and Dunstone, 1967) and selective precipitation with antisera (e.g., Heidelberger et af., 1955), neither of which are based directly on differences in polymer size or charge. V. Summary and Conclusions

Carbohydrates represent 5 to 25 percent of soil organic materials. They consist of a wide range of monosaccharides, such as hexoses, pentoses, deoxy- and 0-methyl sugars, uronic acids, and amino sugars. Such monosaccharides exist in polymeric molecules of various sizes and degrees of complexity, which are associated more or less strongly with inorganic colloids in soils. Large proportions of carbohydrates in many soils are present in partly decomposed plant and animal remains. Glucose, presumably in the form of cellulose, is dominant in such materials. Plant litter and roots, either living or dead, are the main primary source of soil carbohydrates, but the composition of soil polysaccharides, apart from obvious plant remains, would suggest a microbial origin, either wholly or in part, e.g., plant materials which have been modified by the soil flora and fauna. Polysaccharides have been extracted from soils by many different chemical reagents, and recently methods have been devised that enable most of the carbohydrates to be isolated from other soil materials. The extracted polysaccharides show a continuum of molecular sizes and contain a wide range of neutral and charged monosaccharides, amino acids, and other unidentified nitrogenous and acid components. Carbohydrates from different soils are similar in chemical composition suggesting that the microbial population of different soils is qualitatively similar. Many methods have been used to fractionate extracted soil polysaccharides usually with limited success. The most successful methods have been based on gel filtration and chromatography on charged supports

230

G . D. SWINCER, J . M. OADES, AND D. J . GREENLAND

such as cellulose. However, fractions obtained are still complex and contain a range of different components. This complexity is not surprising in view of the wide range of substrates, organisms, and metabolic products of organisms that are subjected to chemical extraction and fractionation procedures. Generally the “turnover” of sugars in soil carbohydrates appears to be rapid, but some microbial polysaccharides are resistant to breakdown by soil organisms. The subject is complicated because of interactions with metal cations and sorption on colloid surfaces. The composition of soil polysaccharides suggests that in soils they may carry charged sites and take part in exchange reactions and act as energy sources for heterotrophic organisms. However, the main stimulus for the study of soil polysaccharides has arisen from repeated indications of their favorable influence on soil physical conditions. Much work has been directed toward this aspect, and it has been shown that microbially produced soil polysaccharides are capable of stabilizing soil aggregates against dispersion in water. N o specific fraction has yet been definitely identified as particularly active, but it is suggested that the larger polysaccharides produced by microorganisms in coarse pores of aggregates are likely to be the most effective. The mechanisms by which these polymers react with inorganic colloids is not understood, but the complex preparations obtained from soils are sorbed from aqueous solution by clay materials, and further work on the fractionation of carbohydrate preparations followed by studies of the sorption of these “purer” characterized fractions will undoubtedly prove to be worthwhile. Methods for the isolation of polysaccharides from other soil materials in good yield are now available and methods for the analysis of the extracted polysaccharides have been developed by carbohydrate chemists. Combinations of these techniques in the future will enable new information about the composition, origin, and function of soil carbohydrates to be obtained. Particularly useful information should arise from the cooperation of chemists and microbiologists using techniques involving isotopically labeled materials. REFERENCES Acton, C. J., Paul, E. A., and Rennie, D. A . I963a. Can. J . Soil Sci. 43,14 I - I 50. Acton, C. J . , Rennie, D. A., and Paul, E. A. 1963b. Can. J . Soil Sci. 43, 201-209. A d a m , G . A. 1965. Methods Carbohydrate Chem. 5,269-276. Alexander, M. 196 I . “Introduction to Soil Microbiology,” Wiley, New York. Allison, L. E. 1947. Soil Sci. 63,439-450.


23 1

Anderson, D. M. W.. and Stoddart, J. F. 1966. Carbohydrate Res. 2,104-1 14. Anderson, D. M. W., Dea, I . C. M.. Rahman, S., and Stoddart, J. F. 1965. Chem. Commun. pp. 145- 146. Antonopoulos, C. A.. Gardell, S., Szirmai, J . A., and D e Tyssonsk, E. R. 1964. Biochim. Biophys.Acta 83,1-19. Antonopoulos, C. A., Fransson, L - k , Heinegird, D., and Gardell, S. 1967. Biochim. BiophyS.ACta 148,158-163. Banks, W., and Greenwood, C. T. 1963.Advan. Carbohydrate Chem. 18,357-398. Barker, S. A. 1963. Comprehensive Biochem. 5,246-26 1. Barker, S. A., Finch, P.. Hayes, M. H. B., Simmonds, R. G., and Stacey, M. 1965. Nature 205,68-69. Barker, S. A., Hayes, M. H. B., Simmonds, R. G., and Stacey, M. 1967. Carbohydrate Res. 5,13-24. Barrett, A. J., and Northcote, D. H. 1965. Biochem.J. 94,617-627. Becker, M. J., and Shefner, A. M. 1964. Nature 202,803. Bernier, B. 1958a. Biochem. J . 70,590-598. Bernier, B. I958b. Can. J . Microbiol. 4, 195-204. Bishop, C. T. 1964.Advan. Carbohydrate Chem. 19,95. Black, W . A. P.. Cornhill, W. J., and Woodward, F. N. 1955. J . Appl. Chem. (London), 5,484-492. B0bbit.J. M. 1956.Advan. Carbohydrate Chem. 11,1-41. Bouveng, H. O., and Lindberg, B. 1960.Advan. CarbohydrateChem. 15,53-89. Bremner, J. M. 1950.J. SoilSci. 1, 198-204. Bremner, J. M. 1954. J. Soil Sci. 5 , 2 14-232. Bremner, J . M., and Harada, T . 1959. J.Agr. Sci. 52, 137- 146. Bremner, J. M., and Lees, H. 1949. J.Agr. Sci. 39,274-279. Bremner, J. M., and Shaw, K. l954.J. Agr. Sci. 44,152-159. Brimacombe, J. S.,and Webber, J . M. 1964. “Mucopolysaccharides,” Elsevier. Amsterdam. Brink, R. H., Dubach, P., and Lynch, D. L. 1960. Soil Sci. 89, 157-166. Burges, A. 1967. I n “Soil Biology”(N. A. Burges and F. Raw, eds.), pp. 479-492. Academic Press, New York. Charlwood, P. A. 1966. Brit. Med. Bull. 22, 121-126. Cheshire, M. V., and Mundie, C . M. 1966. J. Soil Sci. 17,372-381. Chesters, G . , Attoe, 0. J., and Allen. 0. N . 1957. Soil Sci. Soc. Am. Proc. 26,466-469. Choudri, M. B., and Stevenson, F. J. 1957. Soil Sci. Soc. Am. Proc. 21,508-5 13. Clapp, C . E. 1957. Ph.D. Thesis, Cornell University, New York. Clapp, C. E., and Emerson, W. W. 1965. Soil Sci. Soc. Am. Proc. 29,127- 134. Clapp. C. E., Davis, R. J., and Waugaman, S. N. 1962. Soil Sci. Soc. A m . Proc. 26,466-469. Daji, J. A. 1932. Biochem. J. 26, 1275-1280. Dalgleish, C. E. 1955. J. Clin. Pathol. 8,73-78. Dart, P. J., and Mercer, F. V. 1964.Arch. Mikrobiol. 47,344-378. Davidson, E. A. 1966. Methods Enzymol. 8,52-60. Dawes, C. J . , and Bowler, E. 1959. Am. J . Botany 46,561-565. Deshpande, T. I-., Greenland. D. J.. and Quirk, J . P. 1968. J . Soil Sci. 19, 108- 122. Deuel, H . , Dubach, P., and Mehta, N. C. 1960. Sci. Proc. Roy. Dublin Soc. A l , I I S 121.

Dische, 2. 1947. J . B i d . Chem. 167, 189-198.

232

G . D. SWINCER, J. M. OADES, A N D D. J. GREENLAND

Dormaar, J. F. 1967. Soil Sci. 103,4 17-423. Dubach, P., and Mehta, N . C. 1963. Soils Fertilizers 26,293-300. Dubach, P., Zweifel, G., Bach, R., and Deuel, H. 1955. Z . Pjianzenernaehr. Dueng. Bodenk. 69,97- 108. Duff, R. B. 1952a.J. Sci. FoodAgr. 3, 140-144. Duff, R. B. 1952b. Chem. & lnd. (London) p. 1104. Duff, R. B. 1954. Chem. & Ind. (London) p. 15 13. Duff, R. B. 1961. J . Sci. Food Agr. 12,826-83 I. Erskine, A. J., and Jones, J . K. N. 1956. Can. J . Chem. 34,821-826. Estermann, E. F., and McLaren, A. D. 196 I.Plant Soil 15,243-260. Evans, L. T. 1959. J . Soil Sci. 10, I10-1 18. Finch, P., Hayes, M. H. B., and Stacey, M. 1967. Intern. Soc. Soil Sci., Trans. Commun. I 1 and I V , Aberdeen, 1966, pp. 19-32. Intern. SOC.Soil Sci. Ford, G. W., Greenland, D. J., and Oades, J. M. 1969. J . Soil Sci. 20, (in press). Forsyth, W. G. C. 1947. Biochem. J . 41, 176-181. Forsyth, W. (3. C. 1950. Biochem. J . 46, 141-146. Forsyth, W. G. C. 1954. Trans. 5th Intern. Congr. Soil Sci., Leopoldville, 1954 Vol. 3. pp. 119-122. Forsyth, W. G . L a n d Webley, D. M. 1949. J . Gen. Microbiol. 3,395-399. Foster, A. B., and Webber, J. M. 1960. Advan. Carbohydrate Chem. 15,37 1-393. Foster, A. B., Newton-Hearn, P. A., and Stacey, M. 1956.5. Chem. Soc. pp. 30-36. Francois, C., Marshall, R. D., and Neuberger, A. 1962. Biochem.1. 83,335-341. Franek, M. D., and Dunstone, J. R. 1967. J . Biol. Chem. 242,3460-3467. Gascoigne, J. A., and Gascoigne, M. M. 1960. “Biological Degradation of Cellulose.” Butterworth, London and Washington, D.C. Geoghegan, M. J . , and Brian, R. C. 1946. Nature 158,837. Geoghegan, M. J., and Brian, R. C. 1948. Biochem. J . 43, 14. Gibbons, R. A. 1966. In “Glycoproteins” (A. Gottschalk, ed.), pp. 19-95. Elsevier, Amsterdam. Gottschalk, A. 1966. In “Glycoproteins” (A. Gottschalk, ed.), pp. 96-1 I I. Elsevier, Amsterdam. Gottschalk, A., and Ada, G. L. 1956. Biochem. J . 62,681-686. Granath, K. A., and Flodin, P. 1961. Makromol. Chem. 48, 160-171. Granath, K. A., and Kvist, B. E. 1967. J . Chromatog. 28,69-81. Graveland, D. N., and Lynch, D. L. 1961. Soil Sci. 91, 162-165. Greenland. D. .I.1956. J . Soil Sci. 7, 319-334. Greenland. D. J., Lindstrom, G. R., and Quirk. J. P. 1961. Natrirr 191, 1283-1284. Greenland, D. J., Lindstrom, G. R.,and Quirk, J. P. 1962. Soil Sci. Soc. A m . Proc. 26, 366-37 I . Greenwood, C. T. 1952. Advan. Carbohydrate Chem. 7,289-332. Greenwood, C. T. 1956. Advan. Carbohydrate Chem. 11,336-393. Griffiths, E. 1965. Biol. Rev. Cambridge Phil. Soc. 40, 129-142. Griffiths, E., and Jones, D. 1965. Plant Soil 23, 17-33. Gupta, U. C. 1967. In “Soil Biochemistry” (A. D. MacLaren and G . H. Petersen, eds.), pp. 91-1 18. Arnold, London. Gupta, U. C., and Sowden, F. J. 1964. Soil Sci. 97,328-333. Gupta, U . C., and Sowden, F. J . 1965. Can. J . Soil Sci. 45,237-240. Gupta, U. C., Sowden, F. J., and Stobbe, P. 1963. Soil Sci. Soc. Am. Proc. 27, 380382.

SOIL P O L Y S A C C H A R I D E S

233

Harmsen, G. W., and Jager, G. 1962. Nature 195, 1 I 19- 1 120. Harris. R. F., Allen, 0. N., and Chesters, G. 1966. Plant Soil 25, 361-371. Harris, R F., Allen, 0. N., Chesters, G.. and Attoe, 0. J. 1963. Soil Sci. Soc. Am. Proc. 27,542-545.

Hartree, E. F. 1964. Anal. Biochem. 7, 103-109. Haworth, W. N.. and Hirst, E. L. 1930. J . Chem. Soc. pp. 2615-2635. Haworth, W. N.. Pinkard, R. W., and Stacey, M. 1946. Nature 158,836-837. Heidelberger, M.. Dische, Z . , Neely, W. B., and Wolfrorn, M. L. 1955. J . Am. Chem. Soc. 77,351 1-3518.

Hocevar, B. J., and Northcote, D. H. 1957. Nature 179,488-489. Horton, D., and Wolfrom, M. L. 1963. ComprehensiveBiochern. 5,188-232. Ivarson, K. C., and Sowden, F. J. 1962. SoilSci. 94,245-250. Jayrne. G., and Lang, F. 1963. Methods Carbohydrate Chem. 3 , 7 5 4 3 . Jeanloz, R. W. 1963. Comprehensive Biochem. 5,262-296. Jenny, H., and Grossenbacher. K. 1963. SoilSci. Soc. A m . Proc. 27,273-277. Jerrnyn, M. A. 1962. Australian J . B i d . Sci. 15,787-79 I. Johansen, P. G., Marshall, R. D., and Neuberger, A. 1960. Biochem. J . 77,239-247. Jones, D.. Morre, D. J.,and Mollenhauer, H. H. 1966.Am. J . Bottrm 53.62 I Keefer, R. J., and Mortensen, J. L. 1963. Soil Sci. Soc. Am. Proc. 27,156- 160. Kenner, J., and Richards, G. N . 1957. J . Chem. SOC. pp. 3019-3024. Kertesz, Z. I . 1963. Comprehensive Biochem. 5, 233-245. Kononova, M. M. 1961. “Soil Organic Matter.” Pergamon Press, Oxford. Lowe, L. E. 1968. Can. J . Soil Sci. 48, 2 15-2 17. Lynch, D. L.. and Cotnoir, L. J., Jr. 1956. Soil Sci. Soc. A m . Proc. 20,367-370. Lynch, D . L., Wright, L. M., and Cotnoir, L. J., Jr. 1956. Soil Sci. Soc. A m . Proc. 20,6-9. Lynch, D. L.. Wright, L. M., and Olney, H. 0. 1957. Soil Sci. 84,405-41 I . Lynch, D . L., Olney, H. O., and Wright, L. M. 1958. J . Sci. Food Agr. 9,56-60. McCalla, T. M. 1943. Soil Sci. Soc. Am. P roc. 7, 209-2 14. McCalla, T. M. 1945. Soil Sci. 59,287-297. Martin, J. K. 1964. New Zealand J . Agr. Res. 7, 723-749. Martin. J. P. 1945a. Soil Sci. 59, 163- 174. Martin, J. P. 1945b. J . Bacteriol. 50, 349-360. Martin, J. P. 1946. Soil Sci. 61, 157-166. Martin, J. P., and Craggs, B. A . 1946. J . Am. Soc. Agron. 38, 332-339. Martin, J. P., and Richards, S. J. 1963. J . Bacteriol. 85, 1288-1294. Martin, J. P., Ervin, J. O., and Shepherd, R. A . 1965. Soil Sci. SOC.Am. Proc. 29,397-400. Martin, J. P., Ervin, J . 0.. and Shepherd, R. A. 1966. Soil Sci. Soc. Am. Proc. 30,196-200. Mayaudon.J.,andSimonart, P. 1959. PlantSoil11,181-192. Mehta, N . C., Streuli. H., Muller, M., and Deuel, H. 1960. J . Sci. Food Agr. 11, 40-47. Mehta, N . C., Dubach, P., and Deuel, H. 1961.Advan. Carbohydrate Chem. 16,335-355. Moggridge, R. C. G . ,and Neuberger, A . 1938. J . Chem. Soc. pp. 745-750. Morre, D. J., Jones, D. D., and Mollenhauer. H. H. 1967. Planta 74,286-30 I . Mortensen, J . L. 1961 Trans. 7th Intern. Congr. Soil Sci.,Madison, Wise., 1960 Vol. 2, pp. 98- 104. Elsevier, Amsterdam. Mortensen, J. L., and Himes, F. L. 1964. In “Chemistry of the Soil” ( F . E. Bear, ed.), 2nd ed., pp. 206-24 1. Reinhold, New York. Mortensen, J. L., and Schwendinger, R. B. 1963. Geochim. Cosmochim. Acta 27,201-208. Muller, M., Mehta. N . C., and Deuel, H. 1960. Z . Pjanzenernaehr. Dueng. Bodenk. 90, 139- 145.

234

G. D. SWINCER, J. M . OADES, A N D D. J. GREENLAND

Neuberger, A., and Marshall, R. D. 1966a. In “Glycoproteins” (A. Gottschalk, ed.), pp. 190-234. Elsevier, Amsterdam. Neuberger, A,, and Marshall, R. D. 1966b. In “Glycoproteins” (A. Gottschalk, ed.), pp. 235-272. Elsevier, Amsterdam. Neuberger, A,, Marshall, R. D., and Gottschalk, A. 1966. In “Glycoproteins” (A. Gottschalk, ed.), pp. 158- 189. Elsevier, Amsterdam. Northcote, D. H. 1954. Biochem. J . 58,353-358. Northcote, D. H . 1966. Brit. Med. Bull. 22,180-184. Oades, J. M. 1967a. J . Chromatography 28, 246-252. Oades, J. M. 1967b. Australian J . Soil Res. 5, 103-1 IS. Oades, J. M., and Swincer, G. D. 1968. Trans. 9th Intern. Congr. Soil Sci., Adelaide, 1968 Vol. 3, pp. 183-192. Intern. Soc. Soil Sci. and Angus & Robertson, Sydney, Australia. Ogston, A. G. 1958. Bi0chem.J. 70,598-599. Overend, W. G., Rees, C. W., and Sequeira, J. S. 1962.J . Chem. Soc. pp. 3429-3440. Parsons, J. W., and Tinsley, J. 196 I .Soil Sci. 92,46-53. Pearce, R. H., and Mathieson, J. M. 1967. Can.J . Biochem. 45,1565- 1576. Percival, E. 1963. Comprehensive Biochem. 5,l-66. Perry, M. B.,and Hulyalkar, R. K. 1965. Can. J . Biochem. 43,573-584. Pigman, W. 1957. In “The Carbohydrates” (W. Pigman, ed.), pp. 1-76 Academic Press, New York. Pigman, W., and Platt, D. 1957. In “The Carbohydrates” (W. Pigman, ed.), pp. 709-732. Academic Press, New York. Reichstein, T., and Weiss, E. 1962.Advan. Carbohydrate Chem. 17,65-120. Rennie, D. A., Truog, E., and Allen, 0. N. 1954. Soil Sci. SOC.A m . Proc. 18, 399-403. Rogers, H. J. 1966. Brit. Med. Bull. 22, 185-189. Roulet, N., Dubach, P., Mehta, N. C., Muller-Vonmoos, M., and Deuel, H. 1963a. Z . Pflanzenernaehr. Dueng. Bodenk. 101,2 10-2 14. Roulet, N., Mehta, N. C . , Dubach, P., and Deuel, H. 1963b. Z . Pjanzenernaehr. Dueng. Bodenk. 103, 1-9. Rovira, A. D. 1956. Plant Soil 7, 178-194. Rovira, A. D. 1962. Soils Fertilizers 25,167-172. Rovira, A. D., and Greacen, E. L. 1957. Australian J . Agr. Res. 8,659-673. Rovira, A. D., and McDougall, B. M. 1967. In “Soil Biochemistry” (A. D. MacLaren and G. H. Petersen, eds.), pp. 417-471. Arnold, London. Russell, E. W. 1961. “Soil Conditions and Plant Growth,” 9th ed. Longmans, Green, New York. Saeman, J. F., Moore, W. E., Mitchell, R. L., and Millett, M. A. 1954. Tappi 37,336-343. Salomon, M. 1962. Soil Sci. SOC.A m . Proc. 26,5 1-54. Salton, M. R. J. 1960. “Microbial Cell Walls.” Wiley, New York. Salton, M. R. J. 1964. “The Bacterial Cell Wall.” Elsevier, Amsterdam. Samtsevich, S. A. 1965. Soviet Plant Physiol. 12,73 1-740. Sanderson, G. W., and Perera, 9.P. M. 1966. Analyst 91,335-336. Schmid, K . 1964. Chimia (Aarau) 18,321-327. Scott, F. M., Hammer, K. C., Baker, E., and Bowler, E. 1958. A m . J . Botany 45,449-46 1 Scott, F. M., Bystrom, 9.G., and Bowler, E. 1963. Science 140,63-64. Scott, J. E. 1960. Methods Biochem. Anal. 8, 145-197. Sevag, M. G . , Lackman, D. B., and Smo1ens.J. 1938.5. B i d . Chem. 121.325-436.


235

Shafizadeh, F. 1958. Advan. Carbohydrate Chem. 13,9-61. Simonart, P. 1964. Ann. Inst. Pasteur 107,7-20. Siu, R. G . H . 195 1 . ”Microbial Decomposition of Cellulose.” Reinhold, New York. Smith, F., and Montgomery, R. 1959. “The Chemistry of Plant Gums and Mucilages.” Reinhold, New York. Sowden, F. J., and Ivarson, K. C . 1962. Soil Sci. 94,340-344. Spiro, R . G . 1966. Methods Enzymol. 8,3-26. Stacey, M., and Barker, S. A. 1960. “Polysaccharides of Microorganisms.” Oxford Univ. Press, London and New York. Stevenson, 1. L. 1964. In “Chemistry of the Soil” (F. E. Bear, ed.), 2nd ed., pp. 242-291. Reinhold, New York. Streuli, H., Mehta, N. C., Muller, M., and Deuel, H. 1958. Mitt. Gebiete Lebensm. Hyg. 49,396 (quoted by Mehta et al., I96 I ). Swaby. R. J. 1949. J . Gen. Microbiol. 3,236-254. Swincer, G. D., Oades, J. M., and Greenland, D. J. 1968a. Australian J . Soil Res. 6,21 1224.

Swincer, G. D., Oades, J. M., and Greenland, D. J . 1968b. AustrulianJ. Soil Res. 6,225235.

Thomas, R. L. 1963. Ph.D. Thesis, Ohio State University. Thomas, R. L., and Lynch, D. L. 1961. SoilSci. 91,312-3 16. Thomas, R. L., Mortensen, J. L., and Himes. F. L. 1967. Soil Sci. Soc. A m . Proc. 31,568570.

Thompson, A,, Wolfrom, M. L., and Pacsu, E. 1963. Methods Carbohydrate Chem. 2, 2 15-220. Tinsley, J., and Salam, A. I96 1. Soils Fertilizers 24, 8 1-84. Toogood, J. A., and Lynch, D . L. 1959. Can. J . Soil Sci. 39, 15 1-156. Walborg, E. F., and Ward, D. N. 1963. Biochim. Biophys. Acta Walhorg. E. F.. and Ward. D. N. 1963. Biochim. Biophys. Acra 78, 304-312. Watson, J. H., and Stojanovic, 9 . J. 1965. Soil Sci. 100, 57-62. Webber, L. R. 1965. SoilSci. Soc. Am. Proc. 29,39-42. Webley, D. M., Duff, R. B., Bacon, .I.S. D.. and Farmer, V. C . 1965. J . Soil Sci. 16, 149157.

Weidel, W., and Pelzer, H. 1964.Advan. Enzymol. 26,193-232. Whelan, W. J. 1960. Ann. Rev. Biochem. 29,111- 130. Whistler, R. L. 1965. Methods Carbohydrate Chem. 5, Arts. 19-54. Whistler, R. L., and BeMiller, J. N. 1958. Advan. Carbohydrate Chem. 13,289-329. Whistler, R. L., and Corbett, W. M. 1957. I n “The Carbohydrates” (W. Pigman, ed.), pp. 641-708. Academic Press, New York. Whistler, R. L.. and Kirby, K. W. 1956.J.Am. Chem. SOC. 78,1755-1759. Whistler, R. L., and Lauterbach, C. E. 1958. Arch. Biochem. Biophys. 77,62-67. Whistler, R. L., and Smart, C. L. 1953. ”Polysaccharide Chemistry.” Academic Press, New York. Whitehead, D. C., and Tinsley, J. 1963. J . Sci. FoodAgr. 14,849-857. Whitehead, D. C., and Tinsley, J. 1964. Soil Sci. 97,34-42. Williams, 9. G., Greenland, D. J., and Quirk, J. P. 1967. Australian J . Soil Res. 5,77-92. Winogradsky, S . 1929. Ann. Inst. Pasteur 43,549-633. Winzler, R. J. 1955. Methods Biochem. Anal. 2,279-3 I I . Wolfrom, M. L., and Thompson, A. 1957. In “The Carbohydrates” (W. Pigman, ed.), pp. 188-240. Academic Press, New York.


FRAGIPAN SOILS OF THE EASTERN UNITED STATES R. B. Groosman and F. J. Carlisle Soil Conservation Service, United States Deportment of Agriculture, Lincoln, Nebraska ond Hyattsville, Maryland

I. 11.

Introduction .. .. .. ... ... . . . . ... .. . ... .. . Horizons of Fragipan Soils ........ B. C.

Horizon Sequences .................................................... Fragipan Expression . ..... . .

111.

IV. Properties of Fragipans ........... ......... ... .. ... ......................... A. Composition ..............

........................................ C. Consistenc V.

r Regime ..................................

................. C . Movement of Low-Tension Water ................................................. v1. Genesis of Fragipans .... ., . ., . .. . .. . .. . ... ........... A . Bonding of the Fragipan B. Development of Fragipan Soils ...... . . . ..... ................................... . .. . VII. Fragipans and Soil U s e ...................................................................... A. Plant Growth ............. ......... ._......... B. Engineering Manipulation . ...... ... . ... . .. . ................... .. .... ... . . ... ......... VIII. Classification of Fragipan Soils ....,..... ... ..... ... ....................... ..... . . ......... 1x. Unresolved Problems ........ ................................................................ X. Summary . ..... ... .... . ... References ........ .. . ...... ...... .. .. . . . ... .. . ... ....... .... .... . . ... ... ....... ... ............... Appendix.. .. ...... .......

Pft@’ 237 240 240 24 1 242 244 246 246 249 25 1 254 254 254 255 256 256 259 263 263 264 265 269 271 212 216

I. Introduction

Twenty years ago Winters and Simonson ( I 95 1) very adequately reviewed fragipans in three pages of this publication. Since then a large number of pertinent papers have appeared, and it seems appropriate to consider the subject anew. This review is restricted to fragipans in soils of the eastern United States. 237

238

R. B. GROSSMAN A N D F. J. CARLISLE

As Figure 1 illustrates, fragipans are prominent features of some soils. Despite their prominence in the soil profile, the definition of fragipans is vague. The vagueness arises from the nature of the defining properties.

FIG. I . Block diagram of a soil with a fragipan that occurs in northeastern Wisconsin. The fragipan horizons are indicated by the suffix x. (Modified from Olson and Hole, 19671968.)

The principal defining property is the type of failure exhibited by moist soil material when pressure is applied. Yield is abrupt; a moist piece of soil breaks into fragments rather than deforming. This has been described as brittle consistence. The failure by some fragipans has been likened (Daniels et al., 1966) to the crumbling of a dry graham cracker. This defining property does not lend itself to laboratory measurement. The natural organization of the soil fabric must be kept, which raises difficulties with the measurement. Also, the failure observed depends sensitively on the water content. Resultantly, field observations are subject to large variability. A body of standardized measurements on the defining property of fragipans does not exist. Many descriptions of fragipans give no indication of the degree of brittleness. Winters and Simonson ( 1 95 1) listed some of the early publications on soils with fragipans. Olson (1962) presented a comprehensive review. As indicated in Section VIII, fragipans were recognized as pans long before they were given their current name. The concept of a pan in soil morphology involves restricted root penetration, occurrence below the

FRAGIPAN SOILS OF THE EASTERN UNITED STATES

239

soil surface, and for most kinds of pans-although fragipans are an exception- the pronounced accumulation of a substance, such as silicate clay, iron oxides, silica, or carbonates. Historically, the concept also has involved the notion of an unusual or extreme expression of a feature or attribute and something apart from the normal or expected sequence of horizons. Reasons for considering fragipans to be genetic soil horizons have been discussed by the Soil Survey Staff ( 1960) and by Carlisle et al. (1 957). Fragipans parallel the soil surface, have upper boundaries at moderate depths (20 to 100 cm.), have a constant spatial relationship to other soil horizons in a given kind of soil, occur in geological materials of widely different origins, and may transgress differing parent materials in a local area. Their range in thickness (20 to 200 cm.) is within generally accepted limits for soil horizons. Cline (1952) states, “The consistency of depth at which it [the fragipan] occurs . . . indicates considerable control by some thing or things related to the land surface, and this more than any one thing is probably responsible for its designation as a genetic horizon.” Fragipans are subsoil horizons unless the soil has been truncated. As subsoil horizons they are subject to accumulation of illuviated substances. They are beneath the depth of maximum root concentration; organic matter accumulation is not high. Mechanical disturbance of the soil fabric is weak. Consequently, fragipans retain the marks of translocation of substances and reorganization of the fabric. Carlisle et al. (1957) have defined fragipans as follows: “Compact horizons (of high bulk density) which are hard to extremely hard when dry and firm to very firm when moist and display the property of ‘brittleness’ when both dry and moist. The term ‘brittleness,’ as used here, embodies a type of physical behavior that is characteristic of but not exclusively associated with fragipan materials. The term is used to characterize a condition in which a fragment of the material sustains increasing pressure without detectable deformation until a critical pressure is reached, at which point the material suddenly shatters.” This definition differs from that in the Soil Survey Manual (Soil Survey Staff, 195 1) by putting more emphasis on the retention of brittleness when moist. The definition in the 7rh Approximation (Soil Survey Staff, 1960) closely parallels that of Carlisle et al. ( 1 957). Fragipans have a number of characteristics, which, though not necessary to the definition, form part of the general description applicable to most but not all fragipans. They include high silt, very fine sand, or fine sand contents: moderate or low clay content: low organic matter content: medium to high bulk density when moist; slow or very slow satu-

240


rated hydraulic conductivity; well expressed mottling; presence of bleached cracks or fracture planes that form a coarse polygonal pattern on a horizontal plane; weak pedological structural expression within the polyhedrons outlined by the bleached cracks; clearly identifiable and planar upper boundaries; presence of bodies of moved clay; and few roots, with those present largely restricted to the cracks between large polyhedrons. 11. Horizons of Fragipan Soils

A. TERMINOLOGY A N D HORIZONNOMENCLATURE Fragipans occur in soils with argillic, cambic, or spodic horizons. The cambic and spodic horizons, if present, occur above the fragipan. The argillic horizon may occur partly or wholely above or be coextensive with the fragipan. An argillic horizon is “an illuvial horizon in which layer-lattice silicate clays have accumulated by illuviation to a significant extent” (Soil Survey Staff, 1967). Cambic horizons may show several kinds of alteration, but domination by accumulation of mineral substances is excluded. The marks of cambic horizons in soils with fragipans include the mottling indicative of translocation of iron and associated with periodic saturation, higher chroma and redder hue associated with coatings of iron oxides released by mineral weathering, and sufficient development of soil structure largely to obliterate the original organization (stratification, for example) of the parent material (Soil Survey Staff, 1967). Cambic horizons may contain illuvial clay but do not meet the requirements for an argillic horizon. Spodic horizons contain appreciable amounts of precipitated amorphous materials “composed of organic matter and aluminum, with or without iron . . .” (Soil Survey Staff, 1967). Figure 1 depicts a soil with a spodic horizon above the fragipan. The suffix x is currently used to designate the fragipan (Soil Survey Staff, 1962); an example would be Bx. The suffix rn has been used earlier (Soil Survey Staff, 1951) but is now restricted to strongly cemented or indurated horizons having a consistence that is not appreciably affected by moistening. The proposals by the International Society of Soil Science ( 1 968) on horizon nomenclature use x and rn in these senses. Many soils with fragipans have two pairs of A and B horizons. Each pair of A and B horizons is referred to as a sequum. Soils with two pairs are referred to as bisequal. The soil shown in Fig. 1 is bisequal. The lower pair of horizons is designated with a prime accent. An example would be B’x. The fragipan, if present, commonly occurs in the lower sequum if the soil is bisequal. Daniels el al. (1968) have suggested the notation Be for a B horizon


24 1

containing discontinuous eluvial parts. The Be notation may find use as a designation for lower eluvial horizons that cannot be designated A'2 or B & A.

B. HORIZONSEQUENCES The uppermost panel of Fig. 2 illustrates the most general statements that may be made about the position of the fragipan. All fragipans occur beneath an eluvial horizon unless the soil has been eroded severely. In Wetter

Eluvial

A'2 Lxl

ond /or I

Ilc

-

IId

A2

IA2

B

IB

Bx

and/or 1

cx

1

Btx

I

FIG.2. Horizon sequences of soils with fragipans in the eastern United States. Elements within brackets indicate characteristics that are not essential to the definition of the sequence. The fragipan horizons are indicated by the suffix x: the suffix t denotes silicate clay sufficient for recognition of an argillic horizon.

some soils, more often the wetter ones, the fragipan occurs immediately beneath the eluvial horizon. In many soils, however, there is an intervening B horizon, which may be a spodic horizon, a cambic horizon, or an argillic horizon. Some soils with fragipans are bisequal. The fragipan may occur in the lower eluvial horizon and not in the underlying B horizon, in the lower B horizon but not in the second eluvial horizon, or in both. These variations form the basis for the four classes of horizon sequences shown in the set of panels designated Ia through Id; these are illustrated by soil descriptions in the Appendix. These four classes are

242

R.

B. GROSSMAN A N D F. J. CARLISLE

subdivided to give the eight classes, IIa through IIh. Accumulation of clay and occurrence of the fragipan in the lower eluvial horizon are the criteria for these subdivisions. Further subdivisions could be made. Panel IIf, for example, could be subdivided into soils with argillic, spodic, or cambic horizons above the fragipan. Brackets are placed around the horizon designations or elements of the horizon designations that are not necessary to the definition. In Fig. 2 the suffix t denotes an accumulation of silicate clay sufficient for recognition of an argillic horizon. This is a more limiting use of the t suffix than is specified by the Soil Survey Staff ( 1 962). The horizon symbols in Fig. 2 are only for convenience in discussing this subject; it is not suggested that any of these symbols should replace horizon designations in current use. Soils with horizon sequences Ha, Ilc, and IIe have fragipans in horizons where the parent material has been weakly altered. Many of the soils with fragipans in the northeastern United States developed in Wisconsin glacial till belong in one of these three classes. In contrast are fragipans in soils that have horizon sequences IIb, IId, IIf, IIg, and IIh. In these soils, alteration of the parent material of the fragipan by eluviation, illuviation, or both, has been strong. Hence, influence of the parent material in determining the properties of the fragipan is less important. Fragipans in soils south of the Wisconsin glacial advance tend to exhibit stronger alteration of the parent material.

C. FRAGIPAN EXPRESSION Fragipan expression involves thickness and depth to its upper boundary. These are considered in subsequent paragraphs of this section. It also involves properties, such as consistence, structure, and pore arrangement, that determine mechanical impedance to root penetration and rate of movement of low-tension water. Objective measurement and integration of these properties has not been achieved. Within a local association of soils, however, evaluation of relative degree of fragipan expression has validity. Maximum degree of expression of the fragipan is commonly reported for the soils of intermediate wetness (Neeley, 1965; A. E. Thomas, 1967; Redmond and Engberg, 1967; Grossman et al., 1959a; Nettleton et al., 1968a). This intermediate wetness is described approximately by the terms somewhat poorly and poorly drained (Soil Survey Staff, 195 1) and by the Aquic subgroups of the soil classification system (Soil Survey Staff, 1967). The relationship is widespread enough to suggest implication in the genesis of the fragipan.


243

Excluding severely eroded soils, the depth to the fragipan ranges from about 10 to 150 cm., with the upper boundary for most between 25 and 100 cm. In landscapes that have been affected by severe erosion, depth to the fragipan often is closely related to past land use and to the slope. The area of occurrence of Grenada soils in Grenada County, Mississippi (A. E. Thomas, 1967) is an example (see Appendix for description). Concepts have centered on the less eroded soils. Consequently, in many areas fragipans are shallower than much of the literature would suggest. On a given landscape the depth is frequently relatable to wetness of the soil. Two of many examples have been selected. Neeley ( 1965), studying fragipans in soils of New York developed in medium-textured glacial till, reported a range of 75 to 30 cm. to the top of the fragipan from the drier to the wetter soils. The soils range from sequence IIe through IIc to I Ia of Fig. 2. A. E. Thomas ( 1 967), working with soils developed in loess in Mississippi, reports a range from 50 to 10 cm. Sequences involved are IIf and IIb. In some associations, the fragipan is not shallower in the wetter soils. Soils of the Lebanon series and related soils in the Missouri Ozarks (Krusekopf, 1942; Scrivner, 1960) have a fragipan immediately beneath and perhaps partly within the loess that covers a buried soil (sequence 1Ig or JIh). Fragipans in this soil association are no shallower in the somewhat wetter soils that occur in slight depressions than in the associated better drained soils. Thickness ranges from about I5 to 125 cm. In some local associations in the northeastern states, thickness remains fairly constant over a considerable range in wetness (Carlisle, 1954; Neeley, 1965). These fragipans are weakly altered and bear a strong imprint of the compact glacial till parent material. The drier soils have sequences IIc and IIe of Fig. 2; the wetter, sequence IIa. Constant thickness over a wide range in wetness is not restricted to fragipans with strong parent material influence. Bailey (1964) reports rather constant thickness for the fragipans in strongly illuvial horizons (sequence Ild, Fig. 2 ) of some soils developed from limestone residuum and mixed sedimentary rocks in Kentucky. In certain loess-derived soils of the middle and lower Mississippi Valley, the fragipans become thicker in the wetter soils of the association (A. E. Thomas, 1967; Bailey, 1964). The fragipans in these soils are strongly affected by illuviation, eluviation, or both. The drier of these soils have sequences IId, IIf, and IIg; the wetter soils, sequence IIb. In other associations of loess-derived soils in the same general area, the fragipan thins with increasing wetness (Grossman el al., 1959a). Nettleton et al. ( 1968a), in their study of fragipans in certain soils of North Carolina de-

244

R. B. GROSSMAN A N D F. J . CARLISLE

veloped from Coastal Plain sediments (sequence IIh, Fig. 2), also report thinning of the fragipan with increasing wetness. 111. Occurrence of Fragipan Soils

The occurrence of soils with fragipans may be considered at several scales. Looking at the country as a whole, fragipans have been recognized in some soils in all states east of the Mississippi River and in adjacent Minnesota, Missouri, Arkansas, Oklahoma, Louisiana, and eastern Texas. Soils with fragipans are the principal soils in some parts of that area, and they are of minor extent in other parts. They have been reported in the western states (Cline, 1952; Whittig et al., 1957) but seem to be of very minor extent and importance there. Figure 3 shows areas in the United States where one or more of the principal kinds of soils have

FIG. 3. Areas in the United States where one or more of the principal kinds of soils have fragipans. The most extensive soils in the areas delineated are: A, Altisols: I , Inceptisols; S , Spodosols: U and U 2 , Ultisols (generalized from Soil Conservation Service, 1969, except U 2 delineations, which are from Bartelli, 1968). Dashed line shows the approximate eastern boundary of the prairie (generalized from Shantz and Zon, 1914).


245

fragipans. They are extensive and may be among the principal soils in areas where the dominant soils are Spodosols, Inceptisols, Alfisols, or Ultisols in the warm humid and the cold humid, central and eastern parts of the country. Fragipans apparently do not occur in soils of the humid prairies, the Great Plains, or the semiarid and arid areas of the west. The following relationships are evident in the broad-scale occurrence of fragipan soils: ( 1 ) Fragipans are restricted to areas where the excess of precipitation over evapotranspiration is sufficient at some time of the year for movement of water down through the soil. (2) They occur in both warm and cold climates. (3) Fragipans seemingly are absent in soils of the extensive natural grasslands of the humid prairies and the Great Plains. (4) Fragipans occur in Spodosols, lnceptisols, Alfisols, and Ultisols. Spodosols with fragipans are so common as to suggest a genetic connection. Other relationships may be evident at the larger scale of soil association maps for counties where fragipans are important. The occurrence of fragipan soils in relation to composition of soil parent materials apparently is complex when viewed generally. But local relationships may be evident, especially if the soils are not old. The surveys of Tompkins County, New York (Neeley, 1965) and of Franklin County, New York (Carlisle, 1958) illustrate the influence of lime content of parent material on occurrence of soils with fragipans. In these areas soils with well expressed fragipans are prevalent in areas of low-lime glacial till and absent in areas of high-lime till. The Franklin County, New York, survey illustrates the influence of texture of soil materials on the occurrence of soils with fragipans. I n that area fragipans are present in Spodosols developed in relatively well-graded glacial till and are absent in Spodosols developed in poorly graded glaciofluvial materials of roughly comparable mineralogical composition. In Tate County, Mississippi (J. S. Huddleston, 1967) the distribution of soils with fragipans on the countywide scale is related in part to the loess distribution pattern and to the pattern of geological erosion. Other illustrations of these kinds of local relationships may be found in published soil surveys of areas shown in Fig. 3. For a multicounty area in southern New York and northern Pennsylvania, Denny and Lyford ( 1963) showed the distribution of soil associations in which soils with fragipans are extensive. They concluded that within the region studied many of the soil differences are primarily related to kind of parent material and its hydrologic characteristics. A third and still larger scale is the detailed soil map showing delineations dominated by a particular kind of soil. This is the scale commonly used in planning farming practices for individual fields and farms. To

246


obtain a sense of the pattern of occurrence of fragipan soils on this scale, one should mentally walk across the detailed map observing the pattern of mapping unit delineations in relation to slope, topography, and other landscape features. Practical considerations do not permit delineation on the usual soil map of the pattern of soil occurrence within distances measured in tens of feet. This scale of observation, however, may be very fruitful in understanding the genesis of the soils. Daniels and associates, for example, have worked at this scale to study certain soils with fragipans of North Carolina that occur on the less dissected part of a pre-Wisconsin geomorphic surface formed in Coastal Plain sediments (among other papers, Daniels et af., 1966; Daniels and Gamble, 1967; Nettleton et af., 1968a,b). Lyford et al. (1963) have reported a study at a similar scale on soils with fragipans in central Massachusetts that are developed in Wisconsinage glacial till. Less extensive studies on such a very local scale have been reported by Bornstein et al. ( I 9 6 3 , Carlisle (1954), Lyford and MacLean ( 1 966), Olson ( 19621, and Denny and Lyford ( I 963). IV. Properties of Fragipans

A. COMPOSITION 1 . Texture Most fragipans are loamy as the term is defined by the Soil Survey Staff (1 967). They may be skeletal, but not fragmental. Few strong fragipans are sandy. Although some fragipans exceed 35 percent clay, no fragipans have been reported with over 60 percent clay. Fragipans in weakly illuvial horizons (IIa, IIc, IIe, Fig. 21, such as are common in the northeastern United States, contain less than 35 percent clay, and most contain less than 25 percent clay (Jha and Cline, 1963). Maximum clay percentage does not appear to be strongly related to the kind and proportions of clay minerals. Fragipans with montmorillonite predominating may exceed 35 percent clay (for example, Hutcheson et al., 1959). The fine earth of fragipans, after allowance for the clay percentage, usually is high in material from 0.2 10 0.02 mm. and low in the range above 0.25 mm. A high content of silt and very fine sand may promote development and expression of fragipans (Jha and Cline, 1963; Carlisle, 1958; Grossman and Cline, 1957). 2. Chemical Chemical properties of fragipans do not seem unique. Fragipans are low in organic matter, have low or moderate levels of extractable iron

FRAGIPAN SOILS O F THE EASTERN UNITED STATES

247

relative to the clay percentage, are rarely if ever calcareous, lack appreciable soluble salts, and have at most moderate exchangeable-sodium levels. Exchange capacity ranges widely, depending on the amount and kind of clays. Base saturation ranges from low to high. Fragipans in strongly illuvial horizons of soils weakly influenced by the Wisconsin glacial advances may have very low base saturation. Base saturation usually rises with increasing wetness. Fragipans in soils with low base saturation in their lower parts may have more extractable magnesium than calcium, and some extractable sodium may be present. Hutcheson et a f . (1 959) and Pettiet ( 1 964) studied fragipans in such soils. They suggest that high extractable magnesium may make clay more susceptible to movement and rearrangement, thereby fostering fragipan formation. Presence of magnesium may be a factor in some soils with fragipans but many fragipans do not have particularly high extractable magnesium. Soil pH values range from 4 to 7. Aluminum extractable with a neutral salt follows the common pattern. Values commonly increase rapidly as the soil pH drops below 5. Soluble silica and aluminum determinations have been reported by a number of workers, either in connection with studies on bonding or incidental to characterization of. the clay mineralogy (Alexander, 1955; Comerma, 1964; Jha and Cline, 1963; Knox, 1957; Olson and Hole, 1967-1 968; Pettiet, 1964; Yassoglou and Whiteside, 1960). No consistent pattern has been observed.

3. Mineralogy Fragipan expression seems largely unrelated to the mineralogy of the sand and silt other than for the control that presence of carbonate may impose. Daniels er af. (1966) found no relationship between fragipan expression and the percentage of feldspar in the very fine sand fraction for certain soils of North Carolina having strongly eluvial fragipans. In some landscapes fragipans are restricted to the older geomorphic surfaces. These soils may have lower weatherable minerals than the associated soils on the younger geomorphic surfaces. T h e difference in mineralogy, however, would not appear responsible for the pattern of fragipan occurrence. Fragipans are common on the Ozark Plateau and on the Coastal Plain. Parent material sources high in quartz and low in weatherable minerals are common to these areas. The high proportion of quartz does not appear to have directly contributed to the prevalence of fragipans in these areas (see Section VI, A). The clay mineralogy of fragipans is similar to the clay mineralogy of horizons in comparable positions in associated soils without fragipans. In the northeastern United States, related to the prevalence of paleozoic

248

R. B. GROSSMAN AND F. J. CARLiSLE

sedimentary rocks as sources, 2:1 lattice clays other than montmorillonite predominate, with mica (illite) a quantitatively important component (among others, Knox, 1957; Jha and Cline, 1963; F. P. Miller, 1965). Moving westward into the northern Middle West, Yassoglou and Whiteside ( 1960) have reported illite, chlorite, and interstratified minerals with some interlayered montmorillonite in the clay fraction of several soils with fragipans in Michigan. Olson ( 1 962) indicated a somewhat similar suite of clay minerals in some soils with fragipans of northeastern Wisconsin. Moving southward, the fragipan in Lebanon-like soils of Missouri studied by Scrivner (1 960) contains randomly stratified montmorillonite, illite, and Vermiculite with some kaolinite; the mineralogy is allied with the surficial loess rather than the limestone residuum beneath the loess. Soils with fragipans of the middle and lower Mississippi Valley developed in loess contain appreciable montmorillonite (Anderson and White, 1958; Glenn, 1960; Grossman et af., 1959b; Hutcheson et af., 1959; Pettiet, 1964). In these soils the fragipans occur in horizons that have undergone moderate or strong illuviation (sequences IIb, IId, IIf, IIg of Fig. 2). The montmorillonite shows evidence of interlayers that do not extend fully on treatment with glycols. Hutcheson et al. (1959) suggested that in the fragipan vermiculite forms at the expense of montmorillonite. Glenn ( 1 960) and Pettiet ( 1964) reported appreciable amorphous material. In the middle Atlantic states, Nikiforoff et al. ( 1948) found significant amounts of kaolinite in the fragipan of a soil that occurs in Maryland. Nettleton et af. (1968b) studied the clay mineralogy of soils in North Carolina with fragipans in highly eluvial horizons (sequence IIg of Fig. 2). Kaolinite or a vermiculite-chlorite intergrade were the most abundant components, with lesser amounts of montmorillonite. Evidence was presented for considerable amorphous material, and the kaolinite had imperfect crystallinity. 4 . Bulk Density

Bulk densities of the moist fine-earth fabric usually exceed 1.4 g./cc. and mostly exceed 1.6 g.lcc. Differences between the moist and dry natural-clod bulk densities are not large. The coefficient of linear extensibility (Grossman er al., 1968) usually does not exceed 0.04. The fragipan rarely has a lower bulk density than overlying horizons. Differences from horizons below are variable. In soils with a strong influence of parent material on the fragipan (horizon sequences Ila, IIc, IIe of Fig. 2 ) and a


249

parent material with a high bulk density, the fragipan may differ little in bulk density from the parent material (examples in Soil Survey Staff, 1968b). If the bulk density of the parent material is moderate, the fragipan horizon may have the highest bulk density in the soil (for example, Jha and Cline, 1963). Fragipans in eluvial horizons that immediately overlie strongly expressed illuvial horizons (horizon sequence IIh) often have the maximum bulk density of the soil (examples in Nettleton et al., 1968a; Scrivner, 1960; Yassoglou and Whitesite, 1960). Rutledge and Horn (1965), Yassoglou and Whiteside (1960), and Pettiet ( 1964) have discussed estimates of the pore-size distribution of fragipans based on water retention determinations. A large body of information is available for the computation of pore-size distribution in the Soil Survey Investigations Reports series published by the U.S. Department of Agriculture. Medium- and fine-textured fragipans can be shown to have a high proportion of the total porosity filled with water held at energies above 15 bar; this porosity must consist of very small pores. The air-filled porosity at 1/3 bar can be shown to be small, but this is also common to many medium-textured soil materials.

B. MORPHOLOGY I . Macroscopic

Many fragipans have roughly vertical planes that delineate large prisms or blocks. Nikiforoff (1 955) vividly described the feature in the Beltsville soil of Maryland: “Throughout its thickness the pan is split into large irregular blocks ranging from about 1 f.i to 2 feet in horizontal diameter. Planes of cleavage are marked by strong bleaching of the walls of fissures which produces conspicuous streaks on the exposures of the hardpan. In vertical planes, these irregular light colored streaks are roughly parallel, whereas in horizontal planes, they form a striking polygonal network . . . . Walls of the cracks are bleached laterally for . . . a few millimeters to more than an inch . . . . Beyond these bleached zones there are yellow to orange oxidized zones so that on cuts the bleached streaks are enclosed between rust-colored bands.” The pattern may be much less distinct in other soils, particularly if the fragipan is coarse textured. The fragipans studied by Daniels et at. ( I 966) and Nettleton et al. (1 968a) lack the polygonal pattern. Information on the composition and organization of the periphery of the large prisms and blocks relative to the interiors was reported by Carlisle (1954), Gile (1958), Jha and Cline (1963), F. P. Miller (1965), Pettiet (1964), and Vanderford and Shaffer (1966). The study by F. P. Miller (1965) is particularly detailed.

250


Structure of the large prisms or blocks is variable. Trends follow those for horizons other than fragipans: eluvial horizons usually are massive or platy; illuvial horizons usually are blocky. For example, the fragipans described by Yassoglou and Whiteside (1960) and by Nettleton et al. (1968a) occur in horizons that overall are eluvial; much of the fabric has massive or platy structure. In contrast stand the more clayey, strongly illuvial fragipans (Bailey, 1964; Rutledge and Horn, 1965). They have subangular or angular blocky structure of moderate expression and medium size. Fragipans that have undergone weak alteration may be strongly influenced by the structure of the parent material. If the parent material is platy (for example, Lyford et al., 1963), the fragipan may be platy. If it is crudely blocky (for example, Neeley, 1965; Carlisle, 1954), the fragipan usually has a crude blocky structure. If massive (for example, Jha and Cline, 1963), the fragipan may tend to massive structure. Large pores that are continuous vertically over the thickness of the fragipan usually are widely spaced. Vertical planar surfaces within the large structural units extend over distances that are a small fraction of the thickness of the fragipan, usually having dimensions of a few centimeters. Some fragipans have platy structure. The plates usually overlap, and consequently vertical pores between peds are tortuous. Fragipans are often described as vesicular, implying that the pores within peds are not interconnected. Lyford et ad. (1 963) have provided an apt description: “The pores in the peds are seldom continuous. They branch and rebranch but tend to end up inside the ped rather than continue from one side to the other. Many of the pores observed in broken peds are short and bulbous; they do not empty at the surface.” Evidence for translocation of silt or clay is common in fragipans. Field descriptions of fragipans generally refer to clay films on ped surfaces and within pores. The clay films may be thin to thick. Continuous clay films on ped surfaces are not reported in fragipans. Clay films, however, may be strongly expressed in pores. Rearrangement of silty material has been observed in many fragipans. Carlisle ( 1958) described the feature as follows: “Gray silty material, which does not occur in the layers above the fragipan, coats the upper surfaces of rock fragments and impregnates the uppermost part of the peds within the fragipan.” Descriptions of fragipans from outside New York and New England, including fragipans developed in Wisconsin glacial till, do not place as much emphasis on rearrangement of silty material.

2 . Thin-Section Observations A partial list of observations include Jha and Cline (1963), Carlisle (1954), Calhoun (19681, F. P. Miller (196.9, Nettleton et at. (1968b),


25 1

Olson ( I 9621, Pettiet ( I964), Yassoglou and Whiteside (1 960), Grossman et al. (1 959c), Horn and Rutledge ( 1965), Nikiforoff et al. (1 948), and McCracken and Weed ( 1 963). The descriptions have one unifying feature. Some of the clay connects between sand and silt grains, and some of it shows optical evidence of preferred orientation; the fabrics would be plasmic and sepic (Brewer, 1964). The clay may be scarce and strongly concentrated at contact points of sand and silt grains with interstices only partially filled; or the clay may be abundant and form a continuous medium within which the sand and silt grains are set. A significant portion of the clay apparently has moved over distances measured at least in millimeters; some may have come from horizons above. Bodies of moved clay are present in strongly eluvial fragipans as well as in fragipans that are illuvial B horizons. Amounts of moved clay are reported by Calhoun (1968), Carlisle (1 954), Horn and Rutledge (1 9 6 3 , F. P. Miller (1965), Nettleton et ai. ( 1968b), and Soil Survey Staff (1968a). Horn and Rutledge ( 1965) attach particular importance to the sepic micromorphology of fragipans in determining brittleness. The fragipan is viewed as “cellular.” Plasma separations that impart rigidity delimit small volumes with lower strength. Several workers have emphasized the importance of closeness of packing of the sand and silt grains. Some fragipans, however, do not show particularly close packing (Horn and Rutledge, 1965). Packing is generally closer in coarser-textured fragipans and may be a more important factor in determining the rigidity in the coarser textures. Horizons above the fragipan may show closer packing of the sand and silt grains and yet have markedly lesser rigidity than the fragipan (Grossman et al., 1959~). C. CONSISTENCE Consistence is the principal defining property of fragipans. But the classes of soil consistence are defined qualitatively and evaluation of consistence is subjective. This results in vagueness in the definition of fragipan and leads to difficulty in achieving uniform application of the fragipan concept. The soil consistence test (Soil Survey Staff, 195 1) involves compressing a piece of soil about 2 to 4 cm. across between thumb and forefinger until failure occurs. It is a kind of unconfined compression test, a subject reviewed recently by Gill and Vanden Berg (1967). If the piece of soil is uniform with no predetermined planes of weakness and the cracks that form at failure are near the center of the bearing surface and parallel the direction the force is applied, then cohesional forces largely determine the resistance to rupture. Tensile strength is therefore measured. Usually, however, the cracks form at an angle to the axis of corn-

252


pression; hence some frictional force contributes to the resistance to rupture. Moreover, the area of contact between finger and soil is appreciable relative to the dimensions of the piece of soil. This tends to cause shearing stresses. Rearrangement may occur before rupture. The rearrangements commonly are such as to blunt the point of the crack, leading to an increase in strength of the soil material. Often the piece of soil has predetermined planes of weakness. Those cracks which most nearly parallel the axis of compression probably determine the strength. 1 . Induration

Fragipans appear cemented when dry, but soften when wetted. Cementation implies little reduction in hardness on moistening (Soil Survey Staff, 195 1). Fragipans, therefore, are not cemented. The expression “reversibly indurated” has been used, but it is not very satisfactory as it seems to be a contradiction in terms. Most fragipans slake when dry pieces are placed in water (Anderson and White, 1958; Knox, 1957; Comerma, 1964; Jha, 1961; Nikiforoff et al., 1948; Olson, 1962; R. M. Smith and Browning, 1946). One of the fragipans studied by Knox ( 1957) did not slake; silica was implicated in the bonding. Field observations of slaking by fragipans are of common occurrence, but few are published. Nettleton et al. (1968b) described the slaking of fragipans in ditch banks. Olson (1962) studied the breakdown of fragipan material on exposure.

2. Resistance to UnconJined Compression When dry, most fragipans are at least hard-“moderately resistant to pressure; can be broken in the hands without difficulty but is barely breakable between thumb and forefinger” (Soil Survey Staff, 195 1). When moist, fragipans are usually at least firm-“crushes under moderate pressure between thumb and forefinger, but resistance is distinctly noticeable” (1951). Grossman and Bartelli (1957) used a hand dynamometer to determine the force applied at the point of discrimination between several consistence classes by a group of soil scientists. The lower limit was 5 kg. for hard consistence and 3 kg. for firm consistence. Grossman and Cline ( 1 957) report a median value of 17 kg./cm.2 and a range of 4 to 25 kg./cm.2 for the resistance to rupture when dry of a number of fragipan horizons from soils of New York. Grossman (1954) measured a resistance to rupture of 4 kg./cm.2 at a water content near 1/3 bar retention for a fragipan having a resistance of 20 kg./cm.2 when dry. Knox ( I 954) assembled information on the resistance to rupture of


253

various particulate materials. Molding sands with 5 to 10 percent bentonite have crushing strength similar to the median value for the fragipans from New York studied by Grossman and Cline ( 1 957). 3 . Miscellaneous Tests

Yassoglou and Whiteside ( 1 960) used a cone penetration test to characterize the hardness of moist fragipan material from certain soils of Michigan. The fragipan had higher resistance than other horizons, and the fragipan subhorizon judged hardest by field examination offered the highest resistance. Rutledge and Horn ( 1 965) employed a needle penetrometer to study fragipans in soils of Arkansas. Penetration at several moisture tensions was somewhat lower for the fragipan than for the horizon immediately above. Grossman et al. (1959a) used a “dropshatter” test to characterize a fragipan soil from Illinois developed in loess. The fragipan resisted shatter more than horizons above, but less than the C horizon beneath. 4 . Brittleness

There are very few laboratory measurements of brittleness. Grossman ( 1954) compared the abruptness of rupture under unconfined compression of two fragipan materials. At a moisture content near that retained against 5 bars, the fragipan with less than 5 percent clay failed abruptly; the one with 15 percent clay showed some plastic deformation. The concept of brittleness as applied to fragipan soil material has not been clearly formulated. The reference state with respect to moisture tension for medium textures probably should be about 1/3 bar. Many moist, coarse-textured soil materials exhibit brittleness. Even some moist clayey materials exhibit brittleness if there are numerous well expressed planes of weakness in the test specimen; an example would be the B horizon of some Oxisols. Most fragipans, however, have incomplete, weak, and commonly widely spaced structural planes. Their brittleness is usually not the result of rapid failure along structural planes. Rather, it would appear to be related to the bonding by clay-size material of the sand and silt grains (the s-matrix as defined by Brewer, 1964). For coarse-textured fragipans, much of the clay occurs as braces between sand and silt grains. Amounts of clay are insufficient for local plastic movement. The leading point of a crack is not subject to blunting. Rather, the stress remains concentrated, leading to rapid propagation of the crack. Such an explanation is not applicable to medium and fine textures. In some of these instances, the highly sepic micromorphology may contribute to brittleness (Rutledge and Horn, 1965).

254

R.

B. GROSSMAN A N D F. J. CARLISLE

V. Fragipans and the Soil Water Regime

A. WATERTABLES Several investigators have reported on the water tables of soils with fragipans (Gile, 1958; Nettleton, 1965; Spaeth and Diebold, 1938; Lyford, 1964; J. H. Huddleston and Olson, 1967). Many data exist in mimeographed reports; these often are available from state soil survey organizations. Some of the studies employ lined wells placed at depths above, within, and below the fragipans. In principle, this arrangement permits detection of a perched water table above the fragipan. Other studies employ unlined wells which do not permit detection of a perched water table. Interiors of the large structural units may be considerably below saturation while free water is present between them. Lined wells that terminate within these structural units may not indicate the presence of this free water. Much of the area of fragipan soils now supports trees. Removal of water by transpiration by trees usually is greater than when the land is cleared. The water table regime under trees may underestimate the wetness of the soil when it is cultivated or used as sites for construction. On the other hand, measurements in cleared areas may indicate shallower water tables than are actually present over most of the area. Soils with fragipans are subject to the rather well defined seasonal pattern in water table depth found over much of the eastern United States. Water tables rise in late fall and remain high until transpiration by plants becomes appreciable. Often the drop in water table occurs about when the trees leaf. Water-holding capacity of the fragipan is usually low (Section V, B), and small additions of water may raise the water table markedly. Soils are placed in wetness classes mainly on inference based on the depth to and expression of mottles or low chroma parts. A significant practical question is the extent to which the water table regime of soils with fragipans differs from associated soils in the same wetness class.

B. WATER-HOLDING CAPACITY Laboratory estimates of field capacity or maximum water retention must be viewed cautiously. The determinations may be on fragmented samples resulting in a significant overestimate of water retained against low tension. As previously indicated, under field conditions the interior of the large structural units of the fragipan may not contain free water, even though free water is present in the cracks between the structural units (Nikiforoff et al., 1948; Carlisle, 1954; Alexander, 1955). Conse-


255

quently, maximum water retention in place would be below that calculated from the total porosity of the moist fabric. R. M. Smith and Browning ( 1946) commented on the low water content of fragipan material after wetting in the laboratory, even under vacuum. They suggested that the fragipan material may have appreciable porosity that does not fill readily with water. Comer and Zimmerman ( 1 969) reported that for a 3-year period the water content in the fragipan of a wet soil in Vermont ranged only from 19 to 23 percent by volume. The soil is not subject to recharge by upward water movement from a regional water table; recharge is by downward moving water with perhaps a lateral component of movement. The constancy of the water content over this period suggests a high degree of isolation from both withdrawal of water by plants and additions from precipitation. The effective contribution of the fragipan to the water holding capacity of the soil at any point in time over this period would appear to have been small. C. MOVEMENT OF LOW-TENSION WATER Laboratory measurements of the saturated hydraulic conductivity have been reported by R. M. Smith and Browning ( 1 946), Grossman et al. ( 1 959a), Yassoglou and Whiteside ( 1 960), and Pettiet (1 964). Values range from 0.01 to 1 inch per hour. Large numbers of field percolation determinations have been made. In some areas, these are required by law in planning small-scale sewage disposal systems. Hill ( 1966), Alexander (1955), and J . H. Huddleston and Olson (1967) have discussed procedures and present values for soils with fragipans. Horizons above the fragipan are usually quite pervious unless altered by man’s activities. Infiltration usually is not limited by a horizon above the fragipan until near-saturated conditions prevail. Fragipans have a lower saturated hydraulic conductivity than do horizons above. Consequently, low-tension water accumulates at the top of the fragipan and moves laterally. Fragipans do not necessarily have lower saturated hydraulic conductivity than the horizons beneath (for example, Yassoglou and Whiteside, 1960; Alexander, 1955). If the underlying soil materials are pervious, then the saturated hydraulic conductivity of the fragipan may be a minimum for the profile. There may be several reasons for the low saturated hydraulic conductivity of fragipans. Lack of vertical continuity of interped pores and isolation of pores within peds may have importance. O’Neal (1952) discussed the relationship between perviousness and vertical continuity

256


of pores. Some fragipans have low total porosity. Those with moderate porosity tend to have appreciable clay (for example, Rutledge and Horn, 1965). The area that actually conducts low-tension water in the field may be quite small, limited to the periphery of the large structural units. Hydrology studies of watersheds reflect the complexity of the total natural condition. Minimum discharge rates are indicative of the integrated amounts of water moving through the soils of the area. Comer and Zimmerman ( 1 969) reported that the minimum discharge rate for a watershed in Vermont where wet soils with fragipans occupy 44 percent of the area is a magnitude lower than for a contiguous watershed where such soils occupy 22 percent of the area. They suggested that the combination of low permeability of the fragipan and the high water-holding capacity of the horizons above the fragipan is largely responsible for the lower minimum discharge rate of the watershed with the greater proportion of wet fragipan soils. VI. Genesis of Fragipans

This section has been written on the underlying assumption that clay is the bonding agent; moreover, it is assumed that this clay ranges widely in mineralogy and surface chemical properties. A. BONDINGO F THE FRAGIPAN 1 . Silicate Clay as the Bonding Agent Several workers have proposed that silicate clay is the principal bonding agent (R. M. Smith and Browning, 1946; Nikiforoff et al., 1948; Carlisle, 1954; Knox, 1957; Jha and Cline, 1963; Yassoglou and Whiteside, 1960; Comerma, 1964; Hutcheson and Bailey, 1964). Dispersing agents for clay have been shown by Comerma (1964), Knox (1957), and Jha and Cline ( 1 963) to disaggregate the fragipans of certain soils more completely than treatments designed to remove silica, hydrous iron oxides, or hydroxy aluminum compounds. There is no evidence from the large number of particle-size analyses using standard dispersing treatments that the clay in fragipans resists disaggregation. Close packing of the sand and silt is thought to contribute to the effectiveness of the clay as a bonding agent. Nikiforoff et al. ( 1 948) placed principal emphasis on the close packing and related interlocking of the sand and silt. Hutcheson and Bailey ( 1 964) also emphasized closeness of packing. They write, “. . . [we] visualize pan horizons as brick and mortar structure, i.e., silt particles acting as bricks held in a dense mass by clay mortar.” The thin-section observations by Jha and Cline ( 1 963) fit


257

the “brick and mortar” model. Other workers have placed greater emphasis on the clay bridges between sand and silt grains (Knox, 1957; Yassoglou and Whiteside, 1960; Grossman and Cline, 1957) with proportionately less on a dense, continuous filling of clay in the interstices. More would seem to be involved than disposition of the clay. In some horizons of clay accumulation, bridges of clay between sand grains are prominent (Soil Survey Staff, 1960, p. 41). Yet, when moist, these horizons do not necessarily exhibit the brittleness and rigidity of fragipans. Several writers (Carlisle et a f . , 1957; Yassoglou and Whiteside, 1960; Grossman and Cline, 1957) have commented on the dual role of clay. At low clay contents the bridging of clay between sand and silt grains lends rigidity. At higher clay contents, volume changes with moisture promote formation of cracks that reduce the rigidity when moist. 2. Other Bonding Agents a. Silica. The earlier literature contains suggestions that silica is the bonding agent (Marbut, 1935; Krusekopf, 1942; Winters, 1942). At the time, total analyses of soils received more emphasis than they do now. Total analyses of fragipans low in clay, with the sand and silt dominated by quartz, indicate high proportions of silica, which would be consistent with the idea of a siliceous bonding agent. Studies were then current on silica cementation of indurated horizons in certain soils of western United States (for example, Nikiforoff and Alexander, 1942). These studies were employed to support bonding by silica in fragipans. There is similar informal conjecture today. The argument runs that as silica does cement some soils, perhaps small amounts -less than that detectable by methods employed to date-may play a similar role in fragipans. Knox ( 1 957) presented the only experimental data for the implication of silica. Extraction of silica with various reagents (Section IV, A, 2) has not shown a consistent maximum in the fragipan. Baker (1967) determined the mineral stability by solubility investigations for the strongly eluvial fragipan horizon of certain soils developed in loess over cherty limestone residuum in Missouri. Kaolinite and quartz were stable. The concentration of silica was well below that supportable by opal. McKeague and Cline ( 1 963) suggested that concentration of silica by evaporation may be important in the surface adsorption of silica from solution on soil particles. Fragipans as a rule, however, are not subject to frequent and pronounced desiccation. R. W. Miller ( 1967) stressed the importance of small differences among horizons in controlling the translocation and deposition of silica. H e studied soils from Utah that are not subject to high

258


precipitation, and one of the soils was formed in materials containing volcanic glass. Such differences from soils with fragipans of the eastern United States must be considered in the application of research on silica solubility in soils of the western United States. Calcite apparently adsorbs little or no silica (McKeague and Cline, 1963), and presence of carbonate appears to increase the adsorption of silica (R. W. Miller, 1967; McKeague and Cline, 1963). The scarcity of fragipans in calcareous soil materials and the influence of carbonates on silica solubility relationships might be related. b. Aluminum and Iron. Hydrated oxides of iron or aluminum have been suggested as bonding agents. Knox ( 1 957) and Comerma ( 1 964) found no evidence for their implication. Alexander (1955) failed to find a maximum in extractable aluminum in the fragipan. Anderson and White (1958) presented evidence for iron oxides contributing to the rigidity in a fragipan. Horn and Rutledge (1965) suggested that segregations of iron oxides in association with silicate clay are important in determining rigidity. Nettleton et al. ( I 968b) suggested that amorphous aluminum compounds may play a role in the hydrogen bonding of fragipans. c. Water Films. Surface tension effects associated with water films lend rigidity to moist soil material (Knox, 1954; Fountaine, 1954). For spherical particles, cohesion increases proportional to the reciprocal of the radius. As a point of reference, the cohesion for spheres with a radius of 10 p ranges roughly from 0.1 to 1 kg./cm.2, depending on the assumptions about packing and whether the pores are partially or entirely filled (Knox, 1954). This range compares with a crushing strength of 4 kg./cm.2 for a moist fragipan soil material of silt loam texture (Grossman, 1954). Water films may contribute appreciably to the rigidity of some moist fragipan material. It is doubtful, however, that the greater rigidity when moist of fragipans than of otherwise comparable soil materials is due to water films.

3 . Mechanisms of Bonding Knox ( 1 954) concluded from a review of the literature that chemical forces rather than mechanical interlocking are probably responsible for the strength of fragipan material. Nettleton et al. (196%) propose that amorphous clays adhere strongly to the disordered surface of quartz grains by hydrogen bonding. Acidity of the soil material fosters this bonding. Surface negative charge of the clay is low and hydroxyl groups exert a strong influence. Infrared analyses indicate the presence of hydrogen-bonded hydroxyls. These are lost on heating to 300°C. Heating to this temperature leaves the fragipan material soft and loose. Brittle-


259

ness was restored by rehydration under moderate relative humidity, Knox (1954) studied the effect of heating to as high as 500°C. on the strength of fragipan material. The strength was not reduced appreciably. His observations would appear to be at variance with those of Nettleton et al. ( 1968b). The fragipans investigated, however, differ markedly. Knox ( 1 954) studied fragipans with strong influence of the parent material (sequences IIa, IIc of Fig. 2); illite dominates the clays. The fragipan studied by Nettleton et al. (1968b) occurs in a strongly eluvial horizon (sequence IIh, Fig. 2) where accumulation of amorphous clay through soil development would be more likely; kaolinite is a prominent clay mineral. The relationship between the properties of the exchange complex and attraction between clay particles has received much attention in the field of soil physical chemistry. Nettleton et al. (1 968b) apply concepts from this field to bonding of fragipan material. Many fragipans have low pH and contain appreciable aluminum extractable with a neutral salt. Calcareous fragipans, moreover, are rare. Applying the ideas presented by Emerson and Dettmann ( 1960), both the low pH through the increase in positive charge and consequently stronger electrostatic attraction, and the presence of trivalent aluminum, would increase attraction between clay particles. Carbonate would reduce the attraction because the resulting pH leads to low positive charge and to precipitation of the aluminum. Such ideas, however, do not provide a general explanation for the rigidity of fragipans. Many kinds of soil horizons are acid and have high extractable aluminum. Moreover, some fragipans have pH values near neutrality.

B. DEVELOPMENT OF FRAGIPAN SOILS I. Inheritance of Properties Fragipans in some areas of the Coastal Plain south of Wisconsin glaciation occur on the older geomorphic surfaces but not on the younger surfaces (Daniels et al., 1966; Nikiforoff et al., 1948; Nikiforoff, 1955). This is evidence that these fragipans may be relicts of an older environment. Restriction of fragipans to the older, more stable parts of the landscape is not limited to the Coastal Plain. This is the pattern of occurrence in parts of the Ozark Plateau, for example. Investigators of fragipans in areas of Wisconsin glaciation do not agree on the extent to which fragipans may be relict features. Denny and Lyford ( 1 963) wrote for the area of Wisconsin glaciation along the southwestern New York-Pennsylvania border: “The soils are relatively young, are in equilibrium with the present environment, and contain few, if any, features acquired during past weathering intervals.” In contrast, Olson and Hole ( 1967- 1968)

260


placed emphasis on the importance of changes in climate in the development of fragipans in soils in northeastern Wisconsin. Some fragipans bear a strong influence of the parent material. Not only composition, but organization of the soil fabric, may be largely determined by the parent material (Section IV, B, 1). In some soils parent material has such a strong influence that the distinction between glacial till and the fragipan may be difficult to establish (Section IX). 2 . Catastrophic Development

Processes common to a periglacial environment have been implicated in the development of fragipans. Fitzpatrick (1956) offered evidence that three features common to some fragipans can be produced by freezing wet soil. These features are platy structure, discontinuous spherical or vesicular pores, and a sheathing of fine material around pebbles. Yassoglou and Whiteside (1960) and Jha and Cline (1963) presented evidence against a periglacial origin being applicable generally and the main agent in fragipan formation. Fragipans occur in areas thought not to have been subject to periglacial influence. This would seem a convincing argument against a periglacial origin for all fragipans. Some fragipans, however, may have strong relict influence of a periglacial environment. The fact that the upper surface of fragipans occurs at predictable depths and generally conforms to the land surface does not rule out relict periglacial influence. As Nikiforoff (1955) and Lyford et al. (1 963) pointed out, the upper boundary of the fragipan could be controlled largely by the lower limit of obliteration of periglacial influence by soil development. Nikiforoff (1955) explored in detail the possibility that the gross prismatic structure of many fragipans has its origin in a periglacial environment. The large prisms would be formed by frost wedges or would be the result of dessication and contraction of soil material having an originally high water content. Jha and Cline (1963), studying a fragipan in lacustrine sediments, suggested that desiccation cracks which formed soon after drainage of the lake may have been the origin of the polygonal pattern. Olson and Hole ( 1 967- 1968) suggested that the large prisms are related to desiccation cracks formed during a dry period after close of the Pleistocene. The top of the fragipan often coincides with discontinuities in composition, organization, or both of the soil material. Evidence for changes in composition are given in the studies by Nikiforoff et af. (19481, Scrivner ( 1960), Rutledge and Horn ( 1 9 6 3 , Bailey ( 1964), Vanderford and Shaffer ( I 966), Beavers ( 1 960), and Calhoun (1968). An example of a

FRAGIPAN SOILS OF THE EASTERN U N I TED STATES

26 1

change in organization would be the shift from ablation till (let down by ice wastage) to denser till thought to have been compacted by the weight of ice. Such changes, either in composition or organization, although not directly responsible for the fragipan themselves, may have lead to conditions favorable to formation of a fragipan.

3. Incremental Development Fragipans occur beneath eluvial horizons. The overlying B horizon, if present, commonly is a type of cambic horizon which appears to have gone through a stage of active clay removal. If an overlying B horizon is an argillic horizon, it commonly is one in which accumulation of illuvial clay does not greatly overbalance removal. Tavernier and Smith (1957) pointed out that Gray-Brown Podzolic soils with fragipans commonly have either heavy bleached silt coatings on the ped surfaces of the B horizon above the fragipan or a distinct bleached horizon of eluviation between the B and the fragipan. Where the latter horizon occurs, it tongues into or interfingers with the overlying B horizon. Tavernier and Smith interpreted the tonguing or mingling of eluvial and illuvial horizons in these soils as evidence of destruction of the B horizon. Fragipans are therefore subject to the accumulation of substances from the horizons above. The substances may move as particles or in solution. Much speculation has centered on whether a precipitated substance is the bonding agent. Wetting and drying is probably the principal agent responsible for mobilization and translocation of silt and clay into the fragipan. Illuvial silt would reduce porosity and perhaps increase mechanical strength. The role of illuvial clay is less clear-cut. Many investigators have suggested that illuvial clay reduces porosity. Addition of clay, however, may increase volume change with change in moisture content and lead to more large planar pores. Furthermore, illuvial clay coats planar pores and thereby increases the prominence of structural planes of weakness. Wetting and drying may also cause translocation of substances within the fragipan. The mottled color pattern and low-chroma parts of many fragipans strongly suggest translocation of hydrous iron oxides. Once the hydrous iron oxide coatings have been removed, the silicate clay may be more subject to movement. In strongly eluvial horizons, removal of silicate clay and hydrous iron oxide coatings may have so weakened the fabric that wetting and drying can lead to extensive rearrangement to produce a denser fabric (Daniels et al., 1966; Nettleton et al., 1968b; Yassoglou and Whiteside, 1960; Pettiet, 1964). Strongly eluvial subsoil horizons tend to be zones of accumulation of free water. This free water

262


commonly extends throughout the fabric; it is not restricted to vertical planes. Such high water contents may weaken the fabric and make it prone to rearrangement. The subject of incremental development leads to the question of time required to form a fragipan. Fragipans have formed since the close of the Pleistocene. Jha and Cline ( 1 963) reported that a fragipan formed in lacustrine sediments deposited in a lake that drained seven to eight thousand years ago. On the other hand, fragipans are not found in very youthful deposits. Lyford er al. ( 1 963) offer evidence that the fragipans in certain soils of Massachusetts must be atleast 500 years old. Many fragipans show the marks of illuviation of clay, its removal, or both. Such alteration does not occur rapidly. Fragipans are not necessarily restricted to old soils, but neither do they form in a matter of hundreds of years.

4 . Weak Disturbance Many fragipans have undergone only weak disturbance. Strong influence of the organization of the parent material implies weak disturbance as does strong expression of pedological features such as gross polygonal structure, a pronounced pattern of accumulation and depletion of hydrous iron oxides, and moved clay bodies. Fragipans are not subject to frequent and large changes in volume. Many fragipans occur deep enough and contain so few roots that intense desiccation is rare. Interiors of the large structural units of some fragipans resist wetting (Section V, B). Many fragipans have a low potential for volume change because of moderate or low clay contents. Some fragipans do have fairly high clay contents. The latter, however, tend to occur at appreciable depths, where frequent and intense desiccation would be less likely. Fragipans do not occur in shallow, clayey horizons unless the soil has been severely eroded. Disturbance by tree throw and by mass movement is greater above than within the fragipan. Lyford et al. (1 963) showed for an area of soils developed in glacial till on steep slopes that the base of small gravity movements coincided with the upper boundary of the fragipan. R. M. Smith and Browning ( 1 946), studying soils with fragipans that occur on steep slopes in West Virginia, suggested that the base of the small slip scars common to the area was at the top of the fragipan. The base of the zone of disturbance due to tree throw commonly coincides roughly with the top of the fragipans (Lyford and MacLean, 1966; Denny and Lyford, 1963; Olson and Hole, 1967-1 968; Mueller and Cline, 1959). Estimates of the rapidity of mixing by tree throw point to its importance. Denny and Lyford ( 1 963) concluded that for soils of the northeastern United States


263

much of the upper 2 feet has been disturbed by tree throw. The study by Mueller and Cline (1959) tends to substantiate this conclusion. Milfred e? al. ( I 967) emphasized the importance of tree throw in northeastern Wisconsin. Fragipans commonly occur at shallower depths in the wetter soils of a local association. Trees may have rooted less deeply in the wetter soils because of the shallower depth to free water. Disturbance related to the presence of the tree roots, such as the volume change resulting from water withdrawal and recharge, root ramification, tree throw, and root movement resulting from tree sway, would all have extended to shallow depths. In this view, the upper boundary of the fragipan would be largely determined by the control exercised by the water table regime on the depth of frequent disturbance. Presence of a fragipan may increase disturbance of horizons above. Mixing by tree throw, for example, is probably more frequent above the fragipan because the fragipan limits the depth of support roots (Section VII, A). Higher rates of disturbance above the fragipan may increase the perviousness of these horizons. Low-tension water consquently may move more rapidly through the horizons above and accumulate at the top of the fragipan. Accumulation of this low-tension water would affect development of the fragipan. Weak disturbance within the fragipan and stronger disturbance above the fragipan are two sides of the same coin. Properties of the soil are a reflection of both. VII. Fragipans and Soil Use

A. PLANTGROWTH Fragipans unfavorably influsnce growth by restricting rooting, either through mechanical impedance or by creating saturated conditions. Soils with shallow fragipans usually have high water tables over extended periods. Effects of mechanical impedance and of the shallow water table on plant growth often are confounded. Improving the plant nutrient status of soils has become increasingly feasible. One result has been to increase the relative importance of a physical limitation such as presence of a fragipan. This is true particularly in the southeast part of the country. The Coastal Plain of the Southeast contains much of our potential new land for farming (Bartelli, 1968). In parts of the Coastal Plain, soils with fragipans are among the principal soils (Fig. 3). Depth to the fragipan partially determines its significance to plant growth. The critical depth depends on the plants. For many plants, the

264


influence diminishes rapidly at depths greater than 50 cm. Erosion brings the fragipan closer to the soil surface and thereby increases its significance. Erosion has greatly reduced the average depth to the fragipan in certain soils, such as the Beltsville and Grenada soils (see Appendix). Thickness of the fragipan may determine its significance to soil use and management. The apparently small influence of the fragipan in the McBride soil (see Appendix) may be largely a reflection of its thinness. Trees are the principal vegetation on much of the area of soils with fragipans. The density of tree roots is low in the fragipan (Olson and Hole, 1967- 1968; Lyford and MacLean, 1966; Mueller and Cline, 1959). The influence of root distribution on tree throw has received much attention. Olson ( 1 962) suggested that the restriction of roots by fragipans is greater for the larger roots that give mechanical support to trees than for the smaller feeder roots. Mueller and Cline ( 1 959) observed that calcareous glacial till of similar bulk density to the fragipan is not as effective a barrier to rooting. Goodlett ( 1960) discussed the confounding influences of the water table regime and depth to the fragipan for an area in central Massachusetts.

B. ENGINEERING MANIPULATION Depth to the fragipan influences the kind of agricultural drainage system needed. If the fragipan is shallow, the system should be designated to remove surface water. If the fragipan is deep, then ditches or tile lines may be feasible. Thickness of the fragipan may also affect suitability of a drainage system. If the fragipan is thin, although shallow, tile or ditches may be feasible. Fragipans increase construction costs in several ways. Difficulty in digging makes small excavations more expensive. The slow movement of low-tension water in the pan causes several problems. Pervious seepage beds for septic tanks may be required (J. H. Huddleston and Olson, 1967). Accumulation of water at the top of the fragipan may incease the time required for drainage after heavy precipitation during earth-moving operations. Fragipans beneath surfaced runways or roads may lead to accumulation of water in the subgrade, which loses strength and is subject to being pumped out under repeated loading and relaxation. Lateral water movement above the fragipan causes problems where cuts are made on sloping land. Road design must provide for the interception of water moving laterally along the top of the fragipan. Cuts must be deep enough and fill thick enough so that the road surface is either well below or above the top of the fragipan. The location and manner in which people live in the United States


265

increase the importance of fragipans. Most of the people live in the East where soils with fragipans are extensive. Suburban living usually involves individual houses. The scale of construction for individual houses is small enough to be influenced by the presence of a fragipan. The largescale construction in the urban core would not be so affected. Many communities in the eastern United States were established in valleys where the soils lack fragipans. Today the communities are expanding out of the valleys into areas where fragipans may be prevalent (Olson, 1966). Moreover, in these upland areas the more favorable sites for construction from the viewpoint of gentleness of slope often have soils with the strongest fragipans at the shallowest depths. C. J. Thomas ( 1 966) discussed the influence of fragipans on planning the development of a community in Massachusetts. She wrote: “The soils present problems when used for commercial, industrial, or highdensity residential purposes where community sewage disposal is not available. The hardpan [fragipan] restricts water from moving downward readily. When the soil above the hardpan is saturated, water tends to move downslope along the top of the hardpan. Hardpan layers have slow to very slow permeability and individual sewage effluent disposal systems do not function satisfactorily.” T o move beyond such a qualitative description to quantiative statements, which often are required by local ordinances, leads into problems. These are well illustrated in the several studies referred to by Olson ( 1966). T o obtain reliable measurements requires expenditure of a great deal of effort. Even the percolation test, simple in principle, is complicated in application (Hill, 1966; Huddleston and Olson, 1967). The water table regime is another example. Measurements are necessary over several years (J. H. Huddleston and Olson, 1967), and even then may not indicate conditions resulting from very infrequent weather events. Furthermore, as construction often results in diversion of water from one place to another, the measurements may not be applicable to the actual soil use. VIII. Classification of Fragipan Soils

The term fragipan is a coined name from the Latin root for brittle. It was proposed by G. D. Smith in 1946 (unpublished working papers of National Conference of the Cooperative Soil Survey) to identify soils having horizons that had been called silica hardpans by Winters ( 1942), silt-pans by R . M. Smith and Browning ( 1946), and simply hardpans by a number of workers in the eastern part of the country (for example, Carr, 1909; Hearn, 1924; Howe er al., 1924). The term was adopted by the

266


Soil Survey Staff of the U.S. Department of Agriculture ( 1 95 1, 1960) and has been in general use for about 20 years. But that is getting ahead of the story. The kinds of horizons that are designated fragipans were recognized as important characteristics of some soils fairly early in the course of the soil survey of the country. Early in this century, Carr (1 909) wrote about the compact “hardpan” in the subsoil of Volusia soils in New York and its importance to plant roots and water relations in the soil. The concept of the Volusia series at that time, however, was based in large part on geology, and the series was not characterized as having a pan (Marbut el al., 1913). A few years later, Carter and Hull (1916) described the Leonardtown soils in Maryland as having a compact hardpan in the subsoil, and subsequently the presence of the hardpan became part of the concept of that series (H. C. Smith and Rose, 1924). A substantial number of soil series were characterized as having a compact, slowly permeable hardpan layer of the same general nature during the two decades that followed. As more detailed field studies of soils have been made and as more kinds of interpretations of soil surveys have been required, the number of kinds of soils in which fragipans have been recognized has increased. At this writing, approximately 250 soil series in that part of the country lying east of the Great Plains are characterized as having fragipans. The presence or absence of fragipans apparently was not a factor in the classification of soils into groups broader than series during the first two decades of the soil survey in this country. During that period groupings of series were primarily in terms of geography, geology, or physiography (Marbut et al., 191 3; Marbut, 1935), an approach that slowly faded during succeeding decades. Concepts of soil development and of “normal soils” that were to affect strongly soil classification in this country in later years were gaining ground in the late teens. Describing the soils of Charles County, Maryland, H. C . Smith and Rose (1 924) wrote that the profile of the Leonardtown series differs from the normal mature soils of the region in the presence of the compacted zone (fragipan) in the illuvial horizon and somewhat lighter than normal color of the overlying horizon. For the same reasons, Perkins and Bacon ( 1 925) considered Leonardtown soils to be “postmature soils.” This was in accord with Marbut’s ideas about “normal,” “mature,” or “fully developed” soils that strongly influenced his soil classification (Marbut, 1928). Seemingly, soils with fragipans and other kinds of pan horizons did not have a clear place in the higher categories of his classification. They were recognized as “overdeveloped soils” in category 111,


267

within which groupings were based on stage in development, and in the lower two categories. Thus, Marbut (1935) discussed the Grenada series under the heading Red and Yellow soils and the Leonardtown series under Gray-Brown Podzolic soils, but these series apparently were not considered bona fide members of those great soil groups because of their fragipans. The logical inconsistencies of Marbut’s classification were corrected to a large extent by the subsequent classification scheme outlined by Baldwin et al. ( I 938). In this system, soils with fragipans seemed to have a legitimate place in the great soil group of Planosols, which had a place in the order of Intrazonal soils. The term, Planosol, was “proposed to cover those soils with claypans and cemented hardpans not included with Solonetz, Ground-Water Podzol, and Ground-Water Laterite.” Thorp and Smith ( I 949) modified the definition as follows: Intrazonal soils having one or more horizons abruptly separated from and sharply contrasting to an adjacent horizon because of cementation, compaction, or high clay content. Soils of the Grenada, Leonardtown, and other series with fragipans were classified as Planosols in this system (for example, see Fox et al., 1958; Leighty and Wyatt, 1950). Many soil scientists recognized that the Planosol group as then constituted was very heterogeneous and that most Planosols had important properties in common with those of the zonal soils with which they were associated (unpublished working papers of the I 947 National Conference of the Cooperative Soil Survey). But agreement on an alternative classification was not forthcoming. Thorp and Smith (1949) noted that an earlier proposal to elevate Planosols to the rank of a suborder of Intrazonal soils and to recognize groups of soils with “silt pans,” clay pan soils, and certain other soils as subdivisions in the great soil group category had not been unanimously accepted by the National Conference of the Cooperative Soil Survey. During the next dozen or so years, several different approaches were followed in classifying soils with fragipans. In a number of soil surveys in the southeastern part of the country soils with fragipans were classified as Planosols according to the system outlined by Baldwin et al. (1938),as modified by Thorp and Smith (1949) (for example, McNutt et al., 1959; Fox et al., 1958; Leighty and Wyatt, 1950). In the latter part of that period some writers classified only the somewhat poorly drained and the poorly drained fragipan soils as Planosols and included the better drained fragipan soils in the zonal great soil groups to which the soils belonged on the bases of their features other than the fragipan (for example, Love et al., 1959; Matthews et al., 1961). In the northeastern states soils with

268


fragipans were not classified as Planosols but were placed in the great soil group to which they would belong on the basis of their other characteristics (for example, Cline, 1955; Taylor, 1960; Pearson and Cline, 1960; Shearin and Hill, 1962). Nevertheless, the presence or absence of a fragipan was emphasized in the series classification of soils in that part of the country. In discussing Fragipan Planosols Cline (1 952) pointed out that, unlike the claypan of the Planosol, which occurs in the position of the B horizon of the associated normal soil, the fragipan occurs beneath, and in addition to, all the essential horizons of the solums of the Red-Yellow Podzolic, Gray-Brown Podzolic, Podzol, Brown Podzolic, and probably some other zonal great soil groups. He wrote, “In one sense, climate and vegetation appear to have exerted little control over its [the fragipan] character in comparison with their influence on horizons above.” Thus, there could be no modal horizon sequence in a great group that included all soils with fragipans. Cline ( 1 952) suggested that a great group of fragipan soils might be defined for the poorly drained soils with fragipans, as the variation in horizon sequence seemed relatively minor in those soils throughout the several soil zones. The imperfectly, moderately well, and well-drained soils with such pans could then be placed in the appropriate zonal great group and be set apart in the next lower category from soils lacking fragipans. The decision to revise the soil classification system used by National Cooperative Soil Survey and the reviews of early drafts of a new system continued to focus attention on how soils with fragipans would be best classified at categorical levels above the soil series (unpublished working papers of the National Cooperative Soil Survey). Some soil scientists wanted to make claypans, fragipans, and indurated pans bases for grouping soils at the great soil group level or above; others wanted to bring them in at the series level. Most reviewers thought the initial approximation of the new classification system overemphasized the importance of Planosols by giving them a separate place high in the system and advised that pan horizons be made diagnostic in lower categories. The idea of a broad group of Planosols, subdivided in the next lower category into soils with claypans, soils with fragipans, soils with indurated hardpans, etc., survived in the 2nd Approximation prepared in 1952, but at a lower level of the system. That was the last time in the development of the current system that soils with the various kinds of pan horizons were placed together in a class resembling the old Planosol great soil group. In subsequent drafts of the system, soils with and without fragipans were placed together in classes that roughly approximated the


269

Podzol, Gray-Brown Podzolic, Red-Yellow Podzolic, and a few other great soil groups. These were subdivided into classes with and without fragipans, among other criteria in the next lower category, which approximated the current great group level. With some variations, and with one exception, this arrangement persisted through the 7th Approximation of the classification system (Soil Survey Staff, 1960) and to the system that is in use today. The exception was a trial in the 5th, 6th, and 7th Approximations ( 1 956-1 960) of a proposal to keep Spodosols with and without fragipans together at the great group and subgroup levels of the system and to make the distinction on the presence or absence of a fragipan in the family category. (The class of Spodosols is approximately equivalent to the Podzol and part of the Brown Podzolic great soil groups.) The rationale for the proposal was the belief that a fragipan or a comparable horizon was normal in Spodosols. Those Spodosols that lack the fragipan were thought to be too coarse textured for a fragipan to form (Soil Survey Staff, 1960). After a few years’ trial, the classification system was amended to set apart Spodosols with and without fragipans in the great group category. Use of the fragipan as a differentia within Spodosols then paralleled its use in the orders of Inceptisols, Alfisols, and Ultisols. In the current soil classification system (Soil Survey Staff, 1967) fragipans are used uniformly to set apart one great group in each of the eleven suborders of the four orders in which soils with fragipans occur. These are listed in Table I. Note that each of the four aquic suborders has a great group of soils with fragipans but the dry or seasonally dry suborders do not have such great groups. The family category of the current soil classification system is designed to group series that are similar in certain properties, such as texture and mineralogy, that have importance in the use and management of soils. The top of the fragipan is the base of the part of the soil that is considered in applying the texture and mineralogy criteria for the classification of series into families. The rationale is that plant roots do not sufficiently exploit the soil within and below the fragipan. IX. Unresolved Problems

There are difficulties in the identification of fragipans. High bulk density and consistence have not been satisfactory criteria for recognition of fragipans in loess-derived soils of the lower Mississippi Valley (personal communication from L. J . Bartelli). Brittleness when moist does not suffice for identification of fragipans and their distinction from soils with plinthite in North Carolina (personal communication from Forrest

270

R. B. GROSSMAN AND F. J. CARLISLE

TABLE I THEGRE \ T GROUPS OF SOILS HAVINGFRAGIPANS, SHOWlNG THEIRDISTRIBUTION OF THE CURRENT SOILCLASSIFICATION SYSTEM AMONGT H I ORDERSA N D SUBORDERS

Order Entisol. Vertiso' Inceptisc '

Aridis, I Molli>,' Spodo.

1

L,uhorcler

Great groups having fragipans"

Order

Suborder

-

-

Alfisols

Aqualfs Boralfs Udalfs Ustalfs Xeralfs Aquults Humults Udults Ustults Xerults

_-

Andepts Aquepts Ochrepts Plaggepts l'ropepts I I mbrepts Iquods kerrods H timods Orthods

Fragiaquepts Fragiochrepts Fragiurnbrepts

Ultisols

-

Fragiaquods Fragihumods Fragiorthods

Oxisols Histosols

-

Great groups having fragipans" Fragiaqualfs Fragiboralfs Fragiudalfs Fragiaquults Fragiudults -

Data for illuslrative profiles for seven of these are in ttre 7rh Approximation (Soil Survey Staff, 1960) as follows: Fragiaquepts-profile Nos. 30, 48; Fragiochrepts-profile No. 24: Fragiorthods-profile No. 29; Fragiaqualfs-profile No. 82; Fragiudalfs-profile Nos. 84, 98: Fragiaquults-profile No. 94; Fragiudult-profile No. 97.

Steele). These problems concern fragipans showing strong alteration of the parent material. There are also problems of identification of fragipans that show weak alteration of the parent material. Brittle zones occur in the upper substratum of some soils developed from compact glacial till. If many feet thick, the zone can be eliminated as a fragipan on scale. In some soils the brittleness decreases in expression within 2 or 3 feet. These zones are difficult to distinguish from fragipans as presently defined. Modifying the definition of fragipans to include certain morphological features, such as bodies of moved clay or expression of a gross polygonal structure, would provide a basis for excluding certain of these zones of compact glacial till. Such a definition, however, might exclude some horizons of Spodosols and Inceptisols that are currently classified as fragipans and thus raise other problems. Fragipans are now a criterion for distinguishing great groups in the Comprehensive Soil Classification System. There is some concern with the categorical level at which fragipans are recognized. The concern arises from two somewhat different problems: (1) that of distinguishing


27 1

fragipans from other kinds of horizons with properties affecting root growth and water movement similar to those of fragipans; and (2) that of determining the lower limit of fragipan expression. From the point of view of soil use, the first problem may not be too serious. If a horizon has similar limitations to use and management of soil as a fragipan, its recognition as a fragipan does no great immediate harm. The question of the lower limit of expression for recognition of a fragipan seems more important. The problem has similarities to an evaluation of engineering test data that results in overdesign. Fragipans are recognized which may have marginal importance to use and management. Once a fragipan is recognized in a soil, however, there is a tendency to change appreciably the interpretive ratings for many kinds of soil use from the ratings assigned to otherwise similar soils. Moreover, the part of the soil that determines the family placement may change, and this may affect the correlation of soil interpretations. The name fragipan in a sense states and brings into focus an underlying problem that affects both identification and classification. A consistence property, relative fragility or brittleness, is the defining characteristic for a horizon that importantly affects soil use and exhibits a wide range in kind and degree of soil development. Importance of the fragipan does not lie in its fragility or any single feature, but rather in a group of attributes which together make it a pan. These attributes in combination cannot be readily measured. Separation of fragipans from other kinds of soil horizons may be improved by modification in the definition and collection of numerical information on consistence. Certainly, if fragipans were shown to have a unique bonding agent, the distinction from other kinds of pans would be more definite. But this would not necessarily resolve the difficulty in specifying the lower limit of expression for the recognition of a pan. X. Summary

Fragipans are subsoil horizons that are brittle and rigid when moist, that restrict root growth, and that transmit low-tension water more slowly than the overlying horizons. They occur in soils subject to net downward water movement. Trees rather than grass were the dominant vegetation a t the time the country was settled. Some fragipans have undergone strong eluviation, illuviation, or both. Others occur in soil materials that exhibit only weak alteration of the parent material. Clay is probably the chief bonding agent. Organization of the clay relative to the sand and silt grains may be of importance. Fragipans are low in organic matter and infrequently, if ever, calcareous. Otherwise, their mineralogical and

212

R. B. GROSSMAN A N D F. J . CARLISLE

chemical characteristics range widely. Extremes in texture are excluded; high proportions of particle size separates in the 0.2- to 0.02-mm. range may be conducive to fragipan formation. A gross polygonal structure is common. Blocky structure within the large structural units is at most moderately expressed; some fragipans have platy structure. Vertical continuity of large voids is restricted. Bodies of moved clay apparently are found in all fragipans. Most fragipan material is at least firm when moist; i.e., a piece held between thumb and forefinger offers moderate or more resistance to rupture. Fragipans occur deep enough to be subject to weak physical disturbance and below maximum influence of current soil development. Some fragipans may be largely relict features; others are not. Periglacial influences may have affected fragipans in the north, but not those in the south. Depth to the fragipan largely determines its influence on plant growth. Soils with shallow fragipans commonly have high water tables which may be a factor in restricting rooting depth. The importance of fragipans to mechanical manipulation of soils is largely relatable to their slow transmission of low-tension water. In the Comprehensive Soil Classification System, fragipans are diagnostic at the great group level and they are recognized in that category in four soil orders. In those orders they are used at a lower categorical level than the argillic, cambic, and spodic horizons but at a higher categorical level than features of similar importance to soil use having less genetic implication, such as the lithic contact. REFERENCES

Alexander, E. B., Jr. 1955. M.S. Thesis, Ohio State University, Columbus, Ohio. Anderson, J. U . , and White, J. L. 1958. Soil Sci. Soc. Am. Proc. 22,450-454. Bailey, H. H. 1964. Soil Sci. Soc. Am. Proc. 28,680-683. Baker, J. C. 1967. M.S. Thesis, University of Missouri, Columbia, Missouri. Baldwin, M., Kellogg, C. E., and Thorp, J. 1938. Yearbook Agr. (U.S. Dept. Agr.) pp. 979- I00 1 . Bartelli, L. J. 1968. Trans. 9th Intern. Congr. Soil Sci., Adelaide, 1966, Vol. 4, pp. 24325 1. Intern. Soc. Soil Sci. and Angus & Robertson, Sydney. Beavers, A. H. 1960. Trans. 7th Intern. Congr. Soil Sci., Madison, Wise., 1960 Vol. 2, pp. 1-9. Intern. SOC.Soil Sci. Bornstein, J . , Bartlett, K. J . , and Howard, M . , Jr. 196.5. Soil Sci. Soc. Am. Proc. 29,201205. Brewer, R. 1964. “Fabric and Mineral Analysis of Soils.” Wiley, New York. Calhoun, F. G . 1968. Ph.D. Thesis, Ohio State University, Columbus, Ohio. Carlisle, F. J. 1954. Ph.D. Thesis, Cornell University, Ithaca, New York. Carlisle, F. J . 1958. “Soil Survey of Franklin County, New York.” U.S. Dept. Agr., Washington, D.C. Carlisle, F. J . , Knox, E. G ., and Grossman, R. B. 1957. Soil Sci. Soc. Am. Proc. 21,320321.

FRAGIPAN SOILS O F T H E EASTERN U N I T E D STATES

273

Carr, M. E. 1909. US.Depr. Agr., Bur. Soils Bull. 60. Carter. W. T.. Jr.. and Hull, J. P. D. 1916. “Soil Survey of Howard County, Maryland,” U.S. Dept. Agr., Bur. Soils Field Operations 1916, Washington, D.C. Cline, M. G . 1952. “Soil Classification in the United States, Observations and Conclusions during Sabbatic Leave,” Mimeo. Rept. Cornell Univ., Ithaca, New York. Cline, M. G . 1955. Cornell Exr. Bull. 930. Comer, G . H., and Zimmerman, R. C. 1969.5. Hydrol. 7,98-108. Comerma, J. A. 1964. M.S. Thesis, North Carolina State University, Raleigh, North Carolina. Daniels, R. B., and Gamble, E. E. 1967. Geoderma 1, I 17- 124. Daniels, R. B., Nettleton, W. D., McCracken, R. J., and Gamble, E. E. 1966. SoilSci. SOC. A m . Proc. 30,376-380. Daniels. R. B., Gamble. E. E., and Bartelli, L. J. 1968. Soil Sci. 106,200-206. Denny, C. S., and Lyford, W. H. 1963. Geol. Surv. Profess. Paper 379. U.S. Dept. Interior, Washington, D.C. Emerson, W. W.. and Dettmann, M. G. 1960.5. Soil Sci. 11, 149- 158. Fitzpatrick, E. A. 1956. 5. Soil Sci. 7,248-255. Fountaine, E. R. 1954. J . Soil Sci. 5,25 1-263. Fox, C. J., Beesley, T. E., Leighty, R. G., Lusk, E., Harmon, A. B., Smith, H. C . , Methvin, C.. and Flowers. R. L. 1958. “Soil Survey of Franklin County, Tennessee,” U.S. Dept. Agr.. Washington, D.C. Gile, L. H..Jr. 1958. SoilSci. Soc.Am. Proc. 22,560-565. Gill, W. R.,and Vanden Berg, G . E. 1967. U S . Dept. A g r . , A g r . Handbook 316. Glenn, R. C. 1960. Truns. 7th Intern. Congr. Soil Sci., Madison, Wisc., 1960 Vol. 4, pp. 523-53 I . Intern. SOC.Soil Sci. Goodlett, J. C . 1960. Harvard Foresf Bull. 28. Grossman, R. B. 1954. M.S. Thesis, Cornell University, Ithaca, New York. Grossman, R. B., and Bartelli, L. J. 1957. Soil Sci. SOC.A m . Proc. 21,661-662. Grossman, R. B.,and Cline, M. G . 1957. SoilSci. Soc.Am. Proc. 21,322-325. Grossman, R. 9.. Fehrenbacher, J. B., and Beavers, A. H . 1959a. SoilSci. Soc. A m . Proc. 23,65-70.

Grossman, R. B., Stephen, I., Fehrenbacher, J . B., Beavers, A. H., and Parker, J. M. I959b. Soil Sci. SOC.A m . Proc. 23,70-73. . Sci. Grossman, R. B., Stephen, I., Fehrenbacher. J. B., and Beavers, A. H. 1 9 5 9 ~ Soil Soc.Am. Proc. 23,73-75. Grossman, R. B., Brasher, B. R., Fransmeier, D. P., and Walker, J. L. 1968. Soil Sci. Soc. A m . Proc. 32,570-573. Hearn, W. E. 1924. A m . Soil Surv. Assoc. Bull. 5 , 18. Hill, D. E. 1966. Connecticut A g r . Expt. Sta., N e w Haven, BUN. 678. Horn, M. E., and Rutledge, E. M. 1965. Soil Sci. Soc. A m . Proc. 29,443-448. Howe, F. B., Buckman. H. O., and Lewis, H. G . 1924. “Soil Survey ofTompkins County, New York,” U.S. Dept. Agr., Bur Soils Field Operations 1920. Washington. D.C. Huddleston,J. H., and Olson, G. W. 1967. SoilSci. 104,401-409. Huddleston, J. S. 1967. “Soil Survey o f T a t e County, Mississippi,” U.S. Dept. Agr., Washington, D.C. Hutcheson. T. B., Jr., and Bailey, H. H. 1964. Soil Sci. Soc. A m . Proc. 28,684-685. Hutcheson, T . B., Jr., Lewis, R. J., and Seay, W. A. 1959. SoilSci. SOC.A m . Proc. 23, 474-478.

International Society of Soil Science. 1968. SoilSci. Sac. A m . Proc. 32, 153-154. Jha, P. P. 1961. Ph.D. Thesis, Cornell University, Ithaca, New York.

274


Jha, P. P., and Cline, M. G. 1963. SoilSci. Soc. Am. Proc. 27,339-344. Knox, E. G. 1954. Ph:D. Thesis, Cornell University, Ithaca, New York. Knox, E. G. 1957. Soil Sci. Soc. Am. Proc. 21,326-330. Kmsekopf, H. H. 1942. SoilSci. Soc. A m . Proc. 7,434-436. Leighty. W. J.. and Wyatt. C. E. 1950. “Soil Survey of Marshall County, Kentucky,” U.S. Dept. Agr., Washington, D.C. Love, T. R., Williams, L. D., Proffitt, W. H., Epley, I. B., and Elder, J. H. 1959. “Soil Survey of Coffee County, Tennessee,” U.S. Dept. Agr., Washington, D.C. Lyford, W. H. 1964. HurvurdForest Paper No. 8. Lyford, W. H.,and MacLean, D. W. 1966. Harvard Forest Paper No. 15. Lyford, W. H., Goodlett, J. C., and Coates, W. H. 1963. Harvard Forest Bull. 30. McCracken, R. J., and Weed, S. B. 1963. Soil Sci. Soc. A m . Proc. 27,330-334. McKeague, J. A., and Cline, M. G. 1963. Can. J . Soil Sci. 43,70-96. McNutt, E. J., Green, T. W., Kahrein, R. B., Galberry, H. S., Thomas, A. E., Tyler, M. C., and Mathews, E. D. 1959. “Soil Survey of De Soto County, Mississippi,” U.S. Dept. Agr., Washington, D.C. Marbu t, C. F. 1928. Proc. 1 s t Intern. Congr. Soil Sci.,1927, Vol. 4, pp. 1-3 I . Marbut, C. F. 1935. I n “Atlas of American Agriculture” (0.E. Baker, ed.), Part 111. U.S. Dept. Agr., Washington, D.C. Marbut, C. F., Bennett, H. H., Lapham, J. E., and Lapham, M. H. 1913. US.Dept. Agr., Bur. Soils Bull. 96. Matthews, E. D., Compy, E. Z. W., and Johnson, J. C. 1961. “Soil Survey of Montgomery County, Maryland,” U.S. Dept. Agr., Washington, D.C. Milfred, C. J., Olson, G. W., and Hole, F. D. 1967. Univ. Wisconsin Bull. 85, Soil Ser. 60. Miller, F. P. 1965. Ph.D. Thesis, Ohio State University, Columbus, Ohio. Miller, R. W. 1967. Soil Sci. Soc. Am. Proc. 31,46-50. Mueller, 0 .P., and Cline, M. G. 1959. Soil Sci. 88,107- 1 I I . Neeley, J. A. 1965. “Soil Survey of Tompkins County, New York,” U.S. Dept. Agr., Washington, D.C. Nettleton, W. D. 1965. Ph.D. Thesis, North Carolina State University, Raleigh, North Carolina. Nettleton, W. D., Daniels, R. B., and McCracken, R. J. 1968a. Soil Sci. S O C . Am. Proc. 32, 577-582. Nettleton, W. D., McCracken, R. J., and Daniels, R. B. 1968b. Soil Sci. Soc. Am. Proc. 32, 582-587. Nikiforoff, C. C. 1955. Geol. Surv. Profess. Paper 267-B. U S .Dept. lnterior, Washington, D.C. Nikiforoff,C. C.,and Alexander, L.T. 1942.SoilSci. 53,157-172. Nikiforoff,C. C., Humbert, R. P.,andCady,J. G. 1948.SoiISci. 65,135-153. Olson, G. W. 1962. Ph.D. Thesis. University of Wisconsin, Madison, Wisconsin. pp. 1 13-125. Soil Sci. SOC.Am. and Am. SOC.Agron., Madison, Wisconsin. Olson, G. W. 1966. In “Soil Surveys and Land Use Planning” (L. J. Bartelli et ul., eds.), pp. 113-125. Soil Sci. SOC. Am. and Am. SOC. Agron., Madison, Wisconsin. Olwn, G. W., and Hole, F. D. 1967-1968. Wisconsin Acad. Sci. Arts Letters 56, 173-184. O’?leal, A. M. 1952. Soil Sci. Soc. Am. Proc. 22,3 12-3 15. Pcarson, C. S., and Cline, M. G. 1960. “Soil Survey of Lewis County, New York,” U.S. Dept. Agr., Washington, D.C. Perkins, S. O., and Bacon, S. R. 1925. “Soil Survey of Prince Georges County, Maryland,” U.S. Dept. Agr., Washington, D.C.


275

Pettiet, J. V. 1964. Ph.D. Thesis. Mississippi State University, State College, Mississippi. Porter, H. C., Derting, J . F.. Elder, J . H., Henry, E. F., and Pendleton, R. F. 1963. “Soil Survey of Failfax County. Virginia,” U . S . Dept. Agr., Washington. D.C. Redmond, C. E.. and Engberg. C. A. 1967. “Soil Survey of Arenac County. Michigan,” U.S. Dept. Agr., Washington, D.C. Rutledge. E. M.,and Horn. M. E. IY65.SoilSci. Soc.Anr. Proc. 29,437-443. Scrivner, C. L. 1960. Ph.D. Thesis, University of Missouri, Columbia, Missouri. Shantz, H . L., and Zon. R. 1924. I n “Atlas of American Agriculture” (0.E. Baker, ed.), Part IV. U.S. Dept. Agr., Washington, D.C. Shearin, A. E., and Hill. D. E. 1962. “Soil Survey of Hartford County, Connecticut,” U.S. Dept. Agr., Washington, D.C. Smith, H. C . , and Rose. R. C. 1924. “Soil Survey of Charles County, Maryland,” U.S. Dept. Agr.. Bur. Soils Field Operations 1918. U.S. Govt. Printing Office, Washington, D.C. Smith. R. M., and Browning, D. R. 1946. Soil Sci. 62, 307-3 17. soil Conservation Service. 196Y. In “National Atlas,” (U.S. Dept. Interior. Geol. Surv., Sheet N o . 86. Soil Survey Staff. 195 I . “Soil Survey Manual,” U . S . Dept. Agr. Handbook 18. Washington, D.C. Soil Survey Staff. 1960. “Soil Classication- A Comprehensive System -7th Approximation.” U.S. Dept. Agr., Washington, D.C. Soil Survey Staff. 1962. “Supplement to Soil Survey Manual,” U.S. Dept. Agr. Handbook 18. Washington, D.C. Soil Survey Staff. 1967. “Supplement to Soil Classification System (7th Approximation). U.S. Dept. Agr., Washington, D.C. Soil Survey Staff. 1968a. U S . Dept. Agr., Soil Surv. Invest. R e p f . 19. Sod Survey Staff. I968b. US.Dept. Agr., Soil Surv. Invest. R e p t . 20. Spaeth, J. N., and Diebold, C. H. 1938. Cornell Univ.,A g r . Expt. Sra. Mem. 213. Tavernier, R.. and Smith, G. D. l9S7. Advan. Agron. 9,2 17-289. Taylor, D. C. 1960. “Soil Survey of Erie County, Pennsylvania,” U.S. Dept. Agr., Washington, D.C. Thomas, A. E. 1967. “Soil Survey of Grenada County, Mississippi,” U.S. Dept. Agr., Washington, D.C. Thomas, C. J. 1966. In “Soil Surveys and Land Use Planning” (L. J . Bartelli e t a / . , eds.), pp. 60-75. Soil Sci. SOC.Am. and Am. SOC.Agron., Madison, Wisconsin. Thorp, J., and Smith, G. D. 1949. Soil Sci. 67, I17- 126. Threlkeld, G., and Alfred, S. 1967. “Soil Survey of lonia County, Michigan,” U.S. Dept. Agr., Washington, D.C. Vanderford, H . B., and Shaffer, M. E. 1966. SoilSci. Soc. Am. Proc. 30,494-498. Whittig, L. D., Kilmer, V. J., Roberts, R. C., and Cady, J . G. 1957. Soil Sci. Soc. A m . Proc. 21,226-232. Winters, t.1942. SoilSci. Sor. A m . Proc. 7,437-440. Winters, E.. and Simonson, R. W. I95 I.Advan. Agron. 3, 1-92, Yassoglou. N. J . , and Whiteside. E. P. 1960. Soil Sci. Soc. A m . Proc. 24,396-407.

276


Appendix

BELTSVILLESOIL The soil has a fragipan in the lower part of an illuvial B horizon and is an example of sequence Ild, Fig. 2. The fragipan shows evidence of appreciable clay accumulation. The description is from the report on soils of Montgomery County, Maryland (Matthews et al., 1961). The information on soil use comes both from this report and the one for Fairfax County, Virginia (Porter et al., 1963). Surface soil AOO- 1 to f.i inch, scattered pine needles. AO- M to 0 inch, loose but felty decomposed leaf mold. A 1-0 to 2 inches, very dark gray (IOYR 3/1) silt loam; moderate, fine, crumb structure: loose; roots abundant; strongly acid; abrupt, wavy boundary. A2-2 to 13 inches, yellow (2.5Y 8/6) silt loam; weak, fine, subangular blocky structure; friable when moist and slightly sticky when wet; roots abundant; strongly acid; gradual, smooth boundary. t Subsoil B I - 13 to 2 1 inches, brownish-yellow (IOYR 6/6), light silty clay loam; few, medium, faint mottles of strong brown (7.5YR S/8);moderate, medium, and coarse, subangular blocky structure; firm when moist; roots fairly abundant; a few rounded pebbles; thin, distinct clayskins on some faces: strongly acid; gradual to clear, wavy boundary. B21-21 to 31 inches, brownish-yellow (IOYR 618) silty clay loam: few, fine, distinct mottles of red (2.5YR 5/6); weak, coarse, platy structure; firm when moist; roots few, between structural elements only: a few pebbles and a few faint clayskins; strongly acid; clear, irregular boundary. B22m-31 to 42 inches, reddish-yellow (7.5YR 6/6) clay loam; many, fine, faint (B22x) mottles of reddish yellow (5YR 6/8) and many vertical channels and horizontal streaks of gray silt and clay; compound structure-moderate, very coarse, platy and moderate, medium to coarse, subangular blocky; very compact and dense; very firm when moist; very few roots; some pebbles and a few clayskins; this is the fragipan, or hardpan; strongly acid: clear, smooth boundary. B3m-42 to 48 inches, reddish-yellow (7.5YR 6/6) clay loam; many, medium, distinct mottles of light yellowish brown (IOYR 6/4), reddish yellow (SYR 6/8), and (B3x) white (2.5Y 8/2): compound structure-moderate, very thick, platy and moderate, coarse, blocky: compact and dense; very firm when moist; practically no roots: some pebbles and some light-gray to white silt coats; strongly acid; this lower hardpan is a transition between the true subsoil and the substratum. Substratum CD-48 to 54 inches t,very pale brown (IOYR 7/3) silty clay loam; abundant, rnediurn, distinct mottles of reddish yellow (7.5YR 6/6) and light gray (IOYR 7/2); massive; very firm; strongly acid: underlain at some depth by gravel. Matthews et a / . (1961) wrote: “Because of the almost impervious fragipan, the Beltsville soils tend to be wet at times. Frequently, they are saturated near the surface, but almost dry within or below the fragipan. The moisture-supplying capacity is moderate. A few depressed areas in the uplands are ponded for short periods after long heavy rains or quick


277

thaws.” On the steeper slopes, due partly to accelerated erosion, the fragipans occur at shallow depths; percolation of water is slow and runoff rapid. The soil is often too wet for cultivation in the spring, and during the summer it becomes dry. Permanent pasture and hay are the better suited crops. The soils are not suitable for septic tanks. Seepage on cut slopes causes problems in road construction.

GRENADASOIL The soil has a fragipan in the illuvial horizons of the lower sequum and is an example of sequence Ilf, Fig. 2. Parent material is loess. The description and information on soil use are from the report on soils of Grenada County, Mississippi (A. E. Thomas, 1967). Ap-0 to 5 inches, grayish-brown (IOYR 5 / 2 ) silt loam with common, coarse, faint, dark grayish-brown (IOYR 4/2) mottles; weak, fine and medium, granular structure; friable: common fine roots; few worm casts; few, fine, black concretions; few iron stains; plow pan in the lower part of horizon; neutral; abrupt, smooth boundary. A2-5 to 7 inches, brown (IOYR 5/3)silt loam with few, medium, faint, grayish-brown (IOYR 5 / 2 ) mottles; weak, medium, granular and weak, fine subangular blocky structure; friable: few fine roots; few worm casts: few iron stains; few, fine, black concretions: some material from the Ap horizon in worm channels; slightly acid; abrupt, smooth boundary. B21-7 to 17 inches, strong-brown (7.SYR 5/61 heavy silt loam; moderate, medium, subangular blocky structure: friable; few fine roots; few worm casts; some material from the A2 horizon in old root channels; few fine and medium concretions; medium acid: clear, smooth boundary. B22- 17 to 22 inches, yellowish-brown (IOYR 5/6) heavy silt loam with few, medium, distinct, light brownish-gray ( IOYR 6/2) mottles; moderate, fine and medium, subangular blocky structure: friable: few fine roots: common, fine and medium, black concretions: some material from the A2 horizon in old root channels: strongly acid: clear, wavy boundary. A’2x&Bx-22 to 28 inches, light yellowish-brown (IOYR 6/4) silt loam with many, medium and distinct, coarse, light brownish-gray ( IOYR 6/2) mottles; weak to moderate, fine and medium, angular and subangular blocky structure: friable; many, fine, medium and large, black concretions; few fine roots in cracks: few fine voids and vesicles; strongly acid; clear, wavey boundary. [Consistence of this horizon does not support use of the symbol x to designate fragipan character.] B’2xg-28 to 45 inches, mottled yellowish-brown (IOYR 5 / 6 ) , light-gray (IOYR 7/1), and pale-brown (IOYR 6 / 3 ) heavy silt loam: coarse prismatic structure that breaks to moderate angular and subangular blocky structure; firm and compact, hard when dry; common, fine and medium, black and brown concretions: common black and brown coatings on peds: few fine voids; few patchy clay films: light-gray (IOYR 7/2) silt coatings on peds and in cracks: strongly acid; gradual, wavy boundary. yellowish-brown (IOYR 5 / 6 ) silt loam with many, coarse, B’3x-45 to 60 inches faint, pale-brown (IOYR 6/3) mottles; coarse prismatic structure that breaks to weak, medium and coarse, subangular blocky structure: friable: few, fine, black and brown concretions: few black and brown coatings; medium acid. Thomas ( I 967) wrote in regard to Grenada soils that are moderately eroded: “At a depth

+,

278


of about 2 feet, these soils have a fragipan that slows the movement of water and restricts the growth of roots. Consequently, the subsoil is waterlogged during rainy periods, especially in winter and early in spring. During dry periods in summer, however, these soils are slightly droughty because the soil above the fragipan is thin.” The influence of the fragipan depends strongly on its depth. On steeper slopes erosion has been greater and the fragipan occurs at shallower depths. Recommendations for various engineering interpretations reflect the presence of the fragipan. About road construction, Thomas (1967) wrote: “In nearly level areas, side ditches should extend below the fragipan and the pavement should be at least 4 feet above the fragipan. In steeper areas, road cuts normally extend below the fragipan, but adequate underdrainage is needed where the construction changes from cut to fill. This underdrainage can be provided by excavating the fragipan and replacing it with a more permeable material.” MCBRIDESOIL The soil has a fragipan restricted to the lower eluvial horizon and is an example of sequence Ilh, Fig. 2. Parent material is glacial till. The description and comments are from the report on soils of lonia County, Michigan (Threlkeld and Alfred, 1967). Ap-0 to 8 inches, dark grayish-brown (10YR 4/2) sandy loam; weak, fine, granular structure: friable: moderately high organic-matter content; medium acid; abrupt, smooth boundary. Bir-8 to 12 inches, yellowish-brown (IOYR 5/4-5/6) sandy loam: weak, fine, granular structure; very friable: medium acid; clear, wavy boundary. A’21- 12 to 14 inches, very pale-brown (IOYR 7/4) light sandy loam: weak, thin, platy structure; slightly compact when moist; medium acid; abrupt, wavy boundary. A’22x- 14 to 17 inches, light-gray (IOYR 7/2) light sandy loam; moderately compact, brittle fragipan; weakly vesicular: medium acid; clear, wavy boundary. B’21t- 17 to 30 inches, strong-brown (7.5YR 5/6) sandy clay loam: light-gray (IOYR 7/1) coats on cleavage faces in the upper part of the horizon: moderate, coarse, subangular blocky structure: firm: very strongly acid; clear, wavy boundary. B‘22t-30 to 38 inches, strong-brown (7.5YR 5/6) heavy sandy loam; weak, coarse, subangular blocky structure: friable: very strongly acid; clear, wavy boundary. B3 -38 to 48 inches, strong-brown (7.5YR 5/6) sandy loam: weak, coarse, subangular blocky structure; friable: medium acid in upper part, grading to slightly acid in lower part; abrupt, irregular boundary. C-48 inches +, brown (7.5YR 5/4) sandy loam: massive: friable; calcareous. The soil is considered to have moderate or moderately slow permeability and moderately low available water. Threlkeld and Alfred ( 1967) wrote: “Permeability depends to a great extent on the thickness and development of the fragipan . . . .” The fragipan in this soil, probably because of its thinness, apparently has only a marginal influence on use and management. VOLUSIASOIL The soil has a fragipan in the B horizon directly beneath the upper eluvial horizon and is an example of sequence Ila, Fig. 2. Other soils in this series have a thin B horizon above the fragipan. Parent material is glacial till. The fragipan shows insufficient evidence of clay accumulation for recognition of an argillic horizon. The description and information on soil use are from the report on soils of Tomkins County, New York (Neeley, 1965).


279

Ap-0 to 8 inches, dark grayish-brown (IOYR 4/2) channery silt loam: weak, fine, crumb structure; friable; p H 4.8; many fine roots; few flagstones on surface; abrupt, smooth boundary; 6 to 9 inches thick. A2-8 to 14 inches, light olive-brown (2.5Y 5/4) and grayish-brown (2.5 Y S/2) channery silt loam; many, coarse, distinct, yellowish-brown ( IOY R S/6-5/8) mottles: weak, thin to medium, platy structure; friable to slightly firm: pH 5.0; few fine roots: abrupt, wavy boundary; 4 to 6 inches thick. B’xlg- 14 to 28 inches, olive-brown (2.5Y 4/41 channery silt loam; few, medium, faint, gray and brown mottles; weak, coarse prisms 10 to 20 inches across, coated with grayish-brown (2.SY 5/2) and olive-brown (2.5Y 4/4) silty material; prisms break into moderate, medium, subangular blocks when disturbed; very firm: pH 5.4: diffuse boundary; 10 to IS inches thick. - B’x2g-28 to 48 inches, olive-brown (2.5Y 4/4) and olive (5Y 5/3)channery silt loam; few, common, gray and brown mottles; weak, coarse prisms I % to 2 feet across coated with thin, grayish-brown (2.5Y 512) to gray (5Y5 / l ) silt: prisms break into moderate, medium and coarse, subangular blocks; very firm: pH 5.8 at 48 inches: diffuse boundary; 20 to 30 inches thick. C-48 inches +, olive (5Y 5/3) channery silt loam; moderate, medium blocks with gray (5Y 511) surfaces; firm; pH 5.8 to 6.2; vertical streaks border prisms; boundaries disappear below 48 inches. Neeley (1965) wrote: “The pan effectively stops downward movement of water. In sloping areas water seeps downslope through permeable layers above the fragipan . . . .” Wetness is the dominant limitation. The soils are best suited for forage crops that can tolerate wetness. Surface drainage is necessary for successful production of cultivated crops in most years. Trafficability is low for extended periods. Cut banks are subject to seep and slump. The soils are good sites for ponds. The fragipan is an important factor in road construction and maintenance.


The Quantitative Relationships between Plant Population and Crop Yield

R. W. Willey*and S. B. Heath University of Reading, Reading, Berkshire, England

Introduction ...................................................... 11. Relationships between Plant Density and Crop Yield A. Biological Relationships ................................... B. Yield/Density Equations ................................................... C . A Further Examination o ciprocal Equations 11 I. The Relationship between Plant Rectangularity and Crop ............... IV. The Variation in Yield of the Individual Plant ............ V . Conclusions ................................................ References .........................................................................................

I.

I.

Prrge 281

286 30 1 3 14 317 319 320

Introduction

I t may often be desirable for the agronomist to define the relationships between plant population and crop yield quantitatively. Probably the simplest reason for wishing to do this is to evaluate such characteristics as optimum population and maximum yield. This can be a useful end in itself, and it can also facilitate comparisom between different cropping situations. The latter aspect is particularly useful when the factors being examined interact with population, for comparisons can clearly be misleading if they are not made between comparable points on the population response curve. Putter et al. ( 1966) have pointed out that even comparisons between calculated parameters which have little or no biological meaning can be of value, for these may still help to pinpoint essential differences. It is more usual, however, that the agronomist wishes to define the relationships between plant population and crop yield so that in any future situation he can predict yield/population curves easily and accurately from the minimum of data. For this purpose at.least, it is desirable that an equation defining these relationships should not be *Presenr address: Makerere University College, Kampala, Uganda.

28 I

282

R. W. WILLEY AND S. 8 . HEATH

merely a mathematical empiricism but should be so far as possible an accurate description of the biological processes of competition that are involved. Some biological validity of this nature clearly provides greater justification for the use of an equation and at the same time probably ensures a wider applicability. However, any attempt to derive such an equation inevitably involves some degree of compromise, for the very nature of the problem is to describe extremely complex biological processes by a mathematical function sufficiently simple to be of use to an agronomist. The object of this review is to examine some of these equations and to discuss how far they fulfill the agronomist’s needs. But first some general points must be discussed. It is important to realize that plant population should be defined not only in terms of the number of plants per unit area (i.e., plan? density) but also in terms of the arrangement of these plants on the ground (spatial arrangement or plant rectangularity). Few workers have distinguished between these two factors and many plant population experiments have in effect studied the combined effects of both; this is certainly the case where different populations have been studied at a constant row width. Since most workers seem to have assumed that the only effect of population is one of density, it would seem logical to try to define the relationship between crop yield and density first, and to incorporate the effects of rectangularity subsequently. For this reason, the rectangularity factor is ignored when the equations in-Section 11 are discussed, even though this factor has not been eliminated in many of the experiments that will be quoted. A possible means of incorporating the rectangularity factor into an equation is discussed in Section 111. Crop yield itself may also require further definition, for in some crops the grower is as much concerned with the yield of the individual plant as with yield per unit area. Many population experiments have examined yield per plant, but it has almost invariably been in terms of mean yield per plant. Yet the degree of variation in yield of the individual plants is a factor that can determine the total yield of plants within a given size grading. Some studies in which an attempt has been made to estimate this variation are briefly discussed in Section IV. Finally, it is pertinent to point out that it is not always easy to decide upon the correct unit of plant population. This was emphasized by Holliday ( 1 960b), who suggested that it should be the basic independent plant unit, whether this was a tiller in a grass crop or perhaps an individual stem in a potato crop. Accurate identification of these population units could be of particular importance in quantitative studies, for without this it may be even more difficult to produce reliable and meaningful equations applicable to a wide range of crops.

PLANT POPULATION A N D CROP YIELD

It.

283

Relationships between Plant Density and Crop Yield

BIOLOGICALRELATIONSHIPS Before examining the different yield/density equations, it is first necessary to decide what are the basic biological relationships that these equations are attempting to describe. The only real attempt to classify these relationships seems to have been that of Holliday ( 1 960b). He suggested that there were essentially two relationships: an asymptotic one where, with increase in density, yield rises to a maximum and is then relatively constant at high densities: and apavabolic one where yield rises to a maximum but then declines at high densities. For the purposes of this present review this suggested classification is a useful one to adopt, but no exact mathematical description is inferred by the terms asymptotic and parabolic. It may well be argued, of course, that these two suggested relationships are merely different degrees of expression of a single relationship. However, in the present context the important fact to realize is that the two situations exist, for it may often happen that a given yield/density equation can describe either an asymptotic situation or a parabolic situation, but not both. Also, since the mathematical description of these two situations can be quite different, it is mathematically convenient to separate them. I t is not within the scope of this review to discuss in detail which specific crops or types of yield may conform to the different biological relationships; nor is it possible to consider how these relationships may be affected by the level of supply of different growth factors. At the same time it may be of use to give an indication of some of the cropping situations in which the different relationships can occur and to illustrate the shapes of yield/density curves which need description.* A.

1. The Asymptotic Relationship Holliday ( I 960b) suggested that total crop dry matter conformed to this relationship, but more recently several workers (de Wit, 1959; Bleasdale, 1966a; Bruinsma, 1966; Campbell and Viets, 1967; Farazdaghi, 1968) have shown that at high densities decreases in this form of yield can occur. Despite these exceptions it is probably reasonable to assume that, for practical purposes, total dry matter yield often conforms to an essentially asymptotic relationship. This situation is illustrated in *The notation used to express the relationships is as follows: y = yield per unit area; p = plant density; s = space available per plant (s = l/p): d , = distance between plants within a row (intrarow spacing): d2 = distance between rows (interrow spacing); w = yield per plant; W = maximum yield attainable by a plant; w, = yield o f a plant part. Any variable or constant with the subscript p . as in wp above, refers to a plant part as opposed to a total plant.

284

R. W. WILLEY A N D S. B. HEATH

Figs. IA and I B by some data for fodder rape (Holliday, 1960a) and for Wimmera ryegrass and subterranean clover (Donald, 195 l), all of which are asymptotic to particularly high densities. Holliday (1 960b) also suggested that those forms of yield which constituted a vegetative part of the crop conformed to an asymptotic relationship. Notable exceptions may occur (see Section II,A,2) but again it is reasonable to assume that such forms of yield often are asymptotic. This situation is illustrated by some data for potato tubers (Saunt, 1960) and root yield of long beet (Warne, 1951) in Figs. 1C and 1 D, respectively.

2 . The Parabolic Relationship Holliday (1960b) suggested that reproductive forms of yield (i.e., grains and seeds) conformed to a parabolic relationship, and the examples

;lk2oF P

P

Total

-7%:

Y 10

Y

OO 5

1

2

3

OO

P

2

4

6

P

FIG. 1. Examples of the asymptotic yield/density relationship. (A) Total dry matter of Essex Giant rape; y = tons/acre, p = 106plants/acre (Holliday, 1960a). (B) Total dry matter of Wimmera ryegrass and subterranean clover; y = g./sq. lk., p = lo2plants/sq. Ik. (Donald, 195 1). ( C ) Fresh weight of potato tubers; y = tonslacre, p = lo4 parent tuberslacre (Saunt, 1960). (D) Fresh weight yields of long beet; y = poundslplot, p = plantslfoot of row (Warne, 195 1).


285

given in Figs. 2A and 2 B for grain yield of maize (Lang et al., 1956) and barley (Willey, 1965) certainly indicate that this can be so. The maize data are of particular interest because this crop usually displays a very distinct decline in yield at high densities, and as such it represents one of the more extreme forms of the parabolic relationship. The barley data, in which the density reaches a particularly high value, are also of interest because they illustrate a point seldom evident in experimental data,

C 30.

P

P

FIG.2. Examples of the parabolic yieldldensity relationships. (A) Mean grain yield of maize for all hybrids, grown at a low level -( ), medium level (----), and a high level ( - - -) of nitrogen: y = bushelslacre, p = lo3 plantslacre (Lang cf al., 1956). (B) Grain yield of barley grown with 0 -( ), 30 (----), and 60 ( - - -) units of nitrogen: y = cwtlacre, p = lo6 plantslacre (Willey, 1965). (C) Root dry weight of globe red beet; y = 10’ kg./acre, p = lo4 plantslacre (unpublished Reading data). (D) Parsnips var. AVONRESISTOR, total fresh weight yield ( - - -), graded yield > 1.5 inches in diameter (----), graded yield > 2.0 inches in diameter (-): y = tonslacre, p = plantslsq. ft. (Bleasdale and Thompson, 1966).

286

R. W. WILLEY AND S. B. HEATH

namely that the parabolic relationship must at some stage begin to flatten off along the density axis. As mentioned in Section II,A, 1, certain forms of vegetative yield may also be parabolic. A notable instance of this seems to be the root yield of globe red beet, and some example data for this crop are given in Fig. 2C. Yet a further situation can exist where yield is parabolic, and this is where yield constitutes only those plants, or parts of plants, that fall within certain size limits, i.e., where some form of “grading” is practiced. Figure 2D illustrates this situation with some parsnip data of Bleasdale and Thompson ( I 966). It can be seen that in this particular instance total yield of roots is asymptotic, but grading produces a parabolic relationship that becomes more acute as the severity of grading is increased. This situation is of considerable importance in many crops. However, it must be emphasized that “graded” yield cannot be regarded as a biological form of yield in the same sense as those forms discussed above. For this reason, the description of this particular relationship may have to remain more empirical than that of other relationships.

B. YIELD/DENSITY EQUATIONS Section II,A indicated the general form of the biological relationships that exist between crop yield and plant density. The object of this section is to describe the different mathematical equations that have been proposed to define these relationships. Some of these yield/density equations propose a relatively simple mathematical relationship directly between yield per unit area and density, but the majority propose a basic relationship between mean yield per plant and density. The general shape of this latter relationship is illustrated in Fig. 3 for both the asymptotic and parabolic yield/density situations.

I. Polymoniul Equations One of the simplest approaches to the description of yield/density relationships has been the use of two polymonial equations applied directly to the relationship between yield per unit area and density. These have been used largely as a convenient means of smoothing experimental data: they have not been seriously proposed as general yield/density equations, and little or no biological validity has been claimed for them. In these respects they are not of any major importance in the present review, but a brief description of their scope and limitations serves as a useful introduction to the use of yield/density equations, particularly where biological validity is lacking.

287


-0.8 0.8

- 0.6 0.6

100

- 0.4Y

W

W

Y

0.4

- 0.2

50

0.2

.-0

0

4

12

P

P

20

FIG.3. The relationship between yield per plant (w)and plant population ( p ) in an asymptotic (A) and a parabolic (B)yield/density situation. (A) Total dry matter of Essex tonlplant, p = loo plantslacre (HolliGiant rape, 1952 experiment; y = tonslacre, w = day, 1960a). (B) Grain yield of maize hybrid WF9 x 38-1 I at medium N ; y = bushelslacre, w= bushel/plant, p = lo3plantslacre (Lang ef a/., 1956).

Hudson ( 1 94 1) attempted to describe the relationship between grain yield and seed rate of winter wheat with a simple quadratic expression: y

=a

+ b p + cp’

(1)

where a, b, and c are constants, c being negative. The general shape of the yield/density curve described by Eq. ( I ) is illustrated in Fig. 4, where

20.

\\ \ I \ \

FIG.4. The quadratic equation (Eq. 1) (-----) and the square root equation (Eq. 2) ) fitted to grain yield of maize hybrid HY2 X OH7 at low N ; y = bushelslacre, p = l o Tplantslacre (Lang et a/., 1956).

-(

288


it is fitted to some maize data of Lang et al. ( 1 956); it is essentially a curve which is symmetrical about a maximum value of yield. Although the degree of curvature may obviously vary, this basic shape offers little flexibility in fitting yield/density relationships. It is clearly not suitable for fitting a truly asymptotic situation, and in a parabolic situation it is likely to give a good fit only where the yield/density curve is reasonably symmetrical. But even in this latter situation, the accuracy of this equation is probably restricted to a relatively narrow range of densities around the point of maximum yield. This is because of the unrealistic implications of the equation at both high and low densities. A t high densities it implies that yield must drop sharply down to zero (see extrapolation in Fig. 4), whereas at the other extreme it implies that a t zero density yield has a value, a (which in practice may turn out to be either positive o r negative). The former implication is a serious limitation on the use of this equation at high densities. The latter implication could be only a minor disadvantage if the value of u was low; in any case, if an accurate fit at low densities was particularly desirable, the omission of a from the equation would ensure that the curve passed through the origin. The disadvantage of the symmetrical nature of the quadratic curve was avoided by Sharpe and Dent (1968) by using a square root form of pol ymonial Eq. 2):

where a, b and c are constants, b being negative. This equation again gives rise to a curve where yield rises to a maximum value and then decreases at higher densities, so it still cannot describe an asymptotic situation. Compared with the quadratic, however, it can follow a slightly more gradual decline in yield at high densities, although this is accompanied by a rather steeper increase at the low densities (see Fig. 4). It still implies that a t zero density yield has a finite value a, and that at the other end of the scale yield declines to zero, although admittedly at a rather higher density than with the quadratic. The apparent lack of any biological validity must also impose limits on the use of these two equations. For example, it would seem unwise to use them where data were not sufficiently comprehensive to give a good initial indication of the general shape of any particular yield/density situation. Also there would seem little justification for using them to extrapolate data. Such extrapolation was carried out by Keller and Li ( 1949), who used the quadratic to estimate optimum density and maximum yield of some hop data, and it is of significance that when Wilcox ( 1 950), with

289


little more justification, extrapolated the same data using the Mitscherlich equation he obtained substantially different values.

2. Exponential Equations Duncan ( I 958), when reviewing experimental data on maize, proposed an exponential equation to describe the relationship between grain yield and density. He derived this by fitting a linear regression of the logarithm of yield per plant on density. The basic relationship was therefore: log w

= log

+ bp

K

(3)

or y = p K 10bp

where K is a constant and 6, negative, is the slope of the regression line (see Fig. 5A). Carmer and Jackobs (1965) used this equation in a slightly different but analogous form:

where A and K are constants. The yieldldensity curve which this type of equation produces is comparable to the polynomials in as much as yield must rise to a maximum value and then decrease at higher densities. It can give a good fit to parabolic yield/density data, but even though it is much more flexible than the polynomials at high densities, it still cannot

100

Y

50

0

P

P

FIG. 5 . The exponential equation (Eq. 3) of Duncan (1958) fitted to a parabolic (A) and an asymptotic (B) yield/density relationship. (A) The regression line of log w against p, and the fitted yield/density curve for grain yield of maize, mean of all hybrids at medium N ; y = bushelslacre, p = lo3 plantslacre, w = bushel (Lang et al., 1956). (B) The fitted yield/density curve for total dry matter of Essex Giant rape, 1952 data; y = tonslacre, p = lofiplantslacre (Holliday, 1960a).

2 90


give a useful practical fit to data that are asymptotic. This is illustrated in Fig. 5 , where it is fitted to some parabolic maize data of Lang et al. ( 1956) and some asymptotic rape data of Holliday ( 1960a). Apart from greater flexibility, this exponential equation has further advantages over the polynomials. At high densities the yield curve does not cut the density axis but, more realistically, only gradually approaches it. Also, this curve now passes through the origin. However, as pointed out by Duncan there may still be a defect at low densities for, as estrapolation of the regression line in Fig. 5A indicates, the equation cannot allow for a leveling off in yield per plant at densities too low for competition to occur. But this is a common defect of yield/density equations, and it is discussed later when considering Holliday’s reciprocal equations (Section 11, B, 5 , b). Duncan also pointed out that, since his equation was based on a linear regression, it was possible to construct the whole yield/density curve from the yields at only two densities. He therefore suggested that in the maize crop the examination of factors that interacted with density might usefully be carried out at two densities; the use of his equation would then allow comparison of the factors at their calculated points of optimum density and maximum yield. This technique can, of course, be used with any yield/density equation derived from some linear regression on density, and its practical potential makes it of considerable interest. Its application calls for some caution, however, for a prerequisite for its use must be a reasonable assurance that the equation used is an accurate description of the particular yield/density relationship that is under study. Duncan’s justification for suggesting its use in the maize crop was the fact that his equation gave a good practical fit to the data he reviewed. This seems reasonable, but in general a better justification would seem to be the knowledge that an equation used in this way had a good deal of biological validity and was not just an empirical one. This could be particularly important, because it was pointed out by Duncan that the farther apart the two densities, the more accurately the regression line would be determined. While this may be mathematically sound, it would seem safer in practice to include a third intermediate density so that the point of calculated maximum yield is not too far from an experimental treatment. 3 . Mitscherlich Equation Mitscherlich proposed a law of physiological relations by which he described the relationship between the yield of a plant and the supply of an essential growth factor, all other factors being held constant. He assumed that as the supply of such a factor increased, yield per plant


29 1

would approach a maximum value, and at any given point the response would depend on how far the plant yield was below this maximum. This can be expressed: - --

dw df

(W-w)c

where f is the level of supply of the factor and c is a constant. On integration this gives Eq. (4):

Mitscherlich termed c his “Wirkungsfactor” and claimed that it was constant for a given growth factor and independent of other conditions. Later, Mitscherlich ( I9 19) suggested that his equation might be applied more generally to the relationship between “space” and plant growth and so serve as a yield/density equation. Thus, substituting space, s, for the growth factor,f, Eq. (4) can be rewritten:

where K is now a general “space” constant or factor. It is evident from the basic assumption about the nature of the plant’s response that this yield/density equation describes an asymptotic situation, but not a parabolic one. Kira et af. ( 1954) examined the constancy of the space factor K . Using the yield/density data of Donald ( I95 I ) for subterranean clover, they were able to define the asymptotic value of yield per plant, W. From this value, and from mean yields per plant at the other densities, they calculated a range of K values (Table I). I t is apparent that the values decreased with increase in the space available per plant and could not be regarded as constant. Kira et a f . ( I 954) obtained similar changes in K values from yield/density data for azuki bean (Phaseofuschrysanrhus), although the trend was not so clear. This change in the value of K could have interesting agronomic implications, for it may perhaps suggest that a change in density may not only change the space available to a plant, but might also bring about some change in the environment-for example, an effect on rooting depth. However, as far as the practical use of the Mitscherlich equation is concerned, a change in K is clearly undesirable, and the value of this expression as a yield/density equation becomes questionable. Kira et al. ( 1954)

292

R. W . WILLEY AND S. B. HEATH

TABLE I APPLIED TO THE MITSCHERLICH’S FORMULA

RESULTSOF AN EXPERIMENT DONALD (195 I)a

WITH SUBTERRANEAN CLOVER OF

61 days from sowing

Density (plants/sq. link) 0.25

1 .oo

5.95-5.93 15.9-1 6. I3 60.6-62.6 241-248 1247-1393

Dry weight per plant (g.1

K

15.6 15.5 15.6 15.8 (W) 14.2 0.154 13.9 0.563 10.6 1.66

I 3 I days from sowing


182 days from sowing

K

528 562 (W) 386 0.0073 0.0178 364 0.02 13 153 0.0370 13 16 0.0430


K

34,080 (W) 0.00020 2 1,280 0.0006 1 4,560 0.0009 1 2,020 0.00 106 0.001 17 600 160 0.00125 29 0.00123

“After Kira et al. (1954).

did in fact point out that they could stabilize K by arbitrarily reducing the value of W, but in this event the equation must lose much of its biological foundation. Despite these criticisms of the Mitscherlich equation, the basic concept of an asymptotic yield per plant is of considerable interest. This at least provides a satisfactory theoretical description of the yield/density curve at very low densities where there is no competition. As several workers have pointed out (Duncan, 1958; Kira et al., 1954; Shinozaki and Kira, 1956; Holliday, 1960a), yield/density curves are usually unable to provide such a description and their validity at low densities is doubtful. It is also of interest that Goodall ( 1 960), examining some mangold data, and Nelder (1963), commenting on some lucerne data of Jarvis (1962), both found that the Mitscherlich equation could give as good a fit as other equations. On the other hand, as would be expected from the results of the examination of his K values by Kira et al. (1954), Donald (195 1) did not obtain a good fit to his data using the Mitscherlich equation.

4 . Geometric Equations Geometric equations were put forward by Warne ( 1 95 1 ) and Kira et al. ( 1953) to describe certain yield/density relationships; the latter workers used the term “power” equation. Essentially this type of equation assumes a linear relationship between the logarithm of yield per plant and the logarithm of density.


293

Warne (195 I ) was studying the effect of density on the yield of root vegetables (beet, parsnips, and carrots), and he proposed a linear relationship between the logarithm of root yield per plant and the logarithm of distance between plants in the row where row width was constant. Since the row width was constant Warne’s equation can be written in the form log w

= log A

+ b log ( s )

or w = A (S)!’

where A and B are constants and s is the space available per plant. On a yield per unit area basis, and including density rather than space, Warne’s equation becomes y = A (p)’-”

Kira et a f . ( 1953) obtained a linear relationship between the logarithm of total yield per plant and the logarithm of density in a soybean experiment. The form of equation they proposed was log w

+ a log p = log K

or wp“ =K

(7)

where a and K are constants, a being termed the competition-density index. This equation is exactly analogous to that of Warne-the a and K of Kira el al. being comparable with Warne’s h and A , respectively. Strictly speaking, the only type of yield/density curve which this equation can describe is one where yield is still rising at the highest density. Such curves are illustrated in Fig. 6, where Kira et af.’sequation is fitted to some of Donald’s data for different harvests of subterranean clover (Donald, 195 I). I t can be seen that as the yield/density curve approaches an asymptotic shape with the passage of time (Fig. 6B), the slope of the regression line becomes steeper and the value of a (the competitiondensity index) increases and approaches a value of 1 (Fig. 6A). However, if an asymptotic shape is reached, then, to describe constant yield at the high densities accurately, the competition-density index has to take the value of 1, and this then implies that yield is constant at all densities, i.e., the yield/density curve becomes a straight horizontal line (with value K ) . Or, from Eq. (7): w p = K constant = Y

(8)

2 94


[The Japanese workers referred to Eq. (8) as the law of constant final yield (Hozumi et al., 1956)l. It is also of interest that a competitiondensity index greater than 1 implies that yield decreases with all in-

P

P

FIG.6. The geometric (“power”) equation (Eq. 7) of Kira ef al. (1953) fitted to the total dry matter yield/density data of subterranean clover (Donald, 195 1 ) at different numbers of days after sowing (0, 6 I , 13 1 , 182); (A) The regression lines of w against p ; dashed lines indicate densities at which there is no competition; w = g., p = plantslsq. Ik. (B) The fitted yield per area/density curves; y = g./sq. Ik.

creases in density. Thus, on theoretical grounds neither the truly asymptotic nor the parabolic yield/density situation can be described. In some circumstances it may be possible to obtain a reasonably satisfactory practical fit to the former situation with a value of a fractionally less than I , although this was not the case with Donald’s data in Fig. 6. Both Warne (1951) and Kira et al. (1953) emphasized the possible significance of their respective power constants b and a. in Eqs. (6) and (7). Warne said that the higher the value of the constant, the more the plant was dependent on the space available to it; whereas Kira el al. (1953) interpreted an increase in value of the constant as indicating a more thorough utilization of the space available to the plant. From the agronomist’s viewpoint, the significance of these constants is most easily appreciated by considering the succession of yield/density curves already referred to in Fig. 6. It can be seen that the greater the value of a the fkrther the yield/density curve has progressed from its initial competition-free situation. This progression is associated with a greater degree of competition, a greater degree of curvature in the relationship and, as Kira et al. said, a more efficient utilization of space. On this basis the


295

agronomist can readily understand and compare relative differences in the values of these power constants that may have arisen due to differences in time, growing conditions, or plant species. A further criticism of the geometric equation, first raised by the Japanese workers themselves (Shinozaki and Kira, 1956), is the failure of the equation to describe the leveling off in yield per plant at densities too low for competition to occur (indicated by the broken horizontal lines in Fig. 6A). This was a criticism of the exponential equations (Section 11, B, 2), and it also applies to the reciprocal equations; its importance is discussed in Section 11, B, 5 , b.

5 . Reciprocal Equations The reciprocal equations are here regarded as those based on some mathematical relationship between the reciprocal of mean yield per plant and density. These represent a very important group of equations, which at present seems to offer the best possibilities of being able to describe yield/density relationships accurately and meaningfully. Since these equations are to be discussed more fully in Section I1 C, the present section serves only to describe their general characteristics. a. Shinozaki and Kira. The first to propose a reciprocal equation were the Japanese workers Shinozaki and Kira (1 956). This equation was derived from a simple logistic growth curve and the law of constant final yield (see Section 11, C, 1, a). It was proposed by the Japanese workers because they had found that their geometric equation could not satisfactorily fit an asymptotic yield/density situation (Section 11, B, 4). The form of equation derived was

'

- _-a+bbp W

(9)

where a and b are constants. This relationship assumes a linear relationship between the reciprocal of yield per plant and density. Shinozaki and Kira (1956) tested this relationship in a number of asymptotic yield/ density situations and found that it appeared to hold true in practice. However, this form of reciprocal equation cannot describe a parabolic yield/density situation. b. Holliday. Holliday (1960b) was the first worker to propose that both an asymptotic and a parabolic form of yield/density relationship existed. Although it was pointed out earlier that his suggestion of the asymptotic form applying to total or vegetative yield and the parabolic form applying to reproductive yield may not be entirely valid, the two

296


equations which he proposed to describe the two forms of relationship are still of considerable interest. Holliday ( 1960a) arrived at his asymptotic equation from yield/density studies on rape, kale, potatoes, and perennial ryegrass in which he observed that the reciprocal of yield per plant was linear with density:

- ‘_- a + b p W

where a and b are constants. Thus, this basic form of Holliday’s asymptotic equation is identical to that of Shinozaki and Kira (1956), and it is of some interest that Holliday produced his largely from experimental observations whereas the Japanese workers derived theirs largely from mathematical considerations. Like other workers, Holliday (1960a) pointed out that his equation could not allow for a constant yield per plant at those densities where there is no competition. He therefore termed the implied yield per plant at zero density the “apparent maximum” yield per plant and designated this A , which is equal to Ila (see Fig. 7). He also suggested that a modified form of equation beginning at the density where competition first starts might be more accurate. Thus, if this density is n, and p-n = m (see Fig. 7), Holliday’s modification can in effect be written 1

--a’+bm

W

The reciprocal of a’ would now equal the true maximum yield per plant (A’). However, Holliday admitted that, although more accurate biologically, this equation might be of limited use in practice because in any given situation A ’ would have to be determined experimentally. Shinozaki and Kira ( 1 956) suggested a rather different way of allowing for a constant yield per plant at densities free of competition. They proposed the inclusion of a factor 6 in their basic relationship, so that

- l_- a + b ( p + G ) W

They suggested that an appropriate value of 6 should be chosen so that at high densities 6 would be negligible compared with p and could therefore be disregarded, whereas at low densities 6 would come into effect and

297


yield per plant would level off to a relatively constant and more realistic value. However, this factor 6 seems to have no biological meaning and is included purely to try to improve the equation’s goodness of fit at low densities. Moreover, since its value must depend on the maximum yield per plant without competition, then, like Holliday’s A ’ , it presumably could be satisfactorily determined only by experimental means in any given situation.

A’

+n+-

m

b

4

P

b

Plant d e n s i t y

FIG. 7. Diagram of yield per plant ( w ) and reciprocal of yield per plant ( I / w ) plotted against density. illustrating the derivation of the asymptotic equation (Eq. 10) of Holliday (1960a) and its modification to allow for the absence of competiton at low densities (Eq. 1 I ) ; n is the density at which competiton starts, A is the “apparent maximum” yield per plant, A ’ is the true maximum yield per plant.

Holliday ( 1960b) proposed that the parabolic yield/density situation, where the relationship between the reciprocal of yield per plant and density is no longer linear, could be described by the quadratic expression 1

-=a W

+ b p + cpz

(12)

where a, 6, and c are constants. Holliday pointed out that the use of the quadratic was empirical, but this form of quadratic equation is a great improvement on the simple quadratic which Hudson ( I94 I ) applied directly to yield per area against density (Eq. 1). I t produces a very flex-

298


ible parabolic yield/density curve which is not symmetrical about its point of maximum yield and which flattens off realistically at high densities. Some examples of this equation fitted to yield/density curves are given in Section 11, C, 2. c. De Wit. From studies of mixtures of barley and oats, de Wit and Ennik (1 958) derived a single species yield/density equation based on a linear relationship between the reciprocal of yield per unit area and row width (where distance between plants in the row was constant). This took the form 1 Y

-=

a

+ bdz

(13)

where a and b are constants, and d z is the row width. Expressed in terms of the reciprocal of yield per plant this equation becomes: -' -b W

+ap

which is exactly analogous to the reciprocal equation (Eq. 9) of Shinozaki and Kira (1956) and the asymptotic equation (Eq. 10)of Holliday (1 960a). Later, de Wit ( 1 960) proposed a slightly different version of this equation based on a linear relationship between the reciprocal of yield per unit area and space available per plant. This is represented diagrammatically in Fig. 8, and the basic relationship is:

or

where s is the space available per plant and 1/P and Q are the points where the regression line cuts the I/y and s axes, respectively. P is therefor equal to the asymptote of yield per area. De Wit's later equation is again more readily compared with the other reciprocal equations if it is transformed to show the relationship between the reciprocal of yield per plant and density as follows: 1

1 PQ

-= - + - p

w

1 P


299

It can be seen that this equation differs slightly from those of Shinozaki and Kira ( 1956) (Eq. 9) and Holliday (1 960a) (Eq. 101, in that the value of I/w at zero density IIPQ is now defined by two constants instead of one, and one of these constants, P , is the asymptote of yield per area. The possible significance of this is discussed in Section 11, C , I , b.

Space per plant

(5)

-

FIG.8. Diagram of the reciprocal of yield per area (I/y) plotted against space available per plant (s) to illustrate the derivation of the reciprocal equation of de Wit (1960). 1/P and Q are the points where the regression line cuts the I/y and s axes, respectively.

Like Shinozaki and Kira’s equation, de Wit’s can describe only an asymptotic yield/density situation; de Wit does not appear to have proposed any modification of his equation to describe a parabolic situation. d. Bleasdale and Nelder. Bleasdale and Nelder (1960) proposed a reciprocal equation which they derived from a generalization of the logistic growth curve described by Richards (1959). This was originally proposed in the following form: 1 -e a + bp’ W

where a, 6, and 0 are constants. This equation describes an asymptotic yield/density situation, but Bleasdale and Nelder pointed out that if the power on p exceeded the power on w the equation could also describe a parabolic situation. Equation 16 was therefore restated as 1 -

@ =

W

where 4 is a constant.

a

+ bp*

300


This is the form of equation found in their subsequent references (Bleasdale, 1966b; Bleasdale and Thompson, 1966; Bleasdale, 1967); where yield is asymptotic, 0 = 4 ; and where it is parabolic, 8 < 4. In the last situation the yield curve flattens off along the density axis similarly to Holliday’s parabolic equation (Eq. 12). The use of Eq. (1 7) was advocated by Bleasdale and Nelder (1960) because they disagreed with the division by Holliday ( I 960b) of yield/density relationships into vegetative and reproductive types. They preferred the use of a single generalized yield/density equation for all situations. However Bleasdale (1 966b) and Bleasdale and Thompson ( 1966) have stated that although there are theoretical reasons for allowing 4 to have a value other than unity, it is the ratio of 8 to 4 which is important. Also, they considered that data are rarely accurate enough to enable specific values of both 0 and 4 to be determined. They therefore suggested that in practice it is sufficient to take the value of 4 as unity. Thus Eq. (17) was restated as 1 _ W e - a + bP

(18)

Equation (1 8) is subsequently referred to as Bleasdale’s simplified equation. e. Farazdughi and Harris. Farazdaghi and Harris (1968) have recently derived a yield/density equation from the same logistic growth curve used by Shinozaki and Kira (1956). However, these workers stressed that the law of constant final yield for total crop dry matter may not always hold true. [Farazdaghi (1 968) has even shown that total dry matter in the sugar beet crop can be either asymptotic or parabolic depending on the environment.] Thus they modified the law of constant final yield to

w pY= K

(19)

and derived an equation:

This equation can describe either an asymptotic or a parabolic yield/ density situation, depending on the value of y ; in the former case y = 1, and in the latter case y > 1.


301

C. A FURTHER EXAMINATION OF THE RECIPROCAL EQUATIONS In Section I1 B the general characteristics of the different yield/density equations were discussed. The object of this section is to examine the reciprocal equations in more detail. These equations are singled out for this further examination for three main reasons: first, they are the only type of equation that can realistically describe both the asymptotic and the parabolic yield/density situations, either by means of different forms of equation (Holliday, 1960b) or by a single generalized equation (Bleasdale and Nelder, 1960); second, a good deal of biological meaningfulness has been claimed for them; and third, they have probably been used more than any other type of equation in recent years. 1 . The Biological Basis of the Reciprocal Equations a . Biological Derivation. ( 1 ) Shinozaki and Kira (I 9513, Bleasdale and Nelder ( 1 960), and Farazdaghi and Harris ( 1 968). Shinozaki and Kira derived their yield/density equations from the following assumptions on the growth of a plant: (i) The growth of a plant can be described by a simple logistic growth curve.

W =

W

1

+ ke-At

where w is the weight of the plant at time t, A is the coefficient of growth, and k is the integration constant. Both W and A are assumed constant independently to time t. (ii) A in the above equation is independent of density. (iii) Final yield per unit area is constant and independent of density after the law of constant final yield (Eq. 8) of Hozumi et al. ( 1956). On combining Eqs. (8) and (21) and determining the value of k when there is no competition at time zero, when the weight per plant is wo, the reciprocal equation can be derived 1 = W

a+bp

where a = e-A'/w,and b = ( 1 - e-")/Y.

302


Shinozaki and Kira (1 956) go on to show that their reciprocal equation can be derived from more general growth curves where A and W are not independent of time, which would be an obvious objection. It was seen earlier that Farazdaghi and Harris (1968) used the same growth function as Shinozaki and Kira but did not assume the law of constant final yield (Section 11, B, 5 , e). On the other hand, Bleasdale and Nelder (1960) stated that they derived their equation from a generalization of the logistic growth curve given by Richards (1959) “with ahalogous arguments to those of Shinozaki and Kira.” The main interest in the biological derivation of these particular reciprocal equations lies in their application to the yield/density relationship of a plant part. Kira et al. (1 956) had observed that the weight of a plant part could be related to the weight of the whole plant in the following way

or log w, = log k

+ h log w

i.e., the logarithm of the weight of the plant part has a linear relationship with the logarithm of the total weight: log K is the intercept and h is the slope of the regression line. This is Huxley’s law of relative growth, or law of allometry. Bleasdale (1967) pointed out that in its original context this law was applied to the relationship between plant part and total where plant size increased with age, whereas in the present context it is used to describe the relative changes brought about by density. From this allometric relationship, Shinozaki and Kira (1956) modified their original equation (Eq. 9) to describe the yield/density relationship of a plant part

This equation is very similar to Bleasdale’s simplified equation (Eq. 18). It can describe a parabolic yield/density curve for a plant part, although, because of Shinozaki and Kira’s initial assumptions of the law of constant final yield, this equation still assumes that the total yield/density curve is asymptotic. However, Shinozaki and Kira do not appear to have tested this equation in practice.


303

Bleasdale ( 1966a, 1967) also made use of the allometric relationship, which he stated in a slightly different form as w = Kw;

(24)

Combining Eq. (24) with his simplified equation (Eq. 18) he was able to derive a similar equation to the latter to apply to a plant part

I

-=aal+

W 0,

brp

where 0, = 0 A . Thus in the situation where 0 = I , where the total yield/ density curve is asymptotic, 0,= A . In this situation 0, can therefore be estimated directly from the allometric relationship as the slope of the regression line. Bleasdale pointed out that this could allow the construction of the whole yield/density curve from only two densities: two densities would enable 0, to be estimated from the allometric relationship, and once this constant was determined Bleasdale’s simplified equation for a plant part could also be fitted on two densities. (The dangers of fitting a yield/density curve on two densities were discussed earlier under Section 11, B, 2.) The estimation of 0, in this way also has another advantage, in practice perhaps a more useful one. As will be seen later, when fitting any reciprocal equation which contains a power it may be difficult to obtain an accurate estimate of this power. Thus, if in this instance 0, can be estimated from the allometric relationship when total yield per area is asymptotic, Eq. (25) can then be fitted more accurately. Although Farazdaghi and Harris ( 1 968) suggested that their basic equation (Eq. 20) could be used directly for a plant part, they also derived a more meaningful equation using the allometric relationship (Eq. 24). This took the form

They said that this described the way plant density affected the distribution of dry matter into plant parts. It is of interest that this equation is very similar to one of the original ones of Bleasdale and Nelder (1 960) (Eq. 17), which had proved difficult to fit in practice. But Farazdaghi and Harris (1968) pointed out that if y was estimated from their basic equation (Eq. 20) by fitting the total yield/density curve, this would then allow

304


the more complicated equation (Eq. 26) to be fitted, since only one power would have to be estimated from the regression analysis. However, it is noteworthy that A could also be determined from the allometric relationship (whether the total yield/density curve is asymptotic or not), although y still has to be determined by fitting either Eq. (20) or Eq. (26). Also, it is of interest that the only situation in which the approach of Farazdaghi and Harris ( I 968) does not entail determining at least one power by directly fitting one of their yield/density equations is again when total yield is asymptotic, for in this case y = 1 and A is obtainable from the allometric relationship. It must be emphasized, however, that to make use of the allometric relationship, or to estimate y in Eq. (20), it is necessary to have data for total plant weight as well as for plant part. In practice this may present difficulties. For example, in the cereal crop it may be difficult to obtain comparable estimates of grain yield and total dry matter since the latter may have declined from its maximum value before the maximum value of the former is achieved. A similar situation exists in the potato crop when total dry matter and final yield of tubers are considered. Also, where the plant part is not present throughout the whole life of the crop, the use of the allometric relationship may require further consideration. In view of these difficulties, the agronomist may frequently find that in practice he is not in a position to predetermine any power which he can substitute in the equations specially derived for fitting plant part data (Eqs. 25 and 26). He must then fall back on the use of either Bleasdale’s simplified equation (Eq. 18) or Farazdaghi and Harris’s basic equation (Eq. 20) and apply these directly to his plant part data. In this situation there seems little to choose between these two equations; they both involve fitting one power and they describe very similar yield/density curves. (2) De Wit (1960). De Wit’s approach to the derivation of his yield/ density equation is of interest because it differs markedly from the approaches seen in the previous section. De Wit termed his equation a “spacing formula,” and he derived it from a consideration of the space available to a plant and the plant’s ability to take up that space. He developed this formula from a consideration of two species grown on a homogeneous field of unit surface which he assumed to be divided into a number of squares of equal size. In a first model he assumed that the growth of one plant was unaffected by the growth of another. Thus if a plant is grown in each square the yields of each species in different mixtures can be represented by Fig. 9A. However, de Wit pointed out that in practice this situation would occur only where the density was so low that there was no competition or where the competitive powers of the two species were equal.

305


D e Wit developed his argument for the more practical situations where plants did compete. This situation is illustrated diagrammatically in Fig. 9B. I t can be seen that the yields of species 2 are higher, and the yields of species 1 are smaller, than would be expected from the first model.

+Species Species

2

4 P

I

tSpecies

I

Species 2-

P

FIG.9. Diagrammatic representation of the yields of each of two species (sp. I , sp. 2) grown in different mixtures: ( A ) where there i s no competition between the species: and (B) where competition exists between the species. At any point on the p axis, all the squares of the homogeneous field contain a seed either from species I or 2.

De Wit said this was because the plants of species 2 crowded the plants of species 1 out of some of the space alloted to them. H e went on to consider the situation where only one species occupied some of the squares and the rest remained empty (i.e., the relationship between a single species and plant density). He suggested that this single species would now behave in a manner similar to the dominant species in a mixed situation in that it would occupy more space than was allocated to it. Thus the relationship between the number of plants and the yield of this species would be comparable with that observed for species 2 in the example above (Fig. 9B). D e Wit assumed this relationship to be asymptotic and, using some oat data of Montgomery (1912), he derived the expression given earlier (Section 11, B, 5, c). It might be questioned whether the now arbitrary choice of the size of the square might not affect the result of the spacing formula: however, de Wit showed that this was not the case. It is of considerable interest that from this sort of approach de Wit developed an equation very similar to the other reciprocal equations. However, since de Wit has not suggested any modification of his equation to describe parabolic yield/density situations, the derivation outlined above is not examined in depth. b. Biological Validity of the Constants. It has been shown previously that for the situation where the yield/density relationship is asymptotic the equations of Shinozaki and Kira ( 1956), Holliday ( I 960a), Bleasdale

306


and Nelder ( 1960),and Farazdaghi and Harris ( 1968) all become identical and can be written 1

-=a+bp W

or, on a yield per area basis y

=-

P a+bp

Thus as density increases, y approaches the value of llb, i.e., the asymptote of yield per area = l / b . If it can be argued that the asymptote of yield per area is a measure of the potential of a given environment, then b is a meaningful factor indicative of environmental potential. In the case of Bleasdale’s simplified equation (Eq. 18), where 8 does not equal unity, the interpretation of the value of b is less obvious, although perhaps it may still give some indication of environmental potential. As the density tends to zero the value of yield per plant tends to lla in Eq. (27). It was seen earlier that this does not represent a very realistic situation for it ignores the fact that yield per plant levels off at densities too low for competition to occur (see Section 11, B, 5 , b). However, assuming that lla gives some indication of yield per plant in a competitionfree situation then, by a similar argument to that employed for b, a can perhaps be regarded as a meaningful factor indicative of genetic potential. It is interesting to pursue this reasoning with de Wit’s spacing formula _1 - _ 1_ 1 w-QP+Pp

In this instance the asymptote of yield per area is P, and yield per plant appears to approach a value Q P as density approaches unity. Thus in this equation maximum yield per plant, QP, is defined partly in terms of an environmental factor, P. This probably represents a more realistic situation. Several workers have examined the time trend in these constants. Shinozaki and Kira (1956) found that the value of b in a soybean density experiment rapidly increased with time in the period just after germination, but thereafter it fell, rapidly at first and then more slowly, toward a constant value of b. Jones (1 968) found a similar time trend in b for the dwarf bean crop. From a biological point of view it is hard to explain the

307


rapid rise in the value of b after germination. However, this may be an effect of the absence of competition affecting the fitting of the equation. Jones (1968) also found that the value of a fell throughout the season although there was a tendency to approach a constant value toward the end. Reestman and de Wit (1959) determined the course of P and Q in their equation (Eq. 14) throughout the latter part of the growing season in an experiment on sugar beet. Both P and Q increased with time and approached a constant value, Q doing so rather more quickly than P . With the exception of the changes in b at the early stages of growth, these time trends in the values of the constants are reasonably in accordance with what might be expected from their suggested biological significance. Bleasdale and his co-workers have examined the effect of variety and environment on the values of a and b. Bleasdale ( 1 966b) analyzed some yield/density data with Eq. (18) for three varieties of onions grown in the same environment. The values of a and b obtained are given in Table 11, and these suggest that the value of a depends upon the variety. HowTABLE 11 THE VALUES OF THE CONSTANTS a A N D b OBTAINED BY FITTING EQ. (18) TO THE Y I E L D ~ D E N S DATA I T Y FOR THREEONION VARIETIES“”’

Variety LANCASTRIAN RIJNBURCER SUTTONA1

b

(I

0.01 117 0.0 I90 I 0.01706

]

0.00263

“After Bleasdale ( 1966b). ”The data were fitted taking a common value of 0 = 0.8.

ever, it must be emphasized that these differences in a were for a common fitted value of b, since Bleasdale did not find significant differences between the individual values of b. In the same paper Bleasdale stated that unpublished results with other crops suggested that the value of b varied according to the soil fertility, but the value of a did not: further that a appeared to be a constant from year to year for a given variety. This hypothesis was also borne out by work of Bleasdale and Thompson ( 1966) on parsnips. This idea was further investigated for some wheat data of Willey (1965) for which one variety was grown under a number of environmental treatments. The values of a and b obtained by fitting Eq. (18) are given in Table 111. It can be seen that for all four treatments, fitted independently for a and b, a appears to be reasonably constant but b changes.

R. W. WILLEY AND

308

THE VALUESO F TO THE

TABLE 111 CONSTANTS u AND b OBTAINED BY FITTING EQ. (18) YIELD~DENSITY DATAFOR A WHEAT VARIETY“’b

THE

Control Treatment I Treatment 2 Treatment 3 ~~

S. B. HEATH

~

U

b

0.109 0.105 0.097 0.108

0.0673 0.0815 0.1092 0.0963

~~~~

“After Willey (1965). *The data were fitted taking a common value of 8=0.5. (Yield data are given inTable 1V.)

Holliday ( 1960a) examined the meaningfulness of the constants in his equations in rather a different way. It was seen earlier (Section 11, B, 5 , b) that he appreciated the significance of the constant a and termed l/a the “apparent maximum” yield per plant (A). Thus, substituting l/A for a , the form of his basic asymptotic equation becomes

which can be written

or, on an area basis

+

He termed the expression l / ( l Ab p ) the “competition function,” and it can be seen that the value of this decreases as density increases. Holliday considered that the definition of yield per plant as A [I/( I Abp)] gave a realistic description of what actually happens in practice, for it indicates how the yield of a plant at any given density is a product of the potential of the plant (A) and the forces of competition that are acting upon it [ I / ( 1 + A b p ) ] . Similarly, Holliday ( 1 960b) expressed his parabolic equation as

+

PLANT POPULATION AND CROP YIELD

+

309

+

In this case the competition function is I/( 1 A b p Ac pz). In conclusion, from the evidence presented it appears that it may be possible to ascribe some biQlogical meaningfulness to the constants in the reciprocal equations. Thus, examination of these constants may help to pinpoint genetic or environmental components of yield/density relationships. However, it seems likely that the inevitable interaction of these two components is a far more complex situation than can be described by a few simple constants. It would therefore seem desirable that these constants should be examined in much more detail before any exact biological meaning is ascribed to them. 2. Statistical Regression Analysis and Goodness of Fit In this section it is proposed first of all to consider some general points about the statistical fitting of yield/density equations. This is followed by a more detailed examination of the fitting of the reciprocal equations and the goodness of fit which they can give. It should be emphasized that it is important to consider why the regression analysis is being carried out. Where the intention is merely to fit a smooth curve to some data points, the analysis can often be satisfactorily carried out on the yield per area/density data, as has been shown by Sharpe and Dent (1 968). However, for reasons already given, for a more satisfactory description of the true yield/density relationship, it is usually more desirable to carry out the regression on the yield per plant/density data. An illustration of this is given in Fig. 10, in which two regressions have been carried out on the grain yield/density data of Pendleton and Dungan ( 1 960). The simple quadratic equation (Eq. 1) is fitted directly to the yield per area data, and Holliday’s parabolic equation (Eq. 12) is fitted to the yield/plant data. It can be seen that there is little difference between the regression lines as far as their ability to smooth the data is concerned but, at low densities particularly, the regression on yield per plant gives a much better description of the yield/density relationships. The method which has usually been used to fit the yield/density equations has been a least squares regression. The reciprocal equations describing an asymptotic yield/density situation can be fitted by a simple linear regression, but with the introduction of powers or the use of the quadratic to describe a parabolic situation, the regression becomes more complicated. One of the assumptions on which the least squares regres-

3 10


sion rests is that the variation about any one point is the same as that about any other. This means that the variance of the yield values, whether it be yield per plant or yield per area, must be constant over all densities.

/. OO

3

6

12

9

15

18

P

FIG. 10. The quadratic equation (Eq. 1 ) fitted directly to yield per unii area (-----) and the parabolic equation (Eq. 12) of Holliday (1960b) -( ) fitted via the reciprocal of yield per plant for the wheat data of Pendleton and Dungan ( I 960) meaned over four N levels and four varieties: y = bushelslacre, p = peckslacre.

Keller and Li (1 949) found thkassumption to hold with their data for a density experiment with hops, but their range of densities was limited. Hozumi et af. ( 1 956), when considering the individual yield of plants at three densities for leaf beet and turnips, found that the standard deviation of the points increased as plant size increased (with decrease in density), as shown in Fig. 1 1. Also, Nelder (1 963) criticized the curve fitting of Jarvis ( 1 962) for some lucerne density experiments and showed that it

...

.a

50

100

150

200

W

FIG. 11. A scatter diagram of the relationship between standard deviation (S.D.) and mean plant fresh weight of leaf beet, w (in grams), where changes in the latter were brought about by changes in plant density. (After Hozumi et al., 1956).

31 I


would be unlikely that the variation of yield per plant was uniform. He suggested that a more accurate assumption in yield/density experiments would be that the variance of the logarithm of yield per plant was constant. This assumption has been adopted by Bleasdale and his colleagues when fitting yield/density data to their equations. However, this assumption involves a more complicated treatment of the least squares regression than does the more usual assumption that the variance of w is constant. The main interest in examining the goodness of fit of the reciprocal equations is in the parabolic yield/density situation since in the asymptotic situation they are identical. In this situation the most useful comparison would seem to be between one of the equations for which some biological validity has been claimed and Holliday's more empirical equation (Eq. 12). Since most workers are probably more familiar with the approach of Bleasdale and his co-workers than with the more recent one of Farazdaghi and Harris ( 1 968), Bleadsale's simplified equation (Eq. 18) is compared with Holliday's. When fitting Bleasdale's simplified equation, the assumption is made that the variance of log (w-@)is constant; thus for a guessed value of 8, approximation to the true squares estimates of constants a and b can be obtained by a weighted regression of w - & with weights 0 % (Nelder ~ ~ 1963: Mead, unpublished). The criterion of the goodness of fit is the residual sum squares divided by 02, and the best value of 8 is that which reduces this to a minimum. An example of some spring wheat data (Willey, 1965) (Table IV) fitted by the above method illustrates this in Fig. 12. For each of the four TABLE I V G R A I NYIELD DATAFOR A WHEAT VARIETY GROWN AT FOURDENSITIES U N D E R FOURENVIRONMENTAL TREATMENTS".*

Grain yield (cwt./acre) Density (1 O6 plants/acre)

Control

Treatment I

Treatment I I

Treatment 111

0.392 1.122 2.432 5.78

20.82 35.32 29.98 23.95

20.60 28.15 30.14 15.85

19.63 23.71 19.51 10.35

19.62 20.57 22.8 I 12.93

After Willey ( 1965,. *Treatments: mean effect of reducing light intensity to 75 percent or 50 percent of full daylight during period of establishment-ear initiation (Treatment I), enr initiation-flowering (Treatment lI), and flowering-harvest (Treatment I 11). "

~

312

R. W. WILLEY AND S . B. HEATH

treatments fitted, a curve shows the goodness of fit (residual sum squares divided by P ) plotted against different 8 values. From the graph it can be seen that there was quite a wide variation between the best-fitting values

’

0.2

’

0.4

’

0.6

’

0:8

’

1lO

e FIG. 12. An example of obtaining the best-fitting value of 0 in Bleasdale’s simplified equation (Eq. 18); the best fit is where the residual sum squares (s.s.)/O* is reduced to a . .-, minimum. Wheat grain yield data of Willey ( 1965); Control (---), Treatment 1 (Treatment 11 (. . -), Treatment I11 (- ) (see Table IV).

-

-

of 6 for each treatment. However, if a common value of 8 of 0.5 was taken the individual 8’s for the four treatments did not give a significantly better fit. The best values of 8 were particularly well defined in these data, but this is not always the case. It may happen that the minimum is not so sharply defined, and in these circumstances it might be better to fit the correct but more complicated least squares regression of log ( w - O ) to avoid the approximation of the simpler weighted regression suggested by Nelder ( I 963); or this could be a situation in which Holliday’s equation is of more practical use. To make the comparison between Bleasdale’s simplified equation (Eq. 18) and Holliday’s parabolic equation (Eq. 12) valid, the latter is also fitted on the assumption that the variance of log (w)is constant. A comparison is made in Fig. 13 and Tables V and VI of the goodness of fit of

313


these equations to some selected data. The data are not meant to be comprehensive, but are chosen to illustrate the fit in two different parabolic situations, i.e., a definite parabolic situation (Table V and Fig. 13C and

70

IB P

ID

90. 80.

70. Y

60.

t 50L

I.,

4

8

12

P

16

20

24

4

.

8

12

16

20

24

P

FIG. 13. Examples of Holliday’s parabolic equation (Eq. 12) (---), and Bleasdale’s simplilied equation (Eq. 18) ( ), fitted to some yield/density data where yield declines only gradually at high densities (A and B) and where yield declines quite sharply at high densities (C and D); (A) Grain yield of wheat (Pendleton and Dungan, 1960) meaned over four N levels and four varieties; y = busheislacre, p = peckslacre. (B) Grain yield of wheat, var. HILGENDORF meaned over four years (Crawford, 1964);y = bushelslacre, p = bushels/ acre. (C and D) Grain yield of maize for hybrids WF9 X 38-1 1 and HY2 x OH7, respectively, at a medium N level (Lang e t a l . , 1 9 5 6 ) , y = bushelslacre, p = lo3p l a d a c r e .

D) and a situation where there is only a slight decrease in yield at high densities (Table VI and Fig. 13A and B). The curves plotted in Fig. 13 are also represented in the respective tables. Although these data are very limited, it can be seen that there may be little difference on average between the two approaches, although one might be better than another for a particular set of data. It can also be seen that the ability of the equations to fit the data can vary considerably (compare Fig. 13A and B with Fig. 13C and D).

314


Ill.

The Relationship between Plant Rectangularity and Crop Yield

It was emphasized in Section I that yield per unit area is dependent not only on the number of plants per unit area (plant density) but also on the spatial arrangement of those plants (plant rectangularity). Plant rectanguTABLE V A COMPARISON OF THE VARIATION REMAINING AFTER FITTING EQS. (18) AND (20)" A. Dumanovic and Penick (1962): Single-Cross Maize Hybrid Grown at 5 Densities and 4 Levels of Nitrogen Equation ( I 8):

Nitrogen level (kg.lha.1

TSS

H"= 0.45, RSSl0'

Equation (20) RSS

0 50 100 150

1.01 0.961 0.916 0.886

0.0 154 0.023 I 0.0275 0.0272

0.0 152 0.0236 0.0278 0.0274

B. Lang et al. (1956): Three Maize Hybrids Grown at 6 Densities and 3 Levels of Nitrogen

Hybrid

Nitrogen level

TSS

Equation (18), Ob=0.45,RSS/02

Equation (20), RSS

HY2 X OH7

Low Medium High Low Medium High Low Medium High

I .26 0.72 I 0.608 2.30 I .04 0.218 2.83 1.48 1.02

0.00809 0.0191 0.0 128 0.0422 0.00878 0.00209 0.0236 0.0055 1 0.00767

0.00420 0.00847 0.0139 0.0341 0.00257 0.00173 0.0200 0.00770 0.00744

WF9 X OH4 I

WF9 X 38-1 1

" RSS = Residual Sum Squares; TSS = Total Sum Squares. Both equations were fitted on the assumption the variance of log (wJ is constant. The RSSIB" obtained from fitting Eq. (1 8) are directly comparable with the R S S obtained from fitting Eq. (20). V a l u e of 0 which for the set of data as a whole reduced RSS/02 to a minimum.

larity can be most easily visualized in a row crop where it can be defined as the ratio of the distance between plants within the row to the distance between the rows. In a broadcast crop it may be more generally defined as a measure of the unevenness of distribution. This rectangularity, or unevenness of distribution, is important because of the unevenness of competition which it produces; competition may be too intense between some plants and insufficiently intense between others.

315


TABLE VI COMPARISON OF THE VARIATION REMAINING AFTER FITTING EQS. (18)

Study

TSS

Donald ( 1954) I . Wimmera ryegrass grown at 5 30.2 densities for seed 23.7 2. Subterranean clover grown at 5 densities for seed Crawford (1 964) Wheat var. HILGENDORF grown at 5 0.72 densities for grain Pendleton and Dungan ( I 960) Wheat grown at 6 densities for grain, 2.33 average over 4 nitrogen levels and 4 varieties Puckridge and Donald ( 1967) Wheat grown at 5 densities for grain 19.3

Bb

Equation ( I 8) RSS/%'

AND

(20)O

Equation (20) RSS

0.90

0.0149

0.00292

0.90

0.0394

0.0536

0.85

0.0000806

0.0000947

0.75

0.0007 1 I

0.000234

0.85

0.0171

0.040 1

RSS = Residual Sum Squares; TSS =Total Sum Squares about the mean. *Value of B which reduced RSS/A2to a minimum for the particular set of data.

The extent to which rectangularity may effect the yield of a crop is clearly dependent on the plasticity of the individual plant, which in turn must be dependent on the plant species. However, the general pattern of effects is illustrated by some winter wheat data of Harvey et al. ( 1 958) reported in Table VII. The treatments of Harvey etal. were not extreme, yet it can be seen that as rectangularity increases, either by increasing seed rate or increasing row width, yield per area gradually declines. Similar effects have been shown by Wiggans (1939) for soybeans, Reynolds ( 1950) for peas, Pendleton and Seif ( I 96 1) for maize, Bleasdale (1963) for peas, and Weber et al. (1966) for soybeans. Reynolds (1 950) also showed that as rectangularity increases the optimum density may decrease (Fig. 14). It would therefore seem desirable that equations describing the relationships between plant population and crop yield should be able to describe the effects of rectangularity as well as those of density. This can be particularly important because in the many population studies where different populations have been established on constant row width, rectangularity is not constant but increases with increase in density. Goodall ( 1 960) attempted to fit the model

316


TABLE VII THE EFFECTOF Row WIDTHA N D SEED RATE ON WINTERWHEATGRAINYIELDS (cwt./acre)" Seed rate, stoneslacre Row width (inches)

5.5

II

17

4 8

43.9 43.0 41.6

43.9 42.5 41.4

43.6 41.4 38.0

12 UAfterHarvey et al. ( 1 958).

w = adpi dJ" or

where dl is the intrarow spacing and d2 is the interrow spacing. Thus d l d z is the space available per plant. Equation (28) is therefore an extension of Eq. (6).

Goodall fitted this model to some soybean data of Wiggans ( 1939) which covered a range of densities and row widths. He found a significant

30

t

'Ot

o . - - o o

10

20

30

P

FIG. 14. The effects of rectangularity on the yield/density relationship in dried peas (Reynolds, 1950): the three curves represent different row widths, 8, 16, and 24 inches: y = cwt./acre, p = wnedacre.


317

difference between bl and b2; he suggested that this was due to row orientation effects. Donald (1963) pointed out, however, that Eq. (28) has the undesirable characteristic that, if either of the power terms is greater than the other, then the optimum rectangularity at a given density would be obtained where the distance between plants was increased in one direction and decreased in the other. Berry ( 1967) criticized Goodall’s fit to Wiggans’ data, not only on account of the poor fit of log w against log d l , but also because the values of d , and dz were not overlapping, and therefore different values of b1 and 6, could be expected. Berry ( 1967) extended the simplified equation (Eq. 18) of Bleasdale and Nelder to take into account plant rectangularity (29)

Since dld2= s, this model has included an extra term proportional to the square root of density. For a given density, w is greatest where dl = d,, i.e., where recta’ngularity is 1 : 1 , since (l/dl) (I/d2) is at a minimum value. This relationship gave a satisfactory fit to Wiggans’ soybean data. Berry considered that for irregularly spaced crops, i.e., where the rectangularity is not constant, Eq. (29) might still be used as a first approximation from Bleasdale’s simplified equation. For example, it could be used where plants are irregularly spaced within the row and rectangularity is defined by the mean intrarow distance and the interrow distance.

+

IV. The Variation in Yield of the individual Plant

It was emphasized in the introduction that the variation in the yield of the individual plant has seldom been examined in yield/density studies. The analysis has been in terms of the mean yield per plant at a particular density with no consideration of the variation about this mean. Yet this variation can be of great importance wherever the size of the individual plant is an attribute of yield. For example, in Fig. 2D the effect of size grading on the marketable yield of parsnips can be seen at each plant density although the latter has little effect on total yield. Kira et al. (1 953), Hozumi et al. ( 1956), and Stern ( 1 965) attempted to examine the effect of density and time on the variation in individual plant weights by calculating the coefficients of variation at each density. Kira et al. and Stern showed that the coefficients of variation increased with time, but the evidence was not consistent as to whether density affected the value of the coefficient of variation at any one time. However, Mead

318


(1967) has stressed that no attempt was made in these studies to test whether the shape of the distribution curves were constant, a necessary condition before coefficients of variation can be compared. Koyama and Kira ( 1956) considered the frequency distributions of pldnt weights at different densities. They found that although -the distribution for seed weight was normal, as the plants grew the distribution became more and more skew. The development of skewness was greatest at high densities. Kira et al. ( 1953) also tried using a correlation coefficient between the weight of an individual and the mean weight of the six plants nearest to it in a soybean experiment sown in a regular hexagonal arrangement. Surprisingly, the correlation coefficients were low and proved positive, apparently suggesting cooperation among plants rather than competition. Mead ( 1967) considered that for small samples the use of this form of the correlation coefficient is a biased estimator of the degree of competition in the population. Mead proposed a measure of the weight relationship between a plant and its six immediate neighbors, grown in a regular hexagonal arrangement, termed the competition coefficient. This is a general measure of the plant neighbor relationship for all the plants in the community depending only on the size of the neighboring plants but assuming regular arrangement. Mead ( 1968) analyzed the results of some experiments laid down to examine this relationship and showed that, for cabbage, carrots, and to a lesser degree, radishes, the competition coefficients are predominantly negative, showing competition between plants rather than cooperation as was indicated by Kira’s results. Mead therefore considered the competition function of more use than the correlation coefficient. Mead ( 1 966) investigated the importance of irregularity of spacing on the variation of individual plant yield within a population. He did this by examining the importance of the size and shape of the space available to the plant in determining the yield of that plant. He suggested a model in which the total ground area under an irregularly spaced crop was divided into polygons, each one being allocated to a single plant. This was done by allocating any given spot of ground to the nearest plant. The polygons could then be characterized by three parameters, the area, the extent to which the polygon was elliptical rather than circular (a measure of rectangularity), and a measure of how far the plant was from the center of the polygon. On examining the relationship between the three parameters and the root diameter of carrots grown at three densities and three row widths, he found that the proportion of the total variation in plant yield attributable to polygon variation increased with time, the largest mean proportion at a final harvest being 20 percent, or for individual plots


319

as much as 55 percent. Mead also found that the area of the polygon was more important than its shape. Furthermore, for an irregularly planted crop the variation in area did not appear to change with change in density, but at a given density it increased with increasing row width. This would seem to indicate that for maximum uniformity within a crop, the rectangularity ratio should be as low as possible. As Mead pointed out, this approach is somewhat unrealistic for it takes no account of the size of the neighboring plants or the fact that plants other than the immediate neighbors defined by the polygons might affect the plant either directly or indirectly. However, this initial approach seems to be a useful one. V.

Conclusions

This review has attempted to examine the usefulness and biological validity of the different mathematical equations that have been proposed to describe the relationships between plant population and crop yield. I t has been seen that even the simplest and most empirical of equations may be useful in certain circumstances, e.g., to smooth data over a limited range of densities. However, to describe the relationships realistically over a wide range of densities, or to construct yield/density curves from a minimum of data, it would clearly seem to be desirable to use those equations that have a better biological foundation and have proved the most satisfactory in practice. In general the reciprocal equations seem to fulfill these requirements best. The satisfactory description which they provide of the asymptotic yield/density situation has been established by many workers. The description of the parabolic yield/density situation is less certain, but at least this group still provides a choice of three basic equations (Eqs. 12, 18, 20), all of which are inherently very flexible and and all of which offer a reasonable possibility of obtaining a satisfactory and realistic fit. A study of the evidence at present available suggested that some biological meaning can perhaps be ascribed to the constants in the reciprocal equations, but further research is needed in this field. An equally worthwhile field would seem to be the incorporation of the effects of plant rectangularity and plant variability into the yield/density equations; relatively few workers have studied these factors, yet their effects on agricultural yield may be considerable.

ACKNOWLEDGMENTS The authors wish to thank Mr. R. Mead, University of Reading, for many helpful discussions concerning the statistics.

3 20

R. W. WILLEY A N D

S. B. HEATH

REFERENCES Berry,G. 1967. Biometries 23,505-5 15. Bleasdale, J. K. A. 1963. I n “Crop Production in a Weed Free Environment” (E. K. Woodford,ed.),pp. 90-101. Bleasdale, J. K. A. I966a. Ann. Appl. Biol. 57,173- 182. Bleasdale, J. K. A. 1966b. J . Hort. Sci. 41, 145- 154. Bleasdale, J. K. A. 1967.J . Hort. Sci. 42,5 1-8. Bleasdale, J. K. A., and Nelder, J. A. 1960. Nature 188,342. Bleasdale, J. K. A., and Thompson, R. 1966.J. Hort. Sci. 41,37 1-378. Bruinsma, J. 1966. Neth. J . Agr. Sci. 14,198-2 14. Campbell, R. E., and Wets, F. G. 1967.Agron. J. 59,349-354. Carmer, S. G., and Jackobs, J. A. 1965.Agron. J . 57,241-244. Crawford, W. R. 1964. N e w ZealandJ. Agr. 108,455-463. de Wit, C. T. 1959. Jaarb. Inst. Biol. Scheik. Onders. LandbGewass. Meded. 83,129-134. de Wit, C. T . 1960. Verslag. Landbouwk. Onderzoek. 66.8,I-8 1. de Wit, C. T., and Ennik, G. C. 1958. Jaarb. Inst. Biol. Scheik. Onderz. LandbGewass. Meded. 50,59-73. Donald, C. M. 195 I . Australian J. Agr. Res. 2,355-376. Donald, C . M. 1954. Australian J . Agr. Res. 5,585-597. Donald, C. M. 1963. Advan. Agron. 15,l- I 18. Dumanovic, J., and Penick, M. 1962. Savremera Polio Priverda 4,240-250. Duncan, W. G. 1958. Agron. J . 50,82-84. Farazdaghi, H. 1968. Ph.D. Thesis, Reading University, England. Farazdaghi, H.,and Harris, P. M. 1968. Nature 217,289-290. Goodall, D. W. 1960. Bull. Res. Council Israel D8, 18 1-192. Harvey, P. N., Whybrew, J. E., Bullen, E. R., and Scragg, W. 1958. Exptl. Husbandry 3, 3 1-43. Holliday, R. 1960a. Nature 186,22-24. Holliday, R. 1960b. Field Crop Abstr. 13,159- 167 and 247-254. Hozumi, K., Asahira, T., and Kira, T. 1956. J . Inst. Polytech., Osaka City Univ.D7, 15-33. Hudson, H. G. 194 I . J . Agr. Sci. 31,138- 144. Jarvis, R. H. 1962. J. Agr. Sci. 59,28 1-286. Jones, L. H. 1968.Agr. Prog. 42,32-62. Keller, K. R., and Li, J. C. R. 1949.Agron. J . 41,569-573. Kira, T., Ogawa, H., and Shinozaki, N. 1953. J . Inst. Polytech., Osaka City Univ.D4,1-16. Kira, T., Ogawa, H., and Hozumi, K. 1954. J . Inst. Polytech., Osaka Ciiy Univ. D5,l-7. Kira, T., Ogawa, H., Hozumi, K., Koyama, H., and Yoda, K. 1956. J. inst. Polytech., Osaka City Univ. D7,l- 14. Koyama, H., and Kira, T. 1956. J . Inst. Polytech., Osaka City Univ. D7,73-94. Lang, A. L., Pendleton, J. W., and Dungan, G. H. 1956. Agron. J . 48,284-289. Mead, R. 1966.Ann. Botany (London) [N.S.] 118,301-309. Mead, R. 1967. Biometrics 23,189-205. Mead, R. 1968. J. Ecol. 56,35-45. Mitscherlich, E. A. 1919. Landwirtsch. Jahrb. 53,341-360. Montgomery, E. 19 12. Nebraska Univ.,Agr. Expt. Sta., Bull. 24(after de Wit, 1960). Nelder, J. A. 1963.5. Agr. Sci. 61,427-429. Pendleton, J. W., and Dungan, G. H. 1960. Agron. J . 52,3 10-3 12. Pendleton, J. W., and Seif, R. D. 196 I . Crop Sci. 1,433-435.


Puckridge, D. W.. and Donald, C. M. 1967. Australiun J . A g r . R e s . 18,193-2 1 1. Putter, J., Yaron, D., and Bielorai, H. 1966. Agron. J . 58,103-104. Reestrnan, A. J . , and de Wit, C. T. 1959. Nerh. . I Agri. . Sci. 7,257-268. Reynolds, J . D. 1950. Agriculture (London) 56,527-537. Richards, F. J . 1959. J . Exptl. Botany 10,290-300. Saunt, J. 1960. M.Sc. Thesis, University of Leeds, England. Sharpe, P. R.,and Dent,J. B. 1968.J.Agr. Sci. 70,123-129. Shinozaki, K., and Kira, T . 1956. J . lnst. Polytech., Osaka Ciry Univ. D7,35-72. Stern, W. R. 1965. Australian J . Agr. R e s . 16,541-555. Warne, L. G. G. 195 1. J . Hort. Sci. 26,84-97. Weber, C . R., Shibles, R. M., and Blyth, D. E. 1966. Agron. J . 58,99- 102. Wiggans, R. G. 1939. J . A m . Soc.Agron. 31,3 14-32 I . Wilcox, 0.W. 1950. Agron. J . 42,4 10-4 12. Willey, R. W. 1965. Ph.D. Thesis, Leeds University, England.

32 1


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

Aurand, L. W., 65, 108

Abou Akkada, A. R.. 50, 97 Acton, C. J., 197, 199, 204, 207, 209, 21 I , 230 Ada, G. L., 224,232 Adams, G . A., 224,230 Ademosum, A. A., 5 , 13, 97 Adler, F. E., 83, 99 Aitken, J. N., 92, 101 Alderman, G., 59, 97 Aldrich, D. T. A., 20, 22, 23, 66, 97, 100 Alexander, E. B., Jr., 247, 254, 255, 258, 2 72 Alexander, L. T., 150, 169, 257,274 Alexander, M., 2 19, 22 I , 230 Alexander, R. H., 13, 15, 97 Alfred, S., 275, 278 Allaway, W. H., 60, 97, 100 Allcroft, R., 56, 97 Allen, 0. N., 196, 197, 198, 199, 204, 209, 2 l9,22 1,222,231,233,234 Allen, P. E., 184, 192 Allison, J . B., 173, 193 Allison, L. E.. 19R. 230 Altschul, A. M.. 172. 173. 178. 180. 191 Anderson, D. M. W., 288.231 Alvarado, G., 189, 193 Alverez-Tostado, M. A,, 189, 193 Alvistur, E.. 180, 193 Anderson, J. U., 248, 252, 2S8, 272 Anderson, M. J., 17, I03 Andrews, E. D., 57, 97 Andrews, 0. N.. 29, 3 I , 97 Andrews, R. P., 36, 52, 102 Annison, E. F.. 41, 50, 5 I , 97 Antonopoulos, C. A,, 229,231 ap Griffith, G., 22, 54, 55, 56, 97, 99, 107 Appelqvist, L.-A,, 190, 192 Armstrong, D. G., I3,39,41,43,44,97 Arnold, G. W., 84,87,88,92,97, 103 Asahira,T., 294, 301, 310, 317,320 Atloe, 0. J., 196, 197, 198, 199, 204, 209, 22 I , 231, 233 Atwood, K. C., 176, 193

B

Babcock, G. E., 188, 194 Bach, R.. 197,204,208,216,218,232 Bacon, J . S. D., 197, 2 19, 235 Bacon, S. R., 266, 274 Baert, L., 5 5 , 69, 100 Bahn, A. V., 62,104 Bailey, H . H., 243, 250, 256, 260. 272, 2 73 Bailey, P. H., 82, 97 Bailey, R. W., 48, 97 Baker, E., 220, 234 Baker, G., 60, 97 Baker, J. C., 257,272 Baker, 0. E., 141, 169 Balch, C. C., 26, 27, 28, 36, 41,44,47, 52, 76, 97, 99, 106 Balch, D. A., 47, 97 Baldwin, M., 267, 272 Bandemer, S. L., 180, 192 Bandet. J., 185, 193 Banks, W.. 227, 228,231 Barclay, A. S., 180, 194 Barker, S. A., 202, 203, 207,208, 2 10, 2 I I , 212, 213, 214, 216, 217, 220, 223, 228, 231, 235 Barnes, R. F., 14, 15, 29, 30, 31, 97, 104 Barnett, L., 175, 192 Barrett, A. J., 2 I I , 231 Barrett, J. F., 62, 97 Barrow, N . J.. 59, 84, 97 Bartelli, L. J., 240, 244, 252, 263,272, 273 Bartlett, R. J.. 246. 272 Bartlett, S.. 47,97 Bartley, E. E., 73, 107 Barton, R. A., 92, 99 Bassette, R., 65, 100 Bates, L. S., 173, 180, 181, 182, 183, 187, 191, 193 Bath, D. L., 87, 90, 97 Bath, I. H., 45, 98 Battacharya, A. N . , 77. 107

323

324

AUTHOR INDEX

Bauer, P. T., 118, 169 Baum, R. R.,177, 192 Baumann, C. A., 60, I04 Baumgardt, B. R.,5 , 1 1 , 13, 31,47,50,97, 98, 106 Beadle, G. W., 180, 192 Beardsley, D. W., 7 1 , 98 Beath, 0. A,, 59, 60, 101, 106 Beavers, A. H., 242, 243, 248, 251, 253, 255,260,272, 273 Beck, A. B., 61, 62, 98, 106 Becker, G., 173, 191 Becker, M. J., 226,231 Bedell, F., 87, 98 Beesley, T. E., 267,273 Beeson, W. M., 183, 192 Beitz, D. C., 45, 46, 47, 99 Bell, J. M., 65, 103 Bell, M. C., 82, 98 Bell, T. A,, 25, 106 BeMiller, J. N., 198, 205, 235 Beene, E. J., 3, 35, 102 Bennett, H. H., 266, 274 Bennett, W. D., 65, 98 Bensadown, A., 73, 104 Bentley, 0. G . , 1 I , 105 Bergman, E. N., 48, 98 Bernier, B., 197, 200, 202, 208, 209, 210, 211,216, 217,218,231 Berry, G., 317,320 Betts, J. E., 62, 104 Bickoff, E. M., 61, 62, 63, 98, 104 Bielorai, H., 281, 321 Bigsby, F. W., 34, 107 Binns, B. O., 135,169 Bishop, C. T., 225,231 Black, W. A. P., 202, 207, 208, 209, 231 Blake, J. T., 89, 90, 99 Bland, B. F., 32, 98 Blaser, R. E., 17, 22, 24, 92, 95, 98, I00 Blaxter, K. L., 27, 28,3 1,32,37,39,41, 43, 44, 51, 70,72, 73, 91, 97, 98, 100 Bleasdale, J. K. A., 283,285,286,299,300, 301, 302, 303,306, 307, 315,320 Blyth, D. E., 3 15,321 Bobbit, J. M., 198, 231 Bogdan, A. V., 54,100

Bogue, D. J., 110, 169 Bohman, V. R.,8 8 , 9 0 , 1 0 6 Bolin, D. W., 13, 106 Bond, J., 53, 104 Booysen, P. de V., 34,107 Bornstein, J., 246, 272 Bosman, M. S. M., 13, 98 Bouveng, H. O., 228,231 Bowler, E., 220, 231, 234 Braden, A. W. H., 61, I04 Bradley, N. W., 81, 103 Brasher, B. R.,248,273 Bratzler, J. W., 106 Bray, A. C., 59,104 Bredon, R. M., 87, 98 Bremner, J. M., 204, 205, 221, 231 Brenner, S., 175, 192 Bressani, R., 178, 180, 184, 191, 192 Brett, D. J., 48, 103 Brewer, R.,251, 253, 272 Brian, R. C., 196,232 Briggs, D. R., 188, 193 Briggs, M. H., 26, 52, 79, 98 Briggs, P. K., 37,45, 98, I05 Brimacombe, J. S., 223, 228,231 Brimhall, B., 180, 192 Brink, R. H., 209,231 Brockway, J. M., 48, 98 Broderick, G. A,, 60, 104 Broekmeijer, M. W. J. M., 110, 169 Brohult, S., 178, 191 Brook, P. J., 65, 98 Broughan, R. W., 92,105 Brouwer, E., 43, 98 Brown, D., 80, 83,98 Brown, L. D., 27, 98 Brown, R. E., 45, 46, 47, 99 Brown, R. H., 17, 22, 98 Brown, S. M., 78, 79, 98 Brown, T. H., 93,106 Brownell, J., 57, 106 Browning, C. B., 12, 103 Browning, D. R., 252, 255, 256, 262, 265, 2 75 Bruinsma, J., 283,320 Brundage, A. L., 84, 105 Bryant, A. M., 47, 98 Bryant, H. T., 95, 98

AUTHOR INDEX

Buchman, D. T., 72, 98 Buckman, H. O., 265,273 Buettner-Janusch. J., 178, 192 Bull, L. B., 58, 98 Bullen, E. R., 3 15, 3 16, 320 Burau, R. G., 57, 106 Burdick, D., 7, 98 Burges, A., 219,231 Burns, J. C., 14, 104 Bums, K. N., 56, 97 Butler, G. W., 59, 92, 98, 105 Butterworth, M. H., 5 , 6, 98 Bystrom, B. G., 220, 234 C

Cady, J. G., 150, 169, 244, 248, 251, 252, 254,256,259,260,274,275 Calhoun, F. G., 250, 251, 260, 272 Call, J. W., 89, 99 Calvo, J. M., 190, 192 Calvo, R. A., 190, 192 Cameron, G. D. T., 33, 92, 98 Campbell, C. M., 87, 98 Campbell, J. R., 53, 98 Campbell, R. E., 283,320 Campling. R. C.. 26. 27, 28, 32, 36, 52, 7 1 , 72,75,76,77,97,99,100, 106 Carlisle, F. J., 239, 243, 245, 246,249,250, 25 1,254,256,257,272 Carlson, I. T., 13, 22, 107 Carmer, S. G., 289, 320 Carr, M. E., 265, 266, 273 Carrillo, B. J., 65, 108 Carroll, P. H., 167, 169 Carter, A. H., 81, 99 Carter, W. T., Jr., 266, 273 Cary, E. E., 60, 100 Cason, J. L., 1 1 , 98 Castle, M. E., 18, 96, 99 Chalmers, M. I., 50, 5 I , 97 Chalupa, W., 28, 52, 53, 99 Champakam, S., 171, 192 Charlwood, P. A,, 229, 231 Charter, C. F., 167, 169 Chen, M.-L., 182, 192 Chenost, M . , 34, 99 Cheshire, M. V., 200, 225,231

325

Chesters, G., 196, 197, 198, 199, 204, 209, 22 1, 222,231, 233 Choudri, M. B., 204, 205,231 Christian, K. R.. 8 I , 99 Clancy, M. J., 7, 99 Clanton, D. C., 87, 101 Clapp, C. E., 196, 197, 1% 208,21 1,217, 23 I Clapperton, J. L., 91,99 Clark, B., 1 I , 105 Clark, H. F., 184,192 Clark, K. W., I 1,99 Clark, V. R., 62,99 Clarke, E. G. C., 61,99 Clarke, M. L., 6 1,99 Clarke, R. T. J., 64,99 Clifford, A. J., 53,99 Cline, M. G., 239, 244, 246, 247, 248, 249, 250, 252, 253, 256, 257, 258, 260, 262, 263, 264, 268,273, 274 Cline, T. R.,53, 99, 185, 192 Coates, W. H., 246, 250,260,262,274 Collazos, C., 180, 194 Comer, G. H., 255, 256,273 Comerma, J. A., 247,252,256,258,273 Compere, R., 84, 101 Compy, E. 2. W., 267, 274, 276 Connolly, J. O., 84, 106 Conrad, H. R., 14, 27, 31, 32, 53, 71, 99, 100 Conway, A., 94, 95, 99 Cook, C. W., 89.90, Y9 Coop, I. E., 62, 91, 99 Cooper, .I.P., 22, 68, 99 Coote, J. N., 64, 102 Corbett, J. I-., 5,48, 83,84,85,91,99,101, 103 Corbett, W. M., 227,235 Cordes, E. H., 174, 192 Cornhill, W. J., 202,207,208,209,231 Cotnoir, L. J., Jr., 221, 233 Couchman, J. F., 70, 101 Cowlishaw, S. J., 83, 99 c o x , c . P., 47, 97 Craggs, 8. A., 22 I , 233 Cramer, D. A., 92, 99 Crampton, E. W., 4, 7, 29, 35, 36, 38, 90, 99,100

326

AUTHOR INDEX

Crawford, W. R., 3 13, 3 15,320 Crick, F. H. C., 175, 192 Cromwell, G. L., 183, 185, 192 Czochanska, Z., 92, 99 D

Daji, J. A., 202, 231 Dalgleish, C. E., 209, 231 Daniels, R. B., 238,240,242,243,246,247, 248, 249, 250, 251, 252, 258, 259, 261, 273,274 Danielson, C. E., 178,192 Dart, P. J., 220,231 da Silva, J. F. C., 29,99 Davey, B. G., 55,99 Davidson, E. A., 223,225,231 Davies, W. E., 54,56,99 Davis, C. L., 45,46,47,48,99 Davis, L. E., 84,99 Davis, R. J., 196, 197,231 Davison, K. L., 64,77,107,108 Dawes, C. J., 220,231 Dea, I. C. M., 228,231 de Freitas, J., 70, I01 de Groot, T., 54,56,99 De Haan, Sj., 80,108 Dehority, B. A., 7, 11, 12, 14, 16, 25, 71, 99,100,101,102 Deijs, W. B., 57, 108 Deinum, B., 23, 24, 100 de Loose, R., 55,69, 100 Demarquilly, C., 17, 18, 29, 70, 7 I , 72,73, 100

De Muelenaere, H. J. H., 182, 192 Denny, C. S., 245, 246, 259, 262,273 Dent, J. B., 288, 309, 321 Dent, J. W., 13, 20, 22, 23, 32, 66, 97, 98, 100

Derbyshire, J. C., 74,100 Deriaz, R. E., 19,88,91,106 Dermine, P., 24, 101 Derting, J. F., 275, 276 Deshpande, T. L., 198,231 Dettmann, M. G., 259,273 De Tyssonsk, E. R., 229, 231 Dew], H., 196, 197, 198, 203, 204, 205, 208, 209, 211, 212, 216, 217, 218, 231, 232,233, 234, 235

Deuel, M., 196, 197, 19H, 204, 208, 212, 216, 217,233 de Wit, C. T., 283,298,299, 304,307,320, 32 I Dias, C., 180, 194 Dick, A. T., 5 8 , 98, 100 Dickson, G. R., 95, 100 Diebold, C. H., 254, 275 Dijkstra, N. D., 6, 100 Dische, Z., 223, 229, 231, 233 Dobie, J. B., 72, 74, 100, 106 Donald, C. M., 284, 291, 292, 293, 294, 315, 317, 320,321 Donefer, E., 4, 29, 35, 36. 38, 99, 100 Donker, J . D., 32, 92, 103 Dormaar, J. F., 199, 204, 207, 209, 232 Dougall, H. W., 54, 100 Doyle, J. J., 90, 101 Drew, K. R., 15, 100 Drysdale, A. D., 18, 96, 99 Dubach, P., 196, 197, 203, 204, 205, 208, 209,212,216,218,231,232,233,234 Dudzinski, M. L., 84,88,92,97,103 Duff, R. B., 197, 207, 208, 209, 214, 216, 217, 219,232,235 Dumanovic, J., 3 14, 320 Duncan, W. G., 289, 292,320 Dungan, G. H., 285, 287, 288, 289, 290, 309, 310, 313, 314, 315.320 Dunham, J. R., 65,100 Dunstone, J. R., 229, 232 Duval, E., 56, 101 E

Earle, F. R., 178, 193 Echols, H., 177, 192 Edlefesn, J., 90, 99 Egan, A. R., 37, 72,100 Eggum, B., 187, 192 Ehlig, C. F., 60, 100 Ekern, A., 5 I , 70, I00 Elder, J. H., 267, 274, 27.5, 276 Ellington, A., 54, 56, 99 Elliott, F. C., 25, 106 Elliott, R. C., 25, 37, 45, / 0 7 Ellis, W. C., 82, 100, 102 El Sayed Osman, H., 50, 97 El-Shazly, K., 11, 25, 100, 104

327

AUTHOR INDEX

Ely, R. A., 7, loo Emerson, R. A.. 180, 192 Emerson, W. W., 198,231, 259, 273 Eng, K. S., 87, 98 Engberg, C. A., 242, 275 Engels, E. A. N., 13, Is, 100 Engleman, E. M., 178, 180, 191 Ennik, G . C., 298, 320 Enzie, F. D., 93, 103 Epley, I. B., 267, 273 Erskine, A. J., 21 I , 232 Ervin, J. O., 196, 197, 220, 221, 233 Essig, H. W., 79, 100 Estermann, E. F., 220, 232 Evans, L. T., 204, 232 Evans, P. S., 34, 100 Evans, R. J., 180, 192 Eveleth, D. F., 13, 106 F Fabry, J., 84, IOZ Faichney,G. J.,45,46,48,52, ZOO Farazdaghi, H., 283, 300, 301, 302, 303, 304, 306, 3 1 I , 320 Farmer, V. C., 197, 219,235 Featherson, J. R., 185, 192 Fehrenbacher, J. B., 242, 243, 248, 251, 253, 255, 273 Finch, P., 202,203,207,208,2 I0,2 1 I , 2 12, 216, 217,221,231, 232 Fisher, L. J., 76, 106 Fitzpatrick, E. A,, 260, 273 Flatt, W. P., 39, ZOO Fleming, G . A,, 54, 56, I00 Flodin, P., 228, 232 Florence, E.. 83, 99 Flowers, R. L., 267, 273 Folkins, L. P., 22, 105 Fontenot, J. P., 17,22,92,98,100 Ford, G. W., 205,206,232 Forsyth, W. G. C., 196, 201,204,208,209, 214, 216, 217, 219, 222,232 Foster, A. B., 2 1 I , 220, 232 Fountaine, E. R., 258,273 Fox, C. J., 267, 273 Fox, C. W., 62, 104 Francis, C. M., 62, 100, 104 Francois, C., 224, 225, 232 Franek, M. D., 229,232 Frankinet, M., 84, 101

Fransmeier, D. P., 248, 273 Fransson, L-A., 229,231 Fraser, A. C., 180, 192 Freer, M., 26, 32, 36, 52, 86, 99, 100, 104 Frey, K. J., 180, 192 Fulkerson, R. S., 22, 104 G

Gaillard, B. D. E., 7, 8, I00 Galberry, H. S., 267, 274 Gallagher, C. H., 63, ZOO Gamble, E. E., 238, 240, 246, 247, 249, 259, 261,273 Gardell, S., 229,231 Garen, A., 174, 177, 192 Garen, S., 177, 192 Garrett, W. N., 74, 100 Garrigus, U. S., 53, 99, 102 Gascoigne, J . A,, 219,232 Gascoigne, M. M., 219, 232 Geoghegan, M. J., 196,232 George, J. M., 56, 62, 97, 102 Gerloff, E. D., 177, 192 Gibbons, R. A., 223, 227,232 Gile, L. H., Jr., 249, 254, 273 Gilfillan, E. W., 172, 179, 192 Gill, W. R., 251, 273 Gladstones, J. S., 58, ZOO Glenday, A. C., 59, 92, 98, 105 Glenn, R. C., 248,273 Gomide, J. A., 29, 99 Gonzalez Gonzalez, V., 87, 100 Goodall, D. W., 292, 315, 320 Goodlett, J. C., 246, 250, 260, 262, 264, 273, 274 Gopalan, C., 171, 189, 192, 193 Goplen, B. P., 65, 103 Gordon, C. H.. 74, 100 Gottschalk, A., 223, 224, 232, 234 Graham, N . McC., 39, 72, 73, 91, 98, 100, 101 Granath, K. A., 228, 232 Graveland, D. N., 200, 204, 232 Gray, F. V., 48, 107 Greacen, E. L., 221, 234 Green, J. O., 66, 67, 101 Green, T. W., 267, 274 Greenhalgh, J. F. D., 5, 33, 83, 85, 92, 99, 101

328

AUTHOR INDEX

Greenhill, W. L., 70, 101 Greenland, D. J., 156, 170, 196, 197, 198, 200, 201, 202, 203, 204, 205, 206, 207, 208, 209, 210, 211, 213, 214, 215, 216, 217,218,220,221,231,232,235 Greenwood, C. T., 227, 228,231, 232 Grieve, C. M., 29, 101 Griffiths, E., 197, 232 Griffiths, T. W., 48, 101 Grimes, R. C., 47, 87, I01 Grossenbacher, K., 220, 233 Grossman, R. B., 239, 242, 243, 246, 248, 25 I , 252,253,255,257,258,272,273 Grunes, D. L., 56, 101 G ~ p t a U. , C., 196,200, 202, 208,232 H

Hagberg, A., 187,192 Hamilton, F. J., 87,103 Hamilton, H. A., 24,101 Hamilton, J. W., 59,60,101 Hammer, K. C., 220,234 Hammes, R. C., 95,98 Handreck, K.A., 10,34,60,102 Hanel, D. J., 14,107 Hanson, C. H., 62,101 Harada, T., 205,231 Hardison, W. A., 84,95,98, I05 Harker, V. G., 5 , 13,105 Harkness,R. D., 17, 18, 19,101 Harlan, J. R., 94,101 Harmon, A. B., 267,273 Harmsen, G. W., 220,233 Harper,A. E., 182,192 Harpstead, D. D., 184,192,193 Harris, C. E., 5 , 13, 17, 18, 19, 20, 24, 28, 29, 38, 66, 75, 76, 79, 80, 84, 85, 101, 103,104,105 Harris, I . F., 179, 193 Harris, L. E., 90, 99 Harris, P. M., 300,301,302,303, 304,306, 3 11,320 Harris, R. F., 196, 198, 22 1, 222, 233 Hartree, E. F., 224, 233 Harvey, P. N., 315, 316,320 Hatfield, E. E., 53,99, 102 Hawkins, G. E., 25, I01 Haworth, W. N., 196, 224, 233

Hayes, M. H. B., 202, 203, 207, 208, 210, 211, 212, 213, 214, 216, 217, 221, 231, 232 Head, M. J., 26, I01 Heady, H. F., 90, I07 Healy, W. B., 60, 101 Heaney, D. P., 28, 29, 38. 71, 101 Hearn, W. E., 265,273 Hedin, L., 56, 101 Heidelberger, M., 229, 23.3 Heinegard, D., 229, 231 Heintz, R., 174, 193 Helinski, D. R., 174, 192 Hemingway, R. G., 58,103 Hemken, R. W., 72.98 Hennaux, L., 84,101 Henry, E. F., 275,276 Henry, J., 163, I70 Hercus, J. M., 91,101 Hibbs, J. W., 3 1,32,53,99 Hight, G. K., 33,101 Hill, D. E., 255,265,268,273,275 Hill, M. K., 91, 99 Hill, R. L., 178,192 Himes, F. L., 201, 202, 207, 208, 209, 2 12,214,216,2 17,233,235 Hinders, R. G., 7 I , 101 Hiridoglow, M., 24,101 Hirst, E. L., 224,233 Hocevar, B.J.,211,233 Hodge, R. W., 90,101 Hodgson, J., 92,101 Hodgson, J. F., 60,97 Hoehne, 0. E., 87, I01 Hoffman, H., 63,100 Hogan, J. P., 29, 37, 45, 48, 50, 98, 101, 105,107 Hogue, D. E., 5,102 Hole, F. D., 238, 247, 259, 260, 262, 263, 264,274 Holley, R. W., 174,192 Holliday, R., 282, 283, 284, 287, 289, 290, 292, 295, 296, 297, 298, 299, 300, 301, 305, 308,3 10,320 Holmes, J. C., 33,101 Holt, E. C., 94,106 Homb, T., 16,101 Honkanen, E., 65,101 Hopson, J. D., 12,101

AUTHOR INDEX

Horn, M. E., 249, 250, 251, 253, 256, 258, 260,273,275 Horton, D., 204,227,228,233 Hoveland, C. S., 2 9 , 3 I , 97 Howard, M., Jr., 246,272 Howe, E. E., 172. 179,192 Howe, F. B., 265,273 Howe, W. M., 53.98 Hozumi, K., 291, 292, 294, 301, 302, 310. 317,320 Huddleston, J. H., 254,255,264,265,273 Huddleston, J. S., 245,273 Hudson. H . G., 287,297,320 Hughes, M., 8 2 , 9 7 Hughes, R., 22,107 Hull, J. P. D., 266,273 Hulyalkar, R. K., 225,234 Humbert, R. P., 248, 251, 252, 254, 256, 259,260,274 Hungate, R. E., 49, 102 Huston, J. E., 81. 102 Hutcheson. T. B., Jr., 246. 247, 248, 256. 2 73 Hutchings, R. J., 63, 104 Hutton, E. M., 64,102 Hutton, J. B., 3 I , 32,85,102 I

Ignatieff, V., 113. 148,169 Ingalls, J. R., 3,35,102 Ingleton, J. W.. 84,104 Ingram, V. M., 175,192 Irvin, H. M., 40,104 Ivarson, K. C., 200,225,233,235 J

Jackobs, J. A., 289,320 Jacob, F., 176, 177.192 Jacobson. W. C., 74,8 I , 82,100,102 Jacoby, Erich H.. 135,169 Jager, G., 220,233 Jahn, J. R., 72,102 James, W. H., 106 Jansen, G. R., 172, 179,192 Jaques, L. B., 65, 103 Jarrige, R., 7,102 Jarvis, R. H., 292,310,320 Jayrne, G., 202,233 Jeanloz, R. W., 223,228,233

329

Jenny, H., 220,233 Jensen, E. H., 88,90,106 Jermyn, M. A., 229,232 Jha, P. P., 246, 247, 248, 249, 250, 252, 256,260,262,273,274 Jimenez,J. R., 179, 185, 186,192 Johansen, P. G., 224,225,233 Johns, A. T., 64,102 Johnson,HarryG., 118, 121, 170 Johnson, J. C., 267,274,276 Johnson, R. R., 7, I I , 12, 14, 16, 25, 52, 53,71,79,99,100,101, 102, 106 Johnston, T. D., 34,102 Jones, D., 197,220,232,233 Jones, D. I. H., 22.59.97.107 Jones, G. B., 48,107 Jones, H. E. H., 62,105 Jones, J. H., 94,106 Jones,J. K. N.,2I 1,232 Jones, J. R., 5,102 Jones, L. H., 306,307,320 Jones, L. H. P., 10,34,60,97, 102, 107 Jones,Q., 178,180, 193,194 Jones, U. I., 3 I , 102 Jordan, R. M., 84.99 Josefsson, E., 190,192 Journet, M.. 7 I , 72, 73, 100 Julen. G., 68, 102 Jung. G. A., 1 I , 14. 29, 33. 47, 105 Jurion, F., 163,170 Jury, K. E., 85, 102 K

Kahrein, R. B., 267, 274 Kamstra, L. D., I I , 72, 102 Kane, E. A,, 17,23,8 1,82,102 Kapelle, D., 80. 108 Karlsson, K.-E., 187,192 Karn, J. F., 1 I , 102 Karr, M. R., 53.102 Kates, K. C., 93,103 Kaufmann, R. W.. 84,105 Kay, M., 36,52, 102 Keane, E., 76,103 Keefer, R. J ., 20 I , 207,208.2 I 6 , 2 19,233 Keller, K. R., 288, 3 10,320 Kelley, E. G.. 177,192 Kellogg, Charles E., 112, 142, 149, 155, 156, 161,170, 267.272

330

AUTHOR INDEX

Kernp, A., 55,56,57,102,108 Kernp,A. W.. 84,85,105 Kemp, C. D.,49,84,85,103,105 Kendall, W. A., 64,102 Kennedy, G. S., 57,102 Kennedy, W. K., 16,17,23,105 Kenner, J., 205,233 Kertesz, Z. I., 228,233 Kewning, J. A., 54,56,99 Kilmer, V. J., 244,275 Kim, T., 291,292,293,294,295,296,298, 299, 300, 301, 302, 305, 306, 310, 317, 3 18,320,321 Kirby, K. W., 197,204,207,208,2 I I , 2 I 6, 217.2 18,220,235 Kivimae, A., 5,102 Knapp, John, 110, 170 Knight, R., 68,102 Knolek, W. F., 180,194 Knox, E. G., 239, 247, 248, 252, 256, 257, 258,259,272, 274 Koch, J. H., 63,100 Kohler, G. O., 62,104 Kononova, M. M., 204,219,233 Korner, A., 174,192 Koyama, H., 302,3 18,320 Kozak, A. S., 53,104 Krusekopf, H. H., 243,257,274 Kubota, J., 60,100 Kvist, 8.E., 228,232 1

Laby, R. H., 63,102 Lacey, J., 70,102 Lackman, D. B., 209,234 Lager, A,, 68,102 Lambourne, L. J., 59, 84, 8 5 , 86, 87, 91, 97,100,102 Lamond, D. R.,61,62,97,104 Lancaster, R. J., 33,84,IOI, 102 Landry, J., 185,193 Lang, A. L., 285, 287, 288, 289, 290, 313, 3 14,320 Lang, F., 202,233 Lang, R. W., 33, I01 Langlands, J. P., 56,84,86,88,89,91,102 Lapham, J. E., 266,274 Lapham, M. H., 266,274 Large, R. V., 92,93,94,102,106 Lauterbach,C. W.,211,235

Leavenworth, C. S., 179,193 Lees, H., 204,231 Lefevre, C. F., 11,102 Leigh, J. H., 90, 102 Leighty, R. G., 267,273 Leighty, W. J . , 267,274 Leitch, I., 5 , 102 LeMare, P. H., 159, 170 Leng, R. A,, 48, 103 Lesperance, A. L., 88,90,106 Levy-Bruhl, L., 168, I70 Lewis, D., 4 t , 9 7 Lewis, H. G., 265,273 Lewis, R. J., 246,247,248.273 Li, J. C. R., 288,3 10,320 Liener, I. E., 189,192 Lima, I. H., 177,192 Linares, F., 184,193 Lindahl, I. L., 93,103 Lindberg, B., 228,231 Lindner, H. R., 61,103 Lindstrom, G. R., 198,232 Linton, J. H., 65,103 Little, C. O., 8 I , 103 Lloyd, L. E., 4,29,35,36,38,99 Lofgreen, G. P., 74,100 Logan, V. S., 5,105 Lombard, P. E., 87,103 Loosii, J. K., 53, 104 Loper, G. M., 62,103 Love,T. R., 267,274 Lowe, C. C., 18,103 Lowe, L. E., 222,233 Ludwig, T. C., 60,101 Lugg,J. W. H., 177,192 Lusk, E., 261,273 Lusk, J. W., 12,103 Lyford, W. H., 245, 246, 250, 254, 259, 260,262,264,273,274 Lynch, D. L., 197,200,202,204,207,209, 211,221,231,232,233,235 Lynch, J. J.. 56,102 M

McArthur, J. M., 63,64,103 McCalla, T. M., 196,233 McCarrick, R. B., 51,76,103 McClure, K. E., 52,53,79,102,106 McCracken, R.J., 238, 242,243.246.247,

AUTHOR INDEX

248, 249, 250, 251. 252. 258, 259, 261, 273,274 McCroskey, J . E., 73, 103 McCullough, M. E.. 28. 45, 46, 47, 76, 99, 103 McDonald, A. N. C., 82.97 McDonald, I., 5,83,85,91,99,101,102 McDonald, I . W.. 41, 49, 50, 52, 55,56,58, 59,60,61,8 1,91,103 McDonald, P., 74,75,103, 105 McDougall, B. M., 220,234 McGowan, M., 13, 15.97 McGuire, R.L., 81,103 Mcllroy, R.J., 24,47,103 McKeague, J. A., 257,258,274 McLaren, A. D., 220,232 MacLean, D. W., 246,262,264,274 MacLeod, N. A,, 36,52,102 McLuhan, M., 168,170 MacLusky, D. S., 92,103 McManus, W. R.,84,87,103 McMeekan, C. P., 94,103 McMichael, S. C., 190,192 McNaughton, M. J., 62,105 McNutt, E. J., 267,274 Macpherson, A., 5 8 , 103 MacPherson, H. T., 75,103 Maguire, M. F., 49,104 Mahler, H. R., 174,192 Malgren, R.C., 167,169 Marbut,C. F., 257,266,267,274 Margoliash, E., 178,192 Marshall, R. D., 204, 223, 224, 225, 232, 233,234 Marshall, S. B. M., 50.5 I , 97 Marten. G. C., 32,84,92,99, 103 Martin, C. M., 84,105 Martin, J. K., 205,233 Martin, J. P., 196, I97,2 19,220.22 I , 233 Martin, Kirk, 110, 170 Martz, F. A., 53.98 Mathieson, J. M., 229,234 Matthews, E. D., 267,274, 276 May, P. F., 87,101 Mayaudon, J., 2 19.233 Maynard, L. A., 7,90,99 Mead, R.,3 18,320 Meek, D. C., 106 Mehta, N. C., 196, 197, 198, 203, 204,205,

33 1

208, 211, 212, 216, 217, 232, 233, 234, 235 Mellin, T. N., 17,103 Melville, J., 2,103 Melvin, J. F., 70,103 Mendel, L., 179,193 Mercer, F. V., 220,231 Meredith, W. R., 18,103 Merilan, C. P., 53.98 Mertz, E. T., 173, 178, 180, 181, 182, 183, 188,191,192,193 Methvin, C., 267,273 Meyer, J. H., 74,100 Meyer, R. M., 73,107 Meyers, S. M., 184,192 Miles, D. G., 29,103 Miles, J. T., 12,103 Milford, R., 17, 18, 19, 20, 28, 29, 36, 66, 103,104 Milfred, C. J., 263,274 Miller, F. P., 248,249,250.25 1,274 Miller, P. A., 190,193 Miller, R. W., 180,194,257,258,274 Miller, T. B., 7,9,103 Millett, M. A., 225,234 Millington, A. J., 62,100 Miltimore, J. E., 63,64,103 Minson, D. J., 5 , 7, 17, 18, 19, 20, 24, 28, 29, 36, 38, 49, 66, 71, 72, 81, 82,85,10I, 102,103,104,105 Mitchell, R. L., 55,58,99, 104,225,234 Mitscherlich, E. A., 291,320 Moe, P. W., 13, 39, 104 Moggridge, R. C. G., 225,233 Mohan, V. S., 189, 193 Moir, R. J., 37, 59, 72, 100, 104 Moisio, T., 65, 101 Mollenhauer, H. H., 220, 233 Monod, J.. 176, 177, 192 Montgomery, E., 305, 320 Montgomery, R., 223, 224,235 Moore, L. A., 7, 8, 17, 23, 40, 71, 74, 75, 76,8 I , 82,100, 102, 104,107 Moore, R. M., 63,104 Moore, W. E., 225, 234 Morley, F. W. H., 62, 104 Morre, D. J., 220, 233 Morrison, F. B., 4, 104 Mortensen, J. L.. 201, 202, 207, 208, 209,

AUTHOR INDEX

332

210, 211, 212, 214, 216, 217, 218, 219, 233, 235 Morton, R. K., 175, I 9 3 Moseman, A. H., 119, 170 Mosse, J., 179, 185, 186, 187, 188, 193 Mott, G. O., 11, 29, 30, 31, 94, 99, 104 Mottershead, B. E., 29, 31, 106 Moule, G. R., 61, 104 Moureaux, T., 185,193 Mowat, D. N., 22, 68, 104 Moxon, A. L., 11,105 Mueller, 0. P., 262, 263, 264, 274 Mulham, W. E., 90, 102 Muller, M., 197, 198, 204, 208. 21 I , 212, 216,217,233, 235 Muller-Vonmoos, M., 205, 234 Munck, L., 173,193 Mundie, C. M., 200, 225,231 Munro, H. N., 173, 193 Murdoch, J. C., 37, 77, 9 9 Murphy, R. P., 16, 17, 23, 105 Murphy, W. E., 54, 56, 100 Murray, M. G., 48,98 Murray, S., 14, 105 Myrdal, G., 129, 170 N

Naga, M. M. A., 11,104 Nagarajan, V., 189,193 Naismith, W. E. F., 188, I 9 3 Nance, W. E., 177,193 Nash, M., 168,170 Nash, M. J., 5 1,70,74,76,79,80,107 Neeley, J. A., 242,243, 245,250,274, 278, 279 Neely, W. B., 229, 233 Nehring, K., 189, 193 Nelder, J. A., 292, 299, 300, 301, 302, 303, 306,310, 311, 312,320 Nelson, A. B., 87, 98 Nelson, 0. E., 173, 180, 181, 182, 183, 185, 193 Nettleton, W. D., 238, 242, 243, 246, 247, 248, 249, 250, 251, 252, 254, 258, 259, 26 1,273, 274 Neuberger, A., 204, 223, 224, 225, 232, 233,234 Neumark, H., 75,104

Newton, J. E., 62,104 Newton-Hearn, P. A., 2 1 1,232 Nijkamp, H. J., 8,43,98,100 Nikiforoff, C. C., 248, 249, 25 I , 252, 254, 256, 257, 259, 260,274 Noller, C. H., 14, 104 Norman, A. G., 7,104 Northcote, D. H., 2 I I , 225, 228, 231, 233, 234 Norton, H. W., 53, 102 Nye, P. H., 156,170 0

Oades, J. M., 196, 197,200,201,202,203, 204, 205, 206, 207, 208, 209, 210, 211, 213, 214, 215, 216, 217, 218, 220, 222, 226,232,234,235 O’Donovan, P. B., 29, 30,31, 104 Ogawa, H., 291, 292, 293, 294, 302, 3 17, 3 18,320 Ogston, A. G., 210, 218, 234 Okamoto, M., 75, 76, 107 Olfield, J. E., 62, 104 Olney, H. O., 200,202,207,2 11,233 Olson, G. W., 238,246, 247,248,25 1,252, 254, 255, 259, 260, 262, 263, 264, 265, 273, 274 Oltjen, R. R., 53, 104 O’Neal, A. M., 255, 274 Ongun, A., 177, 193 Oram, L. R. N., 63,104 Orth, A,, 77, 104 Ory, R. L., 178, 180,191, Osborne, T. B., 178, 179,193 Osbourn, D. F., 29, 30, 32, 34, 36, 77, 78, 101, 104 Oser, B. L., 189, 193 O’Shea, J., 13, 49, 104 Oslage, H. J., 80, 106 O’Sullivan, M., 57, 104 Overend, W. G., 224, 234 Owen, F. G., 71, 101 Owen, J. B., 84, 104 P

Packett, L. V., 29, 30, 31, I04 Pacsu, E., 235 Page, H. J., 113, 148, 169

AUTHOR INDEX

Paladines, 0. L., 73, 104 Parsons, J. W., 200,203,207,209,2 I I , 2 I 4, 216, 217, 222,234 Pauker, G. L., 169, 170 Paul, E. A., 197, 199, 204, 207, 209, 21 I , 230 Paulson, G. D., 60, 104 Pearce, G. R., 86, 104 Pearce, R. H., 229, 234 Pearson, C. S., 268, 274 Pelzer, H., 220, 235 Pendleton, J. W., 285, 287, 288, 289, 290, 309, 310, 313, 314, 315,320 Pendleton, R. F., 275, 276 Penick, M., 314, 320 Penny, D. H., 129,170 Percival, E., 225, 234 Perera, B. P. M., 209, 234 Perez, C. B., 53,104 Perkins, S. O., 266,274 Perold, 1. S., 5 8 , 107 Perry, M. B., 225.234 Peters, J. E., 180, 194 Peterson, P. J., 60, 105 Pettiet, J. V., 247, 249, 25 1, 255, 26 I , 274 Pfost, H. B., 73, 107 Phillips, P. H.. 73. 108 Pickett, R. A., 183, 184, 192, 193 Pigden, W. J., 5 . 22, 28, 19, 38, 71, 101, I05 Pigman, W., 224,228,234 Pilgrim, A. F., 48,107 Pinkard, R. W., 196,233 Piva,G., 183,193 Platt, D., 228,234 Playne, M. J., 74,105 Plice, M. J., 32,105 Plumlee, M. P., 29, 30.3 I , 104 Pope, A. L..60,73,104,108 Pope, G . S., 62, I05 Pope, L. S., 73,87,98,103 Porter, H. C., 275,276 Poulton, B. R., 17,103 Pradilla, A., 184,192, I 9 3 Pratt, A. D., 3 1,32,99 Prestes, P. J., 14,104 Presthegge, K., 74,75,105 Pritchard, G. I., 5 , 22, 28, 29, 71, 101, 105

333

Proffitt, W. H., 267, 274 Puckridge, D. W., 3 15, 321 Putter, J., 281, 321 Q

Quicke, G. V., 1 I , 105 Quirk, J . P., 197, 198,231,232,235 R

Rae, A. L., 92,105 Rahman, S., 228,231 Raison, J. K., 175,193 Raymond, W. F., 2, 3, 5 , 13, 14, IS, 17, 18, 19, 20, 22, 24, 27, 28, 29, 31, 38,44,45, 48, 66, 68, 69, 74, 75, 76, 79, 80, 81, 82, 84, 85, 86, 87, 94, 95, 96, 99, 101, 103, 104, 105 Reardon, T. F., 85, 86, 91, 102 Reddy, M. C., 6 5 , 1 0 0 Redmond, C. E., 242,275 Reed, W. D. C., 25, 3 7 , 4 5 , 1 0 7 Rees, C. W., 224,234 Reestman, A. J., 307,321 Reichstein, T., 224, 234 Reid, C. S. W., 63, 105 Reid, G. W., 33, 91, 92, 101, 102 Reid, J. T., 13, 16, 17, 18, 23, 39, 73, 80, 8l,84,103,104,105 Reid, R. L., 11, 12, 14, 33, 37, 45, 47, 98, 105, 106 Reid, R. S., 48, 98 Reissig, H., 189, 193 Reith, J. W. S., 58, 104, 105 Rennie, D. A., 197, 199,204,207,209,2 I 1, 2 18,230,234 Reynolds, J . D., 315, 316, 321 Rhykerd, C. L., 14, 104 Rhyne, C. L., 190, 193 Richards, C. R., 84, 106 Richards, F. J., 299, 302, 321 Richards, G. N., 205,233 Richards, S. J., 196, 197, 220, 233 Ridley, J. R., 88, 90, 106 Riewe, M. E., 94, 106 Ritossa, F. M., 176, 193 Robards, G. E., 87, 106 Roberts, R. C., 188, 193, 244,275 Roe, R., 29, 3 1, 106

334

AUTHOR INDEX

Rogers, H. H., 66, 68, 106 Rogers, H. J., 220, 223,234 Rogler, J. C., 185, 192 Rohr, K., 46,47, 107 Ronning, M., 72, 106 Rook, J. A. F., 41,44,45,76,98,106 Rose, R. C., 266, 275 Rosenfeld, I., 59, 106 Rossiter, R. C., 62, 106 Roulet, N., 205, 209, 2 12, 234 Rovira, A. D., 220,22 1,234 Rowland, S. J., 47, 97 Rudman, J. E., 20, 106 Rumsey, T. S., 14, 104 Russell, E. W., 219, 234 Rutledge, E. M., 249, 250, 251, 253, 256, 258, 260,273,275 S

Sadgopal, A., 174,193 Saeman, J. F., 225,234 Salam, A., 204, 205, 235 Salamini, F., 183, 193 Salomon, M., 199, 209, 234 Salton, M. R. J., 220, 234 Samtsevich, S. A., 219, 220, 234 Sandegren, E., 178, I 9 1 Sanderson, G. W., 209,234 Santi, E., 183,193 Sarria, D., 184, 192 Sastry, L. V. S., 187, 194 Satter, L. D., 50, I06 Saunt, J., 284, 321 Sawers, D., 5 1, 70, I00 Schaadt, H., 52, 53, 106 Schillinger, J. A., 25, 106 Schmid, K., 223,234 Schneider, 9. H., 5, 106 Scholl, J. M., 5 , 13, 97 Schukking, S., 80, 108 Schultz, I. W., 170 Schulz, E., 80, 106 Schuphan, W., 173,193 Schweet, R., 174, 193 Schwendinger, R. B., 201, 202, 208, 210, 218,233 Schwerdtfeger, E., 189, 1 9 3 Scott, F. M., 220,234

Scott, H. W., 11, 16, 100, 105 Scott, J. E., 2 1 I , 228, 234 Scragg, W., 315, 316,320 Scrivner, C. L., 243,248,249,260,275 Seay, W. A., 246,247,248,273 Seif, R. D., 3 15,320 Sevag, M. G., 209,234 Sequeira, J. S., 224,234 ShalTer, M. E., 249,260,275 Shafizadeh, F., 224,234 Shantz, H. L., 244, 275 Sharpe, P. R., 288, 309,321 Shaw, J. C., 40,104 Shaw, K., 221,231 Shearin, A. E., 268, 275 Shefner, A. M., 226,231 Shelton, D. C., 11, 12, 105, 106 Shepherd, R. A., 196, 197. 220, 221,233 Sheppard, D. E., 190, 193 Shepperson, G., 70, 106 Shibles, R. M., 315, 321 Shinozaki, N., 292, 293,294,295,296,298, 299, 300, 301, 302, 305, 306, 317, 318, 320,321 Shorland, F. B., 92, 99 Shumard, R. F., 13,106 Simmonds, R. G., 202,203, 207, 208, 210, 21 I , 2 12,2 13,214,2 16,217,232 Simonart, P., 219,233,235 Simonson, R. W., 237, 238,275 Sinclair, D. P., 33, 101 Siu, R. G . H., 21 9,235 Slack, S. T., 16, 17, 23, 105 Sly, D. A., 188, 194 Slyter, L. L., 53, 104 Smart, C. L., 227,235 Smart, V. W., 190, 193 Smart, W. W. G., 25, 45, 46, 47, 203,106 Smith, A. K., 188, 194 Smith, C. A., 19, 24, 106 Smith, C. R., Jr., 178, 193 Smith, E. L., 178, 192 Smith, F., 223, 224, 235 Smith, F. H.,190, 193 Smith, G. D., 261, 267,275 Smith, H. C., 266, 267, 273, 275 Smith, J. C., 94, 106 Smith, R. M., 252, 255, 256, 262, 265,275

335

AUTHOR INDEX

Smolens, J., 209,234 Snow, C. P., 168,170 Sonneveld, A,, 33, 106 Sowden, F. J., 196, 200, 202, 208, 225, 232, 233, 235 Spaeth, J. N., 254,275 Spedding, C. R. W., 3, 13, 24, 92, 93, 94, 102, 105,106 Spedding, D. J., 60, 105 Spiegelman, S., 176, 193 Spiro, R. G., 223, 225, 235 Sprague, G. F., 180, 192 Srikantia, S. G . , 171, 192 Stacey, M.,196, 202, 203, 207, 208, 210, 211, 212, 213, 214, 216, 217, 220, 221, 223,231, 232, 233, 235 Stahmann, M. A,, 172, 177, 192, 193 Stanley, N. W., 25, 106 Stephen, I., 248, 25 1, 273 Stephens, D. F., 73, 103 Stern, W. R., 3 17,321 Stevenson, F. J.. 204,205,231 Stevenson, 1. L., 219, 235 Stewart, D. R. M., 91, 106 Stobbe, P., 200, 232 Stocking, C. R., 177, 193 Stoddart, J. F., 228,231 Stojanovic, B. J., 197, 198, 235 Stout, P. R., 56, 57, 101, 106 Streeter, C. L., 87, 101 Streuli, H., 197, I98,2 I I , 233,235 Sullivan, J. T.. 6, 7, 9, 98, 106 Swaby, R. J., 196,235 Swift, R. W., 106 Swincer, G. D., 196, 197, 200, 201, 202, 203, 204, 205, 206, 207, 208, 209, 210, 211, 213, 214, 215, 216, 217, 218, 220, 222,234, 235 Sykes, J. F., 74, 75, 76, 104, 107 Synge,R. L. M., 50, 51,97, 106 Szirmai, J . A., 229, 231 T

Tadmor, A., 75, 104 Tavernier, R., 26 1 , 275 Tayler, J. C.. 19.20,88.91,106 Taylor, D. C., 268, 275 Taylor, M. W., I I, 98

Tello, F., 189, 193 Terry, R. A., 8, 12, 14, 15, 19, 20, 21, 22, 25, 26, 29, 30, 31, 32, 34, 36, 37,42,45, 4 8 , 6 8 , 7 1,99, 104, 105, 106, 107 Tesar, M.,3,35,102 Theron, E. P., 34,107 Thomas, A. E., 242, 243, 267, 274, 275, 277,278 Thomas, C. J., 265,275 Thomas, G. D., 73,107 Thomas, J. W., 3, 35, 14, 75, 76, 102, 104, 107 Thomas, R. L., 200, 201, 207, 208, 209, 2 1 I, 2 12,214,2 16,2 17,235 Thompson, A., 224,235 Thompson, R., 285,286,300,307,'320 Thomson, D. J., 29, 30, 32, 34, 36, 42, 47, 104, 107 Thorne, J. L., 90, 99 Thornton, R. F., 24, 107 Thorp, J., 267,272, 275 Threlkeld, G., 275, 278 Tilley, J. M. A., 8, 12, 14, 15, 19,20,21,22, 25, 26, 3 I , 36, 37, 42, 45, 68, 7 I , 99, 106, 107 Tillman, A. D., 53,99,106 Tinsley, J., 200, 203, 204, 205, 207, 209, 21 I , 2 14,2 16,217,220,222,234,235 Tobin, J., 76,103 Toogood, J. A,, 197,235 Tookey, H. L., 63,108 Topps, J. H., 25,37,45,107 Torell, D. T., 87,90,97,98,107 Torriani, A., 177, I92 Tossell, W. E., 22, 104 Trimberger, G. W., 16, 17, 23, 105 Troelsen, J. E., 14, 24, 34, 102, 107 Truog, E., 197, 199, 204, 209, 218, 234 Tuckett, S. E., 184, 192 Turk, K. L., 16, 17, 23, 105 Turner, J. H., 93, 103 Tyler, M. C., 267, 274 Tyrrell, H. F., 13, 39, 104 U

Ulyatt, M. J., 47, 76, 98, 107 Underwood, E. J., 54, 57, 5 8 , 107

336

AUTHOR INDEX

V

Vanden Berg, G. E., 25 1, 273 Vanderford, H. B., 249, 260,275 Van der Merwe, F. J., 13, 15, 58, 100, 107 Van Dyne, G. M., 8 7 , 9 0 , 1 0 7 van Es, A. J. H., 23, 24, 43, 98, I00 VanEtten, C. H., 180,194 Van Niekerk, B. D. H., 73, 104 van Schalkwyk, A., 87,103 Van Soest, P. J., 6, 8, 9, 10, 13, 23, 24, 25, 29, 30, 32, 34, 49, 100, 107 Vercoe, J. E., 86, 104 Veron, 0. A., 181, 183, 185, 193, 194 Vetter, R. L., 13, 22, 107 Viets, F. G., 283, 320 Vinas, E., 180, 194 Virtanen, A. I., 53, 107 Virupaksha, T. K., 187, 194 Vogel, H. J., 176, 194 Vogel, R. H., 176, 194 von Kaufmann, W., 46, 47, 77, 104, 107

W

Wainman, F. W., 27, 28, 32, 43, 98 Waite, R., 39, 97 Walborg, E. F., 224, 235 Walker, J. L., 248,273 Walker, T., 36, 52, 102 Waller, G., 73, 103 Walshe, M. J., 94, 103 Walters, R. J. K., 22, 29, 55, 97, 103, 107 Ward, D. N., 224,235 Ward, G., 65, I00 Wardrop, I. D., 60, 9 7 Warne, L. G. G., 284, 292, 293, 294, 321 Warner, A. C. I., 41, 48, 107 Warner, R. G.,53,77,104,107 Watkin, B. R., 87,101 Watson,J. H., 197, 198,235 Watson, J. N., 18, 96, 99 Watson, S. J., 51, 70, 74, 76, 79, 80, 107 Watt, J. A., 75, 103 Watts-Tobin, R. J., 175, 192 Waugaman, S. N., 196, 197,231 Weaver, H. S., 84, 106 Webber, J. M., 223, 228,231

Webber, L. R., 197,235 Weber, C. R., 315,321 Webley, D. M., 196, 197, 219, 232, 235 Wedin, W. F., 13, 22, 107 Weed, S. B., 25 1,274 Weenink, R. O., 63, 102 Weidel, W., 220, 235 Weir, W. C., 87, 90, 97 Weiss, E., 224, 234 Welch, J. A., I I , 105 Weller, R. A,, 48, 107 Weston, R. H., 29, 36, 48, 1 0 7 Whelan, W. J., 224, 235 Whistler, R. L., 197, 198, 204, 205, 207, 208, 211, 216, 217, 218, 220, 227, 228, 235 White, H. S., 180, 194 White, J. L., 248, 252, 258,272 White, P. L., 180, 194 Whitehead, D. C., 54, 108, 200, 203, 207, 209, 220,235 Whitelaw, F. G., 48, 98 Whiteside, E. P., 247, 248, 249, 250, 251, 253,255,256,257,260,261,275 Whitmore, E. T., 68, 106 Whitmore, G. E., 93, 103 Whittig, L. D., 244, 275 Whybrew, J. E., 315, 316,320 Wichser, W. R., 184, 194 Wieringa, G. W., 80, 108 Wiggans, R. G., 3 15, 3 16,321 Wilcox, 0. W., 288,321 Wilkins, R. J., 89, 108 Willey, R. W., 285, 307, 308, 31 I , 312,321 Williams, 9. G., 197, 235 Williams, E. E., 53, 104 Williams, J. D., 63, 104 Williams, L. D., 267, 274 Williams, V. J., 81, 99 Wilson, A. D., 87, 106 Wilson, C. M., 175, I 9 4 Wilson, I. A. N., 93, 106 Wilson, R. F., 76, 80, 101 Wilson, R. K., 7, 13, 99, 104 Wilson, R. S., 27, 28, 31. 32, 37, 98 Winch, J. E., 22, 104 Wind, J., 57, 108 Winogradzsky, S., 196,235

337

AUTHOR INDEX

Winter, K. A., 5 , 105 Winters, E., 237, 238, 257, 265,275 Winzler, R. J., 223, 235 Wiseman, H. G., 74, 100 Wolf, W. J., 188, 194 Wolff, I. A., 178, 180, 193, 194 Wolfrom, M. L., 204, 224, 227, 228, 229, 233, 235 Woods, A. E., 65, 108 Woodward. F. N.. 202, 207, 208, 209,231 Woolfolk, P. G., 84, 105 Worker, N . A., 6 5 , 108 Wright, L. M., 200, 202, 207, 2 I I , 22 I , 233 Wright, M. J., 64,108 Wright, P. L., 73,108 Wyatt, C. E., 267,274

Y

Yamamura, Y ., 184, I 9 2 Yanofsky, C., 174,192 Yaron, D., 281,321 Yassoglou, N . J . , 247, 248, 249, 250, 25 1, 253,255,256,257,260,261,275 Yates, N . G., 24, 63, 68, 102, 107, 108 Yatsu, L. Y., 178, 180, 191 Yoda, K., 302,320 Young, M. C., 90, 99 Z

Zimmerman, R. C., 255, 256, 273 Zon, R.,244,275 Zuckerkandl, E., 177, 194 Zweifel, G., 197, 204, 208, 216, 218, 232

Subject Index

A

D

Dactylis glomerata, 17 Daucus carota, 180 Digitaria decumbens, 36

Alfalfa, 156 nutritive value, 7,45,46 Alfisol, 245 Alluvial soil, 152-153 Aluminum, 258 Avena sativa, 179 Azotobacter indicus, 220 Azuki bean, 291

E

Ensilage, 74-80 Estrogenic compounds, forage crops, 6 1-63 Feed, digestibility coefficient, 4 supplements, 25-27,37-38 voluntary intake, 27-38,7 I , 74-79 Festuca arundinacea, 17, 29, 33 Festuca pratense, 17, 5 5 Forage crop, nutritive value, 1-108 Fragipans, 237-279

6

Bacteria, soil, 2 19 Barley, 179, 187, 188,298 Birdsfoot trevoil, 34,64 Brassica campestris, 190 Bromus inermis, 32

G C

Calcium, 5 5 , 57, 65, 149, 155, 161, 245 Carrot, 3 18 Cellulose, nutritive value, 8, 13, 14,77,89-90 soil, 197,219 Chenopodium pallidiculae, 180 Chenopodium quinoa, 180 Chernozem soil, 145-146 Chestnut soil, 146 Chitin, 221 Chlorophora excelsia, 149 Chromobacterium violaceum, 220 Clay, 198,247 Clover, 16,89 Cobalt, 57,63 Cocksfoot, nutritive value, 17, 18, 20-21, 23, 24, 32, 68 Copper, 54, 58 Corn, 179 see also maize Cornilla varia, 64 Corn silage, 46, 79 Cottonseed meal, 190 Crambe abyssinica, 180 Crownvetch, 64

Genetics, plant protein, 17 I - 194 GIycine max, 180 H

Hordeum vulgare, I79 I

lmperata cylindrica, I58 Inceptisol, 245 Iodine, 59 Iron, 247, 258 1

Ladino clover, 65 Lathyrus sativus, 189 Latosols, 148-151, 155 Lead, 58-59 Leaf proteins, 177-178 Legume, 180, 189 nutritive value, 12, 34, 5 1, 5 5 , 56, 62, 7 1, 74 Lespedeza cuneata, 25 Lignin, 7, 8, 13, 14, 89 Lithosols, 148, 152, 153 L o h m spp., 17, 18 Lotus corniculatus, 34, 64

338

SUBJECT INDEX

Lucerne, nutritive value, 17, 22,25, 30, 34. 36,53,62,63,7 1-72

M

Magnesium, 56-57. 247 Maize, see also corn, 22, 179, 180-187, 188.285 Marah gilensis, I80 Marrow-stem kale, 22 Meadow fescue, 55 Medicago sativa, 17, 20 Melilorus alba, 65 Millet, 179, 188 Molybdenum, 58 Montrnorillonite, 246, 248 N

Narrow-stem kale, 22 Nitrate, 64-65 Nitrogen, 15, 20, 23, 24, 32, 36-37, 47, 50-5 I , 56 0

Oats, 179, 187, 298 Oat straw, 36 Onobrychis vicifolia, 22 Oryza sativa, 179 P

Pangola grass, 36 Panicum milliaceum, 179 Pennisetum purpureum, 156 Phalaris arundinacea, 29, 3 I , 34 Phalaris tuberosa, 3 1, 63, 69 Phaseolus chrysanthus, 29 I Phelum nodosum, 29 Phleum pratense, 17, 29, 55 Phosphate, 65 Phosphorus, 55, 155, 159 Planosol, 267 Plant density, yield, 281-32 1 Plant protein, genetics, 171- I94 Poa pratensis, 55 Podzol, 145-148, 151, 268 Potassium, 56 Potato, 189, 304

339 R

Red clover, 22, 34, 65 Reddish Chestnut soil, 145-146 Regosols, 149, 153 Rice, 179, 187 Rye, 179, 188 Rye grass, nutritive value, 7, 17, 18, 2021, 22, 23, 24, 30.43.47, 54, 55,65, 66, 70, 89, 93 S

Sainfoin, 22, 34 Secale cereale, 179 Seed, storage proteins, 178- 180 Selenium, 59-60 Sericea lespedem, 25 Sierozem soil, 146- 147 Silica, 10-1 1, 60-61, 257 Sodium, 55 Solannum malacoxylon, 65 Soil, classification, 265-269 erosion control, 138-140 fragipans, 237-279 group descriptions, 145-1 53 horizons, 240-242 polysaccharides, 195-235 potentially arable, 109-170 surveys, 123- 130 use, 112-122, 156-163 Sorghum, 179, 187 Sorghum vulgare, 179 Soybean, 180, 188, 3 16, 3 I8 Spodosol, 245, 269 Subterranean clover, 6 1-62, 69, 293 Sulfur, 52, 59 Sweet clover, 65, 69, 156 7

Tall fescue, 17, 18, 29, 63 Terra Rossa soil, 151-152 Timothy, nutritive value, 16, 17, 29, 30, 55

Trqolium pratense, 22 Trfolium repens, 17 Trifolium subterraneum, 63 Triticum vulgare, 179, 187

SUBJECT INDEX

Tropical soil, 156- I63 Turnip, 3 10 U

control, 130-135 fragipans, 254-256, 258, 264 Wheat, 179, 188, 285, 31 1 White clover, nutritive value, 17, 47, 65

Ultisol, 245 Urea, 26, 36, 5 1, 52, 53 V

Y

Yield, plant density, 281-321

Vetch, 25 2 W

Water, 24

Zea mays, 22, 179 Zein, 179, 180-187

Advances in Agronomy, Volume 21

Advances in Agronomy, Volume 60

Advances in Agronomy, Volume 50

Advances in Agronomy, Volume 68

Advances in Agronomy, Volume 54

Advances in Agronomy, Volume 11

Advances in Agronomy, Volume 31

Advances in Agronomy, Volume 33

Advances in Agronomy, Volume 65

Advances in Agronomy, Volume 64

Advances in Agronomy, Volume 105

Advances in Agronomy, Volume 20

Advances in Agronomy, Volume 100

Advances in Agronomy, Volume 108

Advances in Agronomy, Volume 69

Advances in Agronomy, Volume 83

Advances in Agronomy, Volume 91

Advances in Agronomy, Volume 46

Advances in Agronomy, Volume 111

Advances in Agronomy, Volume 62

Advances in Agronomy, Volume 56

Advances in Agronomy, Volume 28

Advances in Agronomy, Volume 53

Advances in Agronomy, Volume 14

Advances in Agronomy, Volume 26

Advances in Agronomy, Volume 103

Advances in Agronomy, Volume 81

Advances in Agronomy, Volume 40

Advances in Agronomy, Volume 45

Advances in Agronomy, Volume 101

Advances in Agronomy, Volume 85

Advances in Agronomy, Volume 21

Advances in Agronomy, Volume 60

Advances in Agronomy, Volume 50

Advances in Agronomy, Volume 68

Advances in Agronomy, Volume 54

Advances in Agronomy, Volume 11

Advances in Agronomy, Volume 31

Advances in Agronomy, Volume 33

Advances in Agronomy, Volume 65

Advances in Agronomy, Volume 64

Advances in Agronomy, Volume 105

Advances in Agronomy, Volume 20

Advances in Agronomy, Volume 100

Advances in Agronomy, Volume 108

Advances in Agronomy, Volume 69

Advances in Agronomy, Volume 83

Advances in Agronomy, Volume 91

Advances in Agronomy, Volume 46

Advances in Agronomy, Volume 111

Advances in Agronomy, Volume 62

Advances in Agronomy, Volume 56

Advances in Agronomy, Volume 28

Advances in Agronomy, Volume 53

Advances in Agronomy, Volume 14

Advances in Agronomy, Volume 26

Advances in Agronomy, Volume 103

Advances in Agronomy, Volume 81

Advances in Agronomy, Volume 40

Advances in Agronomy, Volume 45

Advances in Agronomy, Volume 101

Advances in Agronomy, Volume 85

Recommend Documents