Advances in Agronomy, Volume 51

Advisory Board Martin Alexander Eugene J. Kamprath Cornell University North Carolina State University Kenneth J. Fr...

Author: Donald L. Sparks

16 downloads 1019 Views 11MB 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

Advisory Board Martin Alexander

Eugene J. Kamprath

Cornell University

North Carolina State University

Kenneth J. Frey

Larry P. Wilding

Iowa State University

Texas A&M University

Prepared in cooperation with the

American Society of Agronomy Monographs Committee S. H. Anderson P. S. Baenziger W. T. Frankenberger,Jr.

M. A. Tabatabai, Chairman D. M. Kral S. E. Lingle R. J. Luxmoore

G. A. Peterson S. R. Yates

S I N

VOLUME 51 Edited by

Donald L. Sparks Department of Plant and Soil Sciences University of Delaware Newark, Delaware

ACADEMIC PRESS, INC. A Division of Harcourt Brace & Company San Diego New York Boston London Sydney Tokyo Toronto

This book is printed on acid-free paper. @

Copyright 0 1993 by ACADEMIC PRESS,INC. All Rights Reserved. No part of this publication may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopy, recording, or any information storage and retrieval system, without permission in writing from the publisher.

Academic Press, Inc.

525 B Street, Suite 1900. San Diego, California 92101-4495 United Kingdom Edifion published by

Academic Press Limited 2 4 2 8 Oval Road, London NW 1 7DX

International Standard Serial Number: 0065-21 13 International Standard Book Number: 0-12-000751-7

PRINTED IN THE UNITED STATES OF AMERICA 9 3 9 4 9 5 9 6 9 1 9 8

QW

9 8 1 6 5 4 3 2 1

Contents CONTRIBUTORS ........................................................ PREFACE ..............................................................

vii ix

GYPSUM AND ACIDSOILS: cz\HE WORLD SCENE Malcolm E. Sumner 1. 11. 111. N. V.

Introduction .................................................... Composition and Properties of Gypsum Materials ................ Gypsum as an Ameliorant for Acid Subsoils ....................... Gypsum as an Ameliorant for Soil Physical Properties ............. Conclusions ..................................................... References ......................................................

CONSERVATION TILLAGE: AN ECOLOGICAL APPROACH T O SOIL MANAGEMENT R. L. Blevins and W. W. Frye I. Introduction .................................................... 11. Conservation Tillage and the Environment ....................... 111. Soil Physical Properties .......................................... N. Soil Chemical Properties ........................................ V. Surface Mulch Management ..................................... VI. Nutrient Management ........................................... VII. Pest Management ............................................... VIII. Conclusions ..................................................... References ......................................................

1 2 3 17 26 27

34 38

45 52 57 65 69 72 73

TRANSPOSABLE ELEMENTS IN MAIZE:

THEIR

ROLEIN CREATING PLANTGENETIC VARIABILITY Peter A. Peterson

I. Introduction: The Heterogeneity Question ....................... 11. Maize Breeding Accomplishments: What the Plant Breeder Has Wrought ................................................... 111. Transposable Elements .......................................... References ......................................................

79 80 83 1 18

CONTENTS

vi

CONCEPTS AND DIRECTIONS INARTHROPODPESTMANAGEMENT Joe Funderburk. Leon Higley. and G. David Buntin I . Introduction

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

I1. Concepts of Integrated Pest Management ........................

111. Components of Integrated Pest Management ..................... Iv. Ecological and Environmental Considerations .................... V. Status and Future Directions of Integrated Pest Management ...... VI. Conclusions ..................................................... References ......................................................

126 127 130 154 158 164 166

ACCUMULATION OF CADMIUM IN CROP PLANTS AND ITS CONSEQUENCES TO HUMAN HEALTH George J . Wagner

I . Introduction .................................................... I1. Occurrence of Cadmium in the Agricultural Environment ......... 111. Accumulation of Cadmium: Whole-Plant Studies ................. Iv. Accumulation of Cadmium: Cellular and Subcellular Aspects ...... V. Contribution of Agricultural Products to Human Intake of Cadmium .................................................... VI. Health Consequences of Cadmium Intake in Humans ............. VII . Attempts to Manipulate Plants to Reduce the Cadmium Content of Crops: Promise and Problems .................................... VIII . Summary ....................................................... References ......................................................

201 204 205

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

213

173 174 179 184

193 196

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

R. L. BLEVINS (33), Department ofAgronomy, University of Kentucky, Lexington, Kentucky 40546 G. DAVID BUNTIN (1 2 5), Department of Entomology, University of Georgia, Gnfin, Georgia 30223 W. W. FRYE (33), Division of Regulatory Services, University oflyentucky, Lexington, Kentucky 40546 JOE FUNDERBURK (1 2 5), North Florida Research and Education Center, University of Florih, Quincy, Florida 32351 LEON HIGLEY (1 2 9 , Department of Entomology, University of Nebraska, Lincoln, Lincoln, Nebraska 68583 PETER A. PETERSON (79), Department of Agronomy, Iowa State University, Ames, Iowa SO01 1 MALCOLM E. SUMNER (l), Department of Crop and Soil Sciences, University of Georgia, Athens, Georgia 30602 GEORGE J. WAGNER (1 73), Plant Physiology/Biochemistry/Molecular Biology Program, Department of Agronomy, University of Kentucky, Lexington, Kentucky 40546

vii

This Page Intentionally Left Blank

Preface Volume 51 brings together some leading crop and soil scientists who have reviewed a number of important topics that are of current interest to agronomists around the world. The first chapter provides a comprehensive review of gypsum and acid soils. Topics in this chapter include the composition and properties of gypsum materials, and gypsum as an ameliorant for acid subsoils and for soil physical properties. The second chapter deals with conservation tillage with an emphasis on ecological approaches to soil management. This chapter discusses the effects of conservation tillage on the environment, soil physical and chemical properties, and surface mulch, nutrient, and pest management. The third chapter discusses transposable elements in maize and their role in creating plant genetic variability. An historical perspective on transposable elements is provided as well as comprehensive discussions on components of transposable elements, transposition, effects on gene expression, and presence in the genome. The fourth chapter provides a complete and contemporary review of integrated pest management. Topics that are discussed include concepts and components of integrated pest management, ecological and environmental considerations, and future directions in this very important area. The fifth chapter addresses various aspects of cadmium, a heavy metal of great environmental concern. The accumulation of cadmium in plants and its consequences to human health are comprehensively reviewed. All of these cutting edge reviews should be of immense interest to the readership. Many thanks to the authors for their fine contributions.

DONALD L. SPARKS

ix

This Page Intentionally Left Blank

GYPSUM AND ACIDSOILS: T H E WORLD SCENE Malcolm E. Sumner Department of Crop and Soil Sciences University of Georgia Athens, Georgia 30602

I. Introduction 11. Composition and Properties of Gypsum Materials 111. Gypsum as an Ameliorant for Acid Subsoils

A. Crop Responses to Gypsum B. Prediction of Crop Responses to Gypsum Applications C. Soil Solution Studies D. Leaching Studies with Gypsum E. Mechanisms Involved in Subsoil Acidity Amelioration IV.Gypsum as an Ameliorant for Soil Physical Properties A. Surface Crusting B. Hardsetting Soils C. Subsurface Hard Layers D. Hydraulic Conductivity V. Conclusions References

I. INTRODUCTION Gypsum is a widely occurring mineral that has been used for many years as a soil conditioner and ameliorant for sodic and heavy clay soils and as a nutrient source of Ca and S for plant growth (Shainberg et al., 1989). However, recent research has shown that the utility of gypsum can be extended to acid, infertile soils as an ameliorant for subsoil acidity, surface crusting, and subsurface hard layers and as a source of Ca for developing peanut pods. (In the context of this review, an acid soil is defined as one having a pH value below 7.0.) All of these applications have resulted in substantial improvements in crop rooting patterns and water infiltration into soils, resulting in improved yields and decreased runoff and erosion. Basically two types of gypsum, namely, mined and industrial by-product materials, are currently being used for amelioration on the acid soils of the world. 1 Adiunrrs in A p m n y . Volumt $1

Copyright 9 1993 by Academic Press. Inc. All rights of reproduction in any form reserved

2

MALCOLM E. SUMNER

The mined variant usually occurs as the dihydrate, CaS0,.2H20, although the hemihydrate (Plaster of Paris), CaSO,.JH,O, and anhydrite, CaSO, , are also found naturally (Doner and Lynn, 1989). Industrial by-product materials are derived from the manufacture of phosphoric acid (phosphogypsum, PG) or the capture of SO, from the flue gases in fossil fuel powered generators (flue gas desulfurization gypsum, FGDG) or the neutralization of sulfuric acid in many chemical processing industries. In terms of tonnages produced, PG is currently by far the most important on a worldwide basis, but with increasing pressure to improve air quality, the quantities of FGDG produced are likely to increase dramatically in the future, particularly in densely populated, industrial areas. In addition, because of the distribution of power plants throughout a particular country, FGDG is likely to have a price advantage over PG because of lower transportation costs. In the United States, the movement of PG from sites of production was temporarily banned as a result of the potential hazard from radon emissions from the material. However, this ban has now been reinstated permanently (Anonymous, 1992) despite data indicating that it is environmentally safe for use on soils (Alcardo and Recheigl, 1993).

II. COMPOSITION AND PROPERTIES OF GYPSUM MATERIALS Because the appropriateness of a particular gypsum to a given application depends on its composition and rate of dissolution, it is important to make comparisons of various sources of gypsum. In terms of total Ca content, both mined and industrial sources are similar but vary in their contents of other elements (Table I) and in particle size (Table 11) and kinetics of dissolution (Fig. 1). Phosphogypsum contains substantial quantities of P, which can be beneficial if large amounts of gypsum are applied. The levels of the other elements in the materials are so low that they are of little importance. Phosphogypsum from Florida contains 226Raand gives off radon (,,,Rn) gas, which poses a potential health hazard (Mays and Mortvedt, 1986). In one study (Miller and Sumner, 1991), phosphogypsum applied at rates of up to 10 t ha-' was found to be within tolerable limits environmentally. The Cd content was also found to be of no concern (Mays and Mortvedt, 1986). Because the speed with which gypsum dissolves is of some importance as far as its use is concerned, it is necessary to have data on particle size analysis (Table 11) and the kinetics of dissolution (Fig. 1). Bolan et al. (1992b) found that the dissolution of gypsum materials in water was a transport-limited phenomenon and followed second-order kinetics, whereas dissolution in the presence of soil was much faster and followed first-order kinetics as illustrated in Fig. 1 .

3

GYPSUM AND ACID SOILS Table I

Total Chemical Composition of a Selection of Mined and By-Product Gypsum Materials" Element Phosphogypsum Florida

Flue gas Flue gas desulfuridesulfurization gypsum zation gypsum Florida

Illinois

Major components Ca

B Mg Si A1

0.109

P Fe

0.21 I 0.082

Minor components As

Cd cu Pb Se

K

Na Sr

Nova Scotia

%

25.08 18.90 0.076 0.002 0. I16

S

Mined gypsum

24.41 18.61 0.088 0.015 0.052 0.059 0.007 0.039

24.29 18.22 0.158 0.100 0.078 0.057

O.OO0

0.053

27.20 16.00 0.042 0.466 0.054 0.328 0.003 0.220

mg kg-' 6.91 0.667 20.35

O.OO0 O.OO0

206.70 203.47 523.89

5.73 O.OO0 1.73 O.OO0 O.OO0 O.OO0 22.61 123.82

4.98 1.050 I .73

O.OO0 O.OO0

103.34 33.92 200.19

18.92

O.OO0 6.93

O.OO0 O.OO0 658.90 63.59 558.46

"C. Ishak, private communication (1991).

These results indicate that the rate of dissolution of gypsum is a function of particle size and surface area. Generally mined gypsum materials have been found to dissolve the slowest (Frenkel and Fey, 1989; Keren and Shainberg, 1981; Shainberg et al., 1989). The importance of such differences will become apparent in the discussion of the benefits of gypsum in reducing surface crusting.

111. GYPSUM As AN AMELIORANT FOR ACID SUBSOILS Acid topsoils can be readily ameliorated by the incorporation of appropriate quantities of limestone. However, in the case of acid subsoils, lime is of little

MALCOLM E. SUMNER

4

Table I1

Particle Size Analysis of a Variety of Gypsum Materials"

Particle size

Phosphogypsum

Flue gas desulfurization gypsum

Flue gas desulfurization gypsum

Mined gypsum

Florida

Florida

Illinois

Great Britain

0.05 12.30 63.41 18.06 6.12

11.99 17.35 17.33

(mm)

%

>0.25 0.25-0. I

0.05 6.70 20.20 68.35 4.10

0.1-0.05

0.05-0.02 <0.02

2.30 1.03 1.35

90.05 5.21

40.05

13.28

"From Bolan ef al. (1992b).

immediate benefit because it does not readily move down the soil profile. Deep incorporation of lime is prohibitively expensive and is unlikely to be used to any extent. Subsoil acidity is a major constraint to the successful production of many important crops (Bouldin, 1973) and there is urgent need for practical methods of ameliorating acid subsoils, particularly in those parts of the world struggling

.-c ~

n

E

- 8

X

11-

1

In Solution

In Soil

5

Y

v

-

4

0

+

: 3 0

<

0 2

c

E

0 0 0

50

100

150

200

5

/

j

0 250 Surface area-s

2

50

K.=0.0005 s"'

I

I

I

I

100

150

200

250

(m )

0

In

Figure 1. Dissolution kinetics of various gypsum materials in relation to surface area. (From Bolan er al.. 1992b.)

GYPSUM AND ACID SOILS

5

to achieve self-sufficiency in food production (Haynes, 1984). Because acid subsoils are usually deficient in Ca and contain toxic levels of Al, root proliferation below the plow layer is often severely limited resulting in great susceptibility to drought stress. Reliable estimates of the area likely to be afflicted by subsoil acidity are difficult to obtain, but conservatively the area is likely to exceed 1 billion ha (Orvedal and Ackerson, 1972). One of the possible strategies that can be used to overcome these limitations is the use of gypsum (mined and by-product) applied on or incorporated into the surface soil as originally demonstrated by Sumner (1970) and Reeve and Sumner (1972). This technology has received the greatest attention in Brazil, South Africa, and the southeastern United States, where it is currently being used on a commercial scale. Comprehensive treatises of gypsum use in agriculture have been published (Shainberg et al., 1989; van Raij, 1988; Alcardo and Recheigl, 1993) and therefore the discussion to follow will not devote much time to repeating the same material but will rather highlight the work that has been conducted subsequently. Nevertheless, the results presented by Shainberg et al. (1989) and van Raij (1988) will be briefly summarized to paint the complete picture.

A. CROPRESPONSESTO GYPSUM Substantial and highly significant yield responses (7-200%) in a number of important crops have been obtained in experiments conducted in Brazil, South Africa, and the southeastern United States to applications of both mined and byproduct gypsums applied at rates of I - 10 t ha- ' on the soil surface or incorporated into the plow layer. The crops studied include maize (Carvalho et al., 1986; Ernani et al., 1992; Farina and Channon, 1988; Hammel et al.. 1985; Malavolta et al., 1986; Quaggio, 1992; Shainberg et al., 1989; Sousa and Ritchey. 1986; Sousa et al., 1992b), soybeans (Carvalho et al., 1986; Hammel et al., 1985; Quaggio, 1992; Sousa er al., 1992b; Vitti et al., 1986), sugarcane (Dematte, 1992; Meyer et al., 1991; Vitti et al., 1992; Morelli et al., 1992), peaches (Shainberg et al., 1989), alfalfa (Odom, 1991; Shainberg et al., 1989; Sumner et al., 1986b; Sumner, 1990), coffee (Chaves et al., 1988, 1991; Guimaries, 1986, 1992; Malavolta, 1992; Pavan et al., 1982), wheat (Guimaries, 1986; Sousa et al., 1992b), rice (Ernani et al., 1992; Oliveira et al., 1986), beans (Ernani et al., 1992; Oliveira et al., 1986; Vitti et al., 1986), apples (Pavan and Bingham, 1986; Pavan et al., 1987; Suzuki et al., 1992), barley (Ernani et al., 1992), leucena (Sousa er al., 1992b), and cotton (Shainberg et al., 1989). In most cases, the yield responses were shown to be related to improvements in the Ca status and/or reductions in the Al status of the subsoils involved. Usually yield responses were obtained only after sufficient time (usually at least 1 year) had lapsed to allow the gypsum to be leached down into the subsoil. However,

MALCOLM E. SUMNER

6 h

0

L

2

12000

w

10000

Y v u)

c 0

E:

8000

W

CK

2

6000

.+ -0 =E

4000

0

n -

.>

2000

1982

1984

1986

1988

1990

Figure 2. Cumulative maize yield response over the period 1982-1991 to 10 t gypsum ha-’ incorporated into a South African Oxisol topsoil in 1982. (From M. P. W. Farina, private communication, 1991.)

in some cases, particularly in Brazil, the responses have been due to S and Ca nutrition (da Silva and van Raij, 1992; Guimarges, 1992; Malavolta, 1992; Vitti er al., 1992). As an illustration of the typical types of responses that have been obtained, data kindly supplied by M. P. W.Farina (private communication, 1991) will be used (Figs. 2 and 3). In Fig. 2, the cumulative maize yield response to 10 t gypsum ha-’ incorpo-

Al

Ca (crnol(+)/kg)

0 0.0

1

2

3

4

I

I

I

J

0.0

0.2

0.2

0.4

0.4

0.6

0.6

0.8

0.8

1 .o

1 .o

0

+ 1

H (crnol(+)/kg) 2

3

4

5

h

5 f 0

Figure 3. Comparisons of exchangeable Ca and A1 t H with depth inT990 after application of 10 t gypsum ha-l to the topsoil of a South African Oxisol in 1982. (From M. P. W. Farina, private

communication, 1991.)

7

GYPSUM AND ACID SOILS

11 la

0-

0

0

y

=

4.04

-

0., 1 1 5 ~

r1 = 0.54

0

20

40

60

A l Saturation

80 100

(X)

0

20

40

60

Ca Saturation

80

100

(X)

Figure 4. Relationship between increase in root length of wheat and (A) A1 saturation and (B) Ca saturation of a large number of soils from the Cerrado region of Brazil. (From Sousa er a / . . 1992a.)

rated in 1982 on a Natal Oxisol over the untreated control is substantial (1 1 t grain ha-I) and highly economic. In contrast, the incorporation of lime with a Wye Double Digger resulted only in a cumulative yield response of about 7 t maize grain ha- I . The corresponding changes in exchangeable Ca and A1 H down the soil profile are illustrated in Fig. 3. Gypsum treatment has increased exchangeable Ca and decreased AI+H to a depth of at least 90 cm, which allows for greater proliferation of roots in the subsoil (Farina and Channon, 1988) and the extraction of water by the crop, which in the absence of gypsum would have been beyond its reach. The relationships between wheat root growth and A1 and Ca saturations between 20 and 120 cm in the profiles of Oxisols from the Cerrado region of Brazil are illustrated in Fig. 4 (Sousa et al., 1992a). Similar results have been reported by other workers for a wide variety of crops (Chaves et al., 1988, 1991; Farina and Channon, 1988; O'Brien and Sumner, 1988; van Raij et al., 1988; Meyer et al., 1991; Malavolta, 1992; Shainberg et al., 1989; Sousa et al., 1992b; Sumner, 1990; Vitti et al., 1992). As a result the crop can better withstand periodic droughts during the season, which translates into increases in yield. In general, the effects of gypsum applications on soil pH in the field have been small or inconsistent, probably as a result of the salt effect in measuring soil pH being ignored or of soil heterogeneity and sampling errors (Shainberg et al., 1989). When these factors are eliminated or controlled, soil pH usually increases after gypsum application (Bolan et al., 1992a; Chaves et al., 1988; Sumner et al., 1986a,b). This phenomenon is very well illustrated by the data of da Silva and van Raij (1992), who compared the effects of continued applications of single and triple superphosphate over a long period in Brazil (Table 111). The effect of the gypsum contained in the single superphosphate can be clearly seen,

+

8

MALCOLM E. SUMNER Table I11 Comparison of the Effects of Cumulative Applications of Single and Triple Superphosphate over 17 Years at 100 kg P 2 0 sha - I year - I " pH in 0.01 M CaCl Sampling depth (cm)

Triple Single superphosphate superphosphate

0-20 20-40 40-60 60-80 800- 100 100- 120

5.2 4.6 4.3 4.2 4.3 4.5

5.4 4.7 4.9 5.0 5.2 4.9

"Difference between treatments equivalent to 4.5 gypsum ha-'. From da Silva and Van Raij (1992).

t

particularly in the subsoil, where the pH in CaCl, has been raised by nearly 1 pH unit to a depth of 1.2 m in the profile. Improved rooting with depth has been measured with a variety of crops in Brazil, South Africa, and the southeastern United States as illustrated in Table IV. Clearly gypsum has had a profound effect on the ability of the roots of various

Table IV Effect of Gypsum Application on the Distribution of Roots of Various Crops Down the Profiles of Highly Weathered Soils Maize South Africa (Farina and Channon, 1988)

Root density (m L - l ) Depth (cm)

Cont

Gyp

Apples Brazil Alfalfa Maize (M. A. Pavan, Georgia Brazil (Sousa and private communica- (Sumner and Carter, tion, 1991) Ritchey, 1986) 1988) Relative root distribution (%)

Cont

Gyp

Root density (cmg-')

Root length (m m - ' )

Cont

Gyp

Cont

Gyp

50 60 18 18 18

119 104 89 89 89

375 40 11 52 4

439 94 96 112 28

~~

0- 15 15-30 30-45 45-60 60-75

3.10 2.85 1.80 0.45 0.08

2.95 I .60 2.00 3.95 2.05

53 27 10

8 2

34 25 12 19 10

GYPSUM AND ACID SOILS

9

Table V Effect of Gypsum Treatments and Days without Irrigation during a Very Dry Season on the Yields of Maize and Soybeans Grown nn an Oxisol in Brazilo soybeans

Maize

Days without irrigation prior to flowering Gypsum rate ( + ha - I ) 0 2

4 6 LSD,,,

0

25

0

21

42

2.1 2.2 2.3 2.4 0.2

1.5 1.6

t/ha 4.7

-

6.8 0.4

3.2 3.6 4.6 5.5

2.2 -

1.0

NS

2.6

1.5

2.2 0.4

OFrom Sousa ei al. (1992b).

crops to penetrate acid subsoils where conditions were previously hostile. In all cases presented in Table IV, there were decreases in exchangeable A1 and increases in exchangeable Ca with depth as a result of the gypsum applications. These improved soil chemical conditions allowed the roots to extract more water from the subsoils, which, under water-limited conditions, will always translate into increased yields (Carvalho et af., 1986; Chaves et af., 1988; Shainberg et al., 1989). This effect is well illustrated by the data of Sousa et af.( 1992b), who studied the interaction between gypsum and irrigation treatments on the yield of soybeans (Table V). The gypsum effect is more marked under the more severe drought stress conditions. In many instances, it has been possible to measure this drought relief due to gypsum in terms of increased leaf water potentials (Carvalho et af., 1986; M. A. Pavan, private communication, 1991). Van Raij er af. (1988) have demonstrated that gypsum promoted sorghum root growth in the presence and absence of acid-forming nitrogenous fertilizers (ammonium sulfate, ammonium nitrate), but in the presence of calcium nitrate the effect was reversed. This resulted in increased uptake of N and increased extraction of K, NH,, and NO, from the subsoil. Over all N carriers, gypsum increased soil pH in CaCI, and reduced exchangeable Al, as illustrated in Fig. 5. The beneficial effects on maize yield of the P contained in PG have been demonstrated by M. P. W. Farina (private communication, 1991) and are illustrated in Table VI. Responses were obtained only at the lowest two P rates to the P contained in the phosphogypsum, which was also reflected in the P contents of leaves and soil. Meyer et al. (1991) obtained a response to phosphogypsum on sugarcane, which they attributed largely to its P content. Van der Watt et al.

MALCOLM E. SUMNER

10

A 4.4

4.3

0

N

o

0

-

1.0

+

0.5

0 0

4.2

> :

I n

4.1

z 4.0

0 wlthout gypsum I

1

-5

0

0 wlth gypsum

4

0 without gypsum

I

5

10

meq(*)/pot

Figure 5. Effect of gypsum on (A) pH in CaC12 and (B)exchangeable Al over a range of cation anion balances induced by acid- and alkali-forming N carriers. (From van Raij et al.. 1988.)

Table VI Effect of P (0.74%)in Pbosphogypum on Maize Yield and Leaf and Soil P Contents" Phosphogypsum applied (t ha-')

P applied (kg ha - I ) 40

80

160

Silage yield (t ha-') 5 0

LSDoos

10.07 9.16

10.68 9.92

1 I .22

11.18

0.51

0.51

0.51

Leaf P (S) 5

0.25

0.29

0 LSDaos

0.22 0.01

0.25 0.01

0.31 0.30 0.01

Olsen extractable soil P (mg L-' )

5 0 LSDam

13 8 4

24

45

18

44

4

4

"From M. P. W. Farina, private communication ( 199 I).

GYPSUM AND ACID SOILS

11

(1991) and van der Watt and Claassens (1990a) found that Bray I1 P was increased significantly by the application of phosphogypsum (0.36% P) at rates between 8 and 30 t ha-'. On an infertile Ultisol, Sumner (1990) found over a period of 3 years a slight beneficial effect (yield increase of about 500 kg alfalfa hay ha-' year-') from the P contained in a 10 t ha-l application of phosphogypsum (0.28% P). Thus the P content of the PG should be taken into account in evaluating the economics of its use. Sousa and Ritchey (1986) and van Raij et al. (1988) have demonstrated a further benefit of PG in the increased recovery of nitrate from subsoil horizons. Furthermore, Sousa et al. (1992b) have demonstrated the beneficial effects of gypsum in promoting the uptake of N, P, K, Ca, S, Cu, and Mn on an Oxisol in Brazil. In some cases, particularly on very sandy infertile soils, heavy gypsum applications (>5 t ha-') have had a negative effect on crop growth due to the preferential removal of Mg from the upper part of the profile but with little change in the K status (Syed-Omar and Sumner, 1991). A h a and Gascho (1991) also demonstrated that K and Mg were leached more easily from sandy than from sandy clay loam soils. Under such circumstances, gypsum applications should be made with extreme care with precautions being taken to apply Mg to the topsoil after the gypsum front has moved down the profile. Chaves et al. (1988), van der Watt et al. (1991), and Vitti et al. (1992) reported severe losses of Mg and K from the upper 60 cm of a red loam and from two Oxisols, whereas Farina and Channon (1988) reported substantial downward movement of Mg on a similar soil. Such downward movement of Mg as a result of gypsum applications has been reported by many other workers (Oates and Caldwell, 1985; O'Brien and Sumner, 1988; Quaggio, 1992; Ritchey et al., 1981; Shainberg eral., 1989; Sumner, 1990; Suzuki et al., 1992; van der Watt and Claassens, 1990a,b; Morelli et al., 1992). On a commercial basis, gypsum is being applied extensively to sugarcane in South Africa (J. Spencer, private communication, 1991), alfalfa in Georgia, sugarcane in Brazil, and coffee in Guatemala with good results in all cases.

B. PREDICTION OF CROPRFSPONSESTO GYPSUM APPLICATIONS Sumner ( 1990) demonstrated that responsive and nonresponsive soils could be separated on the basis of the soil's ability to remove both Ca and SO, from a dilute solution (0.005 M ) of gypsum in combination with the difference in pH measured in 0.005 M CaSO, and 0.005 M CaCl, (ApH). Soils that respond to gypsum lie in the quadrant where both of these values are positive (Fig. 6). Sousa et al. (1992a) have suggested that the quantity of gypsum required to obtain a yield response should be based on measurements of the amounts of Ca and S

12

MALCOLM E. SUMNER 0.75 h

-N

u

0

0

0.50

I

In 0

9

0

'*

$

0.25

0

0

I

In 0

8

v

0.00

I

n

a

-0.25 -1000

0

1000

Gypsum sorbed (pprn)

Figure 6. Relationship between ApH and gypsum sorbed for responsive ( 0 ) and nonresponsive Sumner, 1990.)

(0) soils. (From

adsorbed by the soil. They found the best predictor to be the amounts of Ca and S retained by the soil relative to those remaining in solution.

C. SOILSOLUTIONSTUDIES Many studies (Pavan et al., 1982, 1984; Ritchey et al., 1981; Shainberg et al., 1989; Sumner et al., 1986b; van Raij, 1988; Furlani and Berton, 1992) have been conducted on the effect of gypsum application on the chemistry of the soil solution. In general, the results have demonstrated that (a) sometimes but not always the solution pH increases slightly, (b) the total concentration of A1 increases usually in the form of Also,+ while the activity of A13+ invariably decreases, and (c) the concentration and activity of Ca2+ increase markedly in all cases. This behavior is well illustrated by the data (Table VII) of Chaves et al. (1991), who grew coffee in a greenhouse experiment on two soils. In the case of both soils, there is a substantial reduction in exchangeable A1 as a result of gypsum application that, in the Latosol Roxo, is reflected in reduced A13+ activity in the soil solution. The CaCI, treatment had the opposite effect. In both cases, gypsum improved root and top growth of the coffee

13

GYPSUM AND ACID SOILS Table VII

Effect of Gypsum and CaCI, Applied at Rates Equivalent to the CaCO, Required to Bring the pH to 6.0 on the Speciation of Al in Solution and the Growth of Roots and Tops of Coffee" Treatment Latosol Roxo Parameter PH (0.01 M CaCI,) Exch. A1 (meq 1OOg-I) Exch. Ca (meq 1OOg-I) Ca2+ (mmol L-I) Total solution A1 (mmol L-I) All+ (mmol L - I ) ALSO,'

(mmol L-I) Root volume (cm'plant-I) Root length (cm plant - I ) Root weight (g plant - I ) Total dry matter (g plant - I )

Dark red Latosol

Cont

Gyp

CaCl,

Cont

Gyp

CaCl

4.6

4.5

4.2

5.0

5.2

4.5

I .so

1.16

1.82

0.10

0.03

0.45

0.80

1.30

5.67

0.37

4.60

I .60

0.247

0.373

0.380

0.139

0.165

0.434

0.096

0.105

0.105

-

-

-

0.056

0.021

0.080

-

-

-

0.004

0.061

-

-

-

-

9.09

6.90

15.37

15.48

7.60

20.61

1980

4888

2855

2303

3143

2576

1430

2480

1387

0.682

2289

2035

8939

1 1.876

6924

2717

83 LO

75 10

"From Chaves er al. (1991).

significantly, whereas CaCl, was much less effective. These data clearly indicate that when gypsum is applied, both exchangeable A1 and A13+ activity are reduced and exchangeable Ca and Ca2+ activity are increased. Although CaCl, increases the Ca components, it also substantially increases the activity of Al'+ in solution. It is difficult to determine which effect is directly responsible for the improved root growth and it is probable that both the increased Ca2+ and reduced AP+ activity are jointly responsible as reported elsewhere (Noble and Sumner, 1988, 1989; Noble e? al., 1988a,b). Similar results have been reported by van Raij (1988). Apparently conflicting results have been presented by Nogueira and Mozeto (1990), who found that gypsum increased exchangeable Al, total A1 in solution, and A13+activity of all but one soil (Latosol

14

MALCOLM E. SUMNER

Roxo), but the method used was not comparable with those used in other studies and is probably part of the reason for the disagreement. In addition, all the soils used contained extremely low levels of exchangeable and solution A1 prior to treatment, which would raise the question of detection limits.

D. LEACHING STUDIESWITH GYPSUM Many leaching studies, using both disturbed and undisturbed soil columns, have been conducted with a wide variety of soils (Chaves er al., 1988; Kotze and Deist, 1975; Lemus-Grob, 1985; Oates and Caldwell, 1985; O’Brien and Sumner, 1988; Pavan er al., 1984; Reeve and Sumner, 1972; Shainberg ef al., 1989) and, in most cases, the pattern of behavior has been the same and similar to that found in the field, namely, a decrease in exchangeable A1 and an increase in exchangeable Ca without A1 appearing in the leachate. However, in some cases (Camargo and van Raij, 1989; Kotze and Deist, 1975; Oates and Caldwell, 1985; O’Brien and Sumner, 1988), the gypsum was reported to move down the columns and appreciable amounts of A1 appeared in the leachates. This apparent conflict may result from the short columns (15-20 cm) used in two of the studies (Camargo and van Raij, 1989; Oates and Caldwell, 1985) and, in the other cases, from the use of an artificially acidified topsoil (Kotze and Deist, 1975) or very sandy soils (O’Brien and Sumner, 1988). In one case (Oates and Caldwell, 1985), the use of hydrofluorogypsum containing 2.3% F resulted in most of the A1 in the short column being removed presumably as an AI-F soluble complex. In most cases, the favorable reactions induced by gypsum have been observed only in subsoil materials. In some soils (Lemus-Grob, 1985; O’Brien and Sumner, 1988), substantial quantities of Si were found in the leachate, which would support the reaction in equation 4 below for the decomposition of kaolinite, which is less stable under acid conditions than some of the A1 hydroxy sulfates. Oates and Caldwell (1985) and’O’Brien and Sumner (1988) found that gypsum promoted the downward movement of Mg and K. Sousa and Ritchey (1986) showed that gypsum readily displaced nitrate from the profile of a dark red Latosol. Camargo and van Raij (1989) have demonstrated that the passage of CaZ+and S02- ions through soils is a function of their electrochemical properties (Fig. 7). In two subsoil horizons of a red-yellow Latosol that, over the pH range 3.7 to 5.4, were net negatively charged, the passage of Sodz-was retarded roughly in inverse proportion to the net negative charge. At a pH of 5.4, this retardation had completely disappeared. The reverse was true for Caz+.However, in a Latowere retarded to a much greater extent over the sol Roxo, both Caz+and Sodzentire pH range 3.7 to 6.9. Under very acid conditions where the soil had a net

15

GYPSUM AND ACID SOILS soq-

Ca2+ low pH

natural pH

”

high pH

low pH

natural pH

high pH

1 3 5 7

1 3 5 7

yellow red Latosol (0-40 cm)

yellow red Latosol (40-80 cm)

0O

f1 3 K5 7 I V&

1Q 3 5L 7 J 1 3 5 7

L

1 3 5 7

geothltlc purple Latosol (60-80 cm)

V/V,

l

Figure 7. Effect of pH on the elution patterns for Ca2+and S042- in three Brazilian soils. (From Camargo and van Raij, 1989.)

positive charge, SO,2- passage was entirely inhibited, and at high pH, the passage of Ca2+was almost completely prevented.

E. MECHANISMS INVOLVED INSUBSOIL ACIDITYAMELIORATION Subsoil amelioration with gypsum involves the concomitant increase in exchangeable and solution Ca and a decrease in exchangeable and solution Al, the relative effects of which are impossible to separate (Shainberg et al., 1989). There is little doubt that both are important in promoting improved root growth. The mechanisms by which A1 is rendered less labile by gypsum treatment is of particular interest because simple cation exchange considerations would predict the opposite. 1. “Self-Liming Effect”

Much evidence has been presented here and elsewhere (Alva et al., 1990; Farina and Channon, 1988; Shainberg et al., 1989; Sumner et al., 1986b) to support the “self-liming” effect originally proposed by Reeve and Sumner

16

MALCOLM E. SUMNER

(1972), which essentially involves the ligand and exchange of OH by SO, on sesquioxide surfaces followed by hydrolysis and precipitation of exchangeable A1 as follows: OH 2 [Fe ,All

/

\ [Fe,Al]

\

+ Ca2 + SO,,-+

/ OH

so,-

/

\

\

/

2[Fe ,All

[Fe,AI]

+ Ca(OH2)

OH 2A13+

+ 3Ca(OH),+

2A1(OH)3

+ 3Ca2+

(2)

2. Precipitation of Solid Phases The precipitation of solid phases can also be explained in terms of the precipitation of one or more basic A1 sulfates (Adams and Rawajfih, 1977; Hue et al., 1985; Sposito, 1985), which are quite stable under very acid (pH < 4.5) soil conditions. The result of applying gypsum to acid subsoil would be exactly the same as proposed in the foregoing when this hypothesis is invoked as follows:

+ CaSO, + AIOHSO, + Ca(OH), + CaSO, + 5H20+ 2AIOHS0, + 2H,SiO, + 2Ca(OH), ZAP+ + 3Ca(OH), + AI(OH)3

Al(OH), Al,Si,O,(OH),

(3) (4)

(5)

Because such small amounts of new solid phases would be formed, it is impossible to obtain incontrovertible evidence for these reactions, but under certain conditions they are thermodynamically feasible.

3. Co-sorption of SO,*- and A13

+

However, in other cases, the soil solution of some gypsum-treated subsoils is undersaturated with respect to all possible solid phases, which led Sumner et al. (1986a) to suggest that sesquioxide surface reactions may exert the ultimate control. When SO4,- is specifically adsorbed to a sesquioxidic surface, the negative surface charge density increases, which would impart a preference for A13 adsorption and thus lowering its activity in solution. +

GYPSUM AND ACID SOILS

17

4. Ion Pair Formation Pavan et al. (1982, 1984) and Chaves et al. (1991) proposed that A1 toxicity is reduced by the formation of the ion pair AISO,' , which has been shown to be less toxic than A13+ (Noble et a l . , 1988a). This mechanism is likely to be of greatest importance in nutrient solutions in the absence of solid phases but may also be of importance in soils provided that the level of SO,2- in the soil solution is maintained at high enough levels to stabilize the Also,+ and if the replenishment of A13+from solid phases in response to the formation of Also4+ is slow enough. However, it is likely to be more short-lived than the mechanisms discussed earlier (111, E, 1 and 111, E, 2).

TV. GYPSUM As AN AMELIORANT FOR SOIL PHYSICAL PROPERTIES In addition to its beneficial effects on the chemical properties of subsoils, gypsum has been shown to markedly improve both the internal and surface physical properties of acid soils in many regions of the world. The reader is referred to a review (Shainberg et al., 1989) that discusses these effects in great detail and will be used as a starting point for the discussion to follow.

A. SURFACE CRUSTING To set the stage for a clear understanding of the potential benefits of gypsum in this application, it is necessary to briefly discuss clay dispersion. Contrary to popular belief, which states that dispersive soils are those containing elevated levels of Na, many highly weathered acid topsoils throughout the world have been found to disperse under conditions of low electrolyte concentration in the soil solution and mechanical energy inputs (Miller and Radcliffe, 1992; Shainberg et al., 1989). This is illustrated in Fig. 8 for 32 Ultisol topsoils from Georgia that were gently shaken with distilled water and allowed to stand. A large proportion of these soils contain appreciable quantities of dispersible clay. Conditions favoring the dispersion of clay are present during every rainstorm at the soil surface. Under the impact of a raindrop, aggregates at the surface are disintegrated and, because the clay is dispersible at low electrolyte concentrations, it separates from the larger particles and moves into the pore space, which becomes blocked, thus lowering infiltration and increasing runoff and erosion. In soils containing appreciable quantities of sand (loams, sandy loams, and loamy sands), this process often leaves a layer of clean sand grains on the soil surface called

18

MALCOLM E. SUMNER

* Q)

30

1

25 -

Q

E 0

20-

v,

+. 0 +

15 10 -

0

I Q)

a

50-t-

0

10

20

30

40

50

60

70

80

90

100

Dispersible clay (%) Figure 8. Frequency distribution of water-dispersible clay in a selection of 32 Georgia Ultisols. (From Miller and Radcliffe, 1992.)

the “washed out” layer, which is a positive clue that a less permeable seal (“washed in” layer) has formed. This is clearly visible in Fig. 9A, where the sand grains are the light-colored areas in the depressions in the field and the crust formed is shown in transverse section (Fig. 9B). At the top of the section the thin seal (ca. 1 mm thick) or “washed in” layer can be clearly seen. The amount of water-dispersible clay in a topsoil is directly related to the reduction in final infiltration rate under rainfall and to the resulting soil loss as illustrated in Fig. 10. These processes of soil degradation can be overcome by either reducing or eliminating the mechanical energy of impacting raindrops using a mulch or other soil cover or by spreading gypsum over the surface after preparing the seedbed and before the next rain to supply electrolyte to promote flocculation. Because the gypsum must dissolve very rapidly to prevent dispersion, only by-product gypsum materials are suitable for this application (Shainberg et al., 1989). The effect of PG in preventing crust formation and subsequent runoff and erosion for a variety of acid soils is presented in Table VIII. In all cases except the Brazilian Typic Haplorthox, the pattern of behavior was the same with surface-applied PG promoting flocculation of the clay at the soil surface and increasing final infiltration rate and decreasing soil loss. For the Oxisol, which had a pH of 4.0, gypsum, in fact, reduced final infiltration rate, which can be attributed to the greater dispersive effect of the added Ca over that of the Al originally present, which would have promoted flocculation (Sumner, 1992). This view is supported by

GYPSUM AND ACID SOILS

19

Figure 9. Appearance of the soil surface (A) of a Georgia Ultisol after a crust (B)has formed. (From Sumner and Miller, 1992.)

20

MALCOLM E. SUMNER

A

-

5 -

m

f. D

4 -

z

- 3 -

h

f

E

5

0

J 4 c

I

3

n

C

s

$ 2

.-

E

2 -

:: 1 -

c

2 1

-E

0 -0

10

20

30

40

50

60

70

0

10

20

30

40

50

60

70

Water Dlsperslble Clay Index

Figure 10. Relationships between water-dispersible clay and (A) final infiltration rate and (B) soil loss for a range of Georgia Ultisols under simulated rainfall. (From Miller and Radcliffe, 1992.)

the fact that, when this soil was limed to pH 6.0, the final infiltration rate was even further reduced and the sediment concentration in the runoff was markedly increased. Roth and Pavan (1991) showed for this soil that the clay concentration in the runoff increased with increasing pH. Unfortunately, gypsum was not applied to the limed topsoil, which would have been expected to restore the infiltration rate. The tendency of an Alfisol and an Oxisol to disperse as their pH values were raised from 3 to 7 has been reported by Rengasamy et al. (1993). In the former case, there was a large increase in the variable negative charge, whereas for the Oxisol, there was a charge reversal of substantial magnitude. The foregoing results emphasize the importance of pH in the management of soils to reduce crusting. Liming acid soils even with divalent ions (Ca2+and Mg2+)will tend to promote dispersion and its consequences. Other research (Miller and Scifres, 1988) shows that even small additions of Na (equivalent to 100 kg N ha-') as NaNO, can have severe deleterious effects on infiltration rate, runoff, and erosion of highly weathered acid soils. Norton (1991) demonstrated that several sulfur-rich by-product materials containing gypsum from power plants were effective in reducing crusting. Miller (1988) demonstrated the beneficial effects of PG at 2 t ha-' applied over the row on the emergence of cotton on a Norfolk sandy loam (Typic Kandiudult) as illustrated in Table IX. Emergence was better at all times up to 12 days on the gypsum treatment and the effect was particularly marked at the earlier stages. The beneficial effect is due to crust prevention above the seed during the first rainfall event after planting. It can thus be concluded that by-product gypsum is a viable remedy for the prevention of soil crusting and its attendant problems on soils containing water-

21

GYPSUM AND ACID SOILS Table VIII

Effect of Phosphogypsumon Final Infiltration Rate, Total Runoff, Sediment Concentration, and Soil Loss for Acid Inceptlsols, Alfisols, and Ultisols" Final infiltration rate (mmh-I) Soil typeand location Cecil (Typic Kanhapludult) Georgia Worsham (Typic Ochraquult) Georgia Wedowee (Typic Hapludult) Georgia Greenville (Rhodic Kandiudult) Georgia Opequon (Lithic Hapludalf) Maryland Shorrocks (Rhodic Paleastalf) South Africa Typic Rhodoxeralf Israel R hodoxeralf Israel Mitra (Lithic Xerochrept) Portugal Formoso (Lithic Rhodoxeralf) Portugal Appling (Typic Kanhapludult) Georgia Rhodic Haplorthox Brazil Typic Haplorthox Brazil

Sediment concentration (g L-1)

Soil loss (kg ha-')

Cont

Gyp

Cont

Gyp

Cont

Gyp

7.3

22.2

266

96

1.28

1.43

0.8

2.2

1315

732

3.96

2.96

2.3

11.0

1135

442

3.43

3.15

11

23

939

50

1.42

0.45

2.3

7.5

250

80

-

-

22.2

40.8

-

-

-

-

11.7

17.5

25800

2000

-

-

4.5

11.7

66800

8000

-

-

4.8

8.8

1700

1260

3.2

2.8

5.6

10.0

1980

1500

3.8

3.5

21

61

3812

1468

4.2

2.2

37.7

25.6

-

-

0.98

1.17

37.7

11.8'

-

-

0.98

2.48

'

"From Agassi er al. (1989). Miller (1987). Miller and Scifres (1988), Norton er al. (1993). Roth et al. (1986). Roth and Pavan (1991), Shainberg er al. (1991), d. Scifres, private communication (1990), van der Watt and Claassens (199Ob), and Warrington er al. (1989). 'Limed to pH 6.0 without gypsum.

22

MALCOLM E. SUMJVER Table IX Effect of Surface Application of Phosphogypsum on Emergence of Cotton Seedlings on a Typic Kandiuduit in Georgia" Days after planting 3

4

Treatment Control Gypsum

5

9

12

56 69

68 79

Emergence (%) 12 37

22 45

35 55

"From Miller (1988).

dispersible clay. Currently, gypsum is being used commercially in Israel, Georgia, and Australia (Howell, 1987) for this purpose.

B. HARDSETTING SOILS In Australia in particular, there have been many reports of soils that have very weak water-stable aggregation and that slake and slump, and the entire topsoil sets hard on drying. A full description of these soils has been presented in a review by Mullins e? al. (1990) to which the interested reader is referred. Although these soils are not always acid by the definition outlined at the beginning of this review, a large proportion are and hence the brief discussion of hardsetting behavior in relation to amelioration by gypsum. There is little question that for hardsetting to occur, the soil must be in the sandy loam textural range, be low in organic matter, and contain dispersible clay and possibly cementing agents such as amorphous silica or other short-range materials (Chartres et al., 1990; Sumner, 1993). There is a parallel between crusting and hardsetting behavior with the major difference being that the structure of hardsetting soils is so unstable that even wetting causes aggregate breakdown and clay movement within the entire A, horizon, whereas in crusting, clay mobility is only manifest in the top few millimeters of soil. A number of studies have demonstrated that gypsum alleviates the hardsetting character of these soils probably primarily as a result of its beneficial effects on physical properties such as clay dispersion, crusting, and hydraulic conductivity (Grierson, 1978; So et al., 1978; Shanmuganathan and Oades, 1983; White and Robson, 1989), promotion of aggregate stability and macroporosity (Loveday and Scotter, 1966; Grierson, 1978; So et al., 1978; Chartres et al., 1985; Taylor and Olsson, 1987). reduction in exchangeable Na (Loveday and Scotter, 1966;

GYPSUM AND ACID SOILS

23

Grierson, 1978; Greene and Ford, 1985; Taylor and Olsson, 1987), and modulus of rupture (Grierson, 1978; So eral., 1978; Alymore and Sills, 1982; White and Robson, 1989). Because leaching removes gypsum from the upper part of the soil profile where the major problem in hardsetting is located, periodic applications are necessary both to maintain adequate electrolyte to prevent dispersion and slumping and to slowly reduce the ESP level (Chartres et al., 1985; Greene and Ford, 1985). Greene and Wilson (1989) demonstrated that over a period of 3.5 years, the beneficial effects of gypsum on clay dispersion were lost as a result of leaching, but because the establishment of pasture protected the surface from the impacting raindrops, no loss in hydraulic conductivity was recorded. This work confirmed that of Quirk (1978), who came to the conclusion that once pasture had been established, the increased level of organic matter stabilized the structure, thus reducing the need for further gypsum. Taylor and Olsson (1987), working with alfalfa on a red-brown earth, came to essentially the same conclusion that initially responses to gypsum are due to the electrolyte effect and once the crop has established a good root system, biological stabilization of structure takes place, allowing continued penetration by water despite reduced EC levels in the upper part of the soil. The gypsum treatments reduced ESP values down the profile to 0.6 m. Cochrane and Alymore (1991) found that dispersive failure was the main mechanism contributing to the structural instability of a brown solonetzic soil from western Australia and demonstrated that gypsum application, which reduced dispersion, was the major determinant of yield responses. What does not seem to be clear is whether hardsetting behavior can be caused by clays that only disperse with mechanical energy inputs. If so one would expect a surface crust or seal to form, which would reduce the likelihood of the soil wetting up under zero tension because intake of water would be drastically reduced. The author is familiar with soils in southern Africa and the southeastern United States that set hard after wetting and drying. None of these soils contains spontaneously dispersible clay but usually clay can be dispersed with inputs of mechanical energy. There is definitely need for more work in this area to resolve this paradox.

C. SUBSURFACEHARDLAYERS In many highly weathered soils, particularly Ultisols and Alfisols, root penetration into the subsoil is limited by subsurface hard layers in addition to the chemical barrier (Al) discussed earlier. Radcliffe et al. (1986) and Sumner et al. (1 990) demonstrated that both mined and PG incorporated into the topsoil were effective in reducing the Cone Index (CI) of subsoil hard layers after sufficient time had elapsed for the gypsum front to move down the profile as illustrated in Fig. 11. (The CI value is a measure of the resistance of the soil to penetration by

Cone I n d e x D e c r e a s e (%) 0

10

20

30

40

50

I

I

I

I

I

e,:

Appling Alfalfa eMined Gypsum

ele

50 60

ee' ee

n 70

E

(a)

80

f

0

Q

10

20

30

40

0

10

30

20

40

50

0

0

0

10

20

30

40

50

I

I

I

I

I

I

10

i 10 -

20

20

30

30

40

40 - . . e ~ ~ '

50

50

60

60

70

70

80

80

50

0

10

30

20

40

ef

0

50

Cecil cotton Phosphogypsum

10

30

20

40

50

0 10

20

Alfalfa Phosphogypsum

20 -

30 40 50

-

-e-

Phorphogypsum

-e '-

40

RO

50

-

80

-

I-

60 -

60 70

20 - - . y e

30

..e-

j '.

€ e.

Fallow Phosphogypsurn

-e

' 2

70 (8)

80

Figure 11. Decrease in cone index (CI) with depth after application of 10 t ha-' of either mined or phosphogypsum to different soillcrop combinations. Measurements [CI values in MPa below 20 cm] were made (a) 48 13.411, (b) 30 13.441, (c) 30 (3.771. (d) 33 [3.321, (e) 21 16.411. and (f) 21 13.811 months after gypsum treatment. From Sumner et al., 1990.)

2s

GYPSUM AND ACID SOILS

a cone-shaped needle that is designed to simulate the behavior of a root.) Substantial reductions in resistance to penetration were recorded in all cases and on the only two soils where roots were measured, there were significant improvements in root distribution down the profile. McCray et al. (1991) demonstrated that the CI value was significantly reduced in packed cores of an Ultisol when a gypsum solution rather than distilled water was present in the pore space. Sumner et al. (1990) suggested that gypsum directly affects flocculation and aggregation of the subsoil and indirectly improves rooting, which leads to greater aggregation. Lehrsch et af. (1993) found that gypsum was better than lime in promoting stable aggregation of a range of acid soils after freezing.

D. HYDRAULIC CONDUCTMTY Chiang et al. (1987) found that undisturbed cores of three Ultisols exhibited severe declines in hydraulic conductivity when leached with distilled water. The percolate contained dispersed clay, which is evidence of the degradation and dispersion taking place within the profile. When leached with a saturated gypsum solution, hydraulic conductivity was maintained at a much higher level and the percolate contained no signs of sediment as illustrated in Fig. 12. A curvilinear relationship was found between the final steady-state hydraulic conductivity of these soils and the peak turbidity of the percolate, suggesting that the decline in

WATER

SATURATED GYPSUM

100 /

60

\

,,

80

\.'\

turbidlty

I I

4

, \

I

hydroulic conductivity

40 z II

201 IL

;

h y d r a u l i c conductivity

turbidity=zero

0

4---T 0

I

5

10 Time (rnin)

15

20

0 1

0

10

20

30

I

I

I

I

40

50

60

70

Time (rnin)

Figure 12. Hydraulic conductivity of a Typic Hapludult leached with distilled water and a saturated gypsum solution and the turbidity of the percolates. (From Chiang et af., 1987.)

26

-;I f

20

MALCOLM E. SUMNER

L

Y=26.2 e

-0.04~

2

(r =0.91)

0 C

4

15

0 10 20 30 40 50 60 70 80 90 100 Dispersible clay (g/kg)

Figure 13. Relationship between dispersible clay content and saturated hydraulic conductivity for a number of British and Samoan soils. (From Bolan er al., 1992a.)

hydraulic conductivity is due to clay dispersion that clogs the pores (Miller, 1988). Bolan et al. (1992a) observed that the saturated hydraulic conductivities of two acid British soils and one ferruginous soil from Western Samoa were much lower when leached with pure water than with a gypsum solution, thus confirming the findings of Chiang et al. (1 987). They found a curvilinear relationship between saturated hydraulic conductivity and dispersible clay for all the soils studied (Fig. 13). Field observations on the British soils indicate that soil tilth was improved by the gypsum treatments. These data indicate that gypsum is beneficial in maintaining the internal physical condition of acid soils.

V. CONCLUSIONS 1. Most gypsum materials available on the market have similar compositions but their rates of dissolution vary with surface area and crystallinity. Usually gypsum dissolves faster in soil than in pure solution. 2. When used as an ameliorant for acid subsoils, gypsum treatment has resulted in substantial yield increases in a wide variety of crops. These yield increases are usually due to increased supply of CaZ+and/or the detoxification of

GYPSUM AND ACID SOILS

27

A13+ in the soil, both of which permit improved root proliferation in the subsoil and increased availability of water. 3. Additions of gypsum to soils usually result in the downward movement of Mg and sometimes K. On very sandy soils, this presents serious problems that would preclude gypsum use under such circumstances. 4. The clay in the topsoil of many acid soils (particularly those that have been limed) is readily dispersible in water provided mechanical energy inputs (raindrops) are available. This leads to crusting, which results in increased runoff and erosion. Application of by-product (rapidly soluble) gypsum on the soil surface prior to rain or irrigation substantially reduces crusting and its consequences. 5 . In Ultisols and Alfisols, which have hard subsurface layers, surface-applied gypsum reduces resistance to penetration by roots, resulting in greater root proliferation and water uptake from the subsoil. 6. Recent research suggests that gypsum applications will be useful in improving the hydraulic properties of acid soils.

REFERENCES Adams, F., and Rawajfih, Z. (1977). Basaluminite and alunite: A possible cause of sulfate retention by acid soils. Soil Sci. Soc. Am. J . 41, 686-692. Agassi, M., Shainberg, I., Warrington, D., and Ben-Hur. M. (1989). Runoff and erosion control in potato fields. Soil Sci. 148, 149- 154. Alcardo, 1. S., and Recheigl, J. E. (1993). Phosphogypsum in agriculture: A review. Adv. Agron. 49,55- 118. Aha, A. K., and Gascho, G. J. (1991). Differential leaching of cations and sulfate in gypsum amended soils. Commun. Soil Sci. PIunr Anal. 22, 1195- 1206. Aha, A. K., Sumner. M. E., and Miller, W. P. (1990). Reactions of gypsum or phosphogypsum in highly weathered acid subsoils. Soil Sci. Soc. Am. J . 54, 993-998. Alymore, L. A. G., and Sills, 1. D. (1982). Characterization of soil structure and stability using modulus of rupture-exchangeable sodium percentage relationships. Aust. J. Soil Res. 20, 2 I 3 - 224. Anonymous (1992). National emission standards for hazardous air pollutants, national emission standards for radon emissions from phosphogypsum stacks. Fed. Reg. 57,23305. Bolan, N. S., Syers, J. K., Adey, M. A., and Sumner, M. E. (l992a). Factors affecting the saturated hydraulic conductivity of non-sodic soils amended with gypsum. J . Soil Sci. (in press). Bolan, N. S . , Syers, 1. K., and Sumner, M. E. (1992b). Dissolution of various sources of gypsum in aqueous solutions and soils. J . Sci. Food Agric. 57, 527-541. Bouldin, D. R. (1973). The influence of subsoil acidity on crop yield potential. Cornell In!. Agric. Bull. 34, 1-17. Camargo, 0 . A., and van Raij, B. (1989). Movimento do gesso em amostras de latossolos com diferentes propiedades eletroquimicas. Rev. Bras. Cien. Solo 13, 275-280. Carvalho, L. J. C. B., Gomide, R. L., Rodriguez, G. C., Souza, D. M. G., and de Freitas, E. ( 1986). Reposta do milho a aplicaqio de gesso e dCficit hidrico em solo de Cerrados. “Anais do 1 SeminLio sobre o Us0 do Fosfogesso na Agricultura,” pp. 61-83. EMBRAPA, Brasilia, D. F., Brazil.

28

MALCOLM E. SUMNER

Chartres, C. 3.. Greene, R. S . , Ford, G. W., and Rengasamy, P. (1985). The effects of gypsum on macroporosity and crusting of two red duplex soils. Aust. J. Soil Res. 23, 467-479. Chartres, C. J., Kirby, J. M., and Raupach, M. (1990). Poorly ordered silica and aluminosilicates as temporary cementing agents in hard-setting soils. Soil Sci. SOC.Am. J . 54, 1060- 1067. Chaves, J. C. D., Pavan. M. A., and Miyazawa, M. (1988). Reduqio da acidez subsuperficial em coluna de solo. Pesqui. Agropecu. Bras. 23,469-476. Chaves. J. C. D., Pavan, M. A,, and Miyazawa, M. (1991). Especiaqio quimica da soluqio do solo para interpretaqio da absoqio de cilcio e aluminio por raizes de cafeeiro. Pesqui. Agropecu. Bras. 26,447-453. Chiang, S. C., Radcliffe, D. E.. Miller. W. P., and Newman, K . D. (1987). Hydraulic conductivity of three southeastern soils as affected by sodium, electrolyte concentration and pH. Soil Sci. Soc. Am. J . 51, 1293-1299. Cochrane, H. R . , and Alymore, L. A. G. (1991). Assessing management-induced changes in the structural stability of hardsetting soils. Soil Tillage Res. 20, 123- 132. da Silva, N. M., and van Raij, B. (1992). 0 us0 do gesso e do superfosfato simples na cultura do algodoeiro. “I1 Seminhrio sobre o Us0 do Gesso na Agricultura,” pp. 159-174. IBRAFOS, SP, Brazil. Dematte, J. L. I. (1992). Aptidio agricola de solos e o us0 do gesso. “I1 Seminhrio sobre o Us0 do Gesso na Agricultura,” pp. 307-324. IBRAFOS, SP, Brazil. Doner, H. E., and Lynn, W. C. (1989). Carbonate, halide, sulfate and sulfide minerals. In “Minerals in Soil Environments” (J. B. Dixon and S . B. Weed, eds.), pp. 279-330. Am. SOC.Agron., Madison, WI. Ernani, P. R., Cassol, P. C., and Peruzo, G. (1992). Eficikncia agronbmica do gesso agricola no sul do B r a d . “I1 Seminhrio sobre o Us0 do Gesso na Agricultura,” pp. 263-276. IBRAFOS, SP, Brazil. Farina, M. P. W., and Channon, P. (1988). Acid-subsoil amelioration: 11. Gypsum effects on growth and subsoil chemical properties. Soil Sci. Soc. Am. J . 52, 175- 180. Frenkel, H., and Fey, M. V. (1989). Rate of dissolution of gypsum from different sources and its effect on water infiltration. S. A f . J. Plant Soil 6 , 191 - 194. Furlani, P. R., and Berton, R. S. (1992). Atividade de cllcio e aluminio e desenvolvimento radicular. “I1 Seminhrio sobre o Us0 do Gesso na Agricultura,” pp. 121 - 138. IBRAFOS, SP, Brazil. Greene, R. S. B., and Ford, G. W. (1985). The effect of gypsum on cation exchange in two red duplex soils. Aust. J . Soil Res. 23, 61 -74. Greene, R. S. B., and Wilson, I. B. (1989). Amelioration of some physical properties and nutrient availability of an exposed B horizon of a red-brown earth. Soil Use Manage. 5 , 66-71. Grierson, I. T. (1978). Gypsum and red-brown earths. In “Modification of Soil Structure’’ (W. W. Emerson, R. D. Bond, and A. R. Dexter, eds.), pp. 315-324. Wiley, New York. Guimaries, P. T. C. (1986). 0 gesso agricola na neutralizagio do aluminio nas camadas subsupeficias do solo: AplicaqBes As cultivars anuais e perenes. “Anais do I Seminhrio sobre o Us0 do Fosfogesso na Agricultura,” pp. 145- 168, EMBRAPA, Brasilia, D. F., Brazil. Guimaries, P. T. C. (1992). 0 us0 do gesso agricola na cultura do cafeeiro. “I1 Seminhrio sobre o Us0 do Gesso na Agricultura,” pp. 175- 190. IBRAFOS, SP, Brazil. Hammel, J. E., Sumner, M. E., and Shahandeh, H. (1985). Effect of physical and chemical profile modification on soybean and corn production. Soil Sci. Soc. Am. J . 49, 1508- 1512. Haynes, R. J. (1984). Lime and phosphate in the soil-plant system. Adv. Agron. 37, 249-315. Howell, M. (1987). Gypsum use in the wheatbelt. J . Agric., West. Aust. 28, 40-43. Hue, N. V., Adams, F., and Evans, C. E. (1985). Sulfate retention by an acid BE horizon of an Ultisol. SoilSci. Sor. Am. J . 49, 1196-1200. Keren, R., and Shainberg, I. (1981). Effect of dissolution rate on the efficiency of industrial and mined gypsum in improving infiltration of a sodic soil. Soil Sci. Sac. Am. J . 45, 103- 107.

GYPSUM AND ACID SOILS

29

Kotze, W. A. G., and Deist, J. (1975). Amelioration of subsurface acidity by leaching of surface applied amendments. A laboratory study. Agrochemophysica 7 , 39-46. Lehrsch, G. A,, Sojka, R. E., and Jolley, P. M. (1993). Freezing effects on aggregate stability of soils amended with lime and gypsum. Catena. Suppl. (in press). Lemus-Grob, F. (1985). Effects and behavior of gypsum in an Ultisol. M.S. Thesis, University of Georgia, Athens. Loveday, J., and Scotter, D. R. (1966). Emergence response of subterranean clover to dissolved gypsum in relation to soil properties and evaporative conditions. Aust. J . Soil Res. 4, 55-68. Malavolta, E. (1992). 0 gesso agncola no ambiente e na nutrieb da planta-Perguntas e respostas. “I1 Seminiirio sobre o Us0 do Gesso na Agricultura,” pp. 41 -66. IBRAFOS, SP, Brazil. Malavolta, E., Guilherme, M. R., and Liem, 7 .H. (1986). Associq6es fosfogesso calchrio: Principios e aplicaq6ees. “Anais do I Seminiirio sobre o Us0 do Fosfogesso na Agricultura,” pp. 177196, EMBRAPA, Brasilia, D.F., Brazil. Mays, D. A,, and Mortvedt, J. J. (1986). Crop response to soil applications of phosphogypsum. J . Environ. Qual. 15, 78-8 I . McCray, J. M., Radcliffe, D. E., and Sumner, M. E. (1991). Influence of solution Ca on water retention and soil strength of Typic Hapludults. Soil Sci. 151, 312-316. Meyer, J. H., Turner, P. E. T.. and Fey, M. V. (1991). Interim evaluation of phosphogypsum as an ameliorant for soil acidity in sugarcane. Proc. Annu. Congr.-S. Ajk Sugar Technol. Assoc. 65,41-46. Miller, W. P. (1987). Infiltration and soil loss of three gypsum-amended Ultisols under simulated rainfall. Soil Sci. SOC.Am. J . 51, 1314-1320. Miller, W. P. (1988). “Use of Gypsum to Improve Physical Properties and Water Relations in Southeastern Soils,” Publ. No. 01-020-082. Florida Institute of Phosphate Research, Bartow. Miller, W. P., and Radcliffe, D. E. (1992). Soil crusting in the southeastern United States. I n “Soil Crusting: Chemical and Physical Processes” (M. E. Sumner and B. A. Stewart, eds.), pp. 233266. Lewis Publ.. Boca Raton, FL. Miller, W. P., and Scifres, J. (1988). Effect of sodium nitrate and gypsum on infiltration and erosion of a highly weathered soil. Soil Sci. 145, 304-309. Miller, W. P., and Sumner, M. E. (1991). Impacts from radionuclides on soil treated with phosphogypsum. USEPA, unpublished report. Morelli, J. L., Dalben, A. E., Almeida, J. 0. C., and Dematte, J. L. 1. (1992). Calciirio e gesso na produtividade da cana-de-aqicar e nas caractensticas qufmicas de um latossolo de textura mkdia Blico. Rev. Eras. Cien. Solo 16, 187-194. Mullins, C. E., MacLeod, D. A., Northcote, K. H., Tisdall. J. M., and Young, I. M. (1990). Hardsetting soils: Behavior, occurrence and management. A h . Soil Sci. 11,38- 108. Noble, A. D.. and Sumner. M. E. (1988). Calcium and Al interactions and soybean growth in nutrient solutions. Commun. Soil Sci. PIanr Anal. 19, 11 19- 1131. Noble, A. D.. and Sumner, M. E. (1989). Growth and nutrient content of soybeans in relation to solution calcium and aluminum. S . Afr. J. Plant Soil 6, 113-1 19. Noble, A. D., Sumner, M. E., and Aha, A. K. (1988a). The pH dependency of aluminum phytotoxicity alleviation by calcium sulfate. Soil Sci. SOC. Am. J . 52, 1398- 1402. Noble, A. D., Fey, M.V.,and Sumner, M. E. (1988b). Calcium-aluminum balance and the growth of soybean roots in nutrient solutions. Soil Sci. SOC. Am. J . 52, 1651-1656. Nogueira, A. R. A , , and Mozeto, A. A. (1990). Interag6es qufmicas do sulfato e carbonato de cdcio em seis solos Paulistas sob vegeteb de cerrado. Rev. Eras. Cien. Solo 14, 1-6. Norton, L. D. (1991). Comparison of the effectiveness of several sulfur rich by-products on reducing surface sealing. Agron. Abstr. 83, 337. Norton, L. D., Shainberg, I., and King, K. W. (1993). Utilization of gypsiferous amendments to reduce surface sealing in some humid soils in the eastern USA. Carena Suppl. (in press).

30

MALCOLM E. SUMNER

Oates, K. M., and Caldwell, A. G. (1985). Use of by-product gypsum to alleviate soil acidity. Soil Sci. SOC. Am. J . 49,915-918. O’Brien, L. 0..and Sumner, M. E. (1988). Effect of phosphogypsum on leachate and soil chemical composition. Commun. Soil Sci. Plant Anal. 19, 7- 12. Odom, J. W. (1991). Alfalfa response to gypsum, boron and subsoiling on an acid Ultisol. Agron. Absrr. 83, 296. Oliveira, 1. P., Kluthcouski, J., and Reynier, F. N. (1986). Efeito de fosfogesso na producao de feijao e arroz e no comportamento de alguns nutrientes. “Anais do I SeminLio sobre o Us0 do Fosfogesso na Agricultura,” pp. 119- 144, EMBRAPA, Brasilia, D.F., Brazil. Orvedal, A. C., and Ackerson, K. T. (1972). “Agricultural Soil Resources of the World.” USDA, Soil Conservation Service, Washington DC. Pavan, M. A., and Bingham, F. T. (1986). Effects of phosphogypsum and lime on yield, root density and fruit and foliar composition of apple in Brazilian Oxisols. Proc. Int. Symp. Phosphogypsum,2nd, pp. 51-58, University of Miami, Miami, FL. Pavan, M. A., Bingham, F. T., and Pratt, P. F. (1982). Toxicity of aluminum to coffee in Ultisols and Oxisols amended with CaCO,, MgCO,, and CaS04.2H20. Soil Sci. SOC. Am. J . 46, 1201-1207.

Pavan, M. A., Bingham, F. T., and Pratt, P. F. (1984). Redistribution of exchangeable calcium, magnesium, and aluminum following lime or gypsum applications to a Brazilian Oxisol. Soil Sci. SOC. Am. J . 48, 33-38. Pavan. M. A., Bingham, F. T., and Peryea. F. J. (1987). Influence of calcium and magnesium salts on acid soil chemistry and calcium nutrition of apple. Soil Sci. SOC. Am. J . 52, 1526- 1530. Quaggio, J. A. (1992). Respostas das culturas de milho e soja, A aplicaqio de calcirio, gesso e movimentaqio de ions em solos do Estado de Sio Paulo. “I1 Seminario sobre o Us0 do Gesso na Agricultura,” pp. 241-262. IBRAFOS, SP, Brazil. Quirk, J. P. (1978). Some physico-chemical aspects of soil structural stability-A review. In “Modification of Soil Structure’’ (W. W. Emerson, R. D. Bond, and A. R. Dexter, eds.), pp. 308314. Wiley, New York. Radcliffe, D. E., Clark, R. L., and Sumner, M. E. (1986). Effect of gypsum and deep-rooting perennials on subsoil mechanical impedance. Soil Sci. SOC.Am. J . 50, 1566- 1570. Reeve, N. G.,and Sumner, M. E. (1972). Amelioration of subsoil acidity in Natal Oxisols by leaching of surface-applied amendments. Agrochemophysica 4, 1-6. Rengasamy, P., Naidu, R.. Beech, T. A., Chan, K. Y.. and Chartres, C. (1993). Crust formation as related to dispersive potential in Australian soils. Catena, Suppl. (in press). Ritchey, K. D., Souza, D. M. G., Lobato, E., and Correa, 0. (1981). Calcium leaching to increase rooting depth in a Brazilian savannah Oxisol. Agron. J . 72,40-44. Roth, C. H., and Pavan, M. A. (1991). Effects of lime and gypsum on clay dispersion and infiltration in samples of a Brazilian Oxisol. Geoderma 48, 351-361. Roth, C. H., Pavan, M. A., Chaves, J. C. D., Meyer, B., and Frede. H. G. (1986). Efeitos das aplicaq6es de calcLio e gesso sobre a estabilidade de agregados e infiltrabilidade de figua em um Latossolo Roxo cultivado com cafeeiros. Rev. Bras. Cien. Solo 10, 163- 166. Shainberg, I., Sumner, M. E., Miller, W. P., Farina, M. P. W., Pavan, M. A., and Fey, M. V. (1989). Use of gypsum on soils: A review. Adv. Soil Sci. 9, I - 1 I I . Shainberg, I . , Gal, M.. Ferreira, A. G., and Goldstein, D. (1991). Effect of water quality and amendments on the hydraulic properties and erosion from several Mediterranean soils. Soil Technol. 4, 135- 146. Shanmuganathan, R. T., and Oades, J. M. (1983). Modification of soil physical properties by addition of calcium compounds. Ausr. J. Soil Res. 21, 285-300. So, H. B., Tayler, D. W., Yates, W. J., and McGarity, J. W. (1978). Amelioration of structurally

GYPSUM AND ACID SOILS

31

unstable grey and brown clays. In “Modification of Soil Structure” (W. W. Emerson, R. D. Bond, and A. R. Dexter, eds.), pp. 325-333. Wiley, New York. Sousa, D. M. G., and Ritchey, K. D. (1986). Us0 do gesso no solo de Cerrado. “Anais do I Seminirio sobre o Us0 do Fosfogesso na Agricultura,” pp. 119- 144. EMBRAPA, Brasilia, D.F., Brazil. Sousa. D. M. G., Rein, T. A.. Lobato, E., and Ritchey, K. D. (1992a). Sugestks para diagnose e recommendaGb de gesso em solos de Cerrado. “11 Seminirio sobre o Us0 do Gesso na Agricultura,” pp. 139- 158. IBRAFOS, SP, Brazil. Sousa, D. M. G., Lobato, E., Ritchey, K. D., and Rein, T. A. (1992b). Resposta de culturas anuais e leucena a gesso no Cerrado. “I1 Seminirio sobre o Us0 do Gesso na Agricultura,” pp. 277306. IBRAFOS, SP, Brazil. Sposito, G. (1985). Chemical models of weathering in soils. In “The Chemistry of Weathering” (J. I. Drever, ed.), pp. 186-198. Reidel Publ., New York. Sumner, M. E. (1970). Aluminum toxicity-A growth limiting factor in some Natal sands. Proc. Annu. Congr.-S. Afr. Sugar Technol. Assoc. 44, 1-6. Sumner, M. E. (1990). “Gypsum as an Ameliorant for the Subsoil Acidity Syndrome,” Publ. No. 01-024-090. Florida Institute of Phosphate Research, Bartow. Sumner, M. E. (1992). The electrical double layer and clay dispersion. In “Soil Crusting: Chemical and Physical Processes” (M. E. Sumner and B. A. Stewart, eds.), pp. 1-31. Lewis Publ., Boca Raton, FL. Sumner, M. E. (1993). Sodic soils: New perspectives. Ausr. J . Soil Res. (in press). Sumner, M. E., and Carter, E. (1988). Amelioration of subsoil acidity. Commun. Soil Sci. Plant Anal. 19, 1309-1318. Sumner, M. E., and Miller, W. P. (1992). Soil crusting in relation to global soil degradation. Am. J. Allern. Agric. 7 , 56-62. Sumner, M. E., Fey, M. V., and Farina, M. P. W. (1986a). “Amelioration of Acid Subsoils with Phosphogypsum,” pp. 41 -45. Florida Institute Phosphate Research, Bartow. Sumner. M. E., Shahandeh, H., Bouton, J., and Hammel, J. E. (1986b). Amelioration of an acid soil profile through deep liming and surface application of gypsum. Soil Sci. SOC.Am. J . 50, 1254- 1258. Sumner, M. E., Radcliffe, D. E., McCray, M., Carter, E., and Clark, R. L. (1990). Gypsum as an ameliorant for subsoil hardpans. Soil Techno/. 3, 253-258. Suzuki, A., Basso, C., and Wilms, F. W. W. (1992). 0 us0 de gesso como fonte complementar de calcio em macieira. “I1 Seminirio sobre o Us0 do Gesso na Agricultura,” pp. 225-240. IBRAFOS, SP, Brazil. Syed, Omar, S. R., and Sumner, M. E. (1991). Effect of gypsum on soil potassium and magnesium status and growth of alfalfa. Commun. Soil Sci. Planf Anal. 22, 2017-2028. Taylor, A. J., and Olsson, K. A. (1987). Effect of gypsum and deep ripping on lucerne (Medicago sariva L.) yields on a red-brown earth under flood and spray irrigation. Ausr. J. Exp. Agric. 27, 84 I - 849. van der Watt, H. v. H., and Claassens, A. S. (1990a). Soil physical and chemical changes resulting from surface applications of phosphogypsum and organic mulch. Trans. Inr. Congr. Soil Sci.. 14fh, Vol. 6, pp. 255-256. van der Watt, H. v. H., and Claassens, A. D. (1990b). Effect of surface treatments on soil crusting and infiltration. Soil Technol. 3, 241 -25 1. van der Watt, H. v. H., Claassens, A. S.. and Croft, G. J. B. (1991). Soil chemical changes resulting from application of phosphogypsum and organic mulch. Unpublished manuscript. van Raij, B. (1988). “Gesso agricola na melhoria do ambiente radicular no subsolo.” AssociaGHo Nacional para Difuslo de Abudos e Corretivos Agricolas, Sio Paulo, Brazil.

32

MALCOLM E. SUMNER

van Raij, B., Cantarella, H., and Furlani, P. R. (1988). Efeito na rea@o do solo da absorGb de amBnio e nitrato pel0 sorgo na preseqa e na audncia de gesso. Rev. Bras. Cien. Solo 12, 131-136.

Vitti, G. C., Ferreira, M. E., and Malavolta, E. (1986). Respostas de culturas anuais e perenes. “Anais do I Seminirio sobre o Us0 do Fosfogesso na Agricultura,” pp. 17-43. EMBRAPA. Brasflia, D.F., Brazil. Vitti, G. C., Mazza, J. A., Pereira, H. S., and Dematte, J. L. I. (1992). Resultados experimentais do us0 de gesso na agricultura-Cana de aqlcar. “I1 Seminirio sobre o Us0 do Gesso na Agricultura,” pp. 191-224. IBRAFOS, SP. Brazil. Warrington, D., Shainberg, I., Agassi, M., and Morin. J. (1989). Slope and phosphogypsum’s effects on runoff and erosion. SoilSci. Soc. Am. J . 53, 1201- 1205. White, P. F., and Robson, A. D. (1989). Emergence of lupins from a hard setting soil compared with peas, wheat and medic. Ausr. J . Agric. Res. 40,529-537.

CONSERVATION TILLAGE: AN ECOLOGICAL APPROACH TO SOIL

MANAGEMENT

R. L. Blevins I and W. W. Frye Department of Agronomy and Division of Regulatory Services University of Kentucky Lexington, Kentucky 40546

I

2

1. Introduction

A. Evolution of Conservation Tillage B. Definitions C. Adoption and Suitability D. Advantages and Disadvantages 11. Conservation Tillage and the Environment A. Soil Erosion B. Water Quality C. Soil Productivity D. Energy Use 111. Soil Physical Properties A. Bulk Density and Compaction B. Soil Aggregation and Infiltration C. Soil Water D. Soil Temperature TV. Soil Chemical Properties A. Soil pH B. Distribution of Nutrients in the Soil C. Soil Organic Matter V. Surface Mulch Management A. Benefits from Crop Residue B. Effects of Tillage on Residue Cover C. Growing Cover Crops for Mulch and Nitrogen VI. Nutrient Management A. Efficient Use of Nitrogen B. Management of Fertilizers VII. Pest Management A. Weed Control B. Diseases and Insects VIII. Conclusions A. Progress in Conservation Tillage B. Future Needs and Direction References 33 Advances in Aqonmny. Volume 51

Copyrighr 0 1993 hy Academic Pres5. Inc. All righrs of reproducrion in any form reserved

34

R. L. BLEVINS AND W. W. FRYE

I. INTRODUCTION Conservation tillage is an ecological approach to seedbed preparation and soil surface management. Conservation tillage, in its broad concept, includes many different management practices. Therefore, the choice of the best-suited tillage system may vary widely among different agroecological environments.

A. EVOLUTION OF CONSERVATION TILLAGE During the 1930s, stubble mulching was developed to deal with the severe wind erosion occurring in the Great Plains of the United States and Canada. Crop residue left on the soil surface was found to be especially effective in keeping the soil in place against wind erosion. Early work on stubble mulching and subsequent conservation tillage is presented in a comprehensive review by McCalla and Army (1961). Keeping the residue anchored at the soil surface proved to be equally effective in preventing soil erosion by water. Researchers in the 1940s (Ellison, 1944; Laws, 1941) showed that the falling raindrop is the main detaching agent in sheet erosion by water. Duley and Kelly (1939) noted the development of a thin compacted layer at the soil surface when unimpeded raindrops hit the bare soil. This layer reduced the infiltration rate, and it was observed that additions of residues on the soil surface improved infiltration and absorbed the energy of the falling raindrops. Pioneering studies on the use of mulch in cropping systems was started in Nebraska by Duley and Russel (1939). As early as the late 1940s, attention turned to development of systems that required less tillage. Minimum tillage researchers advocated reduced amounts of tillage in the late 1940s and began to use plant growth regulators for postemergence weed control. The development of new herbicides in the 1960s allowed conservation tillage systems to successfully evolve. In the 1960s, M. A. Sprague in New Jersey reported successful pasture renovation using chemicals as a substitute for tillage. The development of nonselective contact weed control material that became commercially available in the mid- 1960s was a major breakthrough in making conservation tillage work. Refinement of many of the conservation tillage systems we use today in the United States can be attributed to early workers such as George McKibben (Illinois), D. B. Shear and W. W. Moschler (Virginia), G. B. Triplett and D. M. Van Doren, Jr. (Ohio), J. F. Freeman, Harry Young, and S. H. Phillips (Kentucky), and W. M. Lewis and A. D. Worsham (North Carolina), to name a few. By the 1970s, there was a wide range of herbicides available and rapid progress was being made to develop machinery to effectively plant in soils with little

CONSERVATIONTILLAGE IN SOIL MANAGEMENT

35

or without tillage and with minimal disturbance of residues remaining at the soil surface. Interest in soil conservation continued to spread throughout the world in the 1970s and 1980s. By the 1990s, there was a wide range of machinery available specifically developed for conservation tillage. This includes planters, disks, sprayers, and tillage equipment especially designed to leave residues at or near the soil surface. The timing of the successful development of conservation tillage technology was ideal for the U.S. farmers. The 1985 and 1990 Farm Bills contained strong conservation tillage provisions that require any farmer growing annual crops on highly erodible land to have a conservation plan in place by 1990 and fully implemented by 1995. One of the provisions of the bill is mandatory compliance with the conservation program. Conservation tillage that relies on reduced tillage and leaves residues at the soil surface is the most acceptable way for bringing highly erodible land into compliance and should be used when possible to avoid more costly conservation techniques. Conservation tillage has been described by conservationists as the greatest soil conservation practice to come along in the twentieth century. As we approach the twenty-first century, technology to successfully grow crops using a wide variety of conservation tillage systems is available to our farmers. The agricultural alliance of innovative farmers, research scientists, conservationists, and agribusiness has transformed crop residue management and conservation tillage from a concept to a future-oriented system of tillage and residue management that effectively reduces erosion and soil degradation, is cost-effective and will remain environmentally acceptable.

B. DEFINITIONS Conservation tillage is a term encompassing many different soil management practices. One commonly used definition (Mannering and Fenster, 1983) is “any tillage system that reduces loss of soil or water relative to conventional tillage; often a form of non-inversion tillage that retains protective amounts of residue mulch on the surface.” More recently, the definition adopted by the Conservation Technology Information Center (CTIC) is commonly used. CTIC (1992) defines conservation tillage as “any tillage and planting system that maintains at least 30% of the soil surface covered by residue after planting to reduce water erosion; or where soil erosion by wind is a primary concern, maintain at least 1,000 pounds of flat, small grain residue equivalent on the surface during the critical wind erosion period.” This definition is more of a working definition that is especially useful in evaluating crop residue management and a quantitative method of determining if a field meets compliance with the 1985 and 1990 Farm Bills. Within the broad umbrella term “conservation tillage” are several contrast-

36

R. L. BLEVINS AND W. W. FRYE

ing tillage systems, but all are designed to protect the soil and water resources. The following are commonly accepted definitions of the most important tillage systems. No-tillage-This involves very minimal mechanical seedbed preparation (narrow band where the seed is placed) and reliance on herbicides or cover crops or both to control weeds. In no-tillage the soil is left undisturbed from harvest to planting, except for possible nutrient injection (CTIC, 1992). Ridge tillage-The soil is left undisturbed from harvest to planting except for nutrient injection. Planting is completed in a seedbed prepared on ridges with sweeps, disk openers, coulters, or row cleaners. The residue is left on the surface between ridges. Weed control is accomplished with herbicides and/or cultivation (CTIC, 1992). Mulch tillage-The soil is disturbed prior to planting. Tillage tools such as chisels, field cultivators, disks, sweeps, or blades are used. Weed control is accomplished with herbicide and/or cultivation (CTIC, 1992). Conventional rillage-Generally refers to moldboard plowing (inversion of the soil) followed by a secondary tillage operation such as disking and/ or harrowing. Weed control may be accomplished by cultivation or through the use of herbicides. A tillage system commonly used in small grain production systems is stubblemulch farming. Stubble-mulch farming is described by Mannering and Fenster (1983) as a system that maintains a protective cover of vegetative residue on the surface at all times. Tillage equipment used in stubble-mulch farming includes those that mix the soil and those that cut beneath the surface without inverting the tilled layer. According to Mannering and Fenster (1983), the stirring and mixing machines include the disk, chisel-plows, field cultivators, and mulch treader. The noninversion subsurface tillage equipment includes sweeps and rotary rod weeders. Another form of small grain tillage is the “direct seeding” or “no-tillage” system. This system allows the small grain to be seeded directly into the residue of the previous crop, and weed control is accomplished with herbicides. Terminology often used in the past, such as “reduced tillage” and “minimum tillage,” may or may not qualify as conservation tillage depending on the level of residue remaining at the soil surface after planting.

C. ADOPTIONAND SUITABILITY The unique properties of individual soils and their ecological environment determine their limitations and suitability for utilizing conservation tillage methods. The rapid technological advances in the development of new and more ef-

CONSERVATION TILLAGE IN SOIL MANAGEMENT

37

fective herbicides along with the development of equipment that can effectively plow or plant through surface residues in relatively undisturbed soils have made it possible for different versions of conservation tillage to be developed around the world. The increasing interest in and rapid growth of conservation tillage are associated with the increasing pressures for food production around the world and the continuing concern about soil degradation by erosion, compaction, and reduced fertility. Conservation tillage is especially well suited for use on sloping lands. These systems are also well suited for well-drained and moderately well-drained soils. The limiting problem of using conservation tillage on wet soils is the inability to remove excess water early enough to allow the soil to warm up in time enough to plant the crop. Furthermore, the wet, nearly level soils do not benefit as much from the lack of tillage and presence of crop residue on the surface. In fact, this system may add to an already existing problem, namely, wetness and cool soil temperatures. As the world population increases, the demand for food and fiber increases. This results in more pressure to use steep, fragile soils that are extremely erosive when cropped. Although it is not prudent to grow crops on steeplands, the use of conservation tillage and structures such as terraces can at least reduce soil erosion losses.

D. ADVANTAGES AND DISADVANTAGES No single tillage system is suitable for all soils and climatic conditions. For environments where conservation tillage is ecologically suitable, it offers advantages over the moldboard plow or other tillage systems that result in soil inversion or extensive mixing of soil. Advantages include effective soil erosion control, water conservation, less use of fossil fuels normally associated with land preparation, reduced labor requirements, more timeliness of operations or greater flexibility in planting and harvesting operations that may facilitate double cropping, more intensive use of sloping soils, and less risk of environmental pollution. Disadvantages that may be associated with conservation tillage systems include higher herbicide costs, more difficulty in controlling certain weed infestation (e.g., Johnsongrass), and, for imperfectly drained soils, conservation tillage may aggravate the existing wetness limitation. This wetness creates a more anaerobic soil environment. Crop residue mulches left at the soil surface cause slower warming of soils during the spring and has been identified as a serious problem in the Corn Belt of the United States (Griffith et al., 1977). In contrast, the surface residue mulch that may delay germination and early growth in temperate regions often becomes a positive factor in the tropics by reducing the high temperatures that adversely affect plant growth, thereby enhancing germination

R.L. BLEVINS AND W. W. FRYE

38

and growth (Lal, 1982). Conservation tillage is an important component of modern agriculture and is aimed at preserving the soil and water resource base.

11. CONSERVATION TILLAGE AND THE ENVIRONMENT The lack of soil disturbance and the presence of plant residue cover characterize conservation tillage and are primarily responsible for its environmental benefits. Ironically, these characteristics are responsible for most of the management problems that have been identified with conservation tillage in the field. Major environmental benefits of conservation tillage include decreased soil erosion; improved water quality, particularly surface water; enhanced soil quality that may result in increased soil productivity; and decreased use of fossil fuels.

A. Son.EROSION Conservation tillage decreases soil erosion almost in direct proportion to the amount of soil cover left following the tillage practice. The literature abounds with evidence of this. An excellent illustration is the research reported by Mannering and Fenster (1977) and shown in Fig. 1. Although the relationship is strong between soil cover and soil loss, there

30 (d

f

r" 20 10

0

20 o/o

40 Cover

60

80

Figure 1. Relationship between soil loss and residue cover using moldboard-plow (0). chiseland no-tillage).( systems. (Adapted from Mannering and Fenster, 1977.) plow

(v),

CONSERVATIONTILLAGE IN SOIL MANAGEMENT

39

appear to be benefits from lack of disturbance per se independent of crop residue. Rasnake er al. (1986) reported soil losses of 426 kg ha-' with no-tillage soybeans [Glycine max (L.) Merr.] without a cover crop compared to 9050 kg ha-I with conventional tillage (moldboard plowing) soybeans without a cover crop. Corresponding values with a winter wheat (Triticum aestivurn L.) cover crop were 269 kg ha - I for no-tillage and 1142 kg ha - I for conventional tillage. Similar results were found with no-tillage and conventional tillage soybeans in Tennessee (Shelton and Bradley, 1987). The decreased soil erodibility associated with no-tillage appears to be caused primarily by the lack of soil disturbance and its ramifications rather than runoff, since runoff is often similar for no-tillage and conventional tillage (Blevins et al., 1990). Runoff volume is affected greatly by surface roughness as well as residue cover left by the different tillage practices (Romkens et al., 1973; Mannering et al., 1975; Lindstrom and Onstad, 1984). Soil loss, on the other hand, is determined largely by the erosivity of the rainfall and runoff and the erodibility of the soil. No-tillage decreases erodibility of the soil in comparison to conventional tillage.

B. WATERQUALITY Public concern is currently focused on the effects of agricultural practices on soil erosion and water quality. Runoff from farmland has the potential to carry large amounts of fertilizers, pesticides, and sediments into lakes and streams. The surface runoff waters may readily affect the groundwater in some landscapes such as karst regions. Much speculation has occurred in recent years about the effects of various tillage systems on water quality. Some researchers (Blevins et al., 1983a; Wendt and Burwell, 1985) suggested that conservation tillage systems create fewer environmental hazards than conventional tillage because it reduces runoff and soil erosion losses. Others have suggested that the increased infiltration resulting from reduced runoff may increase the potential for groundwater pollution with agricultural chemicals.

1. Surface Water Tillage systems affects the amount of water moving both over the surface and through the soil. Moldboard or other inversion types of plowing increase the rate at which water moves into soil over the short term. However, after several rainfall events, a crust often forms at the surface, reducing infiltration rate. Conservation tillage reduces soil losses (Blevins et al., 1990; Mostaghimi et al., 1988), but does not always reduce the volume of runoff as effectively as it reduces sediment losses.

40

R. L. BLEVINS AND W. W. FRYE

No-tillage usually decreases soil erosion and chemical losses in comparison with conventional tillage (Angle et al., 1984; Mostaghimi et al., 1988; Blevins et al., 1990). In contrast, studies by Lindstrom and Onstad (1984) showed a higher runoff volume for no-tillage as compared to conventional tillage. Because fertilizer and herbicides are commonly applied on the soil surface in no-tillage systems, there may be a greater threat of losses of agricultural chemicals if rainfall occurs soon after application. Johnson ef al. (1979) reported that, even though concentrations of agricultural chemicals were much higher in sediments than in water components, 75 to 99% of chemical losses occurred in runoff water. Although the reduction of soil sediment losses under conservation tillage systems strongly influences total loss of chemicals into surface water, the reduced runoff volumes are the overriding factor determining the lower total chemical losses from the field. Studies were conducted in Kentucky under natural rainfall using conventional tillage, no-tillage, and chisel-plow without secondary tillage (Blevins et al., 1990) to evaluate tillage-water quality relationships. The highest runoff and sediment losses were observed for conventional tillage and no differences occurred between the no-tillage and chisel-plow systems. The ranking of these tillage systems from highest to lowest in amount of NO-3, P, and atrazine losses was conventional tillage > no tillage > chisel-plow tillage. In this study, chiselplowing across the slope without secondary tillage was as effective as no-tillage in reducing runoff and chemical losses. These findings further support the notion that a combination of surface roughness and residue left at or near the surface is an effective method of reducing chemical losses to surface water. The lower runoff volumes from no-tillage or conservation tillage methods produce lower total chemical losses, not only because soil losses are far less, but also because the amount of pesticide transported offsite is proportional to total runoff volume (Laflen et al., 1978).

2. Groundwater Quality Today, potential chemical movement from agricultural land into groundwater are a concern to our society. The agricultural community is being asked to develop management practices that minimize losses to groundwater. Most interest on nutrient contamination of groundwater is focused on nitrate; however, phosphate is not without scrutiny also. The tendency for nitrate to leach rather than being absorbed to soils is due to most soils having a net negative charge. Consequently, nitrate is repelled from the soil surface and readily leached. Although NO - 3 transport through soils has been extensively studied, research on the effect of tillage practices under field-scale conditions is seriously lacking. Work in Kentucky (Thomas et al., 1973; Tyler and Thomas, 1977)demonstrated greater NO - 3 leaching with no-tillage than with conventional tillage sys-

CONSERVATION TILLAGE IN SOIL MANAGEMENT

41

tems.They concluded that no-tillage enhanced the preferential leaching of NO - 3 through macropores. W. M. Edwards er al. (1992) reported that the intensity of rainfall following fertilizer or pesticide applications plays a significant role in leaching losses. A light rainfall following chemical application promotes movement into soil micropores, thus partially immobilizing the chemical for subsequent rainfall. In this scenario according to Kanwar et af. (1983, the macropores facilitate the bypassing of water-filled micropores, thus reducing the leaching of NO - 3 and other mobile compounds relative to displacement of NO -]-rich micropore water. Tyler et af. (1992), reporting on a joint field study from Tennessee and Kentucky, found a high preferential flow out of the root zone in winter and early spring when NO -3-N and herbicide concentrations were low. No-tillage appeared to reduce NO-3 leaching under corn and under soybean-wheat-corn rotation during the wheat-corn period of the rotation. Infiltration rates on the Tennessee site were relatively low with insignificant differences between no-tillage and conventional tillage. The environmental issues of pollution of natural water are a major concern in the United States where the land area under conservation tillage has dramatically increased. Consequently, the use of pesticides, especially herbicides, has also increased. Herbicide use is especially heavy for corn and soybean production regardless of the tillage system that is practiced. Although conservation tillage usually decreases surface runoff and erosion, it is known to increase leaching and subsurface movement of water in some soils and climatic zones. This potentially higher rate of infiltration and higher subsurface drainage in conservation tillage creates a situation that may result in deeper leaching of agricultural chemicals applied to the soil. The presence of large numbers of earthworm channels in no-tillage soil environments serves as passageways that conduct water (Edwards et d., 1988). Water moving through such macropores carries dissolved chemicals in aqueous phase to subsurface horizons and may pollute the groundwater. Consequently, water leaching through conservation tillage managed soils has the potential to transport more NO - 3 and pesticides than water from plowed treatments. How conservation tillage affects the movement of fertilizers and pesticides is related to other factors as well, such as time of their application in relation to time of land preparation, occurrence of rainfall, and nature of the rainfall. Highintensity rainfall often results in rapid surface losses and deep leaching by preferential flow. Volatilization of herbicides is also influenced by tillage systems that leave residues at the surface. Glotfelty (1987) reported that larger quantities of pesticides are lost by volatilization than by deep leaching or runoff. According to Wagenet (1987), leaching is the third major source of pesticide losses from the root zone. The surface soil in conservation tillage contains relatively more organic matter than that in conventional tillage. Therefore, the retention of

42

R. L. BLEVINS AND W. W. FRYE

herbicides such as paraquat, atrazine, and metolachlor is likely to be greater. This partly offsets the effects of macropores that expedite deep leaching (Wagenet, 1987). There is substantial research that supports the conclusion that conservation tillage systems that leave substantial residues at the soil surface in conjunction with surface roughness will reduce the total losses of agricultural chemicals from a field by way of surface runoff because of both less sediment losses and lower runoff volume. How conservation tillage affects our groundwater as compared to potential pollution from conventional tillage is less clear at this time. It is difficult to determine the full effects of deep leaching with the limited research now available. This is a research area receiving high priority today with numerous in situ leaching research studies in progress.

C. SOILPRODUCTIVITY The tillage system affects soil productivity mainly through its influence on soil organic matter, soil erosion, and soil water supply. The relationships between tillage and these three soil properties are discussed elsewhere in this chapter. Here we will discuss how these soil properties affect soil productivity.

1. Soil Organic Matter Soil organic matter contains plant nutrients that are mineralized as it decomposes. What effect these nutrients might have on soil productivity is clouded by the fact that some of the plant nutrient requirements are usually supplied by commercial fertilizers and other inorganic amendments and some are supplied by soil minerals. Therefore, most farmers are not dependent on nutrients in soil organic matter to grow their crops. Certainly, increasing soil organic matter increases the nutrient-supplying capacity of the soil and decreases the need for inorganic amendments. 2. Soil Erosion

Soil productivity factors that are usually diminished by soil erosion include direct loss of soil fertility, loss of soil organic matter, deterioration of soil structure, and decreased water-supplying capacity (capacity to provide water to growing plants). The primary seat of fertility of many soils is the topsoil. Direct loss of soil fertility occurs when surface-applied fertilizers or available plant nutrients attached to soil particles are removed during runoff and erosion. Indirect loss of soil fertility occurs in the organic matter that is lost when topsoil erodes. Burwell et al. (1975) found that sediment transport accounted for more

CONSERVATIONTILLAGE IN SOIL MANAGEMENT

43

than 95% of the N and P lost and most of the K lost from fallow, continuous corn, and rotational corn treatments. A classic study in Missouri (Miller and Krusekopf, 1932) showed that nutrients removed by erosion, expressed as a percentage of the amounts removed by continuous corn, were N, 55%;P, 90%; K, 605%;Ca, 550%; and Mg, 290%. A rotational cropping system of cornwheat-clover markedly decreased soil erosion and reduced nutrient losses to 22, 36, 214, 212, and 97%, respectively, of the N, P, K, Ca, and Mg removed by the crops. As these data illustrate, loss of soil organic matter by erosion causes a disproportionate loss of several nutrients. Organic matter is the most abundant indigenous source of N and some secondary nutrients and micronutrients, especially in strongly weathered soils. Generally, eroded soils have higher acidity and lime requirements than uneroded soils (Frye et al., 1982). Soil erosion often results in replacement of topsoil with more-acid subsoil, selective removal of the baseforming elements (K, Ca, and Mg) from the topsoil, and removal of applied lime before it reacts to neutralize soil acidity (Frye et al., 1985). Soil fertility and lime requirement can be corrected so readily by soil amendments that it is difficult to claim soil productivity benefit through these parameters directly from controlling soil erosion. However, the added costs of maintaining productivity with purchased amendments is an obvious disadvantage. Furthermore, substantial productivity is often sacrificed before nutrient and lime deficiencies are discovered. Deteriorating soil structure during soil erosion contributes to diminished productivity in at least two important ways-decreased infiltration and storage of available water and impaired root exploitation of the soil. Lower organic matter and higher clay contents associated with erosion make soil more susceptible to compaction. If it does not actually decrease the available water-holding capacity of the soil, compaction reduces infiltration and restricts plant root growth. Both decrease the water-supplying capacity of the soil and diminish the soil’s yield capacity. Erosion decreases the soil’s water-supplying capacity in still another way. Accelerated erosion removes soil from the surface much more rapidly than the soil profile deepens. Thus, the rooting depth rapidly decreases in soils that are shallow to a root-restricting layer such as a bedrock or hardpans (Frye et al., 1983).

3. Water-Supplying Capacity The effects of water-supplying capacity on soil productivity are highly dependent on the amount and distribution of rainfall during the growing season. The water-supplying capacity has little effect on crop yields when rainfall is abundant and well distributed, but has a profound effect during droughty seasons. Soil water is discussed more thoroughly in Section II1,C.

44

R. L. BLEVINS AND W. W. FRYE

D. ENERGYUSE Conservation tillage, especially no-tillage, along with N fertilizer management, offers farmers one of their greatest opportunities to conserve energy in crop production. Moldboard plowing to a 20-cm depth requires an estimated 17 liters ha-' of diesel-fuel equivalents (DFE = 41 MJ per liter) of energy. In contrast, chisel-plowing to 20 cm requires 1.18 and disking requires 0.64 liters ha-l DFE (Frye, 1984). In no-tillage, of course, tillage is eliminated, and this energy is conserved. Some, but not all, of the energy conservation is offset by slightly greater need for herbicides in no-tillage. Frye (1984) assigned energy values for corn production under conventional moldboard-plow tillage, chisel-plow tillage, and no-tillage based on inputs and practices outlined by Wittmuss and Yazar (1981) for those tillage systems. These values are shown in Table I. Tillage operations used 7% of the total production energy required in the conventional plow system and 4% of that re-

Table I Energy Analysis of Corn under Three Tillage Systems in Nebraska" Tillage system Moldboard plow Operation or input Moldboard plow Chisel-plow Disk Apply herbicides Plant corn Herbicides (manufacturing) Machinery (manufacturing and repair) Fertilizer (168 kg N h a - ' ) Harvest grain Dry grain Chop stalks Production energy (total of above) Energy in grain Energy output : input ratio

Chiselplow

DFE (liters ha

Notillage ~

')

17 II

12 I 4 22

6 I 4 22

I7

15

238 15 80 7 413 2030 4.9

238 15 80 7 399 2030

22 6 238 15 80 7 374 2030

5.1

5.4

1 5

"From Frye ( 1984). 'Based on 16 Megajoule kg - I of grain and 41 Megajoule liter I of diesel fuel. Grain yields (3-year average) were approximately 5200 kg ha I for all tillage systems. ~

CONSERVATIONTILLAGE IN SOIL MANAGMENT

45

quired in the chisel-plow system. Total production energy values for chisel-plow and no-tillage, respectively, were about 97 and 91% that of the conventional system.

111. SOIL PHYSICAL, PROPERTIES For a crop production system to be widely accepted, it must maintain the soil’s physical quality. Soil physical factors that influence water infiltration and soil erodibility include organic matter content, aggregation, porosity, and density. The amount of mechanical mixing and levels of residue left on or near the soil surface directly influence these soil properties.

A. BULK DENSITY AND COMPACTION Soil physical properties and conservation tillage are influenced by surface and internal drainage, nature and amount of clay, climate, drainage, physiography, vehicular traffic, and soil and crop management systems. Because of the variation in climate and soils, it should be no surprise that contradicting data appear in the literature, particularly regarding the relationships of tillage and cropping systems to bulk density and compaction. In Kentucky, for example, we found no significant effect on bulk density after 20 yr of corn production comparing no-tillage and moldboard-plow tillage (Table 11). The surface 0 to 5 cm of the no-tillage soil had slightly lower bulk density than the surface of the moldboard-plow system. These plots were sampled in April prior to tillage. The moldboard-plow plots had reconsolidated following the tillage operations performed 11 months earlier. In contrast, Gantzer and Blake (1978) in Minnesota found significantly higher bulk densities of a clay loam soil in no-tillage than in plow tillage. A similar study on a poorly drained Haplaquoll in Minnesota reported a higher proportion of macropores in untracked zones of no-till as compared to conventionally tilled plots. Hill and Cruse (1985) reported no significant effects of tillage methods (no-tillage, plowingtillage, and minimum tillage) on bulk density of a loess-derived Iowa soil. Contradicting results have been reported also for wheel traffic impact on soil properties. Wheel traffic in no-tillage has been demonstrated in some cases to cause compaction, resulting in high bulk density and low infiltration in tracked soil zones (Voorhees, 1983). In contrast, Lindstrom et al. (1981) observed no significant effects of wheel tracks on infiltration rate in no-tillage treatment. Regardless of the ecological region, compaction and crusting are major production constraints to intensive row-crop agriculture (Lal, 1989). Some soils in

46

R. L. BLEVINS AND W. W. FRYE Table I1

Bulk Density of Soil after 20 Years of Continuous Conventional Tillage (CT) and No-Tillage (NT) Corn"

Bulk density Soil depth (cm) 0-5b 5-156 15-30

CT

NT

Undisturbed soil

1.27 1.27 1.47"

1.17 1.36' 1.47'

1.15 1.28 1.41

'Adapted from Ismail et al., 1993. "Sampled as single increment in CT. 'Significantly higher than value above in same column by LSD ( p < 0.05).

Africa and India in aridhemiarid regions are naturally compacted and restrict root growth. Mechanical loosening as a part of conservation tillage may be required to improve crop growth. Soil compaction is caused by wheel traffic at the soil surface and by formation of a plow-pan in subsurface layers (Lal, 1985). One strategy to deal with traffic-induced compaction is the adoption of notillage or reduced tillage (conservation tillage) methods so the number of operations with equipment is reduced. Guided traffic is another approach for dealing with compaction, in which farming operations are organized such that tractor wheels always follow the same tracks, thus reducing the volume of soil influenced by wheels. Soils susceptible to compaction occur in nearly all agroecological zones. However, upland soils in the tropics may be more susceptible to compaction than equivalent soils in the temperate zone (Lal, 1989). This is due to the antecedent low level of organic matter, lack of ameliorative effect of freezing and thawing, and predominance of low-activity clays.

B. SOILAGGREGATION AND INFILTRATION Soil aggregation involves the binding together of several soil particles into secondary units (Unger and McCalla, 1980). Soil aggregates, especially water stable aggregates, are of special importance for high water infiltration and good soil structure. These properties help determine soil quality and directly influence soil and water conservation. For soils and environments where conservation till-

CONSERVATION TILLAGE IN SOIL MANAGEMENT

47

age is ecologically suitable, the system usually maintains or improves the physical quality of the soils. Water-stable aggregation is often used as a measure of soil structure and is a suitable index of soil resistance to dispersion and compaction. Plant emergence, water infiltration, and soil erosion are directly influenced by aggregate stability. Results from soil aggregation studies on four Indiana soils by Mannering et al. (1975) are given in Table 111. As tillage intensity increased, soil aggregation decreased. Aggregation was highest in the 0- to 5-cm layer of notillage-treated soil. Tillage studies on silty soils in Germany (Ehlers, 1979) showed that no-tillage improved aggregate stability of surface soil under no-tillage. They attributed the improved structure to increased concentrations of organic matter in the surface, resulting in less slaking during heavy rains. Even though total porosity was increased by tillage, the macropores connecting the soil surface to the subsoil were enhanced, thus improving infiltration. Douglas and Goss (1982) found that after repeated direct seeding in Britain, aggregate stability of the topsoil was improved. Again, the improved structure was thought to be related to an increase in organic matter at the soil surface. Research on a poorly drained soil in northern Ohio (La1 et al., 1989) showed that median aggregate size tended to be higher (about 22%) for no-tillage treatments than for plow-till treatments. They concluded that higher organic matter content and lack of mechanical soil disturbance in no-tillage may be responsible for relatively higher percentage aggregation, larger median aggregate size, and more kinetic energy required to disrupt aggregates as compared to plow-tillage treatments. Generally, much of the observed improvements in soil structure observed in conservation tillage are attributed to high biotic activity, especially earthworm

Table Ill

Effect of Tillage on Soil Aggregation of Four Indiana Soils after 5 Years" Aggregate index Tillage system Moldboard plow Chisel-plow Till-plow No-tillage

0- to 5-cm depth

5- to 15-cm

0.35

0.47

0.46 0.47 0.77

0.56

"Adapted from Mannering er al. (1975).

depth

0.56 0.70

48

R. L. BLEVINS AND W. W. FRYE

activity. Earthworms flourish in a no-tillage or residue-managed system. Ehlers (1975) attributed the higher infiltration rate of loess soil in Germany to the greater number of worm channels and to their continuity, which was better in notilled soil than in plowed soil. Edwards et d . (1988) concluded that no-tillage effectively preserve the macropores during the intercrop period, whereas tillage disrupts many of them.

C. SOILWATER The capacity of a soil to supply water to plants during periods of water stress (i.e., its water-supplying capacity) is determined by the soil’s available waterholding capacity, infiltration and percolation rates, evaporation rate, effective rooting depth, position on the landscape, and depth to the water table (if within the rooting zone). Of these, tillage can significantly affect infiltration and evaporation in essentially all soils and affect available water-holding capacity and effective rooting in some soils.

I. Infiltration Tillage loosens soil, creating large openings that water can move into rapidly. It also leaves the surface rough with many microbasins that retain water on the soil surface, allowing more time for infiltration before runoff occurs (Blevins et al., 1990). Thus, a recently tilled soil has higher infiltration rate than the same soil that has not been tilled recently. However, in the case of moldboard plowing, the enhanced infiltration may be relatively short-lived. Rainfall may cause a crust to form at the surface and that crust reduces the infiltration rate. Infiltration rates with chisel-plow tillage and no-tillage are more likely to be sustained throughout the growing season, thereby enhancing the water-supplying capacity. Two major factors contributing to the sustained infiltration rates with conservation tillage are improved soil aggregate stability mentioned earlier and the effects of plant residue cover. The plant residues left on the soil surface with conservation tillage protect the soil from raindrop impact, thereby preventing crusting. They also slow the flow of surface runoff, permitting greater infiltration. Tillage systems, then, affect the amount of water moving both over the surface and into and through the soil. That is largely why the tillage system has such a great influence on soil erosion and water-supplying capacity. It also has implications in leaching and potential groundwater contamination as discussed in Section II,B above.

CONSERVATION TILLAGE IN SOIL MANAGEMENT

49

2. Evaporation The plant residues left on the soil surface slow the rate of water evaporation under conservation tillage relative to conventional tillage. Plant residues shade the soil from solar radiation, insulate the soil from heat in the air, and impede movement of water vapor from soil to air, thus developing high humidity and permitting condensation within the residue mulch (Bond and Willis, 1969; Phillips, 1984). The mulch slows the rate of evaporation but does not change the amount of water that can ultimately be removed from the soil. Therefore, the mulch provides protection against short-term but not long-term droughts. According to Bond and Willis (1969), the protection lasts 7 to 14 days. It is most effective during the early part of the growing season when evaporation rather than transpiration is the primary avenue of water loss from the soil. Utomo (1986) found that the effect of a mulch from a killed hairy vetch (Vicia viffosa Roth) cover crop on soil water in the 0- to 15-cm depth was apparent virtually throughout the entire 1985 corn growing season (Table IV). Clearly, no-tillage was superior to conventional tillage (Table V). Part of that effect may have arisen from enhanced infiltration, resulting in more water in the soil. When fertility, particularly N, is adequate, the improved soil water status usually results in higher crop yields for no-tillage than for conventional tillage (Phillips et al., 1980; Frye er al., 1988). Below optimum N levels, however, Table IV Gravimetric Soil Water at 0- to 15-cm Soil Depth in 1985 as Affected by Hairy Vetch Cover Crop with No-Tillage Corn Soil water (kg kg-') Sampling date

Fallow

14 May 31 May 14 June 20 June 27 June 8 July 20 July 5 Sept.

0.23 0.27 0.28 0.23 0.22 0.21 0.15

0.20

Hairy vetch

0.21 0.27 NSd 0.27 NS 0.28 0.27' 0.26' 0.19b 0.23 NS

"Adapted from Utomo (1986). "Significantly higher at 5% level. Significantly higher at 10% level for that date. "NS, not significantly different.

50

R. L. BLEVINS AND W. W. FRYE Table V Gravimetric Soil Water at 0- to 15-em Soil Depth in 1985 under Conventional Tillage and No-Tillage Corn with Hairy Vetch Cover Crop" Soil water (kg kg - I ) Sampling date

Conv. tillage

14 May 31 May 14 June 20 June 27 June 8 July 20 July 5 Sept.

0.20 0.24 0.23 0.22 0.19 0.18 0.13 0.18

No-tillage

0.21 NS" 0.27' 0.27E 0.28' 0.27' 0.26' 0.19' 0.23'

OAdapted from Utomo (1986). "NS, not significantly different ( p > 0.05). 'Significantly greater ( p < 0.05).

conventional tillage usually outyields no-tillage because the plowing speeds up N mineralization from soil organic matter.

D. SOILTEMPERATURE Colder soil temperature in the early spring can be a serious drawback to the adoption of conservation tillage. The less the soil disturbance, the more residue that is left on the surface and the colder the soil in the spring. Of course, the greatest soil temperature problem is likely to be with no-tillage. The problem occurs mostly with early planted crops when the soil is still cold. The few degrees colder temperature under conservation tillage can make a profound difference in seed germination and plant growth. A common observation is conventional tillage corn, for example, emerging sooner and growing more vigorously than no-tillage corn early in the season (Phillips, 1984). Later in the season, the lower soil temperature turns beneficial, especially as it affects water-supplying capacity. Moody et al. (1963) in Virginia found that soil water in the 0- to 46-cm depth averaged 2.1 cm more under 6.8 Mg ha-' of straw mulch than in a bare soil. Maximum soil temperatures were 1.0 to 3.0"C cooler under the mulch in May and 3.2 to 3.8"C cooler in June. Corn yields averaged for the 3 yr of the study were 4.15 Mg ha-' for bare soil and 7.02 Mg ha-l for the mulched soil, a clear benefit from the additional soil water. The additional soil water is, of course, only partly attributable to lower soil temperature.

CONSERVATION TILLAGE IN SOIL MANAGEMENT

51

Another advantage of cooler soil later in the season is the slower rate of organic matter decomposition (Utomo, 1986). With slower decomposition, the mulch remains effective longer, and soil organic matter content increases with time. Utomo (1986) measured soil temperatures at 5-cm depth daily from corn planting to 1 mo. later. His measurements, averaged weekly, are shown in Table VI. Differences between conventional tillage and no-tillage ranged from 1.5 to 4.9"C cooler under no-tillage without a cover crop (corn residue only), 1.7 to 5.5"C cooler for no-tillage under a killed hairy vetch mulch, and 1.9 to 3.9"C cooler for no-tillage with a mulch from a rye (Secule cereale L.) cover crop. Differences between no-tillage and conventional tillage grew larger with time. Utomo's measurements were made between the rows, thus under plant residues in no-tillage and bare soil in conventional tillage. When readings were taken in the row of no-tillage directly in the slit made by the no-tillage planter, the temperature values were about halfway between values from between the rows of no-tillage and values from conventional tillage. Thus, Utomo (1986) concluded that the temperature of the environment of the seeds and seedlings in notillage was not as different from that in conventional tillage as had previously been thought based on between-the-row measurements. A number of techniques have been developed to ameliorate the early-season soil temperature problem in no-tillage. Most of these techniques involve removing or redistributing some of the plant residue to permit radiation energy to reach the soil and facilitate drying and warming. Munawar er ul. (1 990) experimented with early-killed versus late-killed rye

Table VI Weekly Averages of Daily Maximum Temperatures at 5-cm Depth in 1985 Soil temperature ("C) Corn residue

CT

Date _

_

~

5/23-5126 5121-612 613 - 619 6/10-6/16

6/11-6123

_

_

22.1 21.3 21.1 21.3 30.4

NT* ~

Hairy vetch

Rye

CT

NT

CT

NT

23.2 28.9 28.3 26.1 29.7

21.5 NS 23.4d 24.9d 24.1 NS 25.9

23.8 28.4 27.8 25.6 28.7

21.9 NS 25.9d 25.gd 24.5 NS 24.gd

~

21.2 NS' 23.1d 24.1d 24.5 NS 25.5d

"Adapted from Utomo (1986). bCT, conventional tillage; NT, no-tillage. 'NS. no significant difference between CT and NT. dSignificant difference ( p < 0.05) between CT and NT on a particular date.

52

R.L. BLEVINS AND W. W. FRYE

cover crop as a means of reducing the amount of mulch and permitting earlier warm-up of the soil. Early kill (killed with herbicides 3 wk before corn planting) resulted in 1 to 2°C higher maximum daily soil temperatures at 5 cm than late kill (killed at corn planting on May 14). By late June, the soil temperatures were essentially the same for the two treatments. They concluded that early killing of the rye had no significant effect on soil temperature. Soil water was generally greater in the early-kill treatment, which would tend to suppress the temperature differences between early-killed and late-killed rye. Row cleaners are designed to attach to no-tillage planters and remove the plant residue from the rows as the crop is planted. Some row cleaners also till a narrow band of soil, In Kentucky, Murdock et al. (1992) used row cleaners to remove residue from a 25-cm strip into which corn was planted no-tillage. Soil temperatures were 0.5 to 1.7"C warmer where row cleaners were used, but corn yields were about the same as for no-tillage without row cleaners and for conventional tillage. These efforts are probably more beneficial on soils that tend to be wet and cold at corn planting time, for example, moderately well or somewhat poorly drained soils, soils with fine-textured A, horizons, and soils farther north. Efforts to warm up the soil earlier are generally effective in raising the soil temperature by a small amount; however, in the end, they seem to make little or no difference in yield, at least not under Kentucky's climatic conditions. That is consistent with observations that no-tillage crops overcome the early season setback and overtake or surpass conventional tillage crops as the season progresses. It is consistent also with the earlier statement that cooler soil temperature becomes an asset as the season progresses.

IV. SOIL CHEMICAL PROPERTIES Long-term effects of tillage on soil chemical properties are of concern to farmers, agronomists, and environmentalists. No-tillage creates a different soil environment and management needs from the plow-tillage. First, in no-tillage there is no mixing of soil amendments. Since most fertilizers, lime, herbicides, and other chemicals are applied to the soil surface, one would expect a change in soil profile distribution because of their concentrations at the soil surface. Also, in conservation tillage systems that maintain residues at the soil surface, the mulch changes the evaporation rate of water from the soil surface. The higher soil water near the soil surface enhances root growth and diffusion of nutrients in this zone where fertilizers were applied. In a continuous no-tillage system, distribution of nonmobile ions will be far from uniform. The surface application of fertilizer may result in more losses of N by ammonia volatilization and more loss of nutrients in surface runoff than

CONSERVATION TILLAGE IN SOIL MANAGEMENT

53

when fertilizers are incorporated annually by plowing under. Plowing under the previous year's crop residue results in more uniform distribution in the plowed layer.

A. S O I L ~ H Numerous studies conducted in temperate climate zones show that no-tillage results in the acidification of the surface layer when continued for several years (3 to 5 ) . Findings from a classic long-term tillage study in Ohio on Wooster silt loam (Typic Fragiudalf) indicated significant acidification of the surface 0 to 7.5 cm under no-tillage (Dick, 1983; Dick et al., 1986). Moschler et al. (1973) also reported increased acidification of the surface layer under no-till. In Kentucky, Blevins et af. (1977) observed that soil pH was lower with no-till than plowtill. The rapid acidification of surface soil under no-tillage is more prevalent in the eastern portion of the United States, where there is usually an excess of rainfall over evapotranspiration for part of the year. The accelerated acidification related to no-tillage has been attributed in part to decomposition of the concentrated layer of organic residues at the surface with subsequent leaching of resultant organic acids into mineral soil, mineralization and subsequent nitrification of organic nitrogen, and losses of basic cations such as Ca and Mg by leaching. However, the major source of acidification in no-tillage corn production is the nitrification of NH,' from acid-forming N fertilizer applied to the soil surface. Soil pH changes after 10 yr no-tillage and conventional tillage corn production in Kentucky are shown in Table VII. These data illustrate the principle that increased increments of acid-forming N fertilizer are strongly correlated with the lowering of soil pH and that no-tillage results in a greater acidification of the surface soil horizon, particularly the 0 to 5 cm. As soil pH declines, a decrease in exchangeable Ca and an increase in exchangeable A1 and Mn can be expected (Blevins ef af.,1983b). In this study, soil pH declined rapidly during the first 5 yr without a significant influence on corn yields. However, between 5 and 10 yr, exchangeable Al more than doubled and crop yields were significantly reduced on unlimed treatments. The decrease in soil pH associated with no-tillage described here does not happen in all ecological regions. In contrast, Alfisols in southwestern Nigeria have shown lower rates of acidification in no-till than in plow-tillage systems (Lal, 1989). How conservation tillage influences soil pH will vary with the system employed. Use of mulch tillage and other systems intermediate to no-tillage versus moldboard plow-tillage will be affected differently. The degree and depth of mixing of soil will influence whether or not an acid surface layer develops. Systems where crop rotations are used and the amount of acid-forming fertilizers is reduced generally retard acidification.

R. L. BLEVINS AND W. W. FRYE

54

Table VII

Effects of 10 Years of Continuous No-Tillage (NT) and Conventional Tillage (CT) Corn on Soil pH of Maury Soil in Kentucky“ Soil pH Unlimed

Limed

Depth (cm)

N rate (kg ha-’)

NT

CT

NT

CT

0-5

0 168 0 I68 0 168

5.8 4.8 6.0 5.6 6.2 6.2

6.4 5.8 6.4 5.9 6.6 6.2

7.3 7.0 6.9 6.6 6.6 6.6

6.9 6.6 7.0 6.6 7.0 6.8

5- I5

15-30

“Adapted from Blevins er 01. (1983b).

It is encouraging that the acidification problem occurs in a thin layer at the soil surface so neutralization is made easier. Application of agricultural lime to the soil surface can ameliorate the problem, as shown in Table VII. The no-tillage treatments receiving periodic lime applications have surface (0-5 cm) pH values greater than those under conventional tillage where the lime gets mixed throughout the upper 0- to 15-cm zone each year during tillage operations.

B. DISTRIBUTION OF NUTRIENTS IN THE SOIL The undisturbed soil in a no-tillage system tends to accumulate the relatively immobile ions at the soil surface. Phosphate and potassium can be expected to concentrate at the surface when fertilizers containing these ions are added regularly. Calcium and Mg introduced as liming materials move to deeper layers (Table VIII), but very slowly compared with mobile ions such as nitrates. Tracy et al. (1990) determined that no-tillage wheat research plots after 16 yr accumulated greater NO,-N, SO,-S, and PO,-P in the 0- to 2.5-cm soil depth than plowed plots. Mineralization of organic N, P, and S can be a major source of plant-available nutrients near the surface of no-tilled soils. The presence of higher levels of Mehlich extractable P (Table VIII) is a common phenomenon observed in no-tillage soils. Because of the high inherent level of P in Maury soils in Kentucky, P fertilizer was not applied during the study. The higher P in the 0- to 5-cm layer of no-tillage is attributed to greater storage and cycling of P in organic matter of no-tilled than conventional tilled soils

CONSERVATION TILLAGE IN SOIL MANAGEMENT

55

Table VIII Mehlich Extractable P and Exchangeable K, Ca, and Mg at Three Soil Depths as Affected by Tillage and N-rate after 20 Years of Continuous Corn Production" Mehlich Ext. P (mg kg-l)

Exch. K (mg kg-')

N rate (kgha-')

CT'

NT"

CT

0 168

I08 I06

150 134

215 224

0 168

I06 101

80 76

140 141

0

I09 I17

95 102

100

I68

89

NT

Exch. Ca (mg kg-')

Exch. Mg (mg kg-l)

CT

NT

CT

NT

0-5 cm 340 26 1

1073 909

1304 1184

175 154

221 226

5-15 cm 129 117

1094 998

1051 974

I78 187

140 136

15-30 cm 81 78

1032 966

923 933

146 159

85 91

'Adapted from Ismail el al., 1993. *CT,conventional tillage (moldboard plowing and disking). CNT,no-tillage.

(Ismail et al., 1993). Also, the solubility of P is known to be enhanced by the presence of organic matter as in the case of no-till. Again, data from Kentucky showed that exchangeable K was not significantly affected by N rates (Table VIII); however, it was greater in the 0- to 5-cm depth of no-tilled than conventional tilled soils. Without mechanical mixing, K continually accumulated near the surface of no-tillage plots, whereas conventional tillage resulted in mixing of K in the surface 20- to 25-cm depth depending on the depth of plowing. Exchangeable Ca and Mg were strongly influenced by fertilizer rates, decreasing with increasing N fertilizer (Table VIII). The Ca and Mg that had accumulated in the surface 0- to 5-cm after 20 yr was probably resulted from lime applications made after 1980. In contrast, unlimed plots (Table IX) after 10 yr had lower Ca and Mg in the 0- to 5-cm layer of no-tillage plots. Below 5 cm, extractable P and exchangeable Ca, Mg, and K were higher for conventionally tilled treatments. The surface layer of long-term no-tilled soils is characterized by higher organic matter and organic N, which can be a valuable N source. Mineralization of organic N in this zone can make significant contributions to nutrition of crops grown on these soils. The availability and N transformations associated with this zone are discussed in detail in Section VI.

R. L. BLEVINS AND W. W. FRYE

56

Table IX

Effects of 10 Years of Continuous No-Tillage (NT) and Conventional Tillage (CT) with Corn on Exchangeable Ca,

Mg, and K of Maury Silt Loam Soil in Kentucky'

Mg (cmol, kg - I )

(cmol, kg ')

NT

CT

NT

CT

NT

CT

5.48 2.61 5.96 4.74 6.61 6.10

7.84 6.13 7.48 6.19 7.49 6.91

0.33 0.47 0.62 0.55 0.58 0.54

0.70 0.59 0.72 0.55 0.71 0.63

0.68 0.44 0.34 0.21 0.21 0.20

0.39 0.34 0.43 0.37 0.29 0.21

Ca

Depth (cm)

N rate (kg ha-')

0-5

0 168 0 168 0 168

5- 15 15-30

K

(cmol, kg - I )

~

"From Blevins et al. (1983a).

C. SOILORGANIC MATTER When conservation tillage and conventional tillage are compared under similar conditions, conservation tillage, especially no-tillage, results in soils having higher soil organic matter contents after a few years. That is not to say that conservation tillage always increases the soil organic matter content of soil or that conventional tillage always decreases it. If the soil organic matter content is high (e.g., long-term grass-legume sod) at the start of the comparison, it may decrease under both tillage systems, but it is likely to decrease faster and to a lower level under conventional tillage. If, on the other hand, the comparison is made on a soil that is initially low in soil organic matter, the organic matter content will usually increase with conservation tillage, but remain fairly constant, or perhaps decrease further, with conventional tillage (Frye et al., 1985). Because there is less soil disturbance with conservation tillage than with conventional tillage, plant residues, fertilizers, and other soil amendments do not get mixed into the soil as much and plant roots tend to proliferate in the top few centimeters. Furthermore, this surface layer is usually wetter, cooler, less oxidative, and more acid (Blevins et al., 1977; Doran, 1980; Rice et al., 1086). These conditions tend to cause the organic matter content to increase or to decrease at a slower rate compared to under conventional tillage. By comparison, conventional tillage increases aeration, mixes plant residues into the soil where they decompose faster, and exposes previously protected soil organic matter. These conditions tend to speed up organic matter decomposition. Additionally, soil erosion may be increased by conventional tillage, removing

CONSERVATION TILLAGE IN SOIL MANAGEMENT

57

Figure 2. Organic C content to 30-cm depth in Maury soil under conventional tillage (CT), notillage (NT), and bluegrass sod in 1975, 1980, and 1989. (From Ismail er d.,1993.)

soil organic matter at an accelerated rate because the organic matter is concentrated in the eroding topsoil. In the long-term (23-year) conventional tillageho-tillage experiment in Kentucky mentioned above, we determined the soil organic matter content after 5, 10, and 20 yr (1975, 1980, and 1989). Figure 2 shows the organic C content in the top 30 cm of soil under conventional tillage and no-tillage in comparison to adjacent plots that remained in bluegrass (Pou prutensis L.) sod. Clearly, the organic matter content decreased with both tillage systems during the first 5 yr out of sod, but no-tillage did not decrease as much as conventional tillage. Both remained stable during the next 5 yr. From 1980 through 1989, the organic matter under both tillage systems returned to about its initial level, the same level as in bluegrass sod. These data are consistent with the theoretical considerations discussed earlier and are corroborated by the work of others. They further suggest that soil organic matter can be maintained or increased at a high level with conservation tillage and a cover crop annually; in this case, it was rye.

V. SURFACE MULCH MANAGEMENT Crop residues are valuable resources that can be used as an important agricultural tool to conserve soil and water, improve the physical properties of soils,

58

R.L. BLEVINS AND W. W. FRYE

improve the quality of water running off agricultural fields, and, in some cases, increase soil fertility. Crop residues are also useful for nonagricultural purposes, such as energy sources, industrial materials, and environmental protection. In the tropics, there is increasing competition for use of crop residues for livestock feed, fencing, roofing, and household fuel (Lal, 1989). Residue management currently involves several approaches, including harvesting as hay or haylage or use by grazing animals. Straw from small grains is commonly baled and sold for bedding. In the semiarid tropics, most of the straw or stover is removed and used for purposes other than a surface soil mulch. In some farming systems in Europe and the southeastern United States, the residue is often burned before establishing the next crop. Plowing under as a green manure crop and chemically burning down any live vegetation before planting the next crop are other approaches commonly used. Considerable technological progress has been made during the past two decades in developing equipment and farming systems that allow more crop residues to remain on the soil surface. In the United States, there is an additional urgency for farmers to use better management strategies in response to the conservation compliance provision of the 1985 Food Security Act.

A. BENEFITSFROM CROP RESIDUES Numerous researchers have documented the dramatic reductions in soil erosion on cropland that can be obtained by leaving portions of the residue from previous crops on the surface of a field (discussed in Section 11). Judicious use of crop residues is the best means of controlling both water and wind erosion in both tropical and temperate regions (Lal, 1989). In the tropics, the benefits from crop residues include water conservation and amelioration of the negative effect of high soil temperatures. On well-drained soils of temperate regions, the utilization of crop residues in conjunction with conservation tillage significantly reduces soil erosion, results in equal or higher crop yields, conserves soil water, reduces labor requirements, conserves nutrients through recycling, improves biological activity, and allows farmers to maintain or improve the productivity of the soil. Advantages for using crop residues as surface mulch are similar over a wide range of soils and climatic zones. For example, use of a mulch resulted in a significant yield increase of maize in the eastern Amazon region (Schoningh and Alkamper, 1985), and under much drier conditions in Texas, significant benefits of mulch on water conservation and grain yields of sorghum were observed (Unger, 1978). Conservation tillage systems utilizing crop residue as a surface mulch are less beneficial to crop yield in regions of high rainfall than in zones of lesser rainfall.

CONSERVATION TILLAGE IN SOIL MANAGEMENT

59

Under certain soil and climatic conditions, yields may be reduced by no-tillage. These areas include soils with slow drainage and northern latitudes that result in slower warming of the soil in spring. Under these conditions yields are often lower for tillage systems that leave residues at the surface as compared to plowing under residues. Even though the cropping/tillage systems that leave residues at or near the surface have been extremely successful, they cannot be used for all soils, crops, and climates. In arid regions, mulch material is not always available in sufficient quantity to effectively control wind erosion and to reduce excessive evaporation losses from the soil surface. It is important to determine the amount of mulch required for different soils and climates in order to select which conservation tillage system is most appropriate for a given environment.

€3. EFFECTS OF TILLAGE ON RESIDUE COVER Choice of tillage method is crucial in managing crop residues. Residue management may be approached by looking at the phases involved. Factors to consider are (1) the quantity and kind of crop residue left on the soil immediately after harvest of the crop; (2) the amount of over-winter reductions of remaining crop residue; (3) soil- or residue-disturbing operations such as removal, chopping, or shredding; and (4)use of tillage equipment that mixes and incorporates residues. These factors determine the amount of residue cover remaining for the next year’s crop. Also, some crops leave fairly sparse levels of residue following harvest. For example, soybean, cotton, and many vegetable crops leave low residue levels as compared to high-yielding wheat or corn. To maintain at least 30% mulch cover at the time of planting the next crop may require no or limited tillage following soybean production. Small grain and corn crops produce residues that decompose more slowly than legume crops. Whether the remaining crop residues are left partially standing, shredded and uniformly distributed on the soil surface, or partially incorporated influences its decomposition rate. Rice (1983) concluded that placement of corn stalk residues had more influence than tillage on decomposition. Residues buried to a depth of 15 cm decomposed twice as fast as residues left at the surface. Wilson and Hargrove (1986) showed that the rate of N disappearance from crimson clover (Trifulium incurnuturn L.) residue was more rapid under conventional tillage than no-tillage. Studies in Kentucky where corn stover was left on the plots and a rye small grain cover crop was seeded in the fall following corn harvest show 95 to 98% cover at planting time for the following year’s crop under no-tillage and 40 to 60% ground cover where spring tillage included either chisel-plowing with straight points or disking with a tandem disk to prepare the soil for planting

60

R. L. BLEVINS AND W. W. FRYE

Tillage method

Figure 3. Percentage of surface covered by residue 3 week after corn was planted at Lexington, Ky. Residue include residue from previous corn crop and a rye winter cover crop. CT, conventional tillage; DT, disk tillage; CH, chisel-plow tillage; and NT, no-tillage.

(Fig. 3). Less than 3% residue remained after moldboard plowing and two diskings. Griffith et al. (1986) presented excellent data showing how different tillage equipment and the number of passes affected residue cover remaining after planting (Table X). These values are lower than those reported in Fig. 3, but the Kentucky data included a rye winter annual cover crop in addition to corn stover from previous years. The data in Table X indicate that, under the soils and climate of the Corn Belt, chisel-plowingfollowed by disking does not leave enough cover to meet the minimum standard of 30% cover at planting to qualify as conservation tillage. Shallow disking (one pass) left 40% cover following corn production and 45% following small grain, but only 20% after soybean. The challenge in surface mulch management is to conduct tillage operations while maintaining surface cover. This may require no-tillage in some situations. Chisel-plows have been used extensively in the Great Plains for many years. They provide shallow primary tillage without inverting the soil. As use of chiselplows moved into higher-rainfall areas, they were modified to till deeper and to accommodate higher amounts of residue (Johnson, 1991). Delaying chiseling until spring rather than employment at fall tillage allows farmers to maintain more soil cover during the winter and spring. Sweeps are the most common ground-engaging tools. Use of sweeps often results in 5 to 15% more cover than twisted shanks on the chisel-plow that partially invert the soil (Johnson, 1991). Field cultivators work better when multiple passes are required. They have the ability to move some of the mulch buried by previous tillage passes back to the

CONSERVATION TILLAGE IN SOIL MANAGEMENT

61

Table X Residue Cover Remaining afler Planting for Typical Tillage versus Previous Crop Combinations" Previous crop Corn Tillage system Moldboard plow, disk, field cultivate Chisel ( 10-cm twisted points) disk twice Chisel (5-cm straight points) disk twice Primary tillage disk (deeper than 10 cm) disk twice (standard tandem) disk once, field cultivate once Shallow disking (less than 10 cm) once (standard tandem) twice (standard tandem) Till-plant in ridge In-row subsoil, plant No-tillage plant

+ +

+ +

Soybeans Grain

Sod

% Surface cover

5

2

5

10

15

2

10

20

20

5

15

25

10

5

15

20

20

10

20

-

40

20

45

50

20 30 70

10

25 -

30 -

50

80

60

90

85 95

80

20

"From Griffith er al. (1986).

surface. Disking cuts mulch material into smaller pieces and buries more of it with each subsequent pass across the field.

C. GROWING COVERCROPSFOR MULCHAND NITROGEN The most effective way to ensure a mulch cover for conservation tillage is to grow a winter cover crop. Cover crops may be leguminous or nonleguminous, depending on their purpose. Generally, nonleguminous crops, that is, mostly small grains, are easier and less costly to grow than legumes. However, legumes have the advantage of providing both mulch and N to the system. 1. Adaptability of Cover Crops

After determining the purpose of a cover crop, selecting the cover crop is the first and most important management decision. The cover crop must be well

62

R. L. BLEVINS AND W. W. FRYE

adapted to the climate and management conditions. Small grains that are most commonly used are rye, wheat, oat (Avena sativa L.), and barley (Hordeum vufgare L.). Annual forage grasses are sometimes used as cover crops also. Hairy vetch and crimson clover are the legumes that seem best suited as cover crops in the middle and southeastern two-thirds of the United States. Hairy vetch is more winter hardy and has performed well in experiments in Delaware, Maryland, Kentucky, Tennessee, and Nebraska (Frye et al., 1988). A growing number of states are being added to that list as research progresses. Crimson clover appears to perform better than hairy vetch under more southern climatic conditions (Touchton et a f . , 1982; Hargrove, 1986). Many other crops, both leguminous and nonleguminous, may be adaptable for various areas as cover crops for conservation tillage (USDAKSRS Sustainable Agriculture Research and Education Program, 1992). 2. Seeding Cover Crops

Seeding is the main cost of growing a cover crop, and an adequate stand that survives the winter is, of course, essential to the effectiveness of the cover crop. Several planting methods have been successful. Aerial interseeding by surface broadcast into a standing summer crop (e.g., corn, sorghum, soybean) just before the beginning of leaf-drop is one of the most successful methods used in Kentucky. Adequate plant residues from leaf-drop or previous crops to cover the seed and adequate soil water or rainfall for germination and seedling growth are crucial elements to the success of this method. A no-tillage drill places the seed into the soil, where the environment is more favorable for germination than on the surface, but it has the disadvantage of delaying planting of the cover crop until after the summer crop is harvested. This decreases the effectiveness of the cover crop and may make it more susceptible to winterkill. On the basis of work in Kentucky, we believe that earlier planting date is a more important factor than seed placement. We compared no-tillage drilling after corn harvest with aerial interseeding 3 wk before corn harvest (Frye et al., 1988) (Table XI). Brown et a f . (1985) did a similar study with cotton (Gossypium hirsutum L.) in Alabama. Both methods were successful in establishing a cover crop stand, but the earlier planting was superior in both dry matter and N production. Differences in Alabama were even more dramatic than in Kentucky (Table XI). Other successful seeding methods include techniques such as seeding behind the last cultivation of the corn (Nanni and Baldwin, 1987) and broadcast seeding after corn grain harvest and disking lightly or shredding stalks to ensure seed cover. In Kentucky, we drilled hairy vetch and crimson clover into wheat stubble in late July with great success. Dry matter and N production before freezing-

CONSERVATIONTILLAGE IN SOIL MANAGEMENT

63

Table XI Effect of Planting Method on Dry Matter Yield and N Production of Cover Crops' Dry matter yield

IS b

D" Mg ha

Cover crop

~

Nitrogen content IS

D kg ha-'

I

Hairy vetch Bigflower vetch Rye

3.16 2.74 2.31

Kentucky 3.06 I33 2.31 98 2.14 21

Hairy vetch Crimson clover Rye

2.96 4.38 2.84

Alabama I .83 133 1.33 133 I .90 27

104

85 19 75

44 21

"Adapted from Frye era/. (1988). blS. interseeded early September before harvest in Kentucky and approximately September 25 before cotton defoliation in Alabama. 'D, drilled after crops harvested-mid-October in Kentucky and early November in Alabama.

down of the plants about November 1 are shown in Table XI1 (unpublished data). The crops were winter-killed completely, leaving an excellent mulch cover in which to plant no-tillage corn the next April, one month earlier than normal for

Table XI1 Fall Production of Dry Matter and Nitrogen from Hairy Vetch and Crimson Clover Planted Late July into Wheat Stubble at Lexington, Kentucky' Dry matter (kg ha - I )

Nitrogen (kg ha-')

Date sampled

H. vetch

C. clover

H. vetch

C. clover

I Nov. 1990 19Oct. 1991

3493 6105

3353 5863

122 195

114 188

"Unpublished data.

64

R. L. BLEVINS AND W. W. FRYE

planting into a legume cover crop. Grain yields without N fertilizer averaged 8.1 and 9.1 Mg ha-l with hairy vetch and crimson clover, respectively, in 1991, and 11.6 and 10.9 Mg ha-', respectively, in 1992.

3. Self-Reseeding Cover Crops Perpetuating an annual legume cover crop by a self-reseeding technique seems to be a very practical way of providing a legume cover crop in conservation tillage. There are disadvantages, however, that should be considered. Crimson clover and bigflower vetch (Vicia grundiJora var. Kitaibeliunu W. Koch) are particularly well adapted to such management (Frye et al., 1988). The most obvious advantage of this practice is to eliminate the need to plant a cover crop each fall, thus saving the high cost of annual seeding. The disadvantages include the need to either delay planting the crop or to spray strips with herbicides, first over the rows at planting, then between the rows after seed maturity. Thus shielded spraying is required. A stand of bigflower vetch has been maintained in this way for about 10 yr in experimental plots in Kentucky. Our experience, however, indicates that the legume stand slowly loses its vigor because of what we believe is loss of inoculation with an effective Rhizobium species.

4. Killing the Cover Crop Where the cover crop is used to provide mulch for conservation tillage, it is necessary to kill it with herbicides to prevent it from competing with the summer crop. We know of no research supporting the practicality of growing summer row crops in a living mulch for a sustained period of time. A number of different herbicides are used in conservation tillage, and the only special requirement that a cover crop imposes is a bum-down herbicide in the mixture, such as glyphosate [N-(phosphonomethyl)glycine]or paraquat (l,l-dimethyl-4,4-bipyridinium ion). Where a triazine-sensitive legume is to be established in the fall for a cover crop, cyanazine (2- [ [4-chloro-6-(ethylaminso)-1,3,5-triazin-2-y1] amino] -2-methylpropaneitrile) is often substituted for at least part (usually one-half to all) of the atrazine [6-chloro-N-ethyl-N-(1-methylethyl)-1,3,5-triazine-2,4-diamine]. Farmers have reported complete success using 2,4-D (2,4-dichlorophenoxy) acetic acid to kill the legume cover crop in the spring (Illinois Sustainable Agricultural Network, personal communication, 1992). Droughty conditions during the spring may result in excessive depletion of the soil water by a cover crop (Utomo et a f . . 1986). An easy way to lessen this problem is to kill the cover crop earlier in the spring, usually 2 to 3 wk before

CONSERVATION TILLAGE IN SOIL MANAGEMENT

65

planting the summer crop. Early killing may conserve soil water and help facilitate soil warm-up, as discussed earlier, but considerable dry matter yield and N production (if a legume) may be sacrificed. In North Carolina, Wagger (1987) found that hairy vetch produced about 2.00 Mg ha-l dry matter with 58 kg N ha-' during about 2 wk, from the third week of April to the first week of May. Corresponding values for crimson clover were 1.57 Mg ha-' dry matter with 26 kg N ha - I . So, the decision to kill early or at planting time should be dictated by weather conditions and stage of development of the cover crop. If dry weather appears imminent, early killing would be advisable. Conversely, during a wet spring, the cover crop can help dry the soil for earlier planting. When a cover crop passes a certain optimum stage of maturity, it rapidly loses value as a mulch and N source. We have observed this problem in crimson clover.

VI. NUTRIENT MANAGEMENT A. EFFICIENTUSEOF NITROGEN One of the most crucial aspects of conservation tillage is N management. The lack of tillage profoundly affects the soil environment as it relates to N transformations. As tillage is decreased, the soil tends to be wetter, cooler, less aerated, less oxidative, and more acid (Blevins et al., 1977; Doran, 1980). Organic matter decomposition slows, thus decreasing nutrient mineralization rate. This is more important in the case of N than for most other nutrients. Also, other N transformations that are 0,-dependent are greatly affected, for example, nitrification, dinitrification, and immobilization. As pointed out by Doran (1980), the microbial biomass is greater under notillage than conventional tillage ( I .58 times greater in the 0- to 7.5-cm depth), but the population is more anaerobic under no-tillage. The expected result of such conditions would be slower mineralization and nitrification and greater immobilization and denitrification. Generally, that is what has been found by several researchers (Doran, 1980; Rice and Smith, 1982, 1984). There are a number of practical implications associated with these conditions. Indigenous and added organic matter decompose more slowly. Soil organic matter tends to increase, as mentioned earlier. Slower decomposition may result in more favorable timing of the release of N from a legume cover crop in conservation tillage (Peterson and Power, 1991), causing the legume to be, in effect, a slow-release N source. Applied fertilizer N is likely to be immobilized to a greater extent under conservation tillage. This could cause a temporary deficiency in the crop,

66

R. L. BLEVINS AND W. W. FRYE

but it may immobilize NO - 3 that might otherwise leach to the ground water (Peterson and Power, 1991). N tends to remain in the NH+4 form longer before being converted to NO-3.

N is more likely to be lost by denitrification, especially in fine-textured soils or when the soil is wet from extended rainfall.

B. MANAGEMENT OF FERTILIZERS Differences in fertilizer management in conservation tillage and in conventional tillage can generally be traced to the lack of soil disturbance and the vegetative mulch cover.

1. Nutrient Distribution in Soil In modem agriculture, most solid and liquid fertilizers are surface-applied. Additionally, nutrient cycling brings nutrients to the soil surface. Conventional tillage with moldboard plowing mixes the surface 15 to 25 cm of soil, whereas conservation tillage mixes much less to essentially none. Thus, fertilizer nutrients and nutrients in organic matter get mixed into the soil annually under conventional tillage but tend to accumulate near the soil surface under conservation tillage. In the latter case, the mixing process is dependent largely on natural mixing forces, such as freezing and thawing, earthworms, and other natural forces of disturbance. Clearly, our results in Kentucky after 20 yr of no-tillage and conventional tillage illustrate these principles (Table VIII). Soil-test P, K, Ca, and Mg are all higher in the 0- to 5-cm depth of no-tillage soil than in the conventional tillage soil, but are higher in conventional tillage at the 5- to 15- and 15- to 30-cm depths. Other research shows similar effects of tillage or no-tillage on soil-test nutrient values (J. H. Edwards et al., 1992; Prasad and Power, 1991). These data tend to lead to the conclusion that surface-applied soil amendments and plant residue would be less effective in conservation tillage than in conventional tillage. But, under most circumstances, the concentration of nutrients and soil organic matter near the surface seems to have little, if any, influence on their effectiveness. In fact, nutrient uptake may be enhanced by conservation tillage compared to conventional tillage. Hargrove (1986) reported enhanced uptake of several nutrients by corn planted no-tillage and no-tillage with in-row chiseling compared to conventional moldboard-plow tillage. Grain yields were also enhanced. This enhanced nutrient efficiency is quite understandable when one considers that the moist, nutrient- and organic matter-rich zone of soil is more conducive

CONSERVATION TILLAGE IN SOIL MANAGEMENT

67

Table XI11 Comparison of Corn Root Density in No-Tillage and Conventional Tillage Soil (Root Samples Taken 15 cm from Center of Rows)” Conventional tillage June 10

Soil depth (cm) 0-5 5-15 15-30 0-30

July 6

No-tillage June 10

July 6

cm of root per cm of soil 1.6 6.9 2.5 3.9

1.9 5.1 8.4 6.2

21.2 5.2 1 .o

5.8

21.0 5.3 2.6 6.6

“From Phillips (1984).

to plant root growth. Research shows that plant roots indeed proliferate there (Table XIII). Therefore, nutrient availability to plants is usually not diminished because of their accumulation near the soil surface, and fertilizer needs and management, with the exception of N discussed earlier, are essentially the same for both tillage systems. 2. Methods of Fertilizer Application

Surface broadcast application of fertilizers, under most circumstances, is a satisfactory practice (Thomas and Frye, 1984; Prasad and Power, 1991). Subsurface band placement of P and K may be somewhat more effective where the soiltest values are low, and subsurface or surface banding of N fertilizer may improve its efficiency. Nitrogen fertilizer efficiency is jeopardized to some extent by surface broadcast application where there is a heavy mulch cover of residue on the soil either from the previous crop or from a cover crop, especially a nonlegume. Efficiency may be lost in several ways. The fertilizer N may be immobilized by the microbial population decomposing the residue, lost by denitrification in the less aerobic environment of conservation tillage, lost by NH3 volatilization from the surface of the residue on soil, or lost by leaching of NO-3. These processes, which are generally more prevalent in conservation tillage than conventional tillage, along with the typical differential yield response to N in the two tillage systems, have led some researchers to conclude that more fertilizer N is required for optimum crop yield with conservation tillage. In general, that appears to be true; however, optimum yield under conservation tillage

68

R. L. BLEVINS AND W. W. FRYE

is, in most cases, greater than that under conventional tillage. This may result in a considerable advantage for conservation tillage at optimum N rates (Frye et af., 1981). The increased yield has been attributed to the enhanced soil water status brought about by conservation tillage.

3. Improving N Efficiency Several practices have been developed to improve the efficiency of fertilizer N under conservation tillage. Denitrification is more likely to occur early in the growing season when rainfall is usually more frequent, the soil is wetter, and the soil environment is less oxidative. The best practice to lessen the potential for denitrification loss of N is to delay application of most, if not all, of the N fertilizer until later in the season (Thomas and Frye, 1984). For corn, this means applying most or all of the N 4 to 6 wk after planting. The N may be banded as a sidedress application or broadcast as a topdress. In areas where irrigation is practiced, the N can be applied in the water as a fertigation application. Delayed or split application of N fertilizer has become so widely adopted that it is almost a general practice for N conservation and efficiency, especially for no-tillage corn (Thomas and Frye, 1984). In addition to decreasing denitrification losses, the practice also should provide benefits in decreasing NO-3 leaching. Since the N is applied when the soil is likely to be drier, and the application more closely coincides with rapid N uptake by plants (NaNagara et al., 1976), this leaves less time for loss before uptake occurs. Immobilization of N can be decreased greatly by placing the fertilizer below the mulch, especially when injected into the soil below the zone of high microbial activity (Bandel, 1986). Such placement of solid urea and urea-ammonium nitrate (UAN) solution also prevents volatilization loss of NH3, a common occurrence during urea hydrolysis in surface applications of these fertilizers. The practicality of subsurface placement of fertilizers in no-tillage, for example, is questionable when the additional fuel, horsepower, equipment, time, and labor requirements are considered in comparison to the requirements for surface broadcast application. Thus, the improvement in N efficiency and the somewhat lower N rates they permit may not offset the loss of convenience and advantages of surface broadcast applications. Research in Kentucky by Murdock and Frye (1985) showed that urea surfaceapplied early May or before was about as effective as ammonium nitrate for fescue (Festuca arundinacea Schreb) fertilization. However, when applied later than early May, a significant reduction in effectiveness resulted compared to ammonium nitrate. So, it appears that NH, volatilization loss would not be of great concern if urea or UAN were applied fairly early in the season while temperatures are still cool and rainfall is fairly frequent. However, N fertilizer has

CONSERVATION TILLAGE IN SOIL MANAGEMENT

69

been shown repeatedly to be more effective when applied 4 to 6 wk after planting no-tillage corn, as discussed earlier. Therefore, delayed application of urea or UAN on the surface would be at a time when the probability of NH, volatilization loss is high. In later research, Frye et al. (1990) eliminated the differences in efficiency between ammonium nitrate and urea by using a urease inhibitor, N-(n-butyl) thiophosphoric triamide (NBFT). With urea, the NBPT increased fescue yields by an average of 13% and no-tillage corn yields by 17% over urea alone. Nitrogen and P losses in surface runoff are ordinarily very small regardless of tillage system. In Kentucky, we measured less than I% of the surface-applied N or P in the runoff from conventional tillage, chisel-plow tillage, and no-tillage (Blevins et al., 1990). Loss tended to be greatest from conventional tillage because runoff was greatest there, but the differences among tillage methods were not significant.

VII. PEST MANAGEMENT A successful pest management program must adopt a multidisciplinary, multicrop approach to be successful. Integrated pest management is an approach that is gaining in popularity and use. This technology must be effective in controlling pests and at the same time be economically and ecologically sound. Use of cover crops and conservation tillage has been shown in some situations to have a beneficial influence on reducing weeds, insects, and pathogens (Bezdicek and Granatstein, 1989). Crop rotation is another useful management tool in controlling weeds and other pests.

A. WEEDCONTROL One of the primary benefits of soil tillage in crop production over the years has been weed control. With the evolution of weed control technology and development of new herbicides, there is less and less need for primary tillage. When producers shift from crop production systems that depend heavily on plowing to conservation tillage systems, there is a greater reliance on herbicides to effectively control weeds. Different ecosystems and the presence of certain weeds may dictate tillage for seedbed preparation. In such cases, conservation tillage may not be the best choice (Lewis and Worsham, 1989). Within the last few years, the arrival of herbicides on the market that selectively control perennial weeds in corn as well

70

R. L. BLEVINS AND W. W. FRYE

as soybean has been a significant breakthrough in weed science. Some of these herbicides require only small quantities of material (few grams per ha) because of their high efficacy. Weed management in crops planted by conservation tillage methods depends largely on foliar- and surface-applied herbicides because seedbed preparation is reduced or eliminated and cultivation may not be practical. A mixture of a “burn-down’’ herbicide plus one or more residual herbicides is normally used. In some cases, a postemergence herbicide may be required for control of certain broadleaf weeds or grass. Any tillage system that leaves a substantial mulch at the surface provides shading that may suppress weeds because the environment is unfavorable for germination of some weeds. Research has shown that rye killed in spring inhibits the growth of weeds (Smeda and Weller, 1988). The effectiveness of rye is decreased if its residues are tilled into the soil rather than left on the surface. No-tillage planting into rye residue maximizes potential allelopathic weed control. Subterranean clover (Trijolium subterraneum L.) has been shown to be effective for controlling weeds when used as a living mulch in no-tillage vegetable production (Ilnicki and Enoche, 1992). as has hairy vetch (Schonbeck and Doherty, 1989). Herbicides such as the sulfonylureas and imidozolinone are now used to control weeds in soybean, since soybean is very tolerant to these herbicides. More recently, compounds such as nicolsulfuron and primisulfuron have made possible postemergence control of troublesome perennials in corn, such as Johnsongrass. These chemicals are effective yet are required in only small quantities. The reduction in amount of herbicide materials applied to a crop should reduce the potential for the material to eventually pollute the groundwater. As mentioned in Section 11, herbicides such as triazines are appearing in groundwater sources because of their widespread use as a residual herbicide for corn. Prudent use and management of herbicides is a challenge that farmers and weed scientists currently face. Postemergence herbicide applications allow the use of an integrated weed management approach and applications can be made on the basis of economic threshold levels of weeds.

€3. DISEASES AND INSECTS As conservation tillage evolved in the 1960s and 1970s, there was concern about how such a revolutionary cultural practice would affect the severity of insect infestation and disease incidence. At that time, most pest control recommendations were based on deep coverage of residues by plowing. There are two differing viewpoints on the overall influence of residue management and conservation tillage on pest management. First, conservation tillage may generate a more favorable habitat for soil- and surface-dwelling insects and pathogens. The residue left at the surface reduces water loss, moderates tem-

CONSERVATIONTILLAGE IN SOIL MANAGEMENT

71

perature extremes, and provides food resources for the pest (House and Crossley, 1987). Second, the burial of residue containing root pathogens does not seriously reduce the levels of inoculum produced. For this reason, conservation tillage that leaves the residue at the soil surface where it is subject to drying out may actually provide a less favorable environment for these organisms to multiply. There is an ongoing effort by researchers to understand the possible interactions of conservation tillage with soil organisms. According to Fortnum and Karlen (1985), crop residue and tillage affected nematode species differently. A Meliodogyne incognita population did not change significantly with tillage, but S. bruchyururn populations were significantly higher in conservation tillage treatments where crop residues remained on the surface. In contrast, S. bruchyurum populations were lowest for conservation tillage plots when 90% of the residue was removed or incorporated. Thus, residue management seems to have more influence on nematodes than does tillage per se. Cook et ul. (1978) described the environment produced by conservation tillage and its influence on incidence of plant diseases. They reported that, for many plant pathogens, residues provide a source of food, a place to live, and a place to reproduce. A significant proportion of root-infecting organisms, for example, depend on crop residues for survival in soil. Moldboard plowing to bury the residue does not necessarily reduce the levels of inoculum produced and the potential for disease incidence in the next cropping system. Root diseases caused by Pythium species are favored by wet soils and cool temperatures, especially in corn production. Root disease damage has been reported to be greater under no-tillage treatments for a fine-textured, poorly drained soil in Ohio (Van Doren et ul., 1976). A beneficial effect of crop residues in conservation tillage systems is the reduction of stalk rot in soybean in Nebraska. Mycophagous springtails (Collembola)have been used as a biocontrol for Rhizoctonia soluni by reducing the inoculum density. Rickerl and Touchton (1989) reported that no-tillage systems resulted in 29% more Collembola than conventional tillage, but populations were not high enough for effective control of Rhizoctonia soluni. Tyler et a f . (1983) found lower nematode counts in no-tillage than conventional tillage soybean in Tennessee. They speculated that higher residue levels at the surface in no-tillage may have increased the population of organisms that feed on nematodes. In the same study, they showed a greater incidence of brown spots on leaves of conventionally tilled than no-tilled soybeans, but stem canker was a greater problem on no-tillage soybeans. Although the soil-insect complex may present a problem in conservation tillage, the problems are manageable. Furrow application of systemic insecticides has shown promise in moderating soil insect stress and certain foliar insects. Many of the insect problems are unaffected by tillage methods. Musick and

72

R. L. BLEVINS AND W. W. FRYE

Beasley (1978) reported that wireworms cause more damage in no-tillage systems, however, in most situations, the previous crop and management may affect insect damage more than tillage methods themselves (All and Musick, 1986). Research on cotton in Georgia showed that infestation of tobacco thrips on seedling cotton is reduced for up to 21 d in no-tillage as compared to surface tillage (All et al., 1992). The lesser cornstalk borer is found in fewer numbers in no-tilled corn than in conventionally tilled corn. Research in North Carolina indicates that biological control of cutworm larvae, corn earworm pupae, and corn rootworm eggs and larvae by predatory mites and beetles occurs more often under no-tilled soybeans (Van Duyn and House, 1989).

VIII. CONCLUSIONS As we approach the twenty-first century, the technology to successfully grow crops using a variety of conservation tillage systems is available to our farmers. The alliance of farmers, scientists, and agribusiness has transformed crop residue management strategies and tillage methods from an idea to a system that effectively reduces erosion, reduces soil degradation, is cost-effective, and is environmentally acceptable. Soil properties and their ecological environment determine the limitations and suitability for using conservation tillage methods.

A. PROGRESS IN CONSERVATION TILLAGE The development of new and more effective herbicides allows flexibility in choice of crops, crop rotations, and tillage methods. As a result, different versions of conservation tillage are being used around the world. The systems vary from large-scale farming operations in the Corn Belt of the United States and grain-producing areas of Argentina and Brazil to the small hill farmers in Central America. These systems range from agroforestry systems in Costa Rica and Indonesia to rice farmers in Bangladesh that direct-plant wheat into their rice fields just before harvesting and at the beginning of the dry season. Variations in cultural practices are necessary because of ecological and socioeconomic factors. Even though conservation tillage has many facets, the basic principles include the management of mulches, a reduction in mechanical tillage, and adoption of cropping combinations that will preserve the soil resource base and still produce satisfactory yields. Rapid progress in technology has been made during the last two decades in developing equipment to till, plant, and spray and yet allow the farmer to leave

CONSERVATION TILLAGE IN SOIL MANAGEMENT

73

crop residues at the soil surface. The use of sweeps, field cultivators, and other equipment that incorporate less residue has made mulch tillage a viable version of conservation tillage that is being successfully used in many areas of the United States. Use of conservation tillage in conjunction with a residue management strategy has become the primary approach and management method to reduce soil erosion.

B. FUTURENEEDS AND DIRECTION An area of current concern is the role that conservation tillage may have in the movement of agricultural chemicals into our natural waters. We still do not have a clear understanding of how herbicides react with the soil, the transport processes involved, and the fate of the herbicides if they reach the groundwater. Long-term multidisciplinary field studies that include direct measurements from surface water and groundwater are needed immediately so the potential problems can be addressed with a research data base. To reduce chemical inputs, we need to develop and refine integrated farming systems that include alternative tillage methods, cropping systems, and pest management strategies. The use of biocontrols for pest management has potential but needs a lot of research and testing before it can be used effectively at the farm level. Any successful approach to conservation tillage must be profitable and sustainable, protect our soil and water resources, and be environmentally sound. Better use of previous crop residues that may include winter annual cover crops when needed will be part of a viable system that gives us the best protection for our soil and water resources and is usually cost-effective. There is an urgent need to determine the amount of mulch required for different soils and climates in order that we may select conservation tillage systems that are most appropriate for given site-specific environments.

REFERENCES All, J. N., and Musick, G. J. (1986). Management of vertebrate and invertebrate pests. In “NoTillage and Surface-Tillage Agriculture: The Tillage Revolution” (M. A. Sprague and G . B . Triplett. eds.), pp. 347-387. Wiley, New York. All, J. N.. Tanner. B . H . , and Roberts, P. M . (1992). Influence of no-tillage practices on tobacco thrips infestation in cotton. I n “Profeedings of the 1992 Southern ConservationTillage Conference” (M. D. Mullen and B. N. Duck, eds.), Spec. Publ. 92-01. Tenn. Agric. Exp. Stn.. Knoxville. Angle, J . S . . McClung. G.. Mclntosh. M. S . . Thomas, P. M., and Wolf, D. C. (1984). Nutrient

74

R.L. BLEVINS AND W. W. FRYE

loses in runoff from conventional and no-till corn watersheds. J . Environ. Qual. 13,431-435. Bandel, V. A. (1986). Nitrogen management for no-tillage corn. In “Proceedings of the Southern Region No-Tillage Conference” (R. E. Phillips, ed.), South. Reg. Ser. Bull. 319, pp. 1-15. University of Kentucky, Lexington. Bezdicek, D. F., and Granatstein, D. (1989). Crop rotation efficiencies and biological diversity in farming systems. Am. J . Altern. Agric. 4, I 1 1- 119. Blevins, R. L., Thomas, G.W., and Cornelius, P. L. (1977). Influence of no-tillage and nitrogen fertilization on certain soil properties after five years of continuous corn. Agron. J . 69, 383-386. Blevins, R. L., Smith, M. S . , Thomas, G. W., and Frye, W. W. (l983a). Influence of conservation tillage on soil properties. J. Soil Water Conserv. 38, 301-205. Blevins, R. L., Thomas, G.W., Smith, M. S . , Frye, W. W., and Cornelius, P. L. (1983b). Changes in soil properties after 10 years continuous non-tilled and conventionally tilled corn. Soil Tillage Res. 3, 135-146. Blevins, R. L., Frye, W. W., Baldwin. P. L., and Robertson, S . D. (1990). Tillage effects on sediment and soluble nutrient losses from a Maury silt loam soil. J . Environ. Qual. 19, 683-686. Bond, J. J., and Willis, W. D. (1969). Soil water evaporation: Surface residue rate and placement effects. Soil Sci. SOC.Am. Proc. 33,445-448. Brown, S . M., Whitewell, T., Touchton. J. T., and Burmester, C. H. (1985). Conservation tillage systems for cotton production. Soil Sci. SOC.Am. J . 49, 1256-1260. Burwell, R. E., Timmons, D. R., and Holt, R. F. (1975). Nutrient transport in surface runoff as influenced by soil cover and seasonal periods. Soil Sci. SOC. Am. Proc. 39,523-528. Conservation Technology Information Center (CTIC) (1992). ‘‘ 1992 National Survey of Conservation Tillage Practices.” CTIC, West Lafayette, IN. Cook, R. J., Bwsalis, M. G..and Doupnik, B . (1978). Influence of crop residues on plant diseases. ASA Spec. Publ. 31, 147-164. Dick, W. A. (1983). Organic carbon, nitrogen and phosphorus concentrations and pH in soil profiles as affected by tillage intensity. Soil Sci. SOC. Am. J . 47, 102- 107. Dick, W. A., Van Doren, D. M., Jr., Triplett, G.B.,Jr., and Henry, J. E. (1986). Influence of long term tillage and rotation combinations on crop yields and selected soil parameters. 11. Results obtained for a Typic Fragiudalf soil. Ohio Agric. Res. Dev. Cent., 1181, 1-34. Doran, J. W. (1980). Soil microbial and biochemical changes associated with reduced tillage. Soil Sci. SOC. Am. J . 44,765-771. Douglas, J. T., and Goss, M. J. (1982). Stability and organic matter content of surface soil aggregates under different methods of cultivation and grassland. Soil Tillage Res. 2, 155- 175. Duley, F. L., and Kelly, L. L. (1939). Effect of soil type, slope, and surface conditions on intake of water. Res. Bull. Nebr. Agric. Exp. S m . 112. Duley, F. L., and Russell, J. C. (1939). The use of crop residues for soil and moisture conservation. J. Am. SOC.Agron. 31,703-709. Edwards, J. H., Wood, C. W., Thurlow, D. L., and Reef, M. E. (1992). Tillage and crop rotation effects on fertility status of a Hapludult soil. Soil Sci. SOC.Am. J . 56, 1577- 1582. Edwards, W. M., Shiptalo, M. J., and Norton, L. D. (1988). Contribution of macroporosity to infiltration into continuous corn no-tilled watershed: Implications for contaminant movement. J . Contam. Hydrol. 3, 193-205. Edwards, W. M., Shipitalo, M. J., Dick, W. A,, and Owens, L. B. (1992). Rainfall intensity affects transport of water and chemicals through macropores in no-till soil. Soil Sci. SOC. Am. J . 56, 52-58. Ehlers, W. (1985). Observations on earthworm channels and infiltration on tilled and untilled loess soil. Soil Sci. 119, 242-249.

CONSERVATION TILLAGE IN SOIL MANAGEMENT

75

Ehlers, W. (1979). Influence of tillage on hydraulic properties of loessial soils in western Germany. I n “Soil Tillage and Crop Production” (R. Lal, ed.). Proc. Ser. No. 2, pp. 33-45. Int. Inst. Trop. Agric., Ibadan. Nigeria. Ellison, W. D. (1944). Studies of raindrop erosion. Agric. Eng. 25, 131-136. Fortnum, B. A. and Karlen, D. L. (1985). Effect of tillage system and irrigation on population densities of plant nematodes in field corn. J. Nematol., 17, 25-28. Frye. W. W.. Blevins, R. L., Murdock, L. W., and Wells, K. L. (1981). Energy conservation in notillage production of corn. In “Crop Production with Conservation in the 80s.” Proceedings of ASAE Conference on Crop Production wirh Conservation in the 80s. pp. 7-81. ASAE Pub]. St. Joseph. Michigcn. Frye. W. W., Ebelhar, S . A., Murdock, L. W., and Blevins, R. L. (1982). Soil erosion effects on properties and productivity of two Kentucky soils. Soil Sci. SOC.Am. J . 46, 1051-1055. Frye, W. W., Murdock, L. W., and Blevins, R. L. (1983). Corn yield-fragipan depth relations on a Zanesville soil. SoiISci. SOC.Am. J . 47, 1043-1045. Frye, W. W. (1984). Energy requirement in no-tillage. I n “No-tillage Agriculture: Principles and Practices” (R. E. Phillips and S. H. Phillips, ed.), pp. 127- 151. Van Nostrand Reinhold, New York . Frye, W. W., Burnett, 0. L., and Buntley, G. J. (1985). Restoration of crop productivity on eroded or degraded soils. In “Soil Erosion and Crop Productivity” (R. F. Follett and B. A. Stewart, ed.), pp. 335-356. American Society of Agronomy, Madison, Wisconsin. Frye, W. W., Varco, J. J., Blevins, R. L., Smith, M. S., and Corak, S . J. (1988). Role of annual legume cover crops in efficient use of water and nitrogen. I n “Cropping Strategies for Efficient Use of Water and Nitrogen” (W. L. Hargrove, ed.), pp. 129- 154. Am. SOC.Agron., Madison, WI. Frye, W. W., Murdock, L. W., and Blevins, R. L. (1990). “Improved Efficiency of Urea Fertilizer with Urease Inhibitor,” 1990 Agron. Res. Rep., p. 25. Dep. Agron., University of Kentucky, Lexington. Gantzer, C. J., and Blake, G. R. (1978). Physical characteristics of a Le Sueur clay loam following no-till and conventional tillage. Agron. J . 70, 853-857. Glotfelty, D. E. (1987). The effects of conservation tillage in practices on pesticide volatilization and degradation. I n “Effects of Conservation Tillage on Groundwater Quality (T.J. Logan er al.. eds.), pp. 169- 177. Lewis Pub]., Chelsea, IN. Griffith, D. R.. Mannering, J. V., and Moldenhauer. W. C. (1977). Conservation tillage in the eastern cornbelt. J. Soil Water Conserv. 32, 20-28. Griffith, D. R., Mannering, J. V., and Box, J. E. (1986). Soil and moisture management with reduced tillage. I n “No-Tillage and Surface-Tillage Agriculture: The Tillage Revolution” (M. A. Sprague and G. B. Triplett, eds.), pp. 19-57. Wiley, New York. Hargrove. W. L. (1986). Winter legumes as a nitrogen source for no-till grain sorghum. Agron. J. 78, 70-74. Hill, R. L., and Cruse, R. M. (1985). Tillage effects on bulk density and soil strength of two Mollisols. Soil Sci. Soc. Am. J. 49, 1270- 1273. House. G. J., and Crossley, D. A., Jr. (1987). Legume cover cropping, no-tillage practices, and soil arthropods: Ecological interactions and agronomic significance. I n “The Role of Legumes in Conservation Tillage” (J. F. Power, ed.). Soil Conserv. SOC.Am., Ankeny, 10. Ilnicki, R. D., and Enoche, A. J. (1992). Subterranean clover living mulch: An alternative method of weed control. Agric. Ecosys. Environ. 40, 249-264. lsmail, Isro, Blevins, R. L., and Frye, W. W. (1993). Long-term no-tillage effects on soil properties and continuous corn yields. Soil Sci. SOC.Am. J., in press. Johnson, H. P., Baker, J. L., Shrader. W. D., and Laflen, J. M. (1979). Tillage system effects on sediment and nutrients in runoff from small watersheds. Trans. ASAE 22, 11 10- 1114.

76

R. L. BLEVINS AND W. W. FRYE

Johnson, R. R. (1991). Residue management with chisel-type implements. Proceedings of national conference on crop residue management for conservation. Soil Water Conserv. Soc., pp. 2 1-23. Kanwar, R. S . , Baker, 1. L., and Laflen, J. M. (1985). Nitrate movement through the soil profile in relation to tillage system and fertilizer application method. Trans. ASAE 28, 1802- 1807. Laflen, J. M., Baker, J. L., Hartwig, R. O., Buchele. W. F., and Johnson, H. P. (1978). Soil and water losses from conservation tillage systems. Trans. ASAE 21, 881 -885. Lal, R. (1982). “No-Till Farming,” Monogr. No. 2. Int. Inst. Trop. Agric., Ibadan, Nigeria. Lal, R. (1985). Mechanized tillage systems effect on soil properties of tropical Alfisol in watersheds cropped to maize. Soil TillageRes. 6, 149-161. Lal, R. ( 1989). Conservation tillage for sustainable agriculture: Tropics vs. temperate environments. Adv. Agron. 42,85-197. Lal, R., Logan, T. J., and Fausey, N. R. (1989). Long-term tillage and wheel-track effects on a poorly drained mollic ochraqualf in northwest Ohio. 1. Soil physical properties, root distribution and grain yield of corn and soybeans. Soil Tillage Res. 14, 34-58. 1.aws, J. 0. (1941). Measurements of fall velocity of water drops and raindrops. Trans. Am. Geophys. Union 22,709-721. Lewis, W. M., and Worsham, A. D. (1989). Weed management. In “Conservation Tillage for Crop Production” (M. G. Cook and W. M. Lewis, eds.), pp. 28-39. North Carolina State Agriculture Extension Service, Raleigh. Lindstrom, J. J., and Onstad, C. A. (1984). Influence of tillage systems on soil physical parameters and infiltration after planting. J. Soil Water Conserv. 39, 149- 152. Lindstrom, M. J., Voorhees, W. R., and Randall, G. W. (1981). Long-term tillage effects on interrow runoff and infiltration. Soil Sci. Soc. Am. J . 45, 945-948. Mannering, J. V., and Fenster, C. R. (1977). Vegetative water erosion control for agricultural areas. I n “Soil Erosion and Sedimentation.” Am. Soc. Agric. Eng., St. Joseph, MI. Mannering, J. V., and Fenster, C. R. (1983). What is conservation tillage? J. Soil Water Conserv. 38, 141-143. Mannering, J. V.. Griffith, D. R., and Richey, C. B. (1975). “Tillage for Moisture Conservation,” Pap. No. 75-2523. Am. SOC.Agric. Eng., St. Joseph, MI. McCalla, T. M., and Army, T. J. (1961). Stubble mulch farming. Adv. Agron. 13, 125-196. Miller, M. F., and Krusekopf, H. H. (1932). The influence of systems of cropping and methods of culture on surface runoff and soil erosion. Res. Bull. Mo.,Agric. Exp. Sm. 177. Moody, J. E., Jones, J. N., and Lillard, J. H. (1963). Influence of straw mulch on soil moisture, soil temperature and growth of corn. Soil Sci. SOC.Am. Proc. 27, 700-703. Moschler, W. W.,Martens, D. C., Rich, C. J., and Shear, G. M. (1973). Comparative lime effects on continuous no-tillage and conventionally tilled corn. Agron. J. 65, 781-783. Mostaghimi, S.. Dillaha, T. A.. and Shanholtz, V. 0. (1988). Influence of tillage systems and residue levels on runoff, sediment, and phosphorus losses. Trans. ASAE 31, 128- 132. Munawar, A., Blevins, R. L., Frye, W. W., and Saul, M. R. (1990). Tillage and cover crop management for soil water conservation. Agron. J. 82,773-777. Murdock, L. W.. and Frye, W. W. (1985). Comparison of urea and urea-ammonium phosphate with ammonium nitrate in production of tall fescue. Agron. J. 77, 630-633. Murdock, L. W., Herbek, J. H.. and Gray, T. (1992). “Row Cleaners in No-Till Corn Production.” 1992 Agron. Res. Rep., pp. 14- 15. Dep. Agron., University of Kentucky, Lexington. Musick, G. J., and Beasley, L. E. (1978). Effect of the crop residue system on pest problems in field corn (Zea Mays L.) production. ASA Spec. Publ. 31, 1733-186. NaNagara, T., Phillips, R. E., and Leggett, J. E. (1976). Diffusion and mass flow of nitrate-nitrogen into corn roots grown under field conditions. Agron. J. 68, 67-72. Nanni. C., and Baldwin, C. S. (1987). Interseeding in corn. I n “The Role of Legumes in Conservation Tillage Systems” (J. F. Power, ed.), pp. 26-27. Soil Conserv. Soc.Am., Ankeny. 1A.

CONSERVATION TILLAGE IN SOIL MANAGEMENT

77

Peterson. G. A., and Power, J. F. (1991). Soil, crop, and water management. I n “Managing Nitrogen for Groundwater Quality and Farm Profitability” (R. F. Follett, D. R. Keeny, and R. M. Cruse, eds.), pp. 189- 198. Soil Sci. SOC.Am., Madison, WI. Phillips, R. E. (1984). Soil moisture. In “No-Tillage Agriculture: Principles and Practices” (R. E. Phillips and S. H. Phillips, eds.), pp. 66-86. Van Nostrand-Reinhold, New York. Phillips, R. E., Blevins, R. L., Thomas, G. W., Frye, W. W., and Phillips, S. H.(1980). No-tillage agriculture. Science 208, 1108- 1113. Prasad, R., and Power, J. F. (1991). Crop residue management. I n “Advances in Agronomy” (B. A. Stewart, ed.), pp. 205-251. Springer-Verlag, New York. Rasnake, M., Frye, W. W., Ditsch, D. C.. and Blevins, R. L. (1986). “Soil Erosion with Different Tillage and Cropping Systems. Soil Science News and Views.” Dep. Agron., University of Kentucky, Lexington. Rice, C. W. (1983). Microbial nitrogen transformations in no-till soils. Unpublished Ph.D. Dissertations. Department of Agronomy. University of Kentucky, Lexington. Rice, C. W., and Smith, M. S. (1982). Denitrification in no-till and plowed soils. SoilSci. SOC.Am. J. 46, 1168- 1173. Rice, C. W., and Smith, M. S. (1984). Short-term immobilization of fertilizer nitrogen at the surface of no-till and plowed soils. Soil Sci. Soc. Am. J. 48, 295-297. Rice, C. W., Smith, M. S., and Blevins, R. L. (1986). Soil nitrogen availability after long-term continuous no-tillage and conventional tillage corn production. Soil Sci. SOC. Am. J . 50, I 206- I 2 I 0. Rickerl, E. A., and Touchton, J. T. (1989). Tillage and rotation effects on collembolla populations and Rhizecronia infestation. Soil Tillage Res. 15, 41 -49. Romkens, M. J. M., Wilson, D. W., and Mannering, J. V. (1973). Nitrogen and phosphorus composition of surface runoff as affected by tillage method. J . Environ. Qual. 2, 292-298. Schonbeck, M., and Doherty, W. (1989). Cover crops for Northeast vegetable farms: A report on research at New Alchemy Institute. NUI.Farmer Spring, pp. 12-13. Schoningh, E., and Alkamper, J. (1985). Effects of different mulch materials on soil properties and yield of maize and cowpea in an eastern Amazon Oxisol. Inr. Symp. Humid Trop., Isr, Manus, Brazil. Shelton. C. H., and Bradley, J. F. (1987). Controlling erosion and sustaining production with notillage systems. Tenn. Farm Home Sci. (winter) 18-23. Smeda. R. J., and Welter, S. C. (1988). Factors influencing the effectiveness of rye for weed management in transplanted tomatoes. Proc.-North Cenr. Weed Control Conf. 43, 12. Thomas, G. W., and Frye, W. W. (1984). Fertilization and liming. In “No-Tillage Agriculture: Principles and Practices’’ (R. E. Phillips and S. H.Phillips, eds.), pp. 87-126. Van NostrandReinhold, New York. Thomas, G. W., Blevins, R. L., Phillips, R. E., and McMahon, M. A. (1973). Effect of a killed sod mulch on nitrate movement and corn yield. Agron. J. 65, 736-739. Touchton, J. T., Gardner, W. A,, Hargrove, W. L., and Duncan, R. R. (1982). Reseeding crimson clover as a N source for no-tillage grain sorghum production. Agron. J. 74, 283-287. Tracy, P. W., Westfall. D. G., Elliott, E. T., Peterson, G. A,, and Cole, C. V. (1990). Carbon, phosphorus. and sulfur mineralization in plow and no-till cultivation. Soil Sci. SOC. Am. J. 54, 457-461. Tyler, D. D., and Thomas, G . W. (1977). Lysimeter measurement of nitrate and chloride losses from soil under conventional and no-tillage corn. J. Environ. Qual. 6,63-66. Tyler, D. D., Overton, J. R., and Chambers, A. Y. (1983). Tillage effects on soil properties, diseases, cyst nematodes, and soybean yield. Soil Water Conserv. J . 38, 374-376. Tyler, D. D., Wilson. G. V., Logan, J., Thomas, G. W., Blevins, R. L., Caldwell, W. E., and Dravillas, M. (1992). Tillage and cover crop effects on nitrate leaching. In “Proceedings of the

78

R. L. BLEVINS AND W. W. FRYE

1992 Southern Conservation Tillage Conference” (M. D. Mullen and B. N. Duck, eds.), Spec. Publ. 92-01. Tenn. Agric. Exp. Stn., Knoxville. Unger, P. L. (1978). Straw-mulch rate effect on soil water storage and sorghum yield. Soil Sci. SOC. Am. J. 42,486-491. Unger, P. W., and McCalla, T. M. (1980). Conservation tillage systems. Adv. Agron. 33, 1-58, Utomo, M. (1986). Role of legume cover crops in no-tillage and conventional tillage corn production. Unpublished Ph.D. Dissertation Dep. Agron., University of Kentucky, Lexington. USDAKSRS Sustainable Agriculture Research and Educational Program ( 1992). “Managing Cover Crops Profitably,” Sustainable Agric. Network, Handb. No. 1. Rodale Institute, Emanaus, PA. Van Doren, D. M., Triplett. G. B., and Henry, J. E. (1976). Influence of long term tillage, crop rotation, and soil type combinations on corn yield. Soil Sci. SOC.Am. J . 40, 100- 105. Van Duyn, J. W., and House, G. J. (1989). Insect management. In “Conservation Tillage for Crop Production” (M. G. Cook and W. M. Lewis, eds.), pp. 46-50. North Carolina State Agricultural Extension Service, Raleigh. Voorhees. W. B. (1983). Relative effectiveness of tillage and natural forces in alleviating wheelinduced soil compaction. Soil Sci. SOC.Am. J . 47, 129- 133. Wagenet, R. J. (1987). Processes influencing pesticide loss with water under conservation tillage. In “Effects of Conservation Tillage on Groundwater Quality” (T. J. Logan e t a / . .eds.), pp. 189204. Lewis Publ., Chelsea, MI. Wagger, M. S. (1987). Timing effects of cover crop desiccation on decomposition rates and subsequent nitrogen uptake by corn. In “The Role of Legumes in Conservation Tillage Systems” (J. F. Power, ed.), pp. 35-37. Soil Conserv. SOC.Am., Ankeny. IA. Wendt, R. C., and Bunvell, R. E. (1985). Runoff and soil losses for conventional, reduced, and notill corn. J . Soil WaferConserv. 40,450-454. Wilson, D. O., and Hargrove, W. L. (1986). Release of nitrogen from crimson clover residue under two tillage systems. SoilSci. SOC. Am. J . 50, 1251- 1254. Wittmuss, H. D., and Yazar, A. (1981). Moisture storage, water use and corn yields for seven tillage systems under water stress. ASAE Publ. 7-81, 66-75.

TRANSPOSABLE ELEMENTS IN MAIZE: THEIRROLEIN CREATING PLANT GENETIC VARIABILITY Peter A. Peterson Department of Agronomy Iowa State University Ames, Iowa 5001 1

I. Introduction: The Heterogeneity Question 11. Maize Breeding Accomplishments: What the Plant Breeder Has Wrought A. The Maize Genome: How Much Has Been Manipulated? B. Yield Components 111. Transposable Elements A. Their Phenotype Variegation B. Transposable Elements: Their Discovery C. Components of Transposable Elements D. Systems E. Genetic Resolution of a Transposable Element F. Transposition G. Effects on Gene Expression H. Presence in the Maize Genome I. Neoteny and Realignment of Resources for More Efficient Networks J. Summary of Effects Definitions of Terms and Symbols References

I. INTRODUCTION:THE HETEROGENEITY QUESTION The focus in this review is on the manipulation of plants by plant breeders that has resulted in a striking improvement of crop plants during the twentieth century and especially in the last three decades. Transposable elements associated with maize (Zea mays L.) breeding are the primary focus, although this subject is applicable to other crop plants. Breeders’ efforts with maize have uncovered a 79 Aduanrrs in A p m y , Vdumr Y J

Copyright 0 1993 by Academic Press, Inc. All rights of r e p d u d o n in any form reserved.

80

PETER A. PETERSON

highly heterogeneous genotype (Goodman and Stuber, 1983a,b) that can be manipulated in most directions to achieve the desired goal (Russell, 1991; Hallauer, 1992). Plant breeders have used numerous breeding schemes with a varied measure of success. When one then examines the Iowa recurrent selection program with the BSSS populations, from which successful inbreds have been uncovered (Hallauer, 1990a,b), an immediate question is where the variability leading to improvement originated. There are three possible options: (a) This heterogeneity is the assemblage of favorable alleles arising from the recombination of existing heterogeneity in the populations during selection protocols. According to this scenario, the existing variability in the genomes of current maize populations is recombined or resorted to fit the breeder’s needs as selection proceeds. (b) In a second option, new variability (new alleles) is continually being generated (Peterson, 1986a,b). According to this scenario, maize is possibly unique in its potential to generate changes at a high rate in the genome, some that are favorable alleles and others that are unfavorable. Thus, breeders’ selection procedures incorporate these new favorable changes and discard the unfavorable types. (c) In the final scenario, the heterogeneous genotype that the maize breeder has available and manipulates to make better inbreds is derived from a combination of (a) and (b). Thus, (a) + (b) = (c). This review is biased toward option (c), with a major emphasis on the genetic input found in (b) in this formula. Specifically, (b) is enhanced by transposons prevalent in most plants but highly visible in maize. These transposons will be described, and their effect on individual genes will be illustrated. Further, their pervasiveness in maize populations will be examined.

II. MAIZE BREEDING ACCOMPLISHMENTS:WHAT THE PLANT BREEDER HAS WROUGHT How MUCHHAS BEEN A. THE MAIZEGENOME: MANIPULATED?

There has been considerable progress in maize improvement. W. A. Russell (1991) has summarized concepts and results of maize breeders during the eras of scientifically driven maize breeding. In this review, Russell examined the genetic gain that resulted in maize improvement. (Genetic gain is that improvement derived from genetic changes. These changes, though, accommodate husbandry practices that advance maize culture.) He reviewed his own extensive research on this problem as well as that of others, notably Duvick (1977, 1984, 1986, 1991) and Castleberry et al. (1984). The reader is referred to his review (Russell, 1991), but a few principles will be summarized.

TRANSPOSABLE ELEMENTS IN MAIZE VARIABILITY

81

1. Some General Concepts a. Exotic Germplasm Exotic germplasm has had a relatively insignificant role in Corn Belt maize breeding programs (Hallauer, 1990a). Though many attempts have been made to incorporate nonadapted germplasm, this resource has not been utilized, despite the multitude of available diversity (Goodman, 1985, 1990; Goodman and Stuber, 1983a,b). This feature will be discussed in a later section. b. Genetic Gain There has been genetic gain over eras of maize breeding: in the decades between 1930 and 1970, there has been a marked increase in yields of maize cultivars at several plant densities but most notably at the higher plant densities. What is very evident is that the 1930s material could not sustain its yield potential at the high plant densities when compared with current materials. It could be concluded from these studies that 79% of the increased yield is attributable to genetic gain. This attribution to genetic gain is substantial. These results are supported in a second study (Russell, 1984) that covered the 1930- 1980 period. The Duvick studies (1977, 1991). using different materials, showed somewhat similar results with genetic gain of 60%. What can be concluded from the genetic gain findings involving higher plant densities is that the maize breeder manipulated the genotype that led to the newer hybrids and, in doing so, accommodated higher plant densities. The plant breeder met the challenge of new husbandry practices of fertilizer, pesticide control, and advances in machinery. Sustaining high yields at higher plant densities includes changes in many plant components that “tailor” or “fine-tune” the genotype for this new level of plant culture. Some changes could include “downgrading” some gene functions. We will return to this aspect in a later section. c. Maize Hybrids and Stress Since final yield is a complex process attributable to a number of components, one significant item is resistance to stress. This is a dominant characteristic of the newer hybrids. The capacity of plants to withstand the debilitating effect of either drought or heat leading to stress is a significant genetic gain. The 1980 hybrids show a decided resistance to a stress environment (Russell, 1984; Duvick, 1991). In the evolutionary development of our progenitor maize plant, the plant accumulated many fail-safe systems such as leaf rolling to avoid moisture loss-a feature used differently in modern maize culture with the numerous amendments. d. Efficiency in Nitrogen Utilization Modern maize growing practices have prompted the selection of hybrids that show a greater capacity to utilize nitrogen by expressing greater nitrogen effi-

82

PETER A. PETERSON

ciencies. This must be a complex trait that would accommodate the more available nitrogen into the maize plant, affecting sink, kernels, leaves, membranes, and other structures. A number of other traits have been improved, such as root and stalk lodging resistance, disease resistance, reduced barrenness, and a simplified smaller tassel. These are changes in the genotype, and in Section I11 the origin of such changes will be discussed. e. Recent Hybrids Are Able to Maintain Yields at High Densities The sustained yield at higher densities of modern-day hybrids is a necessary characteristic. How, then, could plant breeders successfully breed this characteristic into their hybrids? Eliminating excess vegetation such as tassel size, leaf excess, and other undiscovered physiological changes would be possible. In tassel size reduction, the limitation of pollen starch development would save resources that could be directed elsewhere, such as to the ear. Molecular investigations on comparative studies of drought-sensitive versus drought-resistant types will likely identify specific genes that distinguish between the two types.

B. YIELDCOMPONENTS Components of the genome that contribute to yield have been investigated in maize. What would qualify as a component that would have an effect on yield? Studies on this theme have been investigated by the Minnesota Agricultural Experiment Station, Geadelmann and Peterson (1978), and the Iowa Experiment Station by Russell’s group (El-Lakanyu and Russell, 1971; Prior and Russell, 1975). The general objective of these experiments is to maximize a “plant’s opportunity” to increase yield. Would the selection and incorporation of deep kernel (D),long ear (L), and multiple ear (M) (prolificacy) components add to the total yield of a plant? Or would combinations of each of these individual units such as DL, DM, or LM when incorporated in a backcross-selection program substantially affect yield? Geadelmann and Peterson ( 1978) concluded that selectively adding these double components in a backcross-selection program did not exceed the normal type at a high plant density-at least, those densities common to those in the Corn Belt. The modified types did perform well at low plant densities as might be expected, for example, prolificacy where the genomes would maximize its potential. This was true for the other single components. In general, according to a cytogenetic view, these results are not a surprise. At best, a backcross program would entail considerable “drag.” The number of genes in a two-component strategy would be numerous and downstream linkages

TRANSPOSABLE ELEMENTS IN MAIZE VARIABILITY

83

would be difficult to dislodge. Further, a two-component genome contribution might be at cross-purposes. Genes for prolificacy (M)might not be positive for those with ear depth. The genome contribution might be considered as a cassette, a subject that will be discussed in Section 111. Thus prolificacy by itself was successfully utilized at low plant densities whereby this trait “capitalized” on the unused resource potential of the maize hybrid.

111. TRANSPOSABLE ELEMENTS

A. THEIR PHENOTYPE VARIEGATION Transposableelements are pervasive in a wide assortment of organisms (Fig. 1) (Peterson, 1987), and they have been found, often unexpectantly, in most genes that have been molecularly analyzed (Schwarz-Sommer et al., 1984; Wessler and Varagona, 1985). When genes are isolated, inserts are often found that are transposable elements. Mobile elements (mobile and transposable will be used interchangeably) transpose (move from one position to another), and when they insert into genes they interrupt gene function (Figs. 2A and B). This results in a null phenotype such as colorless kernel (disrupts gene transcription at a stop codon or out of frame sequence). This disruption continues until an active element excises the insert (Fig. 2B). The subsequent excision of these inserts, namely, their transposition out of the gene, restores gene function, and this activity on different tissues leads to variegation (Fig. 1A). This is most prominently seen in kernels of maize as kernel color variegation or starch differentials. However, it can also be seen in leaves and in other plant parts (Figs. 1A-D). Though variegation is a prominent part of the expression of mobile elements, these elements can only be seen as variegation when they are inserted into genes that are visually expressed, such as those controlling color or starch type in kernels of maize. That is, there must be a ”reporter” that can be expressed. If these transposons are in an anthocyanin gene controlling kernel color and the plant genotype is not expressing kernel coloration, they will not be evident to the observer. Commercial corn, for example, lacks anthocyanin coloration (two of the genes necessary for color are almost always recessive; consequently, color cannot be observed), and an insert in one of the other color genes is not expressed. 1. Transposons since the Beginning of the Origin of Maize

Though transposons may be prominent either in a genotype or in any population, they are not recognizable to the observer because they have not inserted into recognizable genes; they are still transposing and affecting functions not

84

PETER A. PETERSON

Figure 1. Expression of transposable elements in four species. (A) Zea mays, a2-m (Peterson, 1978). (B) Petunia hybridu (Wijsman, 1986). ( C ) Dahlia, horticultural variety. (D) Glycine m u . 4-m18 (Peterson and Weber, 1969). (FromPeterson, 1987, with permission.)

easily discernible. Transposons have been in maize populations since the beginning of prescientifically driven maize breeding (Blumberg vel Spalve et al., 1990), but they escaped detection in populations of maize because transposable elements themselves do not have an observable phenotype. Transposable elements can only be recognized when they interrupt a gene function. For this, it is necessary to have prominent observable gene functions impaired, such as those affecting coloration in flowers (Fig. lB), leaves (Fig. IC),and kernels (Fig. 1A). These genes with an insert would act as reporter alleles. An insert in a gene for a trait that is part of a complex set of genes (e.g., stem elongation) would not easily be recognizable. Surprisingly, one of the first genes to be experimentally

A

sc

I- - - - - - - - - -

.........xxx xxxxxxxxxx ~llllllll/ll)ll/llllllllllllll ~xxxxxxxxxxxxxxxxxxxxxx.. ........ 1

B

C

C

W

w

m m E E

m m E E

+

+

+

x

I

x

I

+

x

x

1 1 3

m E

I

I

I

x

x

X

3 1 1

3

--

probe A B C ~1111111111111111111llllllllllllllllllllJlllUllllllllllllllllllllllllllllllllllllll~~

C

wx+

P

probo RNA

wx-m8

polyA

- A

P

P O ~ Y A

-c

2.4

-

6

3.2 2

Figure 2. (A) The gene xxx is transcribed (dashed line indicates mRNA) but is interrupted in the insert (diagonal lines) by an out-of-frame sequence or stop codon (sc) terminating the transcript, leading to a null phenotype until the insert is excised. (B) Northern experiment that includes RNA from three different genotypes. The three panels were probed with probes A, B, and C whose origins are indicated in Part C. The A probe identifies the Wx gene; the B probe identifies the I - 8 insert; and the C probe identifies the insert plus the Wx gene. (C) Schematic of wx+ and wxm-8 alleles. The open box represents wild type and the solid box the Spm-18 insertion. (From Gierl et al., 1985, with permission.)

86

PETER A. PETERSON

examined in the beginnings of the science of genetics, namely, the wrinkled peas of Mendel, was identified more than 100 years later (after Mendel’s work) to be caused by an insert (Bhattacharyya er al., 1990). Thus, the absence of transposons in certain crop species may be a consequence of the lack of observable options. A transposable element phenotype has been uncovered in rice (Oryza sariva) (Reddy and Reddy, 1992) and soybean (Glycine m a ) (Groose et al., 1988). One only has to reflect that 50 years of intensive maize genetics preceded McClintock’s (1947) genetic studies before transposons became resolved (McClintock, 1948). These concepts will be demonstrated subsequently.

B. TRANSPOSABLE ELEMENTS: THEIR DISCOVERY 1. A Short Synopsis of the McClintock Experiments

Barbara McClintock arrived at Cold Spring Harbor in 1941. Before going to the Carnegie Laboratory, her studies included chromosome aberrations and rings in maize (McClintock, 1931a,b, 1944, 1945), and before that, she had singlehandedly established the chromosome ideotype of maize that gave maize genetics a sound foundation early in the science of genetics (McClintock, 1929). In the period just preceding these transposon studies, she induced homozygous minute deficiencies by the induction of crossing-over (step A in Fig. 3A) in certain aberrations that led to bridges that yielded mutants (McClintock, 1932, 1944). The breakage events that yielded a number of these instabilities are illustrated in steps B and C of Fig. 3A. The aberration that McClintock used was on chromosome 9 (Fig. 3B). Because chromosome 9 was being used for numerous other studies, McClintock was very familiar with its genes, namely, the color ( C l ) locus, shrunken (Shl) locus, and waxy (wx)locus. She made a detailed analysis of the yellow-green locus at the end of the short arm of chromosome 9. In proceeding with this bridge-breakage-fusion cycle (Fig. 3A), it was clear that the chromosome breaks were at random on the short arm of chromosome 9. Sometimes the Cl and S h l shrunken genes were lost and other times the loss included the Wx gene. Or it was sequential, with the Cl gene lost first, followed by the Shl and then the Wx gene. This random breakage event was expected. However, in the pursuit of these studies, she observed that in one case the breaks were always occurring at a position proximal (direction of Wx toward centromere) to the wx locus (Fig. 3B). Why did the chromosome 9 break at this specific point? She concluded that this specific chromosome site was weakened and had the property to dissociate, breaking rather readily at that specific point. Thus, she named that site Ds for dissociation. As she progressed with crosses in further studies, she observed that this breakage event segregated. That is, this site showed breakage

TRANSPOSABLE ELEMENTS IN MAIZE VARIABILITY

A

87

Bridge-Breakage Fusion Cycles Origin of newly broken ends of chromosomes c wx/c wx x C Wx/C Wx Wx C (Dupl) Crossover

Al

4

cC

1

i \ W x

Wx

A

X : B

BMI

colored

Microspore 4*

X-

cb;;I. , 5*

endosperm enters endkpermw x

Endo 1

w.x/

colorless sector if break at y colorless-waxy

IIIIIIIIIIIIIIIIIIIIIIIIIIIII1111111111111IIIIIIIIIIIII111IIIIIIIIIII

Figure 3. (A) The bridge-breakage-fusion cycle and illustration of some of the events during meiosis leading to transmission of the products of this meiosis to the fertilization event. The crossover from the duplication arrangement (1) leads to a bridge and breakage (2) with a rejoining and bridge and breakage and microspore products (3, 4, 5 ) . (From Peterson, 1987, with permission.) (B) Chromosome 9 in maize showing a break at Ds.

88

PETER A. PETERSON

because the site segregated like a Mendelian event. In reconstitution crosses, she uncovered a second factor that was needed for breakage to occur, indicating that the breakage event required two factors-a breakage site Ds and a second factor, Ac (McClintock, 1945). This is what was suggested. She reisolated these lines in appropriate crosses and reconstituted the experiment. Starting with a line (the Ds line) that did not show breaks, she introduced the second factor to this line. From this cross, breakage again resulted. She was able to demonstrate that, indeed, there was a second factor that activated the Ds breakage event. She called the second factor Ac for activator. There were more surprises to come. In subsequent crosses, McClintock found that this breakage event now was located at a position different from the original one on chromosome 9. Instead of losing all the material distal to wx (Fig. 3B) (chromosome 9), the site was now located in a place whereby the Shl and the C1 loci were lost (McClintock, 1948). Could it be that the Ds locus moved from one position to another and was now located just proximal to Shl? Or was a new site activated to become Ds while the other site was deactivated? Further experiments would clarify this observation. The unexpected observation came in the cross illustrated in Fig. 4A, in which an exceptional kernel appeared (part C of Fig. 4A). The two expected phenotypes resulting from the cross are shown in part B of Fig. 4A. The expected genotype C l cl cJ in the aleurone of the cross ( c l lcl X C1lC1) should show the loss of the C1 locus to yield a colorless region when the cross was made against a recessive c line. However, there was an exceptional kernel in that the pattern of the coloration was reversed. Instead of colorless sectors on a colored background (loss of C), the pattern showed colored spots on a colorless background. This particular kernel was identified as cl -ml . A dominant C1 gene had seemingly changed to cl -m. When pursued further in backcrosses, c l -ml lcl x c l I c l , she found that the variegated phenotype was dependent on a second factor (McClintock, 1948). The segregation told her that. Further, in making crosses to Ds lines with the same second factor, she found that this factor indeed was Ac (Fig. 4B). Thus, if Ac was activating this variegated phenotype, it must mean that Ds had inserted itself into the C1 gene, changing it to a cl-ml allele. This was truly transposition from one point of chromosome 9 to the C1 gene. This was also creditable deductive reasoning, reasonable in hindsight but difficult to envisage. There were yet more surprises. Many other loci became unstable and were shown to be controlled by Ac. It seemed that Ds elements exploded and were transposing into genes (a significant novel concept), inactivating many genes, and responding to Ac. Thus, with these observations over a number of years and with many intricate and delicate experiments, there was established the notion of transposition of genetic material that resulted in the control of the expression of genes (McClintock, 195 1). Variegation, therefore, was caused by elements in the genomes that move.

TRANSPOSABLE ELEMENTS IN MAIZE VARIABILITY

89

A

iB = CC I

c-m-1 Discovered

--

Yg C Sh wx Ds = Colored wx

yg c sh wx

=

1

shrunken

No Ac

on AC tester (CDsShWx)

Ac line

=

Regulator-Receptor Interactions

I

I

t Round-%=C-c

t

4

Ds line

Receptor

I

I I

cm-l

csh wx t’ Round :(shrunken)

I I

:

sh

-

I I

Reg2

-

I

I

I I

I

Jq case (Others)

I I I I

2

l?kIkkl

1 Reg2

-

+

i n AC tester (CDsShWx)

*-50:50*

I I I I I

-

I

I

Regl t H Regl

-H

I

c-c

I

c sh wx c sh wx

kernel CShwx csh wx

kernels

I

I

- x -

-, Variegated

Ac

Yg C Sh wx Ds

4 Ilt-colorless

Expected

&

I

‘ WAC

2

Ds x E Ds c

-

c-c

CI

0 I

0

Figure 4. (A) Origin of cf-mf ( I ) . From the expected phenotypes shown in (Z),an exceptional spotted kernel appeared (3). The test on a CDs line to test for the Ac relation in (4) indicates a c-rnf relation to the Ac-Ds family. (From Peterson, 1987, with permission.) (B) Cross of a Ds line (CDs) with an Ac line ( c Ac) yields variegation by the induced breakage at Ds, causing the loss of C. (C) Assorted segregation types of regulators such as Ac and En and receptors such as CDs and arnl(1). ( + , mutability; - , no mutability). (From Muszynski, M. G., and Peterson, P. A., 1990).

Although variegation has been identified in historical references since botanists first reported “eversporting” varieties in their study of plants (e.g., Darwin, 1868; Lecoq, 1862; Knight, 1808), these transposable elements were not available for analysis until McClintock (1951) found them in her Cold Spring Harbor nursery. And not because new tools were available in the 1940s. This could equally have been analyzed in the 1920s if the same experiments were conducted with the bridge-breakage-fusion cycle. The first clue was the observation that breaks were at a specific site and regulated by a Mendelizing second factor. This was followed by the observation that these elements could be found at a new site

II

90

PETER A. PETERSON

and then in a gene cl -ml that showed the variegation that resembled the eversporting varieties of an earlier era. Now transposition of genetic material can be associated with variegation.

2. Concepts of Unstable Genes in the Pre-McClintock Period Such observations require a concentration on detail and precision in experiments, truly a McClintock trademark. But what image did geneticists have of unstable genes during the 1940s and early 1950s? During that period, unstable genes (as they were called in the 1930s) had a different image than they did later. Some of the early history on the concepts of mutable genes in a variety of plant species has been reviewed by Demerec (1935). At that time, they were considered to be unstable genes. Thus, up to the time of McClintock’s experiments, this variegation of plants was attributed to unstable, mutable, or sick genes. The early studies in maize by Emerson, and subsequent studies in the decades of 1910 to 1930, came very close to a genetic analysis. The key component of demonstrating transposition evaded Emerson’s studies, as well as later studies by Rhoades, because of the lack of evidence for transposition. Given more extensive crosses, especially by Emerson (1914, 1918, 1929) with his unstable pericarp, he would have been mystified by the results if transposition was observed, especially at a time when concepts in genetics were being established. If a reporter allele was not present or available, there would have been a difficulty and, especially working with the P allele, most of the possibilities would have evaded Emerson because the P-vv locus (unstable pericarp color) would have covered up some of the mutants that might have occurred in various aleurone genes. Thus, the discovery waited until the mid- 1940s when McClintock experiments with another goal unexpectedly uncovered mobile elements. It is appropriate to set the stage for the mid-1940s when McClintock, working at the Carnegie Laboratory at Cold Spring Harbor, uncovered a large number of unstable mutants while studying the bridge-breakage-fusion cycle (BBF). The gene at that time was vaguely conceptualized as beads on a string. This was amplified by the consideration of the Drosophila banding patterns. Further, McClintock (1942, 1944) had shown in previous work with the BBF cycle that some of the maize chromosomes contained chromomeres that appeared like beads on a string, which could be “fractured,” and the resulting products showed mutations of various genes on that chromosome.

3. The Concept of Genes at the Time of the McClintock Experiments How did McClintock conceptualize gene material when she conducted her experiments? Actually, it didn’t matter because she was driven by her own acute observations, and this led her in pursuit of further experiments. Her colleagues

TRANSPOSABLEELEMENTS IN MAIZE VARIABILITY

91

at Cold Spring Harbor on Long Island, New York, were very active in the forefront of genetics. Demerec was pursuing Drosophila and Salmonella genes, Kaufmann and McDonald were studying Drosophila, and MacDowell was studying mice (Mus) (Demerec, 1946, 1947). Evelyn Witkin was working with repair mechanisms in bacteria. This was the core group at the Camegie laboratories. Bruce Wallace with his Drosophila populations joined them in the late 1940s. McClintock let her experimental observations drive her research rather than be biased by preconceived notions. Yet, in the genetic community, and especially in the New York area, there was extensive speculation on the genetic material, although the experiments of Avery et al. (1944) had not yet been coupled with those of Hershey and Chase (1952). Also the Watson and Crick (1953) model for the molecular structure of nucleic acids was not available. What was the prevailing understanding of genes there in the late 1940s? Visual observation of the Drosophila chromosome indicated a banding pattern and maize chromosomes showed chromomeres, suggesting beads on a string. Furthermore, the Drosophila work in the late 1930s and 1940s uncovered a phenomenon called position effect. This concept was strongly supported by GoldSchmidt (1953, who proposed that mutations were caused by interruptions in the chromosome by some rearrangement. Why did he support this notion? A number of genes were found, such as the genes of the Notch locus in Drosophila, that by all observable features were an unchanged banding pattern and seemed to be normal. Thus, it could be concluded that there must be some rearrangement of the chromosome material. The banding pattern in Drosophila was the key to gene arrangement. Further, other genes were found to cause mutation by a rearrangement that was not at the gene affected but at a short distance away from the gene site. How could gene expression appear and disappear in development? One could not assume changed genetic material, but one could assume affected genetic material. Thus, there was strong support for position effect, meaning that the normal locus, near a rearrangement break, seems to have changed expression to a mutant action, making the chromosome as a whole the primary focus of gene action. There were many cases of this type. The yellow locus in Drosophila was an example. The BAR duplication (Bridges, 1936) had a quantitative effect when located in a specific arrangement as visualized with the banding pattern. Other significant research was Beadle’s (1932) work with the sticky gene. Here, all the chromosomes were affected by a mutation in only one gene. Trans effects were not part of the dialog. The sticky gene by its action led to an increase in both chromosome rearrangements and point mutations. Thus, it appeared from the phenotype of the sticky gene that it, too, was caused by rearrangements that resulted in a position effect. These observations by Goldschmidt related to his concepts of chromosome structure and integrity. This was similar to thephl gene in wheat (Triticum vulgare) that dominates chromosome pairing between genomes (Gill and Gill, 1991).

92

PETER A. PETERSON

Since position effects could be reversed, it seemed that the genes were unchanged in content but rearranged, and thus could be re-rearranged. There was no support for the idea that there was any difference between position effects and point mutations. Now we come to the so-called repeats. This was a case where mutants (two or more) behaved like multiple alleles, for example, the Drosophila studies of Bithorax by Lewis (1945), lozenge (lz) by Green and Green (1949), and, in maize, the Ab locus by Laughnan (1949). These fit Goldschmidt’s theory of genes and mutation, and although there was limited support for his views, they could not be ignored. Thus, Goldschmidt hypothesized that mutants are rearrangements in the chromosome that cause mutation. This hypothesis supported the idea that the entire chromosome was the basis of genic action, and the interruption in this chromosome led to these mutant loci. This, then, was the status of genes when McClintock hypothesized about her observations.

c. COMPONENTS OF TRANSPOSABLE ELEMENTS 1. Genetic When a gene controlling color, such as a l - m , shows variegation, and is outcrossed to a standard full color line and the subsequent F1 (Allal-m) is backcrossed to a standard recessive ( a l - o ) , two possibilities may be visualized (see Fig. 12). If all or most of the noncolored progeny are variegated (Fig. 12A), one could conclude that the element is expected to reside at the gene locus, and the unstable allele would be considered autonomously mutable. This would indicate that mutability is coupled with the a l - m allele. If, on the other hand, only half of the noncolored kernels are variegated (Fig. 12B), then the control of mutability is independent of the al-m allele. In this second category, independent control, the insert at the locus (such as McClintock’s Ds in our previously described example) was nonfunctional but could respond to an active element that was segregating independently. Specificity: When the second component (identified as Ac) is introduced to the line with a gene with an inserted component (Ds), variegation in the form of spotting results (Fig. 4B). This resulting variegation is derived from a very specific interaction. For example, when this colorless kernel (al-Ds, Ds is an insert in the A1 gene, eliminating A1 expression and therefore a1 -Ds)was crossed to a previously described unstable element, namely, the Dt element (Rhoades, 1936, 1938), nothing happened. There was no instability and the kernels remained colorless. Thus Dt induced spotting on the al-dt allele but showed no spotting with the al-Ds allele, though the inserts are at the same locus. Similarly, when this same al-dt allele that responded to Dt was now exposed to Ac, again

TRANSPOSABLEELEMENTS IN MAIZE VARIABILITY

93

nothing happened, the kernels remained colorless (Fig. 4C). Thus, though these two alleles resided at the same locus, there was something unique about the interaction of each of the elements with the introduced functional element. The resolution of this specificity of interaction awaited molecular investigations (Fedoroff et al., 1983; Pereira et al., 1986). This was further substantiated as new instabilities were uncovered. For example, when another a1 allele, one of the En system, was tested with Dt, again no variegation resulted although it expressed variegation with En (part A in Fig. 4C ). Similarly with Ac, no variegation resulted (Peterson, 1961). Thus there appeared to be a specificity of the active element such as Ac, Dr, or En with something specifically unique to it at the locus. What was unique at each allele could be recognized only by the specific element. From a one-component autonomously mutable allele, a two-component element system arose. Thus, there was a relationship with the initial insert and the resulting allele when a onecomponent mutability changed to a two-component mutability (independent control). As was earlier indicated by Peterson (1970), the visitation at the locus by a functional element caused a change to an allele with resulting loss of mutability. In this change from a functional element to a receptor element, something was “left behind.” What was “left behind at the allele” awaited molecular investigation. 2. Molecular TNPA and TNPD With the uncovering of the Ac transposable element by the Fedoroff group (Fedoroff et al., 1983; Pohlman et al., 1984), these mysteries regarding specificity quickly dissipated. It was clear that with the investigation of the wx-m9 allele, a comparison was possible with the derivative wx-m9 receptor allele. When the one-component wx-m9 autonomous mutable wx was compared with the wx-m9-derived Ds-containing unit, it could be seen that the wx-m9 Ds was derived directly from the wx-m9 Ac by deletion of part of Ac in the original allele that incapacitated its function. What Ds lacked was part of Ac (now we know it as the transposase-determiningportion of Ac), which is an active component that can recognize the insert that it left behind, namely, the Ds. All the other parts remained intact. Research on the elements uncovered that they had a structure and definition that could be analyzed. The Ac element consisted of 4000+ nucleotides (bp unit), and it was found that there was a definitive structure to the insert. From sequence analysis, a terminal inverted repeat (TIR) of a certain length could be seen as well as a target-site duplication (TSD) (GTT in Fig. 5 ) . Was there a pattern to element structure? Were TIRs and TSDs part of other elements? When EnlSpm was uncovered and molecularly analyzed, the TIR and TSD were also there but were different in length and content. This fit the geneticists’ receptor

PETER A. PETERSON

94

I I 5 5 I I

49

i\ A

;:“ T

A

C A T C -C wxm8 5’ C G T G G T C A A m wxt

CGTGGTCAA

m C A A C G C G G C 3’ CAACGCGGC

Figure 5. Part of a transposable element illustrating the TIR (13 bp) and the TSD (GTT) (3 bp). The motifs (the lines between 2 5 , 29, etc.,) are aligned and continue for 200 bp.

element as a fractionated part of the active element when a comparison was made with the w x - d element (Schwarz-Sommer et af., 1984; Pereira et al., 1986). It was also clear that the identifying mark of the element had a specific TIR and TSD like the En. Thus, what was determined genetically on the specificity of interaction became clear. The elements of a specific group fit into a specific category, namely, identifiable TIR and an identifiable TSD. The functional element Ac or En only recognized its own motifs, and the next several years were needed to verify this. These TIRs and TSDs are illustrated in Table I. So, what was “left behind” as conjectured in 1970 (Peterson, 1970) was a fractional part of the original element. These transposable elements have been dissected and reexamined in more extensive detail with more definitive molecular examination in the last decade. Since the EnlSpm is most fully investigated, it is appropriate to examine this element. The EnlSpm transposable element has 8237 bp and contains a definitive structure (Figs. 6A and B). By examining the full-size sequence of the element and comparing it to the cDNA, it was evident that there were 11 exons and 2

TRANSPOSABLE ELEMENTS IN MAIZE VARIABILITY

95

Table I Molecular Characteristics of Some Examples of Plant Transposable Elements“

Element Dsl ACl En I Spm-I8 MuI-MuS BR rbR rcy:Mu7 Mu RI Tam1 Tam2

NAU Au Au NAu NAu Au

NAu NAu Au NAu

Plant Zea mays L . Zea mays L. Zea mays L. Zea mays L. Zea mays L . Zea mays L. Zea mays L. Zea mays L. Zea mays L. Antirrhinum Anrirrhinum

kb 0.405 4.563 8.287 2.241 4.869 2.2 4.0 17. Varies

DuplicaTerminal in- tion of verted repeat target (bp) site I1 11 13 13 + 200 5 5 200 220 13 13

+ +

8 8 3 3 9 8 8 9 9 3 3

“From Peterson (1987). Hartings e r a / . (1991). and Chomet er al. (1991) bNonautonomous. ‘Autonomous.

open-reading frames. It was also evident that there were TIRs on the end and, of course, a TSD of 3 bp. In subsequent genetic studies following the original description of En at the p g locus (Peterson, 1953, 1960), an al-rnl allele was found to be colorless in the absence of EnlSprn but colorless with spots in the presence of EnlSprn. McClintock found an al-rnl(5729)allele that was colored in the absence of En/ Spm but colorless with spots in its presence. This was different. It is readily evident that the functional EnlSprn suppressed coloration with a Suppressor (S) function and induced spots (excision) with a Mutator (M) function. This is what made EnlSpm so attractive for study. It showed two clear functions, S and M. But, how did these genetic functions relate to the molecular structure? Two functions could be determined in genetic investigations. These interactions are illustrated in Fig. 7. It is now necessary to dissect the molecular sequences and determine the boundaries of the functional components. In Southern blotting and probing with En, two bands are recognized, a heavystaining 2.5-kb (Fig. 2B) and a lighter 6.0-kb band. These two transcripts were related to the En structure (Fig. 6A). And, from the sequences, two proteins could be visualized, TNPA and TNPD (Fig. 6A). Several studies demonstrated the functional domains of these two proteins (Frey et al., 1990a,b; Masson et al., 1991a,b). TNPA is a DNA binding protein and recognizes the IZbp sequence motif of

96

PETER A. PETERSON

P

AUG

P

B

UGA

4

En1

C

D

P

AUG

4

4

UGA 4590 bp

1-102

u

PMW-1ERMWAL EXONS

CMBOXY-TERMINAL EXONS

Figure 6. Structure of the En1 element. The ORFs (dotted in A) and 1 1 exons (open boxes, shaded in A) are shown in B. The origins of TNPD and TNPA are also shown in A. A derivative En2 is illustrated in C showing the deletion of OW2 and part of ORFl. The alm(r) deletion derivative (1102) is shown in D.

the TIR and adjacent motifs. The S function was genetically defined as the suppressor function of the EnlSpm system (Fig. 7 , bottom) by McClintock (1961). It should be emphasized that this suppressor component can only be visualized in alleles such as al-m1(5719),where the insert allows gene expression to occur, and this gene expression can be suppressed by the introduction of an EnlSpm. Other alleles such as al-m(r) (Fig. 6D) (Peterson, 1961) are colorless and, therefore, not S expressive. It has finally been shown that this suppressor function is

TRANSPOSABLE ELEMENTS IN MAIZE VARIABILITY

97

Figure 7. The a-ml allele (a-m157l9A-I) without (top) and with (bottom)En (colorless with spot). It is shown with the linked sh2 gene and a weak M action but a strong S (fully suppressed).

caused by the TNPA protein that blocks transcription read-through (Gierl et al., 1985, 1988a,b). At this point, there are two genetic functions, and they relate to the two transcripts tnpA and tnpD. Also known is the structure of En (Fig. 6). What is not known is how the structure relates to the genetic and molecular observation. What is the extent of the En sequences that govern these functions? How many of the exons (Fig. 6A) are related to the S function, and how many are part of the M component? An attack on this problem would come from a deletion analysis. Two mutants were available, En2 (Gierl er al., 1988a) and Spm-w8011 (Masson et al., 1987), to determine the sequences governing these functions. The En2 mutant was molecularly analyzed and shown to be a fractured En and to have all the exons intact (Fig. 6C). There was, however, a 1126-bp dele-

98

PETER A. PETERSON

tion that included all of ORF2 (open-reading frame) and part of ORFl . In genetic tests, En2 had a full S function. Because the S function was expressed in genetic tests, it could be concluded that the exons were responsible for this function. But the M function was impaired. Only a few spots appeared in genetics tests with appropriate reporter alleles (Fig. 8). The M function must therefore be controlled by the ORFs, which showed a deletion. The few spots could come from residual En activity in the genome according to these investigators (Gierl er al., 1988a). Yet, given another reporter allele with En2, a high expression of M is observed (lower ear in Fig. 8). It is clear from this observation that the deletion in En2 did not eliminate the M function. What of the rest of the structure of the En element? The 13-bp perfect TIRs in the ends of the element and the sequence motifs in the subterminal regions of EnlSpm (Figs. 5 and 6) define the critical cis-determinants needed for excision of IIdSPM elements. Further, the subterminal 12-bp TNPA binding motifs (Fig. 3,though quite similar, must be in an ordered orientation, that is, head to head or tail to tail with a prescribed distribution and repetition for proper excision. Nature is very precise. The trans-effect is derived from the TNPA protein (the S function) that brings the ends of the element together like a zipper such that the

Figure 8. The En2 element with a l m 15719 on top and alm(Au)pule-(mr) on bottom. The difference in spotting is due to the receptivity of the reporter alleles.

TRANSPOSABLE ELEMENTS IN MAIZE VARIABILITY

99

TNPD protein (the M effect) could cut the element where the 13-bp TIRs intersect with the TSD (Fig. 5 ) . These two products of EnlSpm, namely, TNPA and TNPD, were critically analyzed by using transgenic systems. Each was inserted into transgenic tobacco (Nicotiana tabucum) with a GUS insert (Frey et al., 1990b; Masson et al., 1991b). By making several definitive constructs, it was found that with TNPA alone, no effect can be seen. Nor with TNPD alone. However, when the two are combined in the same transgenic tobacco plant, GUS expression (expression of P-glucuronidase) is evident, indicating that excisions are taking place. An I element (dSpm) is inserted in the GUS coding sequence and thus inhibits GUS enzyme activity. Excision of the I element insert permits GUS activity, which is expressed as blue-staining tissue upon histological staining with X-Gluc (Jefferson et al., 1987). With these experiments it was clearly obvious that the S function was associated with TNPA and M was associated with TNPD. The previous concepts of the interaction of elements in causing excision are confirmed as earlier envisioned, namely, that TNPA binds and brings the inverted motifs together (Fig. 5 ) and the TNPD along with it will cause excisions to take place. Thus, with these experiments, it is shown that TNPA is necessary for TNPD to function.

D. SYSTEMS Early in the genetic investigations of mobile elements, it was obvious that these various elements fit into systems. This was based on the exclusion of interaction in some cases and the inclusion in other cases. This is illustrated in Fig. 4C.Here the Ds or the I element is identified as the receptor, and the Ac or the En element is the regulator. When receptor 1 shows a response to regulator 1, it is part of the same system. However, it does not show response to regulator 2. This is the case of an En element with a Ds receptor. Similarly, receptor 2 does not respond to regulator 1, but it does respond to regulator 2. This would then account for the Dr not activating the Ds elements as well as the En not activating the a-dr allele. Relating back to the molecular investigations, the Dt protein does not recognize the /-element motif nor does the En protein (TNPA) recognize the Ds motifs. In past decades with genetic studies of transposable elements, eight systems (Fig. 9) have been identified. Each system shows a specific genetic interaction, and those analyzed genetically show a distinctive molecular DNA sequence. These systems are identified with an exclusion pattern as well as an inclusion as indicated, and when their molecular definition is determined, they show unique TIR and TSD configurations (Table I). A question that will be considered is why

100

+

*

PETER A. PETERSON

= responds as mutability; absence of = Au = Autonomous element

+

indicates no response.

Figure 9. Transposable element systems. Regulatory elements are specific in their activation receptor elements. There are three cases of overlap in activity. Ac activates all Ds’s, but Llq activates only Dsl. Fcu activates rcu and R-r#2 but Spfis limited to R-r#2. Cy and MulR appear to have homologous activity. ( + ) of

have these various elements survived in maize populations and why are there so many diverse systems?

E. GENETIC RESOLUTIONOF A TRANSPOSABLE ELEMENT First, let us examine the nature of their survival in maize populations. Variegation in flowers and leaves as well as in kernels was seen in botanical investigations long before the geneticists realized they were associated with elements (transposable) in the genome that transpose. Further, there was clear evidence that newly arisen variegated loci arose in experiments conducted with mutable loci (Demerec, 1935). Why, then, were these variegated phenomena not assignable to transposable elements? The clue to this has already been given. Reporter alleles had not been isolated that would have provided the tangential evidence needed to establish a relationship. Such a case will be given of a newly arisen variegated locus. In northwestern Colombia, in the swampy region near Panama, the Cuna In-

TRANSPOSABLEELEMENTS IN MAIZE VARIABILITY

101

Figure 10. Cuna tribal corn. (FromGonella and Peterson, 1977.)

dians grow a maize with a striking wine-red variegation symptomatic of a transposable element (as opposed to a mottling phenotype often seen in some crosses) (Fig. 10). If such maize seed were brought into a laboratory, one would ask several questions (Peterson, 1978). What gene is involved? To what transposable element system does it belong? And what is its inheritance, whether it is autonomously mutable or whether it is independently controlled? A crossing protocol as illustrated in Fig. 11 would be developed to answer these questions. The crosses in the Y pathway in Fig. 11 would establish that an anthocyanincontrolling locus was involved. In the Cuna example, the variegation was assignable to the r locus and therefore is r-m(Cuna) (Gonella and Peterson, 1977). The X pathway in Fig. 1 1 is followed to answer two of our other questions-the nature of inheritance and system relationship. The results seen in X-1versus X2 illustrate the inheritance pattern. Because the segregation does not agree with results expected for X-1 but was like X-2,variegation is controlled by an independent factor. Using the variegated kernels from the x-2cross progeny, a system relationship could then be tested. If any variegation appeared in the crosses (X-2I ) , then r-m(Cuna) is assignable to the system represented by that reporter allele. Because none of the reporter alleles showed any variegation in the crossing protocol, it could be concluded that r-m(Cuna) is a new transposable element system and is controlled by an independent factor. This was finally identified as the Factor-Cuna system (Fcu) (Gonella and Peterson, 1977, 1978).

F. TRANSPOSITION 1. Genetic Detection McClintock (1948) established that Ds had moved. From a position proximal to the wx locus in her initial studies, Ds was now near the CI-ShI region. In a

PETER A. PETERSON

102 Variegated Cuna

x

V pathway tester lines of anthocyanin genes. al, a2, c, r, c2, etc.

X pathway

only the r tester uncovers the mutant. All the other Fl’s are colored. Conclusion: rlt Is r-m (Cuna)

on to color line &

colored F1 seifed 4

w X-1 - if co1or:variegated -

X-2 if co1or:variegated:colorless

=

31 variegation is autonomously controlled

=

12:3:1 variegation is controlled by an independent factor

determlnatlon X-2-1 Cross r-m (Cuna) to assorted reporter alleles c-m(r), a-rcy, a-Ds., etc. r-m(Cuna) x

reported alleles--ex c-m(r)

b

Colored F1 X-2-2 R/r-m (Cuna) C/c-m(r)

x b

No varieaation

reporter alleles-ex c-m(r) if variegation appeared with one of the reporter alleles, then r-m (Cuna) would be of that system. Result is negative for c-m(r).

-,

Conclusion: r€n is not implicated. The same with all other reporter alleles. A new system Is estabtlshed, Fcu-rcu

Figure 11. Determination of the gene locus, heritability, and transposable element system of a newly discovered variegated allele. Described in text.

strict sense, one could question whether this was the transposition of material. All evidence suggested that it was. For the purist, however, there was a nagging question. Could the events observed be a correlative deactivation followed by activation elsewhere? The geneticists of that period were willing to consider any option but concede that genetic elements moved. That is, where Ds was found in the position proximal to cl, could that be an activation of an element near the cl position followed by deactivation of the one proximal to wx? Similarly, with the origin of c1-mf, could that be from an activation of an element subsequent to deactivation elsewhere? Although this seemed like a remote possibility, it was a legitimate question at the time, especially if one was not prone to accept the possibility that genetic elements can transpose.

TRANSPOSABLE ELEMENTS IN MAIZE VARIABILITY

103

This could readily be obviated if there was a correlated event. If it could be clearly established that the loss at one place was followed by the discovery of the element at a new position, especially in the same tissue, this would be most appropriate. Such a demonstration was possible with a variegated P locus ( P RR-Mp), where a variegated (coarse) phenotype was associated with the original mutable pericarp. This variegation is conditioned by an insert of the modulator element (Mp) (now recognized as Ac) in the P locus (Emerson, 1914; Brink and Nilan, 1952). When a second Mp element is present, the phenotype changes. The new phenotype has a distinctly fine type of variegation as opposed to the coarse type of variegation with the original mutable pericarp locus. Coincident with this change in phenotype, the excision of the M p leads to a colored phenotype. In crosses where medium pericarp variegation (the c o m e type) was observed, there were twin sectors of colored and fine type. This suggested that the fine and the red co-twin were the produce of one event. Numerous twin sectors were found. When it was analyzed, it was shown that the red sector lacked the Mp and the fine sector had the extra Mp identified as rrMp (transposed Mp). This would indicate that a loss of Mp at one position (coarse type) yielding a red full-colored phenotype meant a gain of the element at another position, resulting in two phenotypes as twins (Greenblatt and Brink, 1962, 1963; Greenblatt, 1966, 1968, 1974). The most important feature is that the various rrMp arising from transposition from the P-RR-Mp had their origin from the P locus (because the red phenotype appeared), indicating that transposition did occur from the P locus. There was another surprise. From one transposition, two trMp arose, and when tested via linkage studies, both indicated that they were at homologous positions on two different chromosomes. This became a useful and revealing method concerning the transposition mechanism (Greenblatt, 1984). McClintock’s studies were reconfirmed by the Greenblatt and Brink studies, and transposition was now established. An example is presented to illustrate the transposition of En from a f -m(papu). The a f -m(papu) allele is an autonomously mutable allele inserted in the second exon of the A 1 locus. From several studies it became clear that the En was located at the locus because of the correlative transmission of al-m(papu) and En in testcrosses (Fig. 12). Here, most of the non-Af (colored) kernels are variegated, indicating that mutability follows the a f -m(papu) allele. This was verified in both a self and a backcross population. The few colorless kernels arising in the backcross and in the self populations are disturbing. A test of these colorless exceptions is followed in Fig. 12B, which will determine the nature of the newly arisen colorless kernels. In this cross, the colorless kernels are crossed to a reporter allele, the a l - m f allele (a reporter allele “reports” the presence of En). This is useful because a l -mf is closely linked to sh2, and one can assume that when the shrunken phenotype is observed, sh2 is following the a l - m f allele. The product of this cross is variegated, which indicates that the newly arisen colorless kernels do carry En, and that these changed

104

PETER A. PETERSON Bl-rn(a@g) X A Z a 1-m(papu) A1 4

A

x

F1 a 1-m 1(papu)

4 112 colored:l/2 variegated: few colorless

48 4 B colorless

m

alsh2

3 colored: 1variegated:few colorless = al-ml?)Shl x 9-m-1 sh alsh2 a-m-7 sh

4

a1-mlnrJShZ

C

x

a 1-m lsh2 variegated

4

m alsh2

112 colorless:l/2 colored sh:few variegated sh

D

I

I

I

I

I

el-m(nr) Sh2 f

En

I

MmBmmHmmHWMmmmmmmmm

f

X f

mmmmmmmmmmmmmmmmmmm f

I

al-ml

I

I

sh2

I

Figure 12. The genetic determination of a transposition event at the A/ gene. At B, the newly originated colorless round kernels originating from al-mlpapu) (A) are an unknown but are proven to be a/m(nr) (nonresponding) (contain an En-see variegated at C ) derivatives from the excision of En. In D, the linkage of the newly excised En to the alm(nr) allele is shown and the variegated shrunken kernels arise by a crossover at X that links the En to the a l - m l reporter allele. (al-mpapu. variegated; A / , colored.) f,continuation of gene sequence.

types are defective derivatives that no longer respond to En. That they do not respond indicates that defectives (excision repair difficulties-see Section 111, F3) do arise and are not responsive. But in this case, they do carry En and therefore are considered to be nonresponding a1 -m(nr) derivatives (Fig. 12C). This could be concluded from the cross indicated in Fig. 12B. When the cross is made as is shown in Fig. 12C, there is an indication that the En transposed from the initial a1 -m(papu) allele is close to its original site. This could be concluded because most of the derivative round (Sh2) colorless kernels from the cross in Fig. 12C are colorless, indicating that the En is in a repulsion phase with reference to the al-ml(sh2) reporter allele. The progeny resulting from the cross indicates the consequences of the transposition of En. In Fig. 12D a diagrarnmatic chromosome is shown to graphically indicate the events that occur in Fig. 12C. It would take a crossover (X)to link the a1 -ml reporter allele with the

TRANSPOSABLE ELEMENTS IN MAIZE VARIABILITY

10s

En that yields the few variegated shrunken kernels in the progeny of the cross. Thus, the transposition event would be established. More extensive data on En transposition were given by Nowick and Peterson (1981). The genetic studies demand that the transposable element be excised cleanly from the originating site. The basis for this is that the transposable element at the new site has all the fidelity of its original composition. Otherwise, the En functions would not be expressed. Thus, the mechanism required for this excision of the element must be precise and definitive in cutting out the element. As indicated previously, the mutable pericarp studies were definitive in describing the transposition mechanism. In later studies by Greenblatt (1984), a model was shown that illustrates the mechanism. In brief, the excision event takes place and the excision of the element occurs from a replicated segment. The new site for this transposing element is an unreplicated chromosome segment. Since it is copied before replication, this indicates that after replication we now have two elements where we had only one transpose. This is illustrated in Fig. 13A (Dash and Peterson, 1993). Such a model was aided and abetted by the availability of the twin sectors that yielded the two events in the same position. Thus, it could be established that from one transposition event, the two phenotypes appeared, namely, the colored and the fine type. When it was finally determined that the colored lack the M p and the fine pattern had two Mp's, it was obvious that one event causing the loss of M p yielded two rrMp's. The original site in the M p case was the P locus. And the P locus could be screened for M p to determine that the M p was no longer at the P locus. This was shown molecularly in later studies by Chen et al. (1987) and by Athma et al. (1992). It could be confirmed by showing that the donor site no longer had the element and that the element now was at a new position. What was also shown is that the new positions were close to the donor site. This indicated that the transposition occurred to nearby sites, which is a favorable feature for attempts to target genes. With this possibility, one could place the target close to the element or rather the element close to the target to maximize targeting efficiency for a gene. Constructs are possible to enhance the process (Dash and Peterson, 1989; Chang and Peterson, 1993). Further details on this are shown in several reviews (e.g., Peterson, 1987).

2. Key Features Leading to Variability:Alteration in Genes There are two components to the feature that contributes to gene changes and subsequent variability induction by transposable elements. Recall that the insertion of a transposable element in a host site induces a target-site duplication. In essence, nucleotides are added to the genome. Each system has its own signature with reference to the number of nucleotides added. The least number of nucleotides in a target-site duplication is the 3 bp contributed by the EnlSpm system. The Ac and the Uq elements contribute eight nucleotides to the target-

106

PETER A. PETERSON

IIIIIIII111IIIIIIIIIIIIIIIIIIIIIIIIIIIII; I' II 11111111111111111111llllllllllllllll

B

1. A transposase generates staggered nicks

m

=C I I

1. A transposase generates staggered nicks

5 &&3 --5

5 - - 3

t

1

< 3

5

t

32. Hypothetical intermediates Integration of a "clean" element

2. Hypothetical intermediates

I

-

3 Possible excision products

I

,

-

I

or

-

I I

I

3 Figure 13. (A) Transposition of En. Like Pvv, E n excises and yields two chromosome strands with an En. (B) 1. Excision of a plant transposable element. The wavy line represents sequences of the termini of an inverted repeat. The arrows indicate the staggered nicks. 2. Exonuclease degradation is indicated by fine dotted arrows; polymerase action is illustrated by bold dashed lines. It is here that the errors occur leading to altered sequences as shown as possible excision products in 3. (C) Integration of a transposable element. Arrows indicate the staggered nicks. D) Same symbols as in B, illustrating the possible origin of one excision event. The nicking occurs as in B. but the element has not been released. The boxed-in sequences (e.g., GTT) represent TSDs. Thin, dashed half-circles represent the extent of exonuclease degradation. Polymerase action is illustrated by a thick. dashed line. The resulting sequences are different from the original. (After Saedler and Nevers, 1985, with permission.)

TRANSPOSABLE ELEMENTS IN MAIZE VARIABILITY

107

site duplication. The Mu element contributes upward of 200 bp. Thus in each insertion event, the nucleotide sequence in the gene at the site where the insertion takes place increases by that number of nucleotides-3, 8, or many more. At some time in the growth of the tissue, the excision of the element takes place. The excision process is not precise. It will be shown that in over 90% of the excisions of an element, the host site is different than it was before insertion. As the transposase acts to cut out the element (recall that the cut in the element is very precise since the new site of insertion has a complete, functional element that includes the 13-bp motif of the En system or 1 1-bp TIR motif in the Ac system), errors are made in the replication and ligation of the host site. The problem lies with the whole sequence of the host site. If the gene locus was left as it was before element insertion, the duplication would not be present. However, in most events the excision process is not precise, and this should be described. First, consider that the excision process involves a protein-DNA complex (Fig. 14B)in which the endonuclease enzymes, during the binding process, cut out the element leading to excision. As hypothesized by the studies of Frey et al.

A

lxxxl -Target

site duplication TSD

11111111111111111111lllllllllllllllllllllll111111111111111111

B

GTC AA

ATA ATA AAC GCG

Figure 14. (A) Parts of a transposable element are identified. The xxx of the TSD (3 bp) and TIR (13 bp) are generalizations of nucleotide sequences. Specific bp numbers are given in Table 1. The functional part (of En) is shown in Fig. 6. (B)Diagram of the DNA-protein complex illustrating the TPNA binding and the TPND cutting event. This protein complex is described in Frey e t a / . (1990b). (After Peterson, 1986a. with permission.)

108

PETER A. PETERSON

(1990a,b), the transposase brings the termini of the element together (Figs. 5 and 14A), allowing the enzymes to commence excision. This is initiated by staggered nicks at the ends of the target-site duplication. As a consequence of these nicks, the 5' ends of the target-site duplication and the 5' end of the TIR are converted into complementary single-stranded fringes after repair synthesis (Fig. 13B). This is the reverse of the integration process. In integration, the mobile element also includes the same enzymatic complex (Fig. 14B) that has an affinity for the ends of the specific mobile element. The insertion event is accompanied by staggered nicks at the target site (Figs. 13B and C). The distance between the staggered nicks characterizes the TSD of each of the individual elements (see Table I). As the cut ends spread apart, the gaps are filled by repair synthesis, resulting in a duplication of the target site (Fig. 13C). It is possible that the templates are switching during repair synthesis (Fig. 13D). During the excision process, there is template switching combined with exonuclease degradation of the single-stranded nucleotides as shown in Fig. 13B. This template change leads to the alteration in the fidelity of the original sequence and the resulting replication of the template can lead to an excision product that is apparent. It is clear from these events that altered nucleotide sequences result from the excision process. Because these altered nucleotide sequences affect the reading frame of the gene, the alteration in the template will change the coding, resulting in altered proteins.

3. Footprints and Altered Products of Excision

,

If the excision process merely cut out the element leaving the target-site duplication intact, an examination of the nucleotide sequence of genes would uncover many instances of short duplicated segments such as ACT ACT in the En/ Spm system, or TACTAGGC TACTAGGC in the Ac system. The appearance of a series of these duplications would cause some interest. The role of transposable elements in contributing these duplicated segments would explain their presence. Yet the excision process is not wholly precise, and the footprints would be readily evident. If one makes a determined effort to isolate revertants and examines the junction site where the element left, changes in the sequence are found. This was done in several studies. In the first study, revertants were isolated from the Adhl locus (Sachs et al., 1983). Sachs et af. (1983) described the flanking sequences of the site of the Ds excision of four revertants and showed that although the duplicated sequences were retained, changes occurred at the junction of the duplicated segments (Fig. 15). In two revertants, RVI and RV2, the nucleotides show strand inversion. In two others, RV3 and RV4, the sequences showed 2-bp deletions in addition to the strand inversion. What surprised these investigators was that the qualitative level of gene action was not affected despite the addition of nucleo-

109

TRANSPOSABLE ELEMENTS IN MAIZE VARIABILITY

.

4

Element, gene and revertant

5' of element

Ds1 in Adhl-Fn335 TSD

S'GGGACTGA

RVl

GGGACTCTCGGACTGA

3' of element

<<

GGGACTGA 3'

References

Sachs e t ai.. 1 9 8 3

Dennis el a/.. 1 9 8 6

-1Sh-mti233 revertant C

I

Pohlman e t a/.. 1 9 8 4

CTTGTCC

Wildtype (Wxl

GTC AAG.TT

Spm-18 in Wx TSD

gtc aaG TT

I

I
.TTGTCCC

Weck er al.. 1 9 8 4

C AAC GCG

SchwarzSommer e t a/., 1985

G TTc aac gcg germinal revertants

wx- 1

gtc aaG TTA

. TTc aac gcg

wx-2

gtc aaG T.

CG TTc aac gcg

gtc aaG T.

ACG TTc aac gcg

arc aaG T T

G TTc aac Qca

wildtype *I AA chanaes

+ leu + ser

Somatic exc iions a1 in leaf DNA

I frameshift + leu

Figure 15. A compilation of revertants originating from different elements and at different loci identified. Boldface represents altered nucleotides; dots indicate missing nucleotides; capital letters represent the TSDs of different elements; lowercase letters refer to host sequences.

110

PETER A. PETERSON

tides relative to the progenitor sequence. In a later study with the same locus, Dennis et af. (1986) reported similar effects of excision, such as nucleotide substitutions (RV10, RV6) and nucleotide losses (RV31, RV43, RV46). There were 1-, 2-, or 3-bp alterations in the duplicated segments. It is noted that the assays used for determining the effects of protein differences might not have been adequate to detect slight differences. Schwarz-Sommer et al. (1985) found that most of the events were imprecise excisions leading to gene alterations at the junction of the excised IldSpm elements from the wx-m8 allele. They uncovered 16 revertants or revertant sectors and only 2 were of the precise sequences of wild type. The other 14 included assorted changes. Some were frameshift and, therefore, nonfunctional alleles. Others added amino acids to the proteins (leucine and serine) and others deleted amino acids. Excision of the Mu1 element generates deletions (Taylor and Walbot, 1985) with the excisions having 34-bp deletions or 5-bp additions (Britt and Walbot, 1991). Similar changes were reported in Antirrhinum with the excision of TAM1 (Martin et af., 1989; Sommer et al., 1988). Imprecise excisions are the key to the genetic variability question. Excisions out of gene sequences are a rich source of changes that have consequences on gene function and expression.

G. EFFECTSON GENEEXPRESSION 1. Phenotypic Changes It has been demonstrated that the footprints left by transposable element excisions lead to altered gene sequences, and it is appropriate to examine the kinds of changes that have been recognized at the phenotypic level. Some of the changes uncovered are novel and have not previously been seen. This is not unexpected because the excision event yields random kinds of changes. Phenotypes that have been uncovered include a wide range of color phenotypes of the aleurone color alleles, Wx-starch types that can be distinguishable by Ikl staining, and effects on the alcohol dehydrogenase gene. Changes have also been realized in the unstable Delphinium alleles originally described by Demerec (1935) and described molecularly with the Tam series in Antirrhinum (Sommer et al., 1988). Rhoades (1936, 1938) uncovered 29 full-colored mutations induced by Dt at the a1 -dr allele, and two different types could be distinguished among these. Of the 29, 27 were the standard A1 type, which was confirmed by the red-colored pericarp and would be considered identical to the original A1 allele. This is surprising because most result in different phenotypes. There is a bias, however,

TRANSPOSABLE ELEMENTS IN MAIZE VARIABILITY

111

because full-colored types were uncovered and others may not have been recognized. Further, definitive assays, such as sensitive enzyme tests, are possible with the now established A1 enzyme (Reddy et al., 1987). The remaining two exceptions of the 29 were interesting in their phenotype and what they showed. One of them gave a recessive brown pericarp color. That is, in the heterozygote Al-srlAl-br, the pericarp was red, but in the homozygote Al-brlAl-br, the pericarp was brown. This recessive brown pericarp of Al-br is in contrast to the dominant red of the A1 allele and dominant brown pericarp allele of Al-b. This particular colored derivative showing a recessive brown phenotype was a novel, previously undescribed allele, and it was designated Al-br. The second exception of the 29 full-colored exceptions was identified as A1 -rb; this was also a new allele. This allele produces a reddish-brown pericarp color in conjunction with the P allele, which is dominant to the brown of A1 but recessive to the red of A l . There was a nonfull color pale aleurone allele that was designated a1 -br. This gave a recessive brown pericarp color instead of the dominant brown produced by the previously isolated a1 -p allele originally described by Emerson and Anderson (1932). This range of phenotypes shown by the excisions out of al-dt illustrate the diverse expressions that can be uncovered with the Dt transposon. If the 27 fullcolored alleles were further examined in other kinds of tests, possibly other kinds of differences in phenotype would have been shown. They were never subjected to molecular analysis, and they could have been perfect excisions, which is possible because the a1 -dt studies have not included molecular analyses to any great extent. In McClintock’s last studies with the cl-ml and the cl-m2 alleles, numerous kinds of expressions from the mutability at those alleles were observed. What was readily seen was a wide range of quantitative expressions associated with pigmentation in the aleurone. These intermediate types were not full-color types and they were clearly distinguishable. The induced changes of the a1 -m allele yielded a large number of variable phenotypic types (Peterson, 1959, 1961). One can say that the Wx revertants from wx-ml were overproducers (McClintock, 1949). These examples illustrate a graded expression arising from these various alleles with transposon inserts that were transactivated by their regulatory elements. Again, the sequence changes caused several different forms, which provides a rich source of variability. 2. Enzyme and Molecular Changes

The first studies to examine possible enzymatic changes were with the Adh-1 gene. This was possible because this is one of the first genes isolated and available for biochemical analysis. Sachs et al. (1983) found reduced levels of mRNA

112

PETER A. PETERSON

in Northern blots of the Ds-induced Adh-1 -Fn 335 allele when compared with either the progenitor allele or the revertant. These authors speculated that these mutants reduced the mRNA levels because of the position of the insert near the start of the transcription or because the processing of the dehydrogenase message may be affected by the insert leading to an inefficiency of the normal transcript. The analysis reported by Sachs et al. (1983) was supported by the Osterman and Schwartz (1981) studies. They showed that the enzyme produced by this allele was more thermolabile than that produced with the progenitor allele. The modified protein from this allele must be a consequence of the effect of the Ds insert (the 1.5- kb Ds insert near the 5' of the Adh-1 gene). Other studies of the Adh-1 gene showed that variation occurred when the particular tissues or pollen were subjected to diverse tests. Osterman and Schwartz (1981) demonstrated that the derivative alleles are definitely altered. Alternative forms are also seen in the starch studies with the Wx locus in maize. With the wx-rn allele in the presence and absence of En, differences can be seen at the mRNA level (Gierl et al., 1985). In the section on footprints, the various isolated wx derivatives with altered amino acid content were described (Schwarz-Sommer er al., 1985). These wx derivatives were never subjected to tests at the biochemical level, but one can readily see that such changes are possible and could be realized phenotypically. Two products after Ds excision at the wx locus were found to have a range of enzyme activities (Wessler et al., 1986). These changes showed lower levels of enzyme activity, indicating changes in amino acid content of the protein. Numerous changes have been isolated from transposon changes in the anthocyanin color series in maize. The case with Dt at the A1 locus has been discussed. Another example is the regulatory gene C1. In the absence of C l action, color is absent in the maize kernel. The structural genes require the CI product that binds to structural gene promoters, thus activating transcription leading to corn kernel anthocyanin coloration. A number of alleles of the C1 locus have been described (Paz-Ares et al., 1987, 1990; Goff er al., 1991). One allele, Cl-S, conditions a more abundant message (mRNA), and it expresses a stronger potency against the effect of the inhibitor C1-I. This is obvious from the observation that the Cl-S allele is less inhibited by the standard Cl-1. How then does the C1-S differ from the C l ? When one examines Box I1 of the promoter (Fig. 16), a GCA sequence is changed to a TAG sequence. The possibility that this alteration is caused by a transposon is noted by a typical 5-bp transposon-induced footprint in the vicinity of this 3-bp change in Box I1 (Fig. 16). This change in sequence at the C l locus promoter results in the overexpression of the C1 allele. Interestingly, this same sequence in Box I1 of the promoter is also present in the C1-I (standard) allele. This overexpression is very likely the basis for the dominance of C1-1 in the heterozygote C1-IICl. The reason for this

TRANSPOSABLE ELEMENTS IN MAIZE VARIABILITY

R

113

BOX II

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

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

Cl lTfAGG TGCAG TGCAGG C1-I......lTfAGG lTAGG TGCAGG C1-S lTfAGG lTAGG TGCAGG

Figure 16. Box I1 in the promoter of the CI locus of three different alleles.

observation is that Cl-1 competes favorably with the C1 product for activation of transcripts of the structural genes in competitively binding to promoters of structural genes. The phenotype is therefore dominant colorless in the heterozygote, indicating the dominance of the Cl-1 allele. Because the truncated C1-I protein lacks a basic domain, transcription cannot proceed in the continuation of the transcript of the structural gene. Yet the transactivation site, either at A1 or at C2, is occupied and, therefore, excludes the C1 product from proper transcript activation. So, how does a transposon explain these differences? It is very likely, in view of the associated footprints, that a transposon-inducedchange caused the 3-bp promoter change that results in a more productive transcript. Further, the change occurred in a promoter of a regulatory gene that is at the beginning of the coloration development process. And this anthocyanin regulator and the downstream structural genes represent a model of the hierarchy of gene activity leading to most gene products, especially those governing quantitative traits. Another case of transposon-induced alteration in the promoter region was reported by Sommer ef al. (1988). These investigators found four solid-color revertants from the highly variegated niv-53 ::Taml mutant in which the Tam1 transposable element is integrated in the promoter region of the chalcone synthase gene. In four derived mutant lines, the Taml element was deleted together with the flanking nucleotides of the chalcone synthase promoter. In one case, the TATA box of the chalcone gene was removed, resulting in an extremely low expression of the gene, and causing gene transcription to initiate at a new position. Among the revertants, the amounts of chalcone synthase mRNA varied (with 100 being the wild type) from 107 to 2. Two of the mutants niv-531 (107) and niv-164 (103) were overexpressive in terms of the wild type. The other four were 2, 25, 35, and 26 (percentage chalcone synthase mRNA) for niv-5311, niv-5312, niv-5313, and niv-5314, respectively. Not only were the effects of the promoter change seen, but the analysis allowed Sommer el al. (1988) to explore the promoter parts necessary for proper promoter activity. The analysis of the Cl gene provides further information following a computer analysis of protein efficiency. Analysis of the various transposons randomly inserted at C l indicates that a broad diversity of available allelic possibilities can be generated. These are generated by transposable element excision products. In a study of a number of sites of the C l gene that had inserts, Franken ef al. (1992), identified sites with effects on protein performance. At these sites, each

114

PETER A. PETERSON

excision illustrated the kind of phenotypic effect that could be generated by this excision. This probably affected the three-dimensional formation of proteins and, consequently, their functional activation and efficiency (Branden and Tooze, 1991). It is evident in this brief survey of transposons that the effects of the excisions of transposons from genes could be on proteins and on the potency of expression of promoters. Though only two promoters were analyzed, it is clear that when other transposons are examined, similar changes will be seen.

H. PRESENCE IN THE MAIZEGENOME In view of the demonstration of transposon effects in the genome, could they be prevalent enough to cause these changes? First, it was emphasized that their presence in the genome could only be observed if there is a reporter allele to register its presence. This was tested by crossing a number of the reporter alleles to various cycles of the Iowa stiff stalk synthetic (BSSS) of maize. The results of these tests made evident the pervasiveness of the U q elements throughout various cycles of the improved BSSS populations (Peterson, 1986a; Peterson and Salamini, 1986; Cormack et al., 1988). Because 15 of the lines used to formulate BSSS were recovered (Stucker and Hallauer, 1992), these were examined, and la159 was found to have the Uq element (Cormack et al., 1988). The Uq element has persisted in BSSS during the course of improving this maize population. U q has been maintained since the initiation of this breeding population because BSSS is a closed population, meaning that no new germplasm has been introduced. Lamkey et al. (1991) examined this material for active Uq in 5 of the 11 completed cycles that have been of reciprocal recurrent selection [BSSS(R)(1I)]. In a parallel study, these same authors examined 10 of the 13 completed cycles of the half-sib and S2 progeny selection [BSSS(S)(l3)] for active Uq. Uq increased from 19% (BSSS) to 91% [BSSS(S)(l3)] at a linear rate after 13 cycles of half-sib and S2 progeny recurrent selection. This increase occurred without any cognizance of Uq because primary emphasis was on the yield performance of the advancing populations. This was not the case for the other population. Active Uq decreased from 15% (BSSS) to 0% [BSSS(R)(Cl l)] after 11 cycles of reciprocal recurrent selection. The extinction of the Uq element occurred between the fifth and sixth cycles of selection. Lamkey et al. (1991) concluded that random genetic drift played a key role in the increase and decrease of Uq in the two populations. A selective advantage of a region linked to Uq was also considered, and such a region, if it existed, would be a “bonanza” for plant breeders. How active is a transposon such as Uq in a breeding population? Uq is a

TRANSPOSABLE ELEMENTS IN MAIZE VARIABILITY

115

modified form of a nonactive Ac (Pisabarro er al., 1991). Whereas Ac can excise all Ds’s, Uq is limited to action only on Dsl (Caldwell and Peterson, 1992). When a reporter allele is crossed to various maize lines, spontaneous activation is observed in the form of sectors (Pan and Peterson, 1988). That these are quiescent U q that are spontaneously activated was verified by the isolation of spontaneous germinal originated Uq’s (Pan and Peterson, 1991a). In a further examination, these Uq’s were found to be clustered in a linkage group (Pan and Peterson, 1991b). One of these clustered Uq’s was later confirmed to be an Ac (P. A. Peterson, unpublished). Whether this Ac was a change from a previous Uq is not known, but given the relation between the two, this is probable. The evidence is plentiful that the Uq element is pervasive. Mrh is also pervasive in a number of maize breeding lines, even more so than Uq (Peterson and Salamini, 1986; Peterson. 1988). One significant feature relative to transposable element activity is that observations are confined to elements that have been genetically identified. Uq escaped detection until Sprague and McKinney (1966) began experiments with virus infection of maize that later led to the genetic isolation and identification of Uq (Friedemann and Peterson, 1982). The Mrh element that is also pervasive was not isolated and identified until 1982 (Rhoades and Dempsey, 1982). Other elements, whether they are or are not transposing, is not known, yet they are likely present in the genome. This includes the Cin elements that are present in a high frequency and are highly heterogeneous (Shepherd et al., 1982, 1984). Transposable element inserts are also prevalent in the maize progenitor, teosinte. Although their transposition activity has not been recorded in teosinte, these elements do have the characteristics of transposable elements (Blumberg et al., 1990; Shepherd et al., 1983).

I. NEOTENY AND REALIGNMENT OF RESOURCES FOR MORE EFFICIENT NETWORKS The plant breeder is well aware of the complexity of the properties associated with yield in plants. This should not be a surprise. From the perusal of any textbook on plant biochemistry (e.g., Goodwin and Mercer, 1988), it becomes evident that there are many processes involved in plant metabolism and growth. Further, pathways are often in parallel, leading from common substrates. This would suggest that a breeder is selecting the most efficient pathway that meets the needs of the program. In a discussion on quantitative inheritance, Peterson (1992) suggested that a plant breeder is selecting different “cassettes” and that these cassettes are most efficiently used as a unit by the plant. Though tropical and temperate maize have an overlap in metabolic usage, the “fine-tuning” of

116

PETER A. PETERSON

each of the maize lines is constrained by the cassettes that best support the growth potential of individual plant types. There is a further point. There must be a certain amount of “downgrading” in plant breeding improvement. The large tassel and robust leaves are not needed in most maize-growing environments. Genes must be bridled or reduced in activity in this downgrading process. Although nucleotide alterations induced by transposons can lead to improved protein performance, many of the changes lead to null alleles. In the wx-m8 case reported by Schwarz-Sommer el al. (1985), most of the derivative products following excision were altered or null products. By this process, altered pathways could develop that lead to more effective “networking” of limited resources. This should lead to a better performance among the selected lines of a breeding population.

J. SUMMARY OF EFFECTS Conclusive proof of the role of transposable elements in the induction of significant variability could come from the isolation of genes known to advance the performance of populations. Because these quantitatively important genes are evasive, this form of proof is remote. Yet all genes probably contribute to performance although conclusive proof on this statement is not readily forthcoming. Given that genes are proven to be altered by transposons, it is likely that the contribution of transposons to altered genes leading to altered proteins and, subsequently, variability is a clear possibility. As more genes are uncovered and maize research gets closer to isolating quantitative trait loci (Lander and Bodstein, 1989), it is hopeful that a clearer picture will emerge on the role of transposons in maize populations.

DEFINITIONS OF TERMS AND SYMBOLS Controlling elements Controlling element allele Controlling element systems Controlled allele

A collective term for transposable elements associated with mutable alleles that affect gene activity. An allele with an inserted controlling element at the locus. Includes two units, receptor and regulatory elements, that express a specific interaction (see Fig. 4C). An allele under the control of a controlling element system; includes an inserted receptor or regulatory element.

TRANSPOSABLE ELEMENTS IN MAIZE VARIABILITY

Receptor elements (rcp) Regulatory elements (Rg)

DS

Z 1-102

a2-ml

a-ml

a' a-m(r)

117

Elements such as dt, Ds,and I that suppress gene activity when in cis position to the locus. Elements such as Dt, Ac, F a , and En that alter or excise receptor elements so that the locus under control becomes functional. Dissociation; a receptor element. In combination with Ac, some states result in chromosome breakage at the site of Ds (McClintock, 1951). Inhibitor; a receptor element of the En system (Peterson, 1960). The insert, specific to the al-m(r) allele. A receptor allele at A2. Without En/Spm, it is full colored; with En/Spm, it is colorless with sectors of color from an inactivating En. (al-ml 5719 contains a 789-bp insert, located between the last two nucleotides of the second exon of the A1 gene. The gene expression of this allele is hypothesized to result from the presence of an intron donor that if used in mRNA splicing, permits formation of a modified exon.) This allele is a derivative of al-ml 6078. It is a recessive allele of A1 and responds to En. In the absence of En, the aleurone is pigmented pale; in its presence, colored spots are produced on a colorless background (Peterson, 1965). A neutral nonresponding allele. al-ml 5719; recessive allele of A and responds to En. In the absence of En, the aleurone is colorless; in its presence, colored spots are produced on a colorless background expressing the change of a1 to A 1 (Peterson, 1961).

Ac

Activator; a regulatory element (that triggers changes at

En

Enhancer; a regulatory element (that triggers changes a t UdSpm). Suppressor function of En/Spm. Can turn on or turn off activity of controlled loci; can be assayed only with certain responding alleles. Mutator function of En/Spm resulting in spots on kernels.

S

M

Ds).

118

Med

FCU

MP

Pr Pr

P”

PETER A. PETERSON

Mediator; required along with En to activate the c2-rn881058Y allele. Associated with a three-element system (Muszynski and Peterson, 1990). Factor Cuna; a regulatory element of the Feu system (Gonella and Peterson, 1977). Modulator; a transposable element that can behave as a regulatory element with Ds (Brink and Nilan, 1952). Purple aleurone with all color genes. Red aleurone with all color genes. The mutable pericarp locus. Associate with Mp in a cis condition. Transposability of Mp is expressed as changes from colorless to colored pericarp (Brink and Nilan, 1952).

wx844:En1 wx W X

wx wx

En at the wx locus. This allele gives purple starch with KI staining. The phenotype of this allele. This allele gives red or brown starch with KI staining. The phenotype of this allele.

REFERENCES Athma, P., Grotewold, E., and Peterson, T. (1992). Insertional mutagenesis of the maize P gene by intragenic transposition of Ac. Generics 131, 199-209. Avery, 0. T., Macleod, M., and McCarty, M. (1944). Studies on the chemical nature of the substance inducing transformation of pneumococcal types; Induction of transformation by a desoxyribonucleic acid fraction isolated from pneumococcus type 111. J . Exp. Med. 79, 137-158. Beadle, G. W. (1932). A gene for sticky chromosomes in Zea mays. Z . Indukr. Absramm. Vererbungsl. 63, 195-217. Bhattacharyya, M. K.. Smith, A. M., Ellis, T. H. N., Hedley, C., and Martin, C. (1990). The wrinkled-seed character of pea described by Mendel is caused by a transposon-like insertion in a gene encoding starch-branchingenzyme. Cell (Cambridge, Mass. j 60, 115- 122. Blumberg vel Spalve, J., Schwarz-Sommer, Zs., Saedler, H., and Peterson, P. A. (1990). The Cin2 and the Cin3 insertion elements of Zea mays ssp. parviglumis might serve as markers to monitor the evolution of Zea mays. Maydica 35, 151- 156. Branden, C., and Tooze, J. (1991). Introduction to protein structure. In “Molecular Biology in Three Dimensions,” p. 302. Garland, New York. Bridges, C. B. (1936). The bar ‘gene’ a duplication. Science 83,210-21 1. Brink, R. A,, and Nilan, R. A. (1952). The relation between light variegated and medium variegated pericarp in maize. Generics 37, 519-544.

TRANSPOSABLE ELEMENTS IN MAIZE VARIABILITY

119

Britt, A. B., and Wdlbot, V. (1991). Germinal and somatic products of Mu1 excision from the Bronze-] gene of Zea mays. Mol. Gen. Genet. 227, 267-276. Caldwell, E. E. 0.. and Peterson, P. A. (1992). The Ac and U9 transposable element systems in maize: Interactions among components. Genetics 131, 723-731. Castleberry, R. M., Crum. C. W., and Krull, C. F. (1984). Genetic yield improvement of U.S. maize cultivars under varying fertility and climatic environments. Crop Sci. 24, 33-36. Chang, R. Y . , and Peterson, P. A. (1993). Chromosome labeling with transposable elements in maize. Accepted for publication. Chen. J., Greenblatt. 1. M., and Dellaporta, S. L. (1987). Transposition of Ac from the P locus of maize into unreplicated chromosomal sites. Genetics 117, 109- 116. Chomet, P., Lisch, D., Hardeman, K. J., Chandler, V. L., and Freeling, M. (1991). Identification of a regulatory transposon that controls the mutator transposable element system in maize. Generics 129, 261 -270. Cormack, J. €3.. Cox, D. F., and Peterson, P. A. (1988). Presence of the transposable element 179 in maize breeding material. Crop Sci. 28, 941 -944. Darwin, C. (1868). “The Variation of Animals and Plants under Domestication,” Vol. I . John Murray, London. Dash, S., and Peterson. P. A. (1989). Chromosome constructs for transposon tagging of desirable genes in different parts of the maize genome. Maydica 34, 247-261. Dash, S.. and Peterson. P. A. (1993). Frequent loss of the En transposable element following excision and its relation to chromosome replication. Genetics (submitted for publication). Demerec, M. (1935). Unstable genes. Bot. Rev. 78,233-248. Demerec. M. (1946). Annual report of the Director of the Department of Genetics. Year BookCarnegie Inst. Washington 45, 139- 192. Demerec, M. (1947). Annual report of the Director of the Department of Genetics. Year BookCartiegie Inst. Washington 46, 123- 170. Dennis, E. S . , Gerlach. W. L., and Peacock, W. J. (1986). Excision of the Ds controlling element from the Adhl gene of maize. Maydica 31,47-57. Duvick, D. N. (1977). Genetic rates of gain in hybrid maize yields during the past 40 years. Maydica 22, 187-196. Duvick, D. N. (1984). Progress in conventional plant breeding In “Gene Manipulation in Plant Breeding” (J. P. Gustafson, ed.), pp. 17-31. Plenum, New York. Duvick, D. N. (1986). Genetic contributions to yield gains of U.S. hybrid maize, 1930 to 1980. In “Genetic Contributions to Yield Gains of Five Major Crop Plants” (W. Fehr, ed.), CSSA Spec. Publ. No. 7, pp. 15-47. Crop Sci. Soc.Am., Madison, WI. Duvick. D. N. (1991). Genetic contributions to advances in yield of US. maize. Maydica 37, 69-79. El-Lakanyu, M. A., and Russell, W. A. (1971). Relationship of maize characters with yield in testcrosses of inbreds at different plant densities. Crop Sci. 11,698-701. Emerson, R. A. (1914). The inheritance of a recurring somatic variation in variegated ears of maize. Am. Nat. 48,87- 115. Emerson, R. A. (1918). A fifth pair of factors, Aa, for aleurone colour in maize and its relation to the Cc and Rr pairs. Mem.-N.Y. Agric. Exp. Stan. (Irhaca) 16, 225-289. Emerson, R. A. (1929).The frequency of somatic mutation in variegated pericarp of maize. Genetics 14,488-51 1 . Emerson, R. A., and Anderson, E. G.(1932). The A series of allelomorphs in relation to pigmentation in maize. Genetics 17, 503-509. Fedoroff, N., Wessler. S . , and Shure, M. (1983). Isolation of the transposable maize controlling elements Ac and Ds. Cell (Cambridge. Mass.) 3S, 235-242.

120

PETER A. PETERSON

Franken, P., Kartzke, S.,Peterson, P. A., Saedler, H.. and Wienand, U. (1992). Analysis of functional domains of the CI encoded protein. Maize Gen. Coop. Newsl. 66, 36. Frey, M., Grant, S., Lutticke, S., Menssen, A., Reinecke, J., Trentmann, S., Peterson, P. A., Saedler, H., and Gierl, A. (1990a). Structure and function of the EnlSpm transposable element of Zen mays. In “Plant Molecular Biology,” p. 67. Schloss Elmau, Bavaria, Germany. Frey, M., Reinecke, J., Grant, S., Saedler, H., and Gierl, A. (1990b). Excision of the EnlSpm transposable element of Zea mays requires two element-encoded proteins. EMEO J. 9, 40374044. Friedemann, P., and Peterson, P. A. (1982). The Uq controlling-element system in maize. Mol. Gen. Genet. 187, 19-29. Geadelmann, J. L., and Peterson, R. H. (1978). Effects of two yield component selection procedures on maize. Crop Sci. 18, 396-398. Gierl, A., Schwarz-Sommer, Zs., and Saedler, H. (1985). Molecular interactions between the components of the En-1 transposable element system of Zea mays. EMBO J . 4(3), 579-583. Gierl, A., Cuypers, H., Lutticke, S., Pereira, A., Schwarz-Sommer,Zs., Dash, S., Peterson, P. A., and Saedler, H. (1988a). Structure and function of the EnlSpm transposable element system of Zea mays: Identification of the suppressorcomponent of En. In “Plant Transposable Elements” (0.Nelson, ed.), pp. 115- 119. Plenum, New York. Gierl, A., Lutticke, S., and Saedler, H. (1988b). TnpA product encoded by the transposable element En-] of Zea mays is a DNA binding protein. EMBO J . 7,4045-4053. Gill, K. S., and Gill, B. S. (1991). A DNA fragment mapped within the submicroscopicdeletion of phi, a chromosome pairing regulator gene in polyploid wheat. Generics 129, 257-259. Goff, S. A., Cone, K. C., and Fromm, M.E. (1991). Identification of functional domains in the maize transcriptional activator CI: Comparison of wildtype and dominant inhibitor proteins. Genes Dev. 5,298-309. Goldschmidt, R. B. (1955). “Theoretical Genetics.” Univ. of California Press, Berkeley and Los Angeles. Gonella, J. A., and Peterson, P. A. (1977). Controlling elements in a tribal maize from Colombia: FCU, a two unit system. Genetics 85, 629-645. Gonella, J. A,, and Peterson, P. A. (1978). The FCU controlling-elementsystem in maize 11. On the possible heterogeneity of controlling elements. Mol. Gen. Genet. 167, 29-36. Goodman, M. M. (1985). Exotic maize germplasm: Status, prospects, and remedies. Iowa State J . Res. 59,497-527. Goodman, M. M. (1990). Genetic and germplasm stocks worth conserving. 1. Hered. 81, 11 - 16. Goodman, M. M., and Stuber, C. W. (1983a). Races of maize. VI. Isozyme variation among races of maize in Bolivia. Maydica 28, 169-187. Goodman, M. M., and Stuber, C. W. (1983b). Maize. I n “Isozymes in Plant Genetics and Breeding” (S. D. Tanksley and T. J. Orton, eds.), Part B, pp. 1-33. Elsevier, Amsterdam. Goodwin, T. W., and Mercer, E. I. (1988). “Introduction to Plant Biochemistry,” 2nd ed. Pergamon, Oxford. Green, M. M., and Green, K. C. (1949). Crossing over between alleles at the lozenge locus in Drosophila melanogaster. Proc. Natl. A d . Sci. U.S.A. 35, 586-59 1. Greenblatt, I. M. (1966). Transposition and replication of modulator in maize. Generics 53, 361-369. Greenblatt, I. M. (1968). The mechanism of modulator transposition in maize. Generics 58, 585-597. Greenblatt, 1. M. (1974). Movement of modulator in maize: A test of an hypothesis. Genetics 77, 67 1-678. Greenblatt, I. M. (1984). A chromosome replication pattern deduced from pericarp phenotypes re-

TRANSPOSABLE ELEMENTS IN MAIZE VARIABILITY

121

sulting from movements of the transposable element, modulator in maize. Genetics 108, 471 -485. Greenblatt, I. M., and Brink, R. A. (1962). Twin mutations in medium variegated pericarp maize. Genetics 47,489-501. Greenblatt, I., and Brink, R. A. (1963). Transposition of modulator in maize into divided and undivided chromosome segments. Nature (London) 197,412-413. Groose, R. W., Weigelt, H. D., and Palmer, R. G. (1988). Somatic analysis of an unstable mutation for anthocyanin pigmentation in soybean. J . Hered. 79, 263-267. Hallauer, A. R. (1990a). Methods used in developing maize inbreds. Maydica 35, I - 16. Hallauer, A. R. (1990b). Germplasm sources and breeding strategies for line development in the 1990s. Proc. 45th Annu. Corn Sorghum Ind. Res. Conf., pp. 64-79. Hallauer. A. R. (1992). Recurrent selection in maize. In “Offprints from Plant Breeding Reviews” (1. Janick, ed.), pp. 115- 179. Wiley, New York. Hartings, H., Spilmont, C., Lazzaroni, N., Rossi, V., Salamini, F., Thompson, R. D., and Motto, M. (1991). Molecular analysis of the Bg-rbg transposable element system of Zea mays L. M o f . Gen. Genet. 227, 91 -96. Hershey, A. D.. and Chase, M. (1952). Independent functions of viral protein and nucleic acid in growth of bacteriophage. J. Gen. Physiol. 36, 39-56. Jefferson, R. A., Kavanagh, T. A., and Bevan, M. W. (1987). GUS fusions: Glucuronidase as a sensitive and versatile gene fusion marker in higher plants. EMBO J . 6, 3901-3907. Knight, T. A. (1808). On the variegation of plants. Trans. Linn. SOC.London 9, 268-271. Lamkey, K. R., Peterson, P. A., and Hallauer, A. R. (1991). Frequency of the transposable element Uq in Iowa Stiff Stalk synthetic maize populations. Genet. Res. 57, 1-9. Lander, E. S.. and Bodstein, D. (1989). Mapping Mendelian factors underlying quantitative traits using RFLP linkage maps. Genetics 121, 185-199. Laughnan, J. R. (1949). The action of allelic forms of the gene A in maize. 11. The relation of crossing over to mutation of Ab. Proc. Natl. Acad. Sci. U.S.A. 35(4), 167- 178. Lecoq, H. (1862). “De La fecondation naturelle et artificielle des vtgttaux et de I’hybridization,” 2nd ed. Librairie Agricole de la Maison Rustique, Paris. Lewis, E. B. (1945). The relation of repeats to position effect in Drosophila mehogaster. Genetics 30, 137-166. Martin, C., Prescott, A., Lister, C., and MacKay, S . (1989). Activity of the transposon Tam3 in Antirrhinum and tobacco: Possible role of DNA methylation. EMBO J. 8, 997- 1004. Masson, P., Banks, J., Surosky, R., Kingsbury, J., and Fedoroff, N. (1987). Structure and regulation of the maize suppressor-mutator transposable element. In “Plant Molecular Biology” (D. V. Wettstein and N. H. Chua, eds.), pp. 589-597. Plenum, New York. Masson, P., Banks, J. A,, and Fedoroff, N. (1991a). Structure and function of the maize Spm transposable element. Biochimie 73, 5-8. Masson, P.. Strem, M.. and Fedoroff, N. (1991b). The tnpA and tnpD gene products of the Spm element are required for transposition in tobacco. Pfant Cell 3, 73-85. McClintock. B. (1929). Chromosome morphology in Zea mays. Science 69,629. McClintock, B. ( I 93 la). Cytological observations of deficiencies involving known genes, translocations and an inversion in Zea mays. Res. Bull. -Mo., Agric. Exp. Sin. 163, 1-30. McClintock. B. (1931b). The order of the genes C . Sh and Wx in Zea mays with reference to a cytologically known point in the chromosome. Proc. Natl. Acad. Sci. U.S.A. 17,485-497. McClintock, B. (1932). A correlation of ring-shaped chromosomes with variegation in Zea mays. Proc. Nail. Acad. Sci. U.S.A. 18, 677-681. McClintock, B. (1942). The fusion of broken ends of chromosomes following nuclear fusion. Proc. Natl. Acad. Sci. U.S.A. 28,458-463.

122

PETER A. PETERSON

McClintock. B. (1944). The relation of homozygous deficiencies to mutations and allelic series in maize. Generics 29,478-502. McClintock, B . (1945). Cytogenetic studies of maize and neurospora. Year Book-Carnegie Insi. Washington 44, 108- 112. McClintock, B. (1947). Cytogenetic studies of maize and neurospora. Year Book-Carnegie Insi. Washington 46, 146- 152. McClintock, B. (1948). Mutable loci in maize. Year Book-Curnegie Inst. Washington 47, 155169. McClintock, B. (1949). Mutable loci in maize. Year Book-Carnegie Inst. Washington 48, 142154. McClintock. 8 . (195 I). Chromosome organization and genic expression. Cold Spring Harbor Symp. Quani. Biol. 16, 13-47. McClintock, B. (1961). Further studies of the suppressor-mutator system of control of gene action in maize. Year Book-Carnegie h i . Washington 60,469-476. Muszynski, M. G., and Peterson, P. A. (1990). c2-m884259Y-A mutable allele under two-factor control. Maize Genet. Coop. Newsl. 64, 9. Nowick, E. M., and Peterson, P. A. (1981). Transposition of the enhancer controlling element system in maize. Mol. Gen. Genet. 183,440-448. Osterman, J. C., and Schwartz, D. (1981). Analysis of a controlling-element mutation at the Adh locus of maize. Genetics 99, 267-273. Pan, Y.-B., and Peterson, P. A. (1988). Spontaneous activation of quiescent Uq transposable elements during endosperm development in Zea mays. Generics 119,457-464. Pan, Y.-B., and Peterson, P. A. (1991a). Spontaneous germinal activation of quiescent Uq transposable elements in Zea mays. Genetics 128, 823-830. Pan, Y.-B., and Peterson, P. A. (1991b). Newly activated germinal Uq elements in maize are clustered on one linkage group independently of the standard Uq element. Mol. Gen. Genet. 229, I6 I - 174. Paz-Ares, J., Ghosal, D., Wienand, U., Peterson, P. A., and Saedler, H. (1987). The regulatory C I locus of Zea mays encodes a protein with homology of myb proto-oncogne products and with structural similarities to transcriptional activators. EMBO J. 6, 3553-3558. Paz-Ares, J., Ghosal, D., and Saedler H. (1990). Molecular analysis of the CI-I allele from Zea mays: A dominant mutant of the regulatory Cl locus. EMBO J. 9,315-321. Pereira, A., Cuypers, H.,Gierl, A., Schwarz-Sommer, Zs., and Saedler, H. (1986). Molecular analysis of the EnlSpm transposable element system of Zea mays. EMBO J . 5, 853-841. Peterson, P. A. (1953). A mutable pale green locus in maize. Generics 38, 682-683. Peterson, P. A. (1959). Pale allele. Maize Genet. Coop. Newsl. 33, 61. Peterson, P. A. (1960). The pale green mutable system in maize. Geneiics 45, 115-133. Peterson, P. A. (1961). Mutable a1 of the En system in maize. Generics 46, 759-771. Peterson, P. A. (1965). A relationship between the Spm and En control systems in maize. Am. Nut. 99, 391-398. Peterson, P. A. (1970). Controlling elements and mutable loci in maize: Their relationship to bacterial episomes. Geneiica 41, 33-56. Peterson, P. A. (1978). Controlling elements: The induction of mutability at the A2 and C loci in maize. In “Maize Breeding and Genetics” (D. B. Walden, ed.), pp. 601-635. Wiley, New York. Peterson, P. A. (1986a). Mobile elements in maize: A force in evolutionary and plant breeding processes. In “Genetics, Development, and Evolution” (J. P. Gustafson, G. L. Stebbins, and F. J. Ayala, eds.), pp. 47-78. Plenum, New York. Peterson, P. A. (1986b). Mobile elements in maize. PlaniBreed. Rev. 4, 81- 122. Peterson, P. A. (1987). Mobile elements in plants. CRC Crit. Rev. Plani Sci. 6, 105-208.

TRANSPOSABLEELEMENTS IN MAIZE VARIABILITY

123

Peterson, P. A. (1988). Transposons in maize and their role in corn breeding progress. Proc. 43rd Annu. Corn Sorghum Ind. Res. Conf., pp. 5 1-71, Peterson, P. A. (1992). Quantitative inheritance in the era of molecular biology. Maydica 37, 7- 18. Peterson, P. A., and Salamini, F. (1986). Distribution of active transposable elements among important corn breeding populations. Maydica 31, 163- 172. Peterson, P. A., and Weber, C. R. (1969). An unstable locus in soybeans. Theor. Appl. Genet. 39, 156- 162. Pisabarro, A. G . , Martin, W. F., Peterson, P. A,, Saedler, H., and Gierl, A. (1991). Molecular analysis of the ubiquitous (Uq) transposable element system of Zen mays. Mol. Gen. Genet. 230,201-208. Pohlman, R. F., Fedoroff, N. V., and Messing, J. (1984). The nucleotide sequence of the maize controlling element activator. Cell (Cambridge, Mass.) 37, 635-643. Prior, C. L . , and Russell, R. A. (1975). Yield performance of nonprolific maize hybrids at six plant densities. Crop Sci. 15,482-486. Reddy, A. R.. Britsch, L., Salamini, F., Saedler, H., andRohde, W. (1987). The A1 (Anthocyanin-1) locus in Zea mays encodes dihydroquercetin reductase. Plant Sci. 52, 7- 14. Reddy. V. D.. and Reddy, G. M. (1992). Genetic instability in rice. J. Genet Breed. 46, 247-252. Rhoades, M. M. (1936). The effect of varying gene dosage on aleurone colour in maize. J. Genet. 33, 347-354. Rhoades, M. M. (1938). Effect of the Dt gene on the mutability of the a1 allele in maize. Genetics 23,377-395. Rhoades, M. M., and Dempsey, E. (1982). The induction of mutable systems in plants with the high-loss mechanism. Maize Genet. Coop. Newsl. 56, 21 -26. Russell, W. A. (1984). Agronomic performance of maize cultivars representing different eras of breeding. Maydica 29, 375-390. Russell, W. A. (1991). Genetic improvement of maize yields. Adv. Agron. 46, 245-298. Sachs, M. M., Peacock, W. J., Dennis, E. S., and Gerlach. W. L. (1983). Maize Ac/Ds controlling elements: A molecular viewpoint. Maydica 28, 289-302. Saedler, H . , and Nevers, P. (1985). Transposition in plants: A molecular model. EMEOJ. 4, 585590.

Schwarz-Sommer, Zs., Gierl, A,, Klosgen, R. B., Wienand, U., Peterson, P. A,, and Saedler. H. (1984). The Spm(En) transposable element controls the excision of a 2-kb DNA insert at the wxm-8 allele of Zea mays. EMBO J . 3, 1021 - 1028. Schwarz-Sommer. Zs., Gierl, A , , Cuypers, H., Peterson, P. A., and Saedler, H. (1985). Plant transposable elements generate the DNA sequence diversity needed in evolution. EMBO J . 4, 591-597. Shepherd, N. S., Schwarz-Sommer, Zs., Wienand, U., Sommer, H., Deumling, B., Peterson, P. A . , and Saedler. H. (1982). Cloning of a genomic fragment carrying the insertion element Cinl of Zea mays. Mol. Gen. Genet. 188, 266-271. Shepherd, N. S . , Deumling. B., Wienand. U.. Blumberg, J., and Saedler, H. (1983). The Teol DNA insert of teosinte guerrero. Maize Gener. Coop. Newl. 57, 159. Shepherd, N. S . , Schwarz-Sommer, Zs., Blumberg vel Spalve, J., Gupta, M.. Wienand, U., and Saedler, H. (1984). Similarity of the Cinl repetitive family of Zea mays to eukaryotic transposable elements. Nature (London) 307, 185-187. Sommer, H . , Bonas. U., and Saedler, H. (1988). Transposon-induced alterations in the promoter region affect transcription of the chalcone synthase gene of Antirrhinum majus. Mol. Gen. Genet. 211,49-55. Sprague. G. F., and McKinney, H. H. (1966). Aberrant ratio: An anomaly in maize associated with virus infection. Genetics 54, 1287- 1296.

124

PETER A. PETERSON

Stucker, D. S., and Hallauer, A. R. (1992). Genetic variability as affected by selection in Iowa Stiff Stalk synthetic maize. J . Hered. 83, 410-418. Taylor, L. P., and Walbot, V. (1985). A deletion adjacent to the maize transposable element Mu-1 accompanies loss of Adhl expression. EMEO J . 4, 869-876. Watson, 1. D., and Crick, F. H.C. (1953). Molecular structure of nucleic acids. Nurure (London) 171,737-738.

Wessler, S . R., and Varagona, M. J. (1985). Molecular basis of mutations at the w q locus of maize: Correlation with the fine structure genetic map. Proc. Narl. Acud. Sci. U.S.A. 82, 4177-4181.

Wessler. S . R., Baran, G.,Varagona, M., and Dellaporta, S. L. (1986). Excision of Ds produces waxy proteins with a range of enzymatic activities. EMEO J. 5, 2427-2432. Wijsman, H. J. W. (1986). Evidence for transposition in petunia. Theor. Appl. Gener. 71,791 -796.

CONCEPTS AND DIRECTIONS IN ARTHROPODPESTMANAGEMENT Joe Funderburk,' Leon Higley,z and G. David Buntin3 I

North Florida Research and Education Center University of Florida Quincy, Florida 3 2 3 5 1 Department of Entomology University of Nebraska-Lincoln Lincoln, Nebraska 68583 'Department of Entomology University of Georgia Griffin, Georgia 30223

I. Introduction A. Pesticide Crisis B. Objectives of Integrated Pest Management 11. Concepts of Integrated Pest Management A. Pests and Pest Status B. Philosophies for Preventing Pest Damage C. Issues in Pest Management 111. Components of Integrated Pest Management A. Crop Tolerance to Injury B. Pest Management Decision Making C. Sampling and Monitoring D. Management Actions IV. Ecological and Environmental Considerations A. Population Regulation B. Selection Pressures C. Impacts on Ecosystems V. Status and Future Directions of Integrated Pest Management A. Current Status of Integrated Pest Management B. Improving Efficiency C. Employing More Integration of Tactics VI. Conclusions References

125 Adunncrs in Agwmmy. Yolumr 51

Copyright 0 1993 by Academic Press,Inc. AII rights of reproduction in any form reserved.

126 JOE FUNDERBURK, LEON HIGLEY, AND G. DAVID BUNTIN

I. INTRODUCTION A. PESTICIDE CRISIS Early insecticides were not very effective, and pest control relied on approaches that combined their use with environmental manipulations such as crop sanitation and timing of planting and harvesting. Highly effective, inexpensive chemical insecticides became widely available shortly after World War 11. Purely insecticidal approaches rapidly were adopted to solve pest problems. Entomologists became preoccupied with identifying pests and finding the most effective chemical pesticides for each pest species. The practice of spraying at regular intervals without regard to pest density and damage potential was widely implemented. Two decades of such indiscriminate pesticide use resulted in an ecological disaster. Many target and nontarget pests became genetically resistant, and some pesticides lost their effectiveness. Contamination of soil and water by long-lived pesticides affected wildlife populations, with some species nearing extinction. Public concerns were aroused over potential health risks in food. As a result of this disaster, stringent laws were passed in the United States regarding pesticide use, and the Environmental Protection Agency (EPA) was established as a regulatory agency to ensure that pesticides do not contaminate the environment or pose an unnecessary health risk. An ecologicallycompatible methodology of pest control was developed called integrated pest management (IPM).

B. OBJECTIVES OF INTEGRATED PESTMANAGEMENT The original IPM concept focused on pest populations and embraced integrating insecticide use with natural enemies (Stem et al., 1959). The system advocated conservation of natural enemies through reduction and selective use of insecticides. Integration of management tactics soon expanded to include plant resistance and cultural management (Geier, 1966). The modern IPM concept incorporates the combined public and private interests related to agricultural production, the environment, and food supply and quality. The objective of IPM, as described by Huffaker and Smith (1980), is “the development of improved, ecologically oriented pest management systems that optimize, on a long-term basis, costs and benefits of crop protection.” Fundamental concepts of ecology and economics are encompassed into IPM theory. Programs are developed from information on pest ecology, crop response to pests, and economics of management and production. Important considerations in IPM theory and practice are the economic sustainability of the agricultural production system, sustainability of management tactics and management programs, and efficient utilization of natural resources. Since the pesticide crisis, industry has been forced to develop new and safer

ARTHROPOD PEST MANAGEMENT

127

pesticides. Available scientific evidence indicates that the food supply produced with these new pesticides is safe (Beall et al., 1991). The development and registration of new pesticides is an elaborate and expensive business, with private costs currently exceeding U.S. $65 million [National Academy of Sciences (NAS), 19861. Considerable public monies are spent developing and evaluating the efficacy and safety of insecticides. Insecticides, when used appropriately, serve both private and public interests. Crop losses due to arthropods, even with the addition of other means of management, would be great without insecticides (e.g., Pimentel, 1986). Elimination of pesticide use would increase the land area needed to produce food, leading to destruction of natural habitat and potentially depleting soil, water, and energy resources. The economics and economic sustainability of agricultural production systems also would be affected (Zilberman el al., 1991). Insecticidal management tactics that are sustainable and benign to the environment have been successfully incorporated into many IPM programs. However, development of IPM programs has been utility driven, and abuse of pesticides by ignoring fundamental ecological and economic concepts frequently has resulted in practice. Although problems with pesticides are well known, other categories of management tactics can be environmentallydisruptive and misused. Pest populations can adapt genetically to overcome resistant crop cultivars, cultural management tactics, and even biological management tactics. Individual management tactics of all types are sustained in pest management programs by integrating their use with other tactics. No tactic or approach, including IPM, can eliminate the impact of pests on agricultural systems. However, IPM continues to gain in acceptance as the optimal approach to provide an economically and environmentally sound method for reducing the deleterious effects of pests. Successful IPM programs have been developed for crops in both temperate and tropical climates. Benefits are widely recognized in Europe, the United States, and other industrialized countries and are gaining in recognition in developing countries (Goodell, 1984). In this chapter, we present an overview of IPM contrasted with other philosophies of pest control. The IPM concept was developed to provide sustainable solutions to pest management problems, and particular attention will be given to misuses of management tactics and to development of sustainable solutions to pest management problems. We identify problem situations and evaluate the status and future direction of IPM.

11. CONCEPTS OF INTEGRATED PEST MANAGEMENT A. PESTSAND PESTSTATUS All approaches for handling pests begin with the concept of a pest. The designation of a species as a pest is entirely anthropocentric and does not directly

128 JOE FUNDERBURK, LEON HIGLEY, AND G. DAVID BUNTIN relate to any intrinsic characteristic of a species. Pests are those species that humans identify as impairing human welfare, property, or activities (Pedigo, 1989). Usually, the designation of a species as a pest and an assessment of its importance reflect both what harm individual pests do and how many of the pests are present. Thus, a consideration of insect activities and insect densities (numbers per unit area) are important in recognizing insect pest species. Because insect numbers are subject to change (and also insect activities, though to a much lesser extent), a species may be regarded as a pest under one set of circumstances and not under another. A more formal system for recognizing pest species and their importance involves the idea of pest status. Pest status is a ranking of a pest species relative to the potential economic impact of that species (Pedigo, 1989; Stern et al., 1959). In a seminal paper, Stern et al. (1959) defined both pest status and an array of other terms and concepts that form the foundation of modern pest management. They proposed different rankings of pests, as severe, perennial, occasional, and subeconomic, based on a hypothetical average population level of a species (termed the general equilibrium position) and its relationship to the economic injury level (EIL), a cost-benefit criterion for evaluating the impact of pests. Perennial and severe pests cause the most serious and difficult problems for pest management and are often referred to as key pests (Pedigo, 1989). We discuss the EIL in more detail later; at this point, the key issue to highlight is that Stern et al. (1959) placed identification of pest status on a formal basis through an evaluation of economic impact of pest species. The EIL indicates where the economic costs of pest activities equal the economic benefits of taking action against pests. The concept of the EIL is important because it provides a mechanism for evaluating pest impact and implies that not all levels of pests require management. In establishing an economic and ecological rationale for addressing pest problems, Stern et al. (1959) provided the framework for a new paradigm for pest control. Although their failure to completely define certain concepts delayed the application of many of their ideas for more than a decade, these omissions do not diminish the impact of their work. Many workers built upon their ideas, ultimately developing modern concepts of pest management. Unfortunately, those concepts are not universally accepted.

B.

PHILOSOPHIES FOR PREVENTING PEST DAMAGE

Pest management differs from previous approaches to pest control because it focuses on tolerating pests. Through the EIL, tolerance is expressed in economic terms based on economic impacts from pest activities and economic costs associated with management action. As a fundamental objective, pest management programs seek to reduce the impact of pests to tolerable levels, not to the lowest levels obtainable.

ARTHROPOD PEST MANAGEMENT

129

Perkins (1982), in his important history of pest management, argues that the postinsecticide era was dominated by conflicting philosophies of total population management, an approach calling for areawide or regionwide pest population suppression, and pest management, an approach focusing on tolerating pest populations. Perkins’ (1982) total population management refers to efforts by E. F. Knipling and other scientists to regulate or eradicate pest populations. However, we believe the underlying philosophical conflict between pest management and alternatives is more fundamental than those that Perkins (1982) highlights. Specifically, there are essential differences between what we identify as management and control philosophies. The control philosophy represents a viewpoint that ignores or minimizes the importance of tolerating pests and looks instead to purposeful regulation of pest populations. Some control approaches may consider economics in taking action, particularly in emphasizing economic risks from pest populations. Others may not. Control is based on the premise that we can and should regulate pest populations to whatever level is desirable, typically a very low level. At its extreme, the control philosophy is expressed in pest eradication efforts. In contrast, the management philosophy focuses on reducing pest impacts while tolerating some level of pests. Because tolerating pests is a fundamental issue for management, so is economics, as it directly relates to what is tolerable. Although management programs may be directed to reduce pest population levels, efforts are not directed toward permanently regulating pest populations through continued human intervention. In choosing between these philosophies, it is important to recognize that management is based on ecological principles, whereas most control is not. Successful eradication of the screwworm [Cochliomyia hominivorax (Coquerel)] from North America was based on ecological realities and represents the outstanding success in pest control (Knipling, 1979). Unfortunately, the more widespread legacy of control has been rampant pest resistance to pesticides resulting from pesticide overuse. Some applications of a control approach are appropriate, even in the context of pest management programs. For example, control efforts to eradicate an introduction of a new pest species are appropriate. However, such applications for control are very limited. Although pest management has become the predominant approach for handling pest problems, the conflict between control and management philosophies does persist. Many practioners continue to use pest control approaches. Similarly, much more research is directed toward developing pest control tactics than fundamental pest management tools. Control approaches often seem to offer magic bullets for pest problems, but it is a false promise given that their use is not sustainable. Unfortunately, the appeal of simple solutions to pest problems that control seems to promise is difficult to counter. For example, deployment of bioengineered genes for plant resistance to pests presently is based on notions of control rather than proper pest management, with the virtually certain prospect

130 JOE FUNDERBURK, LEON HIGLEY, AND G. DAVID BUNTIN

that such deployment approaches will encourage pest resistance (Funderburk and Higley, 1994).

C. ISSUES IN PESTMANAGEMENT Pest management has three broad objectives: to maintain profitability when managing pests by taking action only if economically justified; to minimize selection pressure on pest populations from management tactics; and to maintain environmental quality, specifically by minimizing the impact of management tactics on the environment. Historically, pest management programs have emphasized the first of these objectives, ensuring that management actions are economically justified. More recently, other criteria, such as environmental quality and reducing resistance, are receiving increased emphasis. Although formal mechanisms exist for examining the economic justification of a management action, corresponding guidelines for considering environmental safety and minimizing selection pressure have not been as forthcoming. Nevertheless, by calling for management actions only when economically justified, pest management programs have greatly reduced the use of pesticides, thereby improving environmental safety and diminishing the prospect of pest resistance development. Only recently has it been recognized that the pests themselves are not of concern, but rather the impact of their activities, and the latter is now receiving sufficient attention with regard to developing management approaches. Control approaches exclusively focus on pest populations; pest management approaches also emphasize pest population management, but not exclusively. The key problem with population approaches is that pest populations frequently evolve to circumvent artificial management pressures. Pesticide resistance is the classic example of this phenomenon. Pest management need not, and should not, be taken to mean pest population management, although this error often is made. The actual guiding issue is to minimize plant stress from pest activities. Most often this will involve efforts to change pest population levels, but other approaches focused on plant stress management are possible and are beginning to be explored.

III. COMPONENTS OF INTEGRATED PEST MANAGEMENT A. CROPTOLERANCE TO INJURY Because tolerating pests is such an essential concept of pest management, it is important to define tolerance. In using the EIL to define what is tolerable, there

ARTHROPOD PEST MANAGEMENT

131

is a strong emphasis on economic criteria. What is somewhat less obvious, but equally important, is the biological basis for tolerating pests-specifically, the relationship between pest activities and yield losses. These relationships vary based on a host of factors, including, but not limited to, pest densities. On a simple level, where the action of pests has little individual impact on marketable yield, more pests can be tolerated. Complexity arises as we try to quantify this relationship between pest numbers and yield and as we examine possible approaches for minimizing the impact of pests on yield. The nature of pest injury and physiological responses of crops to injury set the boundaries within which management actions are possible. Consequently, understanding the physiological responses of plants to pest injury is an essential undertaking for developing pest management programs. These responses to pest injury should be considered in the context of plant stress. Fundamentally, pest management is one approach for avoiding or reducing the impact of biotic stress; it is a type of plant stress management. Although most emphasis on plant stress has focused on abiotic stressors, insects and other biotic agents are equally important as possible plant stressors. Unfortunately, plant stress from most biotic agents is poorly understood. Because plants integrate the effects of all types of stressors, predicting the impact of those stressors and developing approaches to mitigate that impact require comparable integration in our understanding of stress. Consequently, finding unifying concepts and terminology for examining plant stress is important in building a more comprehensive understanding of stress. Higley et al. (1993) discuss some of the constraints to more unified understandings of plant stress and offer some new definitions for more encompassing views of stress and related parameters. They recognize plant stress as an adverse reaction to the effect of environmental factors, biotic and abiotic, and define stress as a departure from optimal physiological conditions. For example, a reduction in net photosynthesis is a stress that can be caused by insect feeding. Other important considerations are agents producing stress and plant responses to stress. Workers on biotic stress have long distinguished between injury, the action of a stressor on a plant, and damage, the plant’s response to injury (Tammes, 1961; Bardner and Fletcher, 1974; Pedigo er al., 1986). Modifying previous usage, Higley er al. (1993) formally define these terms as: injury-a stimulus producing an abnormal change in a physiological process; and damage-a measurable reduction in plant growth, development, or reproduction resulting from injury. Injury, stress, and damage are terms that provide a common language for addressing all types of stress. Moreover, they indicate the specific events associated with stress: stimulus, physiological change, and reaction. Injury is the specific stimulus producing stress. Stress is the deleterious alteration of one or more physiological processes. Damage is the measure of how stress impacts

132 JOE FUNDERBURK, LEON HIGLEY, AND G. DAVID BUNTIN plant fitness; in agronomic terms, it is a measure of yield loss or quality. Consequently, damage is especially useful as a practical expression of plant response to injury. Our understanding of arthropod injury and resulting stress and damage is poorly developed. Despite the importance of the topic, surprisingly few reviews of plant responses to arthropod injury are available. Bardner and Fletcher (1974), Fenemore (1982), Pedigo et al. (1986), and Welter (1989) reviewed aspects of plant response to arthropod injury. Although a large body of literature addresses insect-plant interactions, almost all of this work focuses on how plants affect insects rather than the reverse. Indeed, many workers have pointed out the need for additional research emphasis on plant responses to arthropod injury (e.g., Bardner and Fletcher, 1974; Boote, 1981; Funderburk and Higley, 1994; Pedigo et al., 1981, 1986; Wintersteen and Higley, 1993). Although details on relationships between specific types of arthropod injury and corresponding plant responses are uncertain in many instances, the theoretical relationship between injury and damage has been known for more than thirty years. Tammes (1961) described the damage curve, which represents the theoretical relationship between injury and damage (Fig. 1). The damage curve describes the relationship between increasing injury and corresponding changes in yield (or other plant responses of interest). In the damage curve, damage is represented by crop yield loss. Subsequently, Pedigo et al. (1986) described specific portions of the damage curve and presented a terminology for plant responses corresponding to different portions of the curve. These responses are: tolerance-no damage per unit injury; overcompensation-negative damage per unit injury (a yield increase for some level of injury); compensation-increasing damage per unit injury; linearity-maximum damage per unit injury; desensitization-decreasing damage per unit injury; and inherent impunity-no damage Overcompensation

I t

injury Figure 1. The damage curve-the from Tarnmes, 1961 .)

theoretical relationship between yield and injury. (Adapted

ARTHROPOD PEST MANAGEMENT

133

per unit injury. These six terms and their corresponding portions of the damage curve describe the theoretical relationships between injury and damage. Not all plants display this entire array of responses to injury; however, all potential responses to injury are encompassed by the damage curve and these individual portions of that curve. The practical importance of the damage curve is in establishing yield loss relationships that are necessary for calculating pest management decision aids, such as EILs (Pedigo et al., 1986). Most often the damage curve is expressed as yield responses of plant populations, but damage curves can be constructed for individual plants or based on biomass or measures of quality rather than seed yield. These alternative forms of the damage curve can improve its usefulness in providing ecological insights. Theoretically, the damage curve and its constituent parts are important in providing a mathematical description of stress. The unique portions of the curve represent identifiable changes in physiological response to injury. Plant responses to injury depend on a number of important factors. Specifically, these are the type of injury, the timing of injury, the plant part injured, the intensity of injury, and environmental factors (Pedigo ef al., 1986). Each of these factors is an important determinant of how a crop will respond to arthropod injury. The injury type is important because it specifies which physiological processes are impaired by the pest activities. Arthropod injury to crops can fit into one of at least seven different categories: stand reduction, leaf mass consumption, assimilate removal, turgor reduction, fruit destruction, architecture modification, and phenological disruption (Boote, 1981; Pedigo ef al., 1986). Injury by a single pest can fall into many of these categories. For instance, seedcorn maggot (Delia plafura Meigen) injury to soybean (Glycine m m L.) produces stand reduction, architecture modification, and phenological disruption (Higley and Pedigo, 1991). Because plant responses to injury are very much a function of physiological processes affected by injury, the type of injury is an important variable in understanding those responses. Injury early in plant development, such as to seedling plants, can have serious repercussions; however, injury to young plants often is of less significance because of greater potential for compensation. Commonly, arthropod injury has the most serious impact on crop yields when it occurs during reproductive development of the plant. For example, the impact of defoliation on soybean yields is much more severe when defoliation occurs during reproductive stages than during vegetative growth (Teigen and Vorst, 1975). The timing of pest injury poses particular challenges to pest management in that management of the same pest may differ greatly depending when the injury by that pest occurs during crop growth. Direct injury, injury to yield-producing organs, typically has more serious consequences than indirect injury, injury to non-yield-producing organs. Distinc-

134 JOE FUNDERBURK, LEON HIGLEY, AND G. DAVID BUNTIN

tions in injury to different ages of plant parts also are possible. Some insects preferentially feed on younger leaves, for example, and plant responses to these tissues are likely to be more severe than to the same levels of injury made to older, less productive leaves. As the damage curve indicates (Fig. l ) , greater amounts of injury (greater injury intensity) cause greater reductions in yield. After sufficient injury, crops may be insensitive to further injury, but typically this only happens after unacceptable losses already have been incurred. For arthropods, the intensity of injury is directly tied to their numbers. consequently, managing most arthropod injury has focused on reducing arthropod numbers in an effort to decrease the intensity of injury. Environmental factors fall into two categories; abiotic and biotic effects. Abiotic factors are temperature, light intensity, CO, concentrations, pollutants, soil factors, and water availability, Of these, water availability is probably the most important single abiotic factor that alters crop responses to injury. For example, the impact of defoliating insects is much more severe on water-stressed plants than on unstressed plants (Hammond and Pedigo, 1982). Among biotic factors that influence crop responses, arthropods, diseases, weeds, and intraspecific competition within the crop all are important variables. Changes in plant populations, for example, may alter responses of a crop to a given arthropod pest. Similarly, changes in cultural practices can lead to changes in intraspecific competition, disease occurrence, soil conditions, and water availability, all of which can have an important impact on crop responses to pest injury. For these reasons, changing cultural practices necessitates a reexamination of relationships between arthropod pests and their associated crop hosts. This discussion of crop responses to arthropod injury illustrates the complexity of these relationships. Details on all aspects of these relationships are valuable in devising effective management programs for arthropod pests. Unfortunately, the complexity of relationships between arthropod pests and crops has prevented their thorough examination for most crop-pest situations. Indeed, relatively little is known about the specific physiological responses of plants to most types of arthropod pest injury. In the absence of such understanding, pest management programs tend to rely heavily on reducing the intensity of injury; in other words, controlling pest numbers. Unfortunately, an emphasis on controlling pest numbers has seriously constrained some IPM programs and has even led to failures in management. Ongoing research efforts to identify general models of crop response to different types of pest injury offer the promise of providing more comprehensive understanding of relationships between pest injury and crop response (Wintersteen and Higley, 1993). Such an understanding is essential for improving existing IPM programs and for directing management efforts into new areas. For example, recent research on insect defoliation of soybean has provided strong evi-

ARTHROPOD PEST MANAGEMENT

135

dence that yield reductions associated with defoliation are a function of changes in the canopy's efficiency in intercepting photosynthetically active radiation (Higley, 1992). This finding provides a useful model for explaining the impact of insect defoliation in many other crop systems. Additionally, it offers the prospect of greatly improved management of soybean defoliators through cultural practices to improve canopy light interception efficiency and through greatly improved estimates of potential yield losses from defoliation by considering soybean canopy size (which is highly correlated to light interception) as well as pest densities. Beyond improvements in existing pest management programs, novel approaches to pest management may arise from greater knowledge of how insects affect plants. Because most pest management efforts are directed at pest populations, these efforts can present a significant selection pressure on pest populations that selects for individuals resistant to a given management tactic. Although resistance to pesticides is the most common expression of this selection, pest resistance occurs to many other management tactics. Improving crop tolerance to pest injury has the extraordinary theoretical advantage of not producing selection pressure on pest populations. In principle, improved tolerance would provide a permanent form of management, unlike virtually all population-based management efforts. Despite this tremendous theoretical advantage, very little research has focused on questions of plant tolerance and compensation to injury. It is well known that most crop species can tolerate at least some levels of pest injury without yield loss, but the mechanisms are largely unknown. As our understanding of those mechanisms improves, it seems likely that new opportunities for developing more pest-tolerant crop varieties will arise. For example, work by Peterson et al. (1992) on alfalfa (Medicago sativa L.) and Higley (1992) on soybean demonstrate that both species compensate for insect defoliation through altered patterns of leaf senescence, including reduced leaf abscission and retention of high photosynthetic activity in older leaves. Genetic variability in these traits would provide one avenue for improving defoliation tolerance based on new understanding of compensatory mechanisms. Undoubtedly, other compensatory mechanisms exist and remain to be discovered.

B. PESTMANAGEMENT DECISION MAKING The foundation of pest management programs is the ability to make objective decisions about managing pests. Stern et al. (1959) presented the concept of the EIL, which is a decision tool for making objective decisions about managing pests. The EIL is that level of pest injury necessary to justify management action against the pest; it is the level of injury at which economic damage occurs.

136 JOE FUNDERBURK, LEON HIGLEY, AND G. DAVlD BUNTIN Another way to view the EIL is as an index that indicates tolerable and intolerable pest levels. Because it is difficult to measure pest injury directly, EILs usually are described in terms of pest densities, with pest density acting as an index of injury. Stern et al. (1959) did not explicitly define economic damage (the underlying basis for an EIL) or show how to calculate EILs, nor did they address the difficult issue of how to define yield losses associated with different levels of pest injury. These omissions greatly delayed the application of their ideas. Indeed, the first calculated EIL for a pest species did not appear until the early 1970s (Stone and Pedigo, 1972). The EIL incorporates economic factors such as cost of control and value of the commodity, as well as biological information on injury and damage relationships between pests and crop plants. Stated simply, the EIL is that level of injury at which control costs equal benefits of the control. For pesticides, control costs include cost of the pesticide and cost of application; benefits are the yield loss that is prevented by use of a pesticide. In its most general form, the EIL equation is EIL = C/VD(TI)IK where C equals control cost, V equals value of the commodity, D(T1) equals damage (yield loss) as a function of total injury (TI), I equals injury per pest, and K equals the proportion of injury prevented by the management tactic. More simply, the EIL can be expressed as EIL

=

Control cost/(Commodity value

X

Yield loss per pest)

(2)

By integrating economic factors regarding crop value and management costs, as well as biological information on relationships between pest injury and yields, the EIL provides a powerful approach for making management decisions regarding pests. Although establishing EILs requires substantial research, particularly with respect to establishing yield losses per unit injury and injury per pest, the EIL is an essential tool for rationalizing pest management decisions. One often overlooked or misunderstood aspect of the EIL is that use of an EIL applies only to preventable injury. The EIL indicates when management is warranted for injury that can still be avoided. Where significant injury has occurred before a management decision is made, in principle a modified EIL could be used to account for the impact of this previous injury, but in practice such modified EILs are not used. Consequently, the use of EILs is based on the expectation that management decisions are made promptly and prior to significant pest impact. As components of the EIL (control costs, market value, and yield loss per pest) change, so will the EIL change. For example, if control costs are higher, it takes more pests to justify management action, so the EIL increases. Similarly, if market values are very low, then more pests can be tolerated and the EIL will

ARTHROPOD PEST MANAGEMENT

137

increase. Alternatively, if market values are very high, then the EIL will decrease. On very high value crops, such as citrus and vegetables, the EILs for many pests are extremely low. Consequently, the EIL functions as an index to how many pests can be tolerated. Typically, low-value commodities can tolerate more pests than can high-value commodities. Although the control cost and market value components of the EIL are straightforward, yield loss per pest is more complicated. The two major problems are establishing the relationship between pest injury and yield loss and variability in that relationship. Considerable research is necessary before relating pest numbers to yield loss is possible. Specifically, these studies require a minimum of two to three years. Additionally, many experimental constraints make conducting and evaluating these studies difficult. Sometimes experimental methods contribute to variability in measured responses, but more often variability occurs because of differences in varieties, cultural practices, or, most importantly, weather. One reason we do not have EILs for every pest is the substantial research requirements of labor, time, and money needed for establishing EILs. Besides the substantial research requirements for establishing EILs, there are a number of other serious limitations to the development and use of EILs (Pedigo et al., 1986; Poston et al., 1983). The EIL is most useful for therapeutic management practices. Because pesticides are the predominant therapeutic practice, EILs have not been used with alternative approaches such as cultural or biological management techniques. This does not represent a limitation in EILs, but rather reflects a limitation in our existing management technologies. The usefulness of EILs also is limited by the degree to which a given commodity can tolerate injury. Typically, very high value crops have extremely low EILs, and management action is very frequently justified. Difficulties in establishing injury-yield loss relationships also have hampered development of EILs in many areas, although new approaches for calculating EILs for livestock insects and for many aesthetic pests of horticultural crops offer much promise (Raupp et al., 1987). A more fundamental problem is that the EIL addresses only economic criteria in pest management decisions and does not directly address the important issues of maintaining environmental quality or of minimizing selection pressure from a management tactic on a pest. Nevertheless, the use of EILs has been extremely important in improving environmental safety and minimizing the development of resistance, because the use of EILs eliminates unnecessary pesticide use. Perhaps most limiting to the use of EILs is the fact that, like many other pest management components, they are knowledge intensive. Considerable research is required to establish them, and significant information (e.g., on crop development, crop status, and pest density) is required from a producer in their use. Various research efforts are under way to expand both the availability and usefulness of EILs. Improving understanding of physiological plant responses to

13 8 JOE FUNDERBURK, LEON HIGLEY, AND G. DAVID BUNTIN insect injury provides the prospect of managing guilds of pests that produce common injuries (Higley et al., 1993; Hutchins and Funderburk, 1990; Hutchins et al., 1988). Indeed, comprehensive yield-loss models and multiple-species economic-injury levels for pest complexes based on homogeneity in physiological response to injury have been developed for weeds (Wilkerson er al., 1991) and defoliating caterpillars (Hutchins et al., 1988) in soybean. These multiple species EILs have the significant advantages of greater accuracy and improved efficiency as compared to combinations of individual species EILs. It seems likely that EILs developed for injury guilds are but an interim step toward multiple-guild EILs, which would allow the consideration of different concurrent stressors in making management decisions. Such multiple-guild EILs would be similar, at least in concept, to some simulation-modelingapproaches to predicting the impact of various plant stressors and would represent a significant advance toward integrated management of plant stress. Another area of refinement in the EIL is research to expand the concept beyond solely economic criteria. In particular, new work has provided a mechanism for incorporating considerationsof environmentalrisks directly into the EIL. Higley and Wintersteen ( 1993) propose a method for developing environmental EILs that assigns a dollar value to environmental risks for various pesticides and incorporates these costs into the EIL equation. Specifically, environmental costs of a given pesticide are added to the control costs in the conventional EIL formula [Eq. (l)]. Their method for assigning cost is based on an objective ranking of relative pesticide risks to different aspects of the environment and assigning a monetary value to avoiding different levels of risk. This monetary value is established through the economic procedure of contingent valuation, which provides a formal mechanism to codify subjective views on pesticide risk. The environmental EIL represents a decision tool that weighs environmental risk with economic risk. Additionally, environmental EILs can be used as an index to relative environmental safety among insecticides, thereby providing a useful aid in pesticide selection. Besides environmental EILs, the EIL concept itself has been proposed as a framework for other approaches to maintaining environmental quality in pest management (Pedigo and Higley, 1992). The EIL is the basis for making management decisions on pests, but it is not used directly. If we wait until a pest population reaches the EIL to make a decision, we will have accrued unnecessary yield or quality losses. Consequently, it is better to decide to take action while a pest population is well below the EIL but when we are still reasonably certain it will reach the EIL. The economic threshold (ET) is used to identify this point. The ET is a pest or injury level set below the EIL that indicates when management action should be taken (Pedigo et d , 1986). The ET is predictive. It predicts when a pest population is going to reach the EIL. Consequently, the practical tool used for making management decisions regarding pests is the ET,

ARTHROPOD PEST MANAGEMENT

139

which is based on a calculated EIL. In part, the ET represents the solution to a dilemma. On the one hand, if a pest is going to reach the EIL, we want to suppress it as early as possible to avoid unnecessary losses. Alternatively, the earlier we take action, the more uncertain we are that pest numbers actually will reach the EIL. The ET represents a compromise between these two objectives. Although the ET usually is defined as a population density, it is actually a measuring stick used to estimate the time at which to take action to suppress a pest. Because the ET is a predictive criterion (predicting that pest numbers will exceed the EIL), determining an ET is complex and requires knowledge of pest phenology, pest population growth rates, management lag time, and management efficiency. Poston et d.(1983) critiqued existing ET development, and Pedigo er af. (1989) identified specific approaches for ET determination. These approaches fall into two broad categories: subjective determinations, ETs based on expert opinion only and without reference to a calculated EIL, and objective determinations, ETs based on a calculated EIL. (A similar distinction can be made with noncalculated EILs, which are subjective estimates, and calculated EILs, which are real EILs based on research data and appropriate calculations.) Among objective determinations, three types of ETs can be recognized: (1) fixed ETs-thresholds set as some fixed percentage (such as 80%) of a calculated EIL; (2) descriptive ETs-thresholds based on models of expected pest population and injury growth rates with reference to a calculated EIL; and (3) dichotomous ETs-thresholds based on statistical procedures to classify a population as being above or below an EIL.

C. SAMPLING AND MONITORING One of the fundamental changes in pest control brought about by the concept of IPM is the way in which decisions are made. Before the advent of IPM, pest control decisions usually were made with little quantitative information about pest status and potential for damage. With IPM, decisions are made with decision rules consisting of at least two primary components, an economic threshold and a procedure for assessing pest population density. A third desired component is a phenological forecast model to determine the appropriate time to assess populations and implement management measures (Binns and Nyrop, 1992). Therefore, sampling is a key element of IPM because current information concerning pest status is essential to make timely and effective decisions about pest management activities. Sampling information typically is used in decisions about therapeutic or curative management actions, such as use of insecticides to prevent pests from causing economic damage. Sampling information also is useful in implementing preventive management tactics that can be desirable to achieve the aims of IPM.

140 JOE FUNDERBURK, LEON HIGLEY, AND G. DAVID BUNTIN Early reviews by Morris (1960) and Strickland (1961) discussed the basic principles of sampling arthropods, but Southwood ( 1978) provided the most comprehensive review of sampling procedures for arthropods. Cochran (1977) reviewed statistical principles of sampling theory. Other advances in arthropod sampling have been presented by Kuno (1991), Shelton and Trumble (1991), Nyrop and Binns (199 I), and Binns and Nyrop (1992). Development of a sampling program consists of the following steps: ( I ) define the sampling universe and target stage, (2) select a sampling technique, (3) select a sample unit, (4)describe population dispersion, (4)develop plans for sampling sample units, ( 5 ) develop guidelines for timing of sampling, and (6) develop rules for decision making. Defining the objective in sampling is important because it often will determine the methods used (Southwood, 1978). For most applications in IPM, the objective of sampling is to provide information on the status (i.e., damaging or nondamaging) of a target pest population to make management decisions. It may not be necessary to accurately or precisely estimate mean density because usually we only need establish if a pest population is above or below some critical density such as an economic threshold. Furthermore, it often is not necessary to express population numbers in terms of unit area. Densities can be expressed in relative terms such as number per leaf per stem or number per unit of a technique. As long as decision rules are expressed in these same units, the sampling information can be used. Another issue in developing sampling procedures is whether decisions will be made on a field-by-field basis or will sampling be aimed at providing regionwide or statewide forecasting. A regionwide network of sampling stations may be used to monitor activity of a pest to time cultural activities such as planting to avoid peak flight periods of a pest. A multistate monitoring system was established in the Pacific Northwest (United States) to monitor fall flight activity of aphids (Homoptera: Aphididae) that can vector barley yellow dwarf virus in small grain crops (Pike et al., 1990). The monitoring system provides information about aphid flight activity so that planting of small grains can be timed to avoid peak aphid flights in the fall. However, most IPM programs are designed to provide field-by-field management decisions. The sampling universe usually consists of the plants within a field or portion of a field. The universe can be further delineated depending on the habits of the pest as a portion of the plant, such as all leaves or main-stem terminals or the root zone. Sampling technique is the method by which information is collected from a sample unit, where a sample unit is the proportion of the habitable space from which counts are taken (Pedigo, 1989). Sample units should be distinct and not overlap, have an equal chance for selection, remain constant at least for each sample event, and should provide a reasonable balance between cost (time, labor, etc.) and variance among sample units (Morris, 1955; Shelton and Trumble, 1991). Because of the cost, sampling techniques that provide an absolute mea-

ARTHROPOD PEST MANAGEMENT

141

sure of density per unit area are not feasible in IPM programs. Instead, suitable techniques for IPM provide population estimates based on a unit of habitat such as a foot of row or on available food supply such as leaves, fruits, or stems. Stem sampling for alfalfa weevil larvae, Hypera postica (Gyllenhal), in alfalfa (Legg et a f . , 1985) and the shake-cloth technique for foliage insects in soybean are examples of these types of techniques (Kogan and Herzog, 1980). Numerous techniques also are suitable that provide relative estimates with no direct relationship to land surface area. Estimates by these relative techniques are referenced to the technique itself (Pedigo, 1989). The most widely used relative technique is the sweep net in which counts are expressed as number per sweep. Various types of traps such as light, pheromone, sticky, or pit-fall traps also provide relative population estimates. Relative techniques are useful in IPM because of ease of collection and low cost. Southwood (1978) provided a very thorough review of various techniques for sampling arthropods, and techniques for sampling arthropods in soybean are reviewed by Kogan and Herzog (1980). In developing a sampling program, it is desirable to determine the number of sample units that must be examined to estimate mean density with a given and acceptable level of precision. Precision usually is expressed as the standard error of the mean or a confidence interval expressed as a percentage of the mean (Southwood, 1978; Karandinos, 1976). The degree of population aggregation influences the optimal sample-unit number in that more sample units must be examined to estimate mean density with a given level of precision as the degree of aggregation increases. Two approaches have been used to describe population dispersion. These are to determine which of a number of theoretical mathematical distributions best describes observed frequency distribution. The most commonly used distributions are the Poisson for random populations and the negative binomial distribution for aggregated populations (Bliss and Fisher, 1953; Southwood, 1978; Binns and Nyrop, 1992). More recently, a number of models based on the relationship between sample variance (crz) to mean (m)have been proposed, with the most useful being Taylor’s (1961) power function (v2= amb) and Iwao’s (1968) regression procedure. Karandinos (1976) and Ruesink (1980) provided equations for calculating optimal sample size for a fixed sample-size sampling program. Considerable time and cost savings can be realized by using sequential sampling procedures as compared with fixed sample-size plans, for example, sampling costs can be reduced by more than 50% (Nyrop and Binns, 1991). In sequential sampling procedures, the number of sample units collected varies, so that only enough sample units are collected to estimate mean density or classify a population with a given level of precision. Sequential decision plans are most useful in IPM programs because these plans collect sample units until it can be determined whether a population is above or below some critical density such as an ET (i.e., the population is classified as damaging or nondamaging).

142 JOE FUNDERBURK, LEON HIGLEY, AND G. DAVID BUNTIN

Two protocols are commonly used; the sequential probability ratio test (SPRT) (Wald, 1947) and a confidence interval method developed by Iwao (1975). To use SPRT, the distribution of sample observations must be described by a probability model such as the binomial, Poisson, or negative binomial distributions. SPRT uses an upper and lower critical density and associated probabilities of error to calculate stop limits. With Iwao’s (1975) method, stop limits are calculated from a sample-size-specificconfidence interval developed about a single critical density (Nyrop and Binns, 1991). Figure 2 shows a typical sequential classification plan. Samples are collected and the cumulative total count is compared with the upper and lower stop-sampling lines. When the cumulative total is outside the indecision zone, the population can be classified with a given probability of error. If the population is above the zone, suppression measures such as an insecticide application are taken. If the population is in the no-treatment area, then no treatment is made and the population is sampled at the next sample date. Although most sequential classification plans require that a minimum number of samples be collected before a decision is made, the population usually can be classified quickly with very few samples if the pest density is substantially greater or less than the critical densities. However, if a pest population density is near the critical density, a large number of samples may be needed to reach a decision (Iwao, 1975). Procedures and equations for constructing sequential decision plans are reviewed in more detail by Wald (1947), Iwao (1975), Fowler and Lynch (1987), and Nyrop and Binns (1991). Although sequential sampling plans can substantially reduce sampling time and cost, complete enumeration of sample units can still be a time-consuming

250 c.

c

3

200-

-

0

.-w 9

CU

-

-

150-

-

z

100-

0

50-

-

3

-

0

I

I

.

,

1

143

ARTHROPOD PEST MANAGEMENT

process. Binomial or presence-absence sampling can greatly reduce counting and handling time compared with standard sampling plans. In binomial sampling, the proportion of sample units containing one or more (T) organisms (pT) is substituted for mean density to classify populations, where T is referred to as the tally threshold. If a curvilinear relationship between pTand mean density can be demonstrated to exist with an acceptable degree of variability, then mean density can be estimated from pT and an economic threshold (i.e., decision rule) can be expressed as a critical pT (Fig. 3). If decision rules can be expressed in terms of pT, sequential classification plans for binomial sampling can be constructed using SPRT (Fowler and Lynch, 1987; Nyrop and Binns, 1991). Binomial sampling with presence or absence of an organism can be effectively used to classify population status, if critical pT is less than a proportion of infested samples of about <0.8 (Nyrop and Binns, 1991). If pT is >0.8, it becomes increasingly difficult to classify populations reliably. This problem can be addressed by changing the size of the sample unit to achieve an appropriate scale. Another approach is to increase the tally threshold to a more appropriate value (Nyrop and Binns, 1991). Although this is less desirable than simple presence or absence because some counting is required, using pT avoids counting when more than T organisms are present in a sample unit thereby reducing sampling effort. One criticism of sequential sampling is that decisions often are reached after sampling without inspecting most of a field. This may cause localized infestations of heterogeneously distributed pests to be missed during sampling. Variable intensity sampling (VIS) minimizes sampling efforts much like sequential sam-

a

1.0

4-

rn

$!

.-C

0.8

+ 0.4

.-0

t:

0 0.2 Q

2

0

2

4

6

8

10

12

14

16

Mean number per sample unit Figure 3. Relationship between mean number of individuals per sample unit and proportion of sample units infested with T or more individuals (pT)for a hypothetical pest.

144 JOE FUNDERBURK, LEON HIGLEY, AND G. DAVID BUNTIN pling procedures while ensuring that samples are collected throughout the entire sampling universe (Hoy et al., 1983; Hoy, 1991). With VIS, the universe is stratified into sites within the field and sample units are collected from all sites. One drawback to VIS is that a programmable calculator or computer must be carried while sampling for determination of updated sample intensity. Another recent notable advance in sampling for the purposes of decision making in IPM was the development of a procedure by Pedigo and van Schaik (1984) for sequentially sampling through time. This procedure, termed time-sequential sampling, uses cumulative results of successive samples in time to classify the trajectory of a pest population as being damaging or nondamaging. This procedure requires extensive research to characterize growth curves of damaging and nondamaging populations of a target pest. Critical upper and lower densities are derived for the two population growth curves for each sample data. These values are used with a modified SPRT to construct stop (decision) limits for each sample date. Once a population can be classified as nondamaging, sampling is discontinued for the remainder of the season. Conversely, if a population is classified as damaging, then suppression measures are initiated and the field is resampled after treatment to verify adequate suppression. Some potential problems with this procedure are that all pests may not exhibit distinct growth curves and the ET must be greater than the densities needed to classify a population’s trajectory. Because of the considerable amount of population research needed to develop time-sequential sampling plans, few have been constructed. Nevertheless, this procedure has the potential of greatly reducing sampling costs by eliminating entire trips to the field for sampling. Despite a quarter century of research and development of sampling procedures for arthropods in agriculture, the absence of reliable, inexpensive, user-friendly sampling procedures is one of the greatest impediments to the implementation of effective IPM programs in many agroecosystems. Sequential sampling has substantially reduced sampling costs in IPM. Currently, sequential binomial sampling plans are receiving much attention because the ease and simplicity of binomial sampling is very attractive for use in IPM. VIS and time-sequential sampling also will permit improved efficiency of sampling resources. Inherent in IPM philosophy is the assumption that using sampling information to make management decisions is more desirable than making decisions without sample information. However, decision rules are economically viable only if using sample information reduces management costs (Nyrop and Binns, 1991). Examination of the value of sample information has received much less attention than procedures for sampling pest populations. Decision rules using sampling information have been found to reduce management costs and losses over no treatment and prophylactic management strategies for the European corn borer, Osfrinia nubilalis (Hubner), in sweet corn (Zea mays L.) (Shelton et al., 1986; Nyrop and Binns, 1991). In contrast, Nyrop ef al. (1989) found that sampling

ARTHROPOD PEST MANAGEMENT

145

did not improve economic returns in control of the leek moth, Acrolepiopsis assectella (Zeller), in leeks (Allium porrum L.) over prophylactic treatments because of the high risk of damaging infestations and high cost of sampling. However, this analysis did not consider possible environmental costs or potential cost of pesticide resistance resulting from routine pesticide use. Generally, the value of sampling information for decision making is reduced when sampling costs are high, treatments are preventive, economic thresholds are low (i.e., little injury can be tolerated), commodity market value is high, or there is a high risk of damaging infestations. The methodology and examples of assessing the value of sample information are discussed by Carlson (1970), Nyrop et al. (1986, 1989), Foster et al. (1986), and Nyrop and Binns (1991).

D. MANAGEMENT ACTIONS Choosing a management tactic is problematic. The more lead time available for responding to a pest problem, the greater the array of options. When a pest problem can be anticipated very early, a variety of management options may be available to prevent or minimize pest injury. When rapid action is needed against a pest population, pesticides often are the only available tactic. Such differences in the ability to anticipate pest problems lead to two types of management actions, therapeutic and preventive. Therapeutic tactics are used to remedy or ameliorate a pest problem, whereas preventive tactics are used to avoid potential problems. Although therapeutic and preventive actions provide different approaches for managing a pest problem, a challenge to IPM programs is that the choice of individual management tactics often is limited, particularly to pesticides. Preventive tactics aim to prohibit establishment, limit growth, and reduce injuriousness of a pest’s population. Development of such tactics may require an in-depth understanding of a pest’s behavior, life cycle, seasonal cycle, habitat ecology, and population dynamics. Ultimately, many preventive tactics, once developed, are relatively inexpensive to implement and afterward may not require as much detailed information on pest status in a crop. Most plant resistance, cultural management, and biological management tactics are employed preventively. 1. Plant Resistance to Arthropods Plant resistance is one of the most effective tactics available in IPM, because management is achieved with little cost to the grower and without potential health risks and environmental contamination from pesticides. Plant resistance also is compatible with other management tactics, including biological management (Boethel and Eikenbary, 1986; Smith, 1989). Even partial resistance may

146 JOE FUNDERBURK, LEON HIGLEY, AND G. DAVID BUNTIN be sufficiently effective in conjunction with cultural or biological management to reduce or avoid the need for therapeutic actions (Starks et al., 1972). Painter (195 1) provided the first comprehensive discussion of plant resistance to arthropods. Russell (1978), Panda (1979), Maxwell and Jennings (1980), and Smith (1989) provide more recent overviews of the subject. Resistance to major pests of selected crops is reviewed by Khush and Brar (1991). Plant resistance is the use of genetically controlled plant defense mechanisms to regulate herbivore populations (Painter, 1951; Smith, 1989). Pseudoresistance, which manipulates plant growth and development to evade arthropod attack (Smith, 1989), and plant associational resistance, which relies on intercropping to disrupt host-finding behavior by a pest (Risch et af., 1983), are discussed in the next section on cultural management. Resistant cultivars traditionally have been developed using natural resistance sources in the same or related plant species. However, recent advances in plant molecular engineering have created a vast potential for artificially incorporating novel resistant mechanisms into plants. Agronomically acceptable engineered crops containing genes for production of the toxin derived from the entomopathogen Bacillus thuringiensis probably will be available in the near future. The mechanisms of plant resistance are antixenosis (nonpreference), antibiosis, and tolerance (Painter, 1951; Smith, 1989). Antixenosis operates by disrupting normal arthropod behavior, whereas antibiosis adversely affects a pest’s physiological processes, thereby impairing arthropod survival, growth, development, and behavior. Antixenosis and antibiosis involve both plant and arthropod characteristics. Tolerance is a plant response to injury where acceptable plant production (i.e., yield) is achieved in spite of injury by a pest. Plant breeders and entomologistshave focused on developing crop cultivars that are resistant to key pests particularly when insecticidal tactics are costly or ineffective. Plant breeders typically have relied on antibiosis as the preferred mechanism for developing resistant cultivars, because antibiotic mechanisms generally are more stable and provide more complete suppression in the field. Furthermore, when antibiosis is conferred by one or a few dominant genes, selection usually can be done easily in the greenhouse or field. Antibiosis can be mediated by toxic allelochemicals, such as various types of alkaloids, arthropod growth inhibitors, reduced plant nutrient levels that inhibit arthropod growth, plant hypersensitivity, and plant structural factors such as glandular trichomes that secrete allelochemicals or adhesive substances that inhibit pest movement. Twenty sources of antibiotic resistance have been identified in small grains to the Hessian fly, Mayetiola destructor (Say) (Patterson et af., 1992). All except two sources are conferred by single dominant genes, although the exact chemical basis for resistance is not known for any source. Virtually all field corn hybrids grown in the Midwest United States produce elevated levels of the allelochemical DIMBOA (2,4-dihydroxy-7-methoxy1,4-benzoxazin-3-0ne), which has antibi-

ARTHROPOD PEST MANAGEMENT

147

otic effects on European corn borer larvae during the whorl stage (Klun er al., 1967). Antixenotic resistance consists of plant morphological or chemical factors that reduce the attractiveness of a plant as a host, resulting in the selection of an alternate host (Smith, 1989). Specific defenses include presence or absence of trichomes and leaf surface waxes, increased thickness of plant tissue, and chemical repellents or feeding deterrents (Smith, 1989). Feeding by the potato leafhopper [Empoasca fabae (Harris)] is deterred by the presence of simple erect trichomes on leaves of soybean (Broersma et al., 1972) and potato (Solanurn tuberosum L.) (Taylor, 1956). Leaves with wax blooms stimulate some insects and deter other insects from feeding on some crucifer crops (Stoner, 1990). Chemically based antixenosis can result from the presence of a chemical deterrent such as saponins, cardiac glycosides, cucurbitacins, condensed tannins, isoflavones, and various alkaloids (Smith, 1989). Antixenosis often provides partial resistance and may be less effective when resistant cultivars are grown over large geographical acres. Tolerance usually is the result of complex physiological changes in plant response to injury that alters the relationship between injury and damage in a plant (Fig. 4).Because of the relative difficulty of identifying and selecting for tolerance, this mechanism has been used the least by plant breeders in developing resistant cultivars. Procedures for screening for tolerance are described by Morgan et al. (1980), Panda and Heinrichs (1983), and Smith (1989). Resistance based on antibiosis and to a lesser extent antixenosis reduces pest populations, thereby creating selective pressure that can result in the development of virulent biotypes. Resistance based on tolerance avoids selective pressure on the target population by allowing the population to develop normally. Therefore, tolerance avoids the development of virulent biotypes, and resistance is genetically stable. However, because tolerance is determined partly by physiological dynamics of

A

9 al

j;

.-----Susceptible -Tolerant

----.. --. -_-. **.

*.

-.-.. --. ..--. ..--. ._ -- --.

Injury or Pest Number

Figure 4. Injury and damage (measured by yield loss) response curve for a susceptible and tolerant plant genotype.

148 JOE FUNDERBURK, LEON HIGLEY, AND G. DAVID BUNTIN plant growth, environmental conditions and plant health affect the expression of tolerance (Smith, 1989). Consequently, tolerance may be compromised when other stresses reduce plant health. A deficiency of plant resistance to arthropods is that the actual biochemical or physiological mechanism of resistance often is not known. Fortunately, ignorance about the exact mechanism does not prevent effective screening and selection for resistance. However, when resistance is based on toxic allelochemicals, caution should be exercised because these compounds also may adversely affect mammalian health. Another issue is the loss of effective resistance by the development of virulent pest biotypes. Biotype development usually is associated with antibiotic, monogenic resistance that kills the pest. Biotype development also seems to occur more frequently in some insects than others, with 64% of the known examples of biotype development occumng in aphids (Smith, 1989). At least 16 biotypes of the Hessian fly have been identified since the widespread use of monogenic, antibiotic resistance (Patterson et al., 1992). The response to biotype development has been the sequential deployment of new monogenic sources as older sources lose effectiveness (Cox and Hatchett, 1986). Pyramiding genes has been proposed as an alternative strategy (Gould, 1986). Polygenic resistance and plant tolerance should be more stable than monogenic resistance and should reduce the potential development of virulent biotypes.

2. Cultural Management Cultural management tactics reduce rates of pest increase and damage through manipulation of the pest’s requisites within the agroecosystem. The favorability of the pest’s niche is reduced by altering the vegetation or by modifying the practices used to produce crops. Development of cultural management requires an understanding of pest ecology as it relates to outbreaks in a crop. Cultural management tactics usually are designed to exploit a vulnerability in the pest’s life cycle. Changing the diversity, abundance, and spatial arrangement of vegetation within an agroecosystem is a major planning decision with other potential economic and ecological ramifications. A few such cultural management strategies are employed, mostly against pests that are very difficult to manage. For example, management of spider mites (Tetranychus spp.) is complicated by difficulty in scouting, rapid reproductive rates, high reproductive potential, and mite sensitivity to pesticides and environmental conditions (Brandenburg and Kennedy, 1982). Spider mites have a wide host range and overwinter on wild plant hosts present in field margins. These hosts and early spring crop hosts such as corn serve as nurseries for population increase and dispersal into adjacent peanut (Aruchis hypoguea L.) fields. Crop rotation away from corn reduces infestations. Not mowing field borders in the summer reduces dispersal into peanut fields.

ARTHROPOD PEST MANAGEMENT

149

Vegetational diversification within fields can be achieved by intercropping, cover cropping, and weedy culture. The presence of grassy weeds in alfalfa reduces abundance of potato leafhopper and is useful as a cultural management strategy (Oloumi-Sadeghi et u f . , 1987). Intercropping includes broadcast mixtures of crops, mixtures of crop plants within rows, and mixtures of alternating rows of crops. Mixed intercropping of alfalfa with oats (Avenu surivu L.) reduces the abundance of potato leafhopper and the need for curative action (Lamp, 1991). Damage to crucifers by the cabbage maggot, Delia rudicum (L.), and the cabbage aphid, Brevicoryne brussicue (L.), is greatly reduced by between-row interplantings of numerous crops and weeds (Coaker, 1980). Cabbage aphid damage declines because of reduced immigration and increased predation; cabbage maggot populations are reduced because of reduced oviposition on the crucifer crop. Early-maturing soybean varieties planted near main soybean plantings attract the bean leaf beetle, Cerutoma trifurcutu (Forster), and southern green stink bug, Nezaru viridufu (L.) (Newsom and Herzog, 1977). Insecticide applications in the early maturing trap crop control the target pests and prevent their spread to the main crop. Natural enemy populations also are conserved in the main crop. A reduction in spread of a viral disease spread by the bean leaf beetle also occurs. This trap cropping system is very effective but has not been widely accepted by producers in regions where the pests are serious economic problems. These pests are easily controlled with insecticides, and growers are unwilling to accept potential reductions in yields in the soybean trap crop. Crop rotation can be an effective management practice against arthropod pest species with long generational cycles. Rotation of host and nonhost crops has been used successfully for managing arthropod pests that become established during one cropping season, overwinter within the field, and are present in an injurious life stage during the next cropping season. Rotating between seasons with a grass crop such as corn reduces problems from whitefringed beetles (Gruphognurhus spp.) in soybean (Ottens and Todd, 1979). Eggs of western corn rootworm (Diubroticu virgiferu virgiferu LeConte) and northern corn rootworm (D.barberi Smith and Lawrence) are almost exclusively laid on corn (Levine and Oloumi-Sadeghi, 1991), and rotation with another crop such as soybean has been used to manage the pests in corn. However, a genetic mechanism of extended egg dispause of more than one year has been documented in northern corn rootworm populations. In some instances where crop rotation has been heavily relied upon as a control tactic, a proportion of rootworms have extended egg dispause, and heavy damage to corn sometimes occurs (Krysan et af., 1986). The selection of which crops to grow within the agroecosystem also is important in determining overall injury from arthropod pests. Outbreaks of soybean looper [Pseudoplusia includens (Walker)] are common in soybean when grown in agroecosystems containing cotton (Gossypium hirsutum L.) (Burleigh, 1972).

150 JOE FUNDERBURK, LEON HIGLEY, AND G. DAVID BUNTIN

Adult feeding on cotton nectar increases oviposition, mating frequency, and longevity (Jensen er a f . ,1974). Discontinuing cotton production would reduce damage but is not utilized because of the economic importance of cotton. Cultural management strategies that operate by modifying the vegetation within an agroecosystem impact the economics of farming operations, which frequently impedes adoption by producers. Risch et al. (1983) examined published studies relating plant species diversification to pest abundance and noted that 53% were less abundant in a diversified system, 18% more abundant in a diversified system, 9% showed no difference, and 20% showed a variable response. The studies also did not examine the impact of plant diversification on crop yields and the economics of production systems. Practices employed to plant, maintain, and harvest a crop can influence the level of injury from arthropod pests, and purposeful modification can help to alleviate crop losses. Pest injury may be avoided by producing a crop at times when pest populations are in nondamaging stages or at low population levels. Modification of some management practices can reduce colonization of pest populations onto crops. Modification of cultural practices often elicits altered microclimatic conditions in crop habitats in such a way that survival or development of pest populations is reduced. Production practices that can sometimes be modified to manage pests include planting date, harvest date, irrigation, fertilization, row spacing or orientation, seeding rate, and tillage. Many are not drastic changes in production practices and such cultural management tactics sometimes are employed in IPM programs. Manipulation of planting and harvest date is a management strategy for important cotton pests (Frisbie and Walker, 1981). Planting early, growing shortseason varieties, and early defoliation enhance maturity of the crop in some regions before boll weevil (Anrhonornusgrandis grandis Boheman) and Heficoverpa spp. populations reach economic levels, whereas in other regions planting late to allow for maximum mortality of overwintered populations of the boll weevil is the better strategy. Movement of populations of adult plant bugs (Lygus and Adelphocoris spp.) and potato leafhoppers from alfalfa to soybean is very damaging when soybean is small (Pedigo et al., 1981). In situations where a soybean field is located in close proximity to alfalfa fields, soybean planting and alfalfa harvest may be timed so that the first cutting occurs before soybean emergence. High plant densities of cotton in narrow rows increase the number of older bolls earlier in the fruiting period, and the practice enhances escape from boll weevil in short-season production schemes (Walker et al., 1977). In south Texas, use of this practice increases cotton earliness, crop yield, and producer profit (Walker et al., 1976). The washing and drowning action of sprinkler irrigation suppresses populations of the Banks grass mite [ Ofigonychuspratensis (Banks)]

ARTHROPOD PEST MANAGEMENT

151

on corn in the Texas High Plains (Chandler er al., 1977). Flooding can be used to manage populations of white grubs (Ligyrus subrropicus Blatchley) in Florida muck-soil sugarcane (Saccharurn oficinarurn L.) fields (Cherry, 1984).

3. Biological Management Arthropod pests are hosts to many natural enemies. Although crop habitats frequently are disturbed by production activities, populations of numerous species of predators, parasitoids, and pathogens can adjust and cause pest mortality. Many predator species are especially adaptable to disturbed crop habitats and provide important natural mortality to arrays of crop pest species. For example, many ant (Hymenoptera:Formicidae) species are important generalist predators. Way and Khoo (1 992) reviewed the role of ants in pest management. Combined abiotic and biotic mortality sometimes is sufficient to prevent pest outbreak. At other times, natural mortality reduces the severity of a pest outbreak. Conservation of natural enemies through the reduction and selective use of pesticides is a primary objective of IPM. Programs that integrate natural enemies with insecticide use have been implemented for many crops in many production situations. Benefits include prevention of pesticide resistance development, reduced induced-pest problems, and lowered production costs. Huffaker (1985) reviewed some early examples, including such an IPM program developed for apples (Malus domestica Backhauser) in Nova Scotia, Canada. Programs that integrate natural enemies with insecticide use, pest-resistant cultivars, and cultural tactics have been developed for many crops in many geographical areas. Pedigo e? af. (1981) described IPM programs for soybean arthropod pests. Funderburk and Brandenburg (1993) reviewed IPM programs for peanut arthropod pests. For most arthropod pests, numerous efficacious chemical or biological insecticides are available, and natural enemy populations can be conserved through careful selection of products and rates. Modifying chemical applications and rates to preserve natural enemies is described by Hull and Beers (1985). Recent techniques have been developed to select for pesticide-resistant beneficial populations. For example, pesticide-resistant mite predator populations have been selected, propagated, and released into orchards and vineyards, which has led to reductions in acaricide use and lessening of damage from spider mite pest species (Hoy et al., 1988; Hoy and Conley, 1989). Classic biological control involves establishing a natural enemy in a new geographical area, frequently to manage an introduced pest that lacks effective natural enemies. Classic biological control projects have been successful for many crop pests, including numerous scale insect and mealybug species (Homoptera: Coccoidea) (DeBach et af., 1971).

I 5 2 JOE FUNDERBURK, LEON HIGLEY, AND G. DAVID BUNTIN 4. Therapeutic Actions Fluctuations in pest populations are likely to cause outbreaks in even the bestmanaged systems. Many pest problems cannot be precisely predicted with current technology. Consequently, pests are inherently destabilizing factors in crop production, and methods are needed to minimize their impact to crop production systems. It is essential that responsive tactics be available and be used to minimize losses caused by pests. Relatively few reliable therapeutic management tactics other than pesticides are available. Widespread use of insecticides for therapeutic management reflects the low cost and rapid short-term curative action of insecticides in preventing damage. For occasional use against sudden pest outbreaks, insecticides can be used quickly with correspondingly rapid suppression, usually with little adverse ecological disruption. No other management tactics provide such flexibility of use and rapid action. Problems with pesticide use usually arise when they are used preventively rather than in a therapeutic role. Because pests rarely require suppression in every field every year, routine preventive use ensures that pesticides will be applied unnecessarily. Not only does this increase costs to the producer, it significantly increases the potential for environmental contamination and the probability of pests developing resistance. A goal of IPM programs is to only use pesticides when necessary to minimize adverse effects on the environment. Nevertheless, growers often feel compelled to use pesticides preventively, particularly when they must be applied before pest populations develop to reduce the risk of catastrophic losses from unpredictable pest outbreaks. Important considerations in pesticide use in IPM are toxicity, persistence, and effectiveness. Ideally a pesticide should be selective, yet effective against the target pest and relatively nonpersistent. Compounds must persist long enough to achieve suppression, but not persist to pose a threat to the environment. Most persistent compounds have been removed from use in the United States. Furthermore, it is ecologically desirable that pesticides kill the target pest without affecting natural enemies and nontarget organisms. However, because of development and registration costs (NAS, 1986), most companies pursue pesticides with a broad spectrum of activity rather than selective materials that will have a more limited market. Indeed, the mode of action of most insecticides is limited to three classes of chemistry: organophosphates, carbamates, and pyrethroids, which often are broadly toxic to arthropods. Most currently available insecticides are not physiologically selective to arthropods (Metcalf, 1982). Selectivity of broad-spectrum materials may be achieved by formulation or selective application procedures. Use of nonpersistent materials may provide selectivity in time by killing only the arthropods present at the time of application. Because pollinators are not active at night, polli-

ARTHROPOD PEST MANAGEMENT

153

nator mortality from use of toxic but nonpersistent insecticides such as parathion (O,O-diethyl-O-4-nitrophenyl phosphorothioate) can be minimized by applying treatments at dusk. Systemic insecticides applied as seed treatments or granular applications at planting may kill phytophagous species with minimal impact on predators and parasites. Unfortunately, granular formulation may be attractive to nontarget vertebrates and may contaminate ground-water in leachable soils. Seed treatments can drastically reduce the amount of pesticide used per unit land area compared with granular treatments, but also may potentially poison birds and mammals attracted to the seed. Partly for this reason, seed treatment with insecticides is not aggressively pursued in the United States, even though it is widely used in Europe. The most promising approach to selective application is using attractant baits or traps impregnated with a nonselective insecticide to deliver insecticide specifically to the target pest. Baits containing carbaryl (1 -naphthyl methylcarbamate) and a potent feeding attractant have been developed to kill adult corn rootworms in corn and may achieve similar levels of suppression as adult foliar sprays with a fraction of the active ingredient per acre (Metcalf et al., 1987; Lance and Sutter, 1990). Pesticide use also may be curtailed by using lower rates (Metcalf, 1982). Because of product liability and other factors, minimum labeled rates often are greater than is actually needed to achieve acceptable suppression. Pedigo and Higley (1992) suggested that in some instances, insecticide rates could be lowered as much as 75%. However, in most instances, rate reductions of 10 to 50% probably are more realistic. Determination and use of minimal effective rates could substantially reduce the amount of pesticides used in agriculture. Improved spray delivery systems also may permit reductions in pesticide rates. Law et al. ( 1992) showed that a reduced-volume, air-assisted electrostatic spray system provided similar pesticide deposition on target plants and efficacy as conventional spray systems with half the standard application rate in greenhouse and field trials. Few therapeutic tactics other than pesticides are available. The most promising is the use of entomopathogens as biological insecticides in which a pathogen is applied much like a conventional insecticide and a subsequent epizootic suppresses the pest population (Falcon, 1985; Gillespie, 1988). Falcon (1985) lists pathogens that are currently available and being considered for development. Another approach is inundative releases of natural enemies, which involves mass rearing and releasing large numbers of a natural enemy such as a predator or parasitoid where the pest is expected to be a problem (Ridgeway and Vinson, 1977). Inundative releases work best when applied before pest populations reach damaging levels. Finally, it is possible to manage pests therapeutically by early crop harvest. A common practice in the northern United States is to harvest alfalfa early if populations of the alfalfa weevil reach damaging levels 1 to 2 weeks before harvest (Onstad and Shoemaker, 1984). Timely harvest of sericea

154 JOE FUNDERBURK, LEON HIGLEY, AND G. DAVID BUNTIN

lespedeza [Lespedeza cuneafa (Dumont) G. Don] also can prevent late summer defoliation by lepidopterous caterpillars (Buntin, 199 1).

IV. ECOLOGICAL AND ENVIRONMENTAL CONSIDERATIONS A. POPULATION REGULATION

Agroecosystems, particularly systems dominated by annual crops, are ecologically less diverse than most natural ecosystems (Liss er al., 1986). Row crops represent the earliest stage of plant succession with simplified habitat heterogeneity, reduced plant species diversity, uniform plant age structure, and limited resource availability. Crops also have a simplified architectural structure that further reduces resource availability. Although the agroecosystem extends beyond the field margins to include adjacent crops, fence rows, and nonagricultural habitats, the habitat within a field is transient and often is disturbed by tillage and other cropping practices. All of these factors reduce diversity resources and niches in field crops. Consequently, arthropod communities in field crops typically are dominated by fewer species than natural systems. Furthermore, arthropods in row crops tend to be highly mobile opportunistic species with high reproductive rates (i.e., r-selected) that can colonize and quickly take advantage of ephemeral resources (Liss et al., 1986). Herbivore species may be specialists adapted to only one or a few hosts or can be generalists that exploit many hosts. However, predators and parasites often are generalists that can more readily colonize and exploit transient habitats. Arthropod populations are regulated by abiotic factors such as temperature, rainfall, humidity, and soil type and by biotic factors such as predators, parasites, pathogens, and intraspecific competition. The effect of abiotic factors is independent of population density, whereas the effect of biotic factors usually is proportionally greater as population density increases. Natural enemies sometimes fail to regulate pest populations, but intraspecific competition as resources are depleted will always regulate pest populations. Arthropod populations are regulated within limits by a combination of density-independent factors (weather) and imperfectly density-dependent factors such as natural enemies (Milne, 1959). Favorable weather that enhances survivorship or failure of a key natural enemy may relax population regulation, thereby permitting a population to increase to outbreak levels. Berryman (198 1) suggested that populations exhibit distinct outbreak and nonoutbreak (endemic) population configurations, each with a different set of regulating factors. This phenomenon is exemplified by the green cloverworm, Plurhypena scubra (F.), which is a migratory pest of soybean in Iowa (Pedigo et al.,

ARTHROPOD PEST MANAGEMENT

155

1983). Outbreak and nonoutbreak populations are determined by the number of immigrant adults, which is a function of the number of favorable weather systems for migration from the southern United States. In outbreak populations, the first generation often exceeds the economic threshold in soybean with the entomogenous fungus Nomurueu rileyi (Farlow) Samson causing an epizootic that ultimately regulates outbreak populations. In nonoutbreak populations, green cloverworm numbers are below economic thresholds, N. rileyi is insignificant, and predators and parasites become important in regulating second-generation populations. In most instances, favorable weather releases populations to outbreak levels and natural enemies or resource depletion regulates populations below outbreak levels (Milne, 1959).

B. SELECTION PRESSURES Use and misuse of management tactics, particularly insecticides, has induced substantial counterresponses by pest populations and adverse effects on arthropod communities. From the point of view of pest management, adverse effects are categorized as pest population resurgence, secondary pest outbreak (replacement), and pest resistance (Ripper, 1956). Pest resurgence occurs when a population, after having been suppressed with a pesticide, rebounds to numbers greater than before treatment. Secondary pest outbreaks occur when a primary pest is suppressed by a tactic, but another pest of minor importance becomes economically important. Resurgence and replacement represent similar phenomena and are caused by reduction of natural enemies along with the pest by pesticides, direct favorable effect of pesticides on arthropod physiology and behavior, or loss of competitive species (Pedigo, 1989). Although favorable effects of pesticides have been documented for a number of pest species, the primary cause of resurgence and secondary pest outbreak usually is the loss of natural enemies (Luck et al., 1977). Resurgence after initial suppression of lepidopterous pests was observed in soybean after treatment with methyl parathion (0,O-dimethyl-o-p-nitrophenylphosphorothioate) or methomyl [S-methyl-N(methylcarbamoy1)oxythioacetimidate] as a result of removal of natural biotic agents by the insecticide (Shepard et ul., 1977). One well-documented example of replacement occurred in California (United States), where an obscure insect, the cotton leafperforator, Bucculutrix thurberiellu Busck, became a major cotton pest after the use of carbaryl to control the pink bollworm, Pecrinophoru gossypiellu (Saunders), virtually eliminated the natural enemies of the cotton leafperforator (Smith, 1970). Once eliminated, natural enemies species generally reestablish much more slowly than herbivore species (Liss et ul., 1986). Resistance usually is associated with pesticide use and is the ability of individuals to tolerate or avoid factors that would be lethal or reproductively degrading

156 JOE FLJNDERBURK, LEON HIGLEY, AND G. DAVID BUNTIN to most individuals in a normal population (Metcalf, 1982). Development of pesticide resistance is reaching crisis proportions and threatens to undermine IPM program in many cropping systems (Pedigo and Higley, 1992). In 1985, over 400 species of insects and mites were resistant to at least one insecticide, with many species being resistant to several insecticides (NAS, 1986). Pesticide resistance was reviewed by Metcalf (1982), Georghiou and Saito (1983), NAS (1986), and Rousch and McKenzie (1987). Frequent and repeated use of a pesticide places enormous pressure on a pest population for selection of a resistant biotype. Historically, the approach to pesticide resistance has been the deployment of new pesticides to replace previously effective compounds (NAS, 1986). This approach, termed the “pesticide treadmill,” can continue only as long as new compounds with novel chemistries continue to be produced. However, effective pesticides should be viewed as a valuable management resource that should be protected from resistance. Consequently, one goal of IPM is to prevent unnecessary pesticide use to reduce the potential for development of pesticide resistance. Procedures for managing resistant populations include use of selective compounds that preserve natural enemies, use of economic thresholds and other tactics to reduce the number of applications, rotation of pesticides, and providing refugia for the maintenance of susceptible populations (Georghiou and Saito, 1983; Gould, 1987). Although sometimes thought of as separate entities, integrated resistance management and IPM are part of the same continuum of a comprehensive management strategy (Metcalf, 1982). Although development of resistant biotypes is usually associated with insecticides, the phenomenon of biotype development is possible for any management tactic that causes extensive mortality in populations. Development of virulent biotypes also occurs with the use of resistant plants, particularly where resistance is monogenic and antibiotic causing a high degree of mortality (Smith, 1989). The development of virulent biotypes may be delayed by reducing selective pressure on target populations by using practices such as growing mixtures of susceptible and resistant lines, by providing refugia for susceptible biotypes, and most importantly by using plant resistance in an integrated program with biological control and cultural practices (Gould, 1987). Greater use of plant tolerance and polygenic sources of antibiosis also would lessen the potential for biotype development. Cultural practices such as crop rotation may cause genetic shifts in pest populations (Krysan et al., 1986).

C. IMPACTS ON ECOSYSTEMS Agriculture is responsible for greatly increased crop yields and has permitted greatly elevated human population levels. However, there are negative trade-offs to the sociological and economic benefits of improvements in agriculture. Agri-

ARTHROPOD PEST MANAGEMENT

157

cultural activities alter the environment, including the soil, water, air, climate, and biota. Scientific and technological advances provide a buffer by increasing yield, improving the efficiency of production, and reducing the negative impacts on the environment. The balance established between the public and private interests related to agriculture, the environment, food safety, and economics is political. Protecting the environment is a major issue in developed countries and is gaining importance in developing countries. Although there is much rhetoric associated with the issues of environmental protection, laws and regulations generally deal with immediate problems and not long-term concerns (Rindos, 1984).

Development of sustainable agricultural production systems also is a political as well as a technical issue. The goal of sustainable agriculture is to have agricultural systems that can be maintained through time without extraordinary inputs and without significant impacts on the environment. Sustainable systems must provide for stability in production and producer profits, while also maintaining environmental stability. Pests present a particular challenge to sustainable agriculture because the effects of pests on agricultural systems are intrinsically destabilizing. These disruptive effects of pests include the direct effect of pests on production and the effects of management tactics used to manage pests. IPM was developed to provide sustainable solutions to pest problems. The concept and practices of IPM predate those of sustainable agriculture. Funderburk and Higley (1994) contrasted IPM and sustainable agriculture and concluded that IPM is compatible and essential to sustainable agricultural systems. IPM provides reliable procedures for managing pests, and IPM programs address issues of economic and environmental sustainability. Economic sustainability is provided by undertaking management actions only when they are economically justified. Criteria such as the EIL and ET were developed for making management decisions that consider the economic merits of action. Environmental sustainability is provided by procedures that minimize environmentally undesirable management tactics, by the use of multiple tactics, and by the development of decision tools that consider environmental factors (such as environmental EILs) (Higley and Wintersteen, 1992). No tactic or approach can eliminate the impact of pests on agricultural systems. Thus, IPM is not a panacea for solving all pest problems; rather, it is the best approach for providing economically and environmentally sound approaches for reducing the deleterious effects of pests. Any type of management tactic can be used inappropriately in an IPM program, resulting in undesirable economic or ecological consequences. Most emphasis is placed on abuses with pesticides, but there is no compelling reason to believe that pesticides, if used according to IPM theory, are incompatible with sustainable agricultural systems. As previously pointed out, crop losses from arthropods would be great if pesticides were eliminated, thereby increasing land area needed to produce food. This would

158 JOE FUNDERBURK, LEON HIGLEY, AND G. DAVID BUNTIN

lead to further destruction of natural habitat and result in increased depletion of soil, water, and energy resources. Economic and environmental consequences of all management tactics (including pesticides) must be considered in IPM programs. Cultural management tactics can lead to water, soil, or air pollution or result in unnecessary depletion of natural resources. Biological management tactics also can have significant environmental impacts. These impacts include enhancement of the target pest, synergistic interaction with other organisms to enhance pest problems, and attacked nontarget organisms, including endangered species. Howarth (199 1) discussed and reviewed the literature on this subject and provided specific examples of negative environmental impacts from biological management tactics.

V. STATUS AND FUTURE DIRECTIONS OF INTEGRATED PEST MANAGEMENT A. CURRENT STATUS OF INTEGRATED PESTMANAGEMENT The principles of IPM are widely accepted by entomologists, and IPM programs have been developed and implemented for many crops in many countries. Generally, IPM programs have been developed from only rudimentary knowledge of pest ecology, crop response to pests, and economics of management and production. The population dynamics of arthropod pests is complex, with many parameters affecting populations of even the simplest species (Gilbert, 1989). Consequently, the interrelationships between host plants, arthropod herbivores, and their natural enemies are poorly understood, although cultural management, pest-resistant cultivars, and biological management tactics sometimes are developed from such information. As an example, consider IPM practices for one soybean pest species. Soybean originates in Asia, but now is grown in numerous geographical areas worldwide. Many herbivore species and their natural enemies have adapted and inhabit the crop in North and South America. Kogan and Turnipseed (1987) review and discuss the ecology and management of soybean arthropods. The velvetbean caterpillar (Anticursiu gernrnutulis Hubner) is an important leaf-mass-consuming pest. Insecticide applications against velvetbean caterpillar are greatly reduced where conservation of predators is carefully considered in soybean IPM programs. Biological and chemical insecticides that have little impact on populations of the important predator species are available to rapidly and reliably suppress velvetbean caterpillars (Funderburk et ul., 1992). A fungal pathogen, N. rileyi, causes natural mortality, and normally prevents velvetbean caterpillar from reaching the ET in tropical production areas. However, fungicide applications must be geared so as not to interfere with mortality from this pathogen

ARTHROPOD PEST MANAGEMENT

159

(Stansly and Orellana, 1990). Epizootics of N. rileyi usually do not occur in temperate and subtropical regions until populations of velvetbean caterpillar have exceeded the EIL. The example of velvetbean caterpillar is typical of most pests in most crops, and it illustrates that the EIL concept primarily has been used for managing individual pest species with the therapeutic approach. Inputs of pesticides usually are reduced, thereby reducing costs and potential environmental contamination. Indigenous natural enemy populations are conserved through reduction and selective use of pesticides. Tolerance of subeconomic pest densities and maximization of mortality from natural enemies and abiotic factors reduce the risk for pesticide resistance in pest populations. Preventive management tactics are employed for a few less important soybean pests. Commercial cultivars resistant to the potato leafhopper have been developed and are widely employed to prevent economic damage from this pest (Turnipseed, 1977). Risks for seedcorn maggot injury in soybean in the Midwest United States are great when green organic matter is incorporated into the soil, but low when only corn or soybean residues are present (Hammond and Stinner, 1987). In situations where soybean must be planted into a field containing green organic matter, the cultural management strategy of planting with no-tillage practices virtually eliminates the damage potential. The need for integration of preventive tactics into a therapeutic management program that maximizes natural mortality is greatly determined by the pest type involved. Occasional pests, which produce outbreaks sporadically, have a low general equilibrium position relative to their EIL because of natural control factors. Examples of occasional pests in the leaf-mass-consumingguild of soybean are the soybean looper and the green cloverworm. Because therapeutic action is required only occasionally for these pests, ecological backlash is unlikely. For severe and perennial pests, multiple tactics are necessary in a sustainable management system. There are no severe or perennial pests of soybean in most production areas.

B. IMPROVING EFFICIENCY The current EIL concept incorporates much of the knowledge needed for minimizing losses to pests but does not allow for separating consumption differences among size classes of the same arthropod species. This is important because seasonal development of injury is very different from seasonal development of arthropod pests (Fig. 5 ) . Further, the collective effect of other pest species causing similar injury to the crop at the same time is not integrated in current management programs. The current focus in IPM is on pest populations. The future trend in IPM will

160 JOE FUNDERBURK, LEON HIGLEY, AND G. DAVID BUNTIN

Population Density or Host lWW

Figure 5. Stylized drawing of lepidopteran insect population development versus host injury development over time. (Adapted from Hutchins and Funderburk, 1991 .)

be to focus on injury, rather than pest species, the surrogates of injury. Categories of injury provide a framework within which to describe in standard terms the impact that each individual or known population density of a pest has upon a particular physiological function of a crop. Focusing on categories of injury will simplify the research process and allow for improved understanding of the influence of multiple pest injuries on crops. Improved understanding of abiotic influences on biotic stress also will be possible. Incorporation of such refinements will result in more efficient and reliable management of pest injury and enhance adoption of IPM principles and practices. A system of consumption equivalents has been proposed that accounts for differential consumption potential of size classes of a pest (Pedigo et al., 1986). In this system, a consumption equivalent is defined as the amount of injury that could be produced by one pest through its complete life cycle, and injury equivalency as the total number of injury equivalents for a population. Hutchins et al. (1988) expanded the system to include not only different size classes of the same species, but also injury guilds or arthropod species that produce the same injury to a crop. A system considering size classes of the leafmass-consuming guild of soybean in the Southeast United States was developed

ARTHROPOD PEST MANAGEMENT

161

and evaluated. By comparison with the conventional approach, the use of injury guilds enhanced the value of the EIL model and the accuracy of management decisions. The limiting factor in pest management decision making is knowledge. This can be illustrated by contrasting the level of knowledge for leaf-mass and fruitfeeding guilds of soybean in the Southeast United States (Hutchins and Funderburk, 1991). Information on age-specific and species-specific survivorship and on management tactic efficacy is available for the leaf-mass-consuming guild and can be incorporated into an injury guild EIL model. The most serious fruitfeeding pests are the southern green stink bug and Helicoverpa spp. Less important are the green stink bug [Acrosternurnhilare (Say)] and the brown stink bug [ Euschistus servus (Say)]. Important informational shortfalls exist concerning injury to soybean by fruit-feeding pests. Consumption rates for individual life stages or for total development have not been determined experimentally for most species, and the effects of their injury on soybean physiology are unknown. EILs exist for fruit-feeding pests, but these were determined by caging different densities directly on soybean and estimating final yield and quality (Hutchins and Funderburk, 1991).Considerably more knowledge is needed before an injury guild for fruit-feeding pests of soybean can be developed, but rectifying the informational shortfalls undoubtedly would result in improved accuracy of management decisions for these pests. The example of fruit-feeding pests illustrates the limitations of focusing on pest populations and relating densities directly to damage without understanding the injuryldamage relationship. In most production situations, crops are attacked by multiple arthropod, pathogen, weed, and nematode pests. Usually, there is a lack of knowledge of crop response to multiple pest injuries and of the impact of multiple pest injuries on decision making in IPM programs. Injury from multiple pests frequently does not result in interactive influences on crop response over the range of acceptable damage. Influence of velvetleaf [Abutilon theophrasti (Medic)] prevalence and leaf-mass consumption are additive on soybeans in the Midwest United States, and individual management decisions for each pest are optimal for the IPM program (Higgins et al., 1984). Significant pest interactions can greatly impact decision making in IPM programs. Injury by thrips (Frankliniella spp.) to leaves of seedling peanuts in the Southeast United States, even when severe, results in little or no economic damage. However, combined effects of alachlor [2-chloro-2’,6’-diethyl-N-(methoxymethyl) acetanilide] and paraquat [ 1 : l-dimethyl-4,4’-bipyridinum(cation) dichloride] herbicides with moderate to severe thrips injury decrease seed yield and prolong maturity, and can result in substantial economic losses (B. K. Brecke, J. E. Funderburk, and 1. D. Teare, unpublished). The EIL for thrips is lowered when these herbicides are needed for effective weed management. Identifying pest interactions is important for pest management decision mak-

162 JOE FUNDERBURK, LEON HIGLEY, AND G. DAVID BUNTIN ing. Appropriate refinements in IPM programs when significant pest interactions occur will increase productivity and efficiency of the production system. This will enhance confidence of producers for IPM programs and result in greater adoption.

C. EMPLOYING MOREINTEGRATION OF TACTICS Therapy in IPM programs is employed after pest assessment predicts that injury has occurred and economic damage is probable and is used when there is a favorable cost-benefit ratio. Insecticides for use in therapeutic management programs are labeled by EPA based in part on issues of public health and environmental safety. Undesirable side effects on target and nontarget pests are not currently considerations for labeling. Based on the history of IPM practice, the therapeutic approach frequently fails over time when employed against severe and perennial pests (i.e., key pests) whose populations may remain above the economic threshold for extended periods of time. Insecticides frequently are employed in such situations without sufficient integration with other tactics, and if natural mortality from abiotic and biotic factors is low, development of insecticide resistance may result in target and nontarget pests. Although they may not be sustainable, therapeutic approaches without sufficient integration with preventive tactics continue to be employed against such pests. Key pests result from a relatively high market value of the crop or characteristically dense populations of the pests (Pedigo, 1989). Examples include the diamondback moth (Plutella xylosfella L.) and armyworms (Spodopfera spp.). Armyworm species in tomato (Lycopersicon esculenfum Mill.) that are severe pests are S . eridania (Cramer), S . ornithogalli (Guenee), and S . exigua (Hubner) (Schuster et al., 1993). These species sometimes inflict substantial injury to leaf mass, but most damage occurs from feeding on fruit. During periods when population densities exceed the ET, repeated insecticide applications are necessary to prevent economic damage. A similar scenario occurs for diamondback moth in cabbage (Brassica oleracea L. var. capitafa), and producers have relied on using insecticides preventively to suppress populations. Consequently, populations in some geographical areas are resistant to virtually all available insecticides (Cheng, 1988). Examples of other key pests in tomato are sweetpotato whitefly [Bemisia tabaci (Gennadius)] (Fig. 6 ) and leafminer (Liriomyza safivae Blanchard). Again, repeated insecticide applications are needed to prevent economic damage during the extended periods when population densities exceed economic thresholds. The therapeutic approach is not appropriate for pests whose populations, once exceeding the economic threshold, cannot be rapidly and reliably suppressed with insecticides. Examples of such difficult-to-suppress pests are the Hessian fly and the southern corn rootworm (D.undecimpunctata howardii Barber).

ARTHROPOD PEST MANAGEMENT

Organochlorine Insecticide

Synethic Pyrethroid Insecticide

163

Organo Phosphate Detergent

Insecticide

Economic Injury Level

ln

Economic Threshold

C

General Equilibrium Position

C

.-0

-ma

c)

n 0

n

Natural Mortality

-

Conserve Natural Enemies within 8 Around Fields Destroy Old & Volunteer Tomato Plants after Harvest

Spatial & Temporal Separation of Old & New Fields

Time Figure 6. The integrated strategy of Stansly er al. (1991) for managing sweetpotato whitefly and the gemini virus vectored by the sweetpotato whitefly. Pesticides are applied only when needed and are applied within a resistance management strategy.

Greater consideration of ecological principles and long-term economic sustainability is needed in the future to manage severe, perennial, and difficult-tosuppress pests. This can be accomplished by continued use of procedures to minimize unnecessary use of a management tactic (such as EILs and ETs), by better integration of multiple tactics including preventive tactics, and when necessary by the use of resistance management strategies. An example of an IPM strategy for a perennial pest that integrates each of the above-mentioned elements is the sweetpotato whitefly in the Southeast United States (Fig. 6) (Stansly et al., 1991). The pest causes direct injury to tomato and other vegetable crops and also is the vector of a serious tomato gemini virus. Cultural management tactics of destroying old and volunteer tomato plants soon after harvest, and spatial and temporal separation of old and new fields as much as possible, help to reduce initial and subsequent populations of the pest and resulting incidence of tomato gemini virus. Pesticides are applied only after populations exceed established ETs, and they are applied within a resistance management strategy by alternating efficacious insecticides from different chemical classes (i.e., organochlorines, organophosphates, synthetic pyrethroids, etc. )

164 JOE FUNDERBURK, LEON HIGLEY, AND G. DAVID BUNTIN with oils and detergents. Mortality from natural enemies can be very great, and care is taken to conserve natural enemy populations within and around fields. Denholm and Rowland (1992) described the theory and practice of resistance management strategies for sweetpotato whitefly and other pests. Deployment of pest-resistant cultivars frequently occurs without sufficient integration with other forms of management, and pest biotypes unaffected by the antibiotic resistance factors are selected from their populations. Resistance management strategies also may be helpful to sustain resistant cultivars, including genetically engineered crops, such as those containing the B. thuringiensis toxin. (Refer to Section 111, C, 1 for a discussion and example.) Southern corn rootworm is a soil pest of peanut in the Atlantic Coast of the United States. Granular insecticides to suppress the pest must be applied preventively before densities reach economic thresholds in order to be reliably effective. Only fields with high organic matter or clay content, however, are at risk, and preventive application is necessary only in such fields (Brandenburgand Herbert, 1991). When integrated with a partially resistant cultivar, reduced rates of insecticide are adequate. Preventive application of an insecticide usually is not considered to be compatible with IPM philosophy, but this example represents use against a difficult-to-suppress pest in high-probability circumstances integrated with another management tactic. Because of the high costs of product development, testing, and marketing, pesticide companies are recognizing the need for resistance management strategies to ensure sustainability of pesticides against severe and perennial pests. Labels for abamectin (Avermectin B ,) and cyromazine (N-cyclopropyl-1,3,5triazine-2,4,6-triamine)against leafminer on lettuce (Lacfucasariva L.), celery (Apiurn graveolens L.), and tomato require alternating these two insecticides as a resistance management strategy. In addition to public health and environmental safety, more emphasis needs to be placed by EPA on regulations that promote sustainability of pesticide effectiveness. Additional regulations that require a pesticide be applied by prescription as verified by a specialist may be necessary to prevent abuse in some cases. Regulation by EPA of pest-resistant crop cultivars, especially genetically engineered crops, is needed to prevent abuse and promote their sustainability for management programs. Higley et al. (1992) discuss laws and regulations related to current pesticide policies, and address inadequacies of policy in relation to development of pest resistance to pesticides and pest-resistant crop cultivars.

VI. CONCLUSIONS The concept of IPM arose out of the pesticide crisis and originally embraced integrated insecticide use with natural enemies. The modern IPM concept has

ARTHROPOD PEST MANAGEMENT

165

broadened and incorporated public and private interests related to agricultural production, the environment, and food supply and quality. Programs are developed from information on pest ecology, crop response to pests, and economics of management and production. Important considerations are the economic sustainability of agricultural production systems, sustainability of management tactics and management programs, and efficient utilization of natural resources. The focus in current IPM programs is on pest populations, and management activities usually are directed against individual pest species. The therapeutic approach is most commonly used for managing pest damage. This approach is cost-effective and has proved successful for most pests, especially occasional pests. These programs generally have been developed from only rudimentary knowledge of pest ecology, crop response to pests, and economics of production and management. However, tolerance of subeconomic pest densities and maximization of mortality from natural enemies and abiotic factors reduce the risk for environmental contamination and development of pesticide resistance in pest populations. The scientific evidence indicates that the food supply produced with new chemical pesticides is safe. IPM is widely accepted as the optimal approach for dealing with the deleterious effects of pests in all production situations. Sustainable, effective IPM strategies have not been successfully employed against many severe, perennial, and difficult-to-suppress pests. There is an overreliance on individual management tactics against these pests in current management programs. Chemical pesticides are most frequently abused, but biological pesticides, pest-resistant cultivars, and even cultural management tactics also sometimes are overused. Future IPM programs for these pests should continue use of EILs and ETs to minimize unnecessary use of a management tactic, but should also employ more integration of multiple tactics, including preventive tactics. When still needed on a frequent basis, pesticides should be used in a manner to reduce the potential for development of pesticide resistance. Although future IPM efforts must continue to be directed at managing pest populations, more emphasis on understanding plant response to pest injury is needed. Categorizing pest injury based on plant physiological response will simplify the research process and allow for improved accuracy of management decisions for pest guilds or species that produce the same injury to a crop. There is a lack of knowledge of the impact of multiple pest injuries on decision making in current IPM programs. Categorizing pest injury also will improve the ability to identify significant interactions between multiple arthropod, pathogen, weed, and nematode pests. This will enhance productivity and efficiency of the production system and will enhance confidence of producers for IPM programs and result in greater adoption. Of all approaches developed to address pest problems, only IPM has the scientific and practical basis to provide long-term, sustainable solutions. This is not to say that IPM is without deficiencies. Viable IPM programs still are lacking for many important pests, and IPM programs are not as widely adopted as is

166 JOE FUNDERBURK, LEON HIGLEY, ~ I N DG. DAVID BUNTIN desirable. In part, these problems reflect the legacy of the pesticide era, and a desire to control pests rather than manage them. Also, the substantial research requirements in developing IPM programs and the knowledge required in using IPM are barriers to its wider acceptance. Nevertheless, with continued pest problems and greater constraints on our ability to manage pest populations (because of pest resistance, fewer pesticides, and more restricted pesticide use), the importance of IPM is likely to increase in the future. Ultimately, management efforts must address the entire array of stressors affecting plants and their interactions. Consequently, greater integration of management tactics in IPM and more integrated understanding of the impact of all types of plant stressors are needed to move IPM toward this goal of sustainable stress management.

REFERENCES Bardner, R., and Fletcher, K. E. (1974). Insect infestations and their effects on the growth and yield of field crops: A review. Bull. Enrornol. Res. 64, 141-160. Beall, G. A., Bruhn, C. M., Craigmill, A. L., and Winter, C. K. (1991). Pesticides and your food: How safe is “safe”? Culif. Agric. 45(4), 4- 11. Bechinski, E. J., Buntin, G. D., Pedigo, L. P., and Thorvilson, H. G. (1983). Sequential count and decision plans for sampling green cloverworm (Lepidoptera: Noctuidae) larvae in soybean. J. Econ. Enromol. 76, 806-812. Berryman, A. A. (198 I). “Population Systems: A General Introduction.” Plenum, New York. Binns, M. R., and Nyrop, J. P. (1992). Sampling insect populations for the purpose of IPM decision making. Annu. Rev. Entomol. 37,427-453. Bliss, C. I., and Fisher, R. A. (1953). Fitting the negative binomial distribution to biological data and a note on the efficient fitting of the negative binomial. Biomerrics 9, 176-200. Boethel, D. J., and Eikenbary, R. D. (1986). “Interactions of Plant Resistance and Parasitoids and Predators of Insects.” Ellis Horwood, Chichester, England. Boote, K. (1981). Concepts for modeling crop response to pest damage. Pup.-Am. SOC.Agric. Eng., 81-4007. Brandenburg, R. L., and Herbert, D. A., Jr. (1991). Effect of timing on prophylactic treatments for southern corn rootworm (Coleoptera: Chrysomelidae) in peanut. J. Econ. Enromol. 84, 1894- 1898.

Brandenburg, R. L., and Kennedy, G. G. (1982). Intercrop relationship and spider mite dispersal in a cordpeanut agroecosystem. Enromol. Exp. Appl. 32,269-271. Broersma, D. B., Bernard, R. L., and Luckman, W. H. (1972). Some effects of soybean pubescence on populations of the potato leafhopper. J. Econ. Enromol. 65, 78-82. Buntin, G. D. (1991). Effect of insect damage on the growth, yield, and quality of sericea lespedeza. J. Econ. Enfomol. 84, 277-284. Burleigh, J. G. (1972). Population dynamics and biotic controls of the soybean looper in Louisiana. Environ. Enromol. 1, 290-294. Carlson, G. A. (1970). A decision theoretic approach to crop disease prediction and control. Am. J . Agric. Econ. 52, 216-223. Chandler, L. D.,Ward, C. R., Lyle W. M., and Bynum, E. D. (1977). Effects of various irrigation schedules on population densities of Banks grass mite (Oligonychus prarensis) and its predatory arthropods in corn, the Texas High Plains. Prog. Rep.-Tex. Agric. Exp. Sfn. 3443c.

ARTHROPOD PEST MANAGEMENT

167

Cheng. E. Y.(1988). Problems of control of insecticide-resistantPlurella xylostella. Pesric. Sci. 23, 177-188. Cherry, R. H. (1984). Flooding to control the grub Ligyrus subrropicas (Coleoptera: Scarabaeidae) in Florida sugarcane. J. Econ. Enromol. 77, 254-257. Coaker, T. H. (1980). Insect pest management in Brassica crops by inter-cropping. I.O.B.C. W.P.R.S. Bull. 3, 117- 125. Cochran, W. G. (1977). “Sampling Techniques,” 3rd ed. Wiley, New York. Cox, T. S., and Hatchett, J. H. (1986). Genetic model for wheatltlessian fly (Diptera: Cecidomyiidae) interactions: Strategies for deployment of resistance genes in wheat cultivars. Environ. Enromol. 15,24-31. DeBach P., Rosen, D., and Kennetth, C. E. (1971). Biological control of coccids by introduced natural enemies. I n “Biological Control” (C. B. Huffaker, ed.), pp. 165-194. Plenum, New York. Denholm, I., and Rowland, M. W. (1992). Tactics for managing pesticide resistance in arthropods: Theory and practice. Annu. Rev. Enromol. 37,91-1 12. Falcon. L. A. (1985). Development and use of microbial insecticides. I n “Biological Control in Agricultural IPM Systems” (M.A. Hoy and D. C. Herzog, eds.), pp. 229-242. Academic Press, Orlando, FL. Fenemore, P. G. (1982). “Plant Pests and Their Control.” Butterworth, Wellington, NZ. Foster, R. E., Tollefson, J. J., Nyrop, J. P., and Hein, G. L. (1986). Value of adult corn rootworm (Coleoptera: Chrysomelidae)population estimates in pest management decision making. J. Econ. Entomol. 79, 303-310. Fowler, G. W.. and Lynch, A. M. (1987). Sampling plans in insect pest management based on Wald’s sequential probability ratio test. Environ. Entomol. 16, 345-354. Frisbie, R. E., and Walker, J. K. (1981). Pest management systems for cotton insects. I n “Handbook of Pest Management in Agriculture” (D. Pimentel, ed.), Vol. 3, pp. 187-202. CRC Press, Boca Raton. FL. Funderburk, J. E., and Brandenburg, R. L. (1993). Management of arthropod pests. I n “Peanut Health Management” (H. A. Melouk and F. S. Shokes, eds.). Am. Phytopathol. SOC. Press, St. Paul, MN (in press). Funderburk, J. E., and Higley, L. G. (1994). Managing arthropod pests. I n “Sustainable Agriculture: The New Conventional Agriculture” (J. L. Hatfield and D. L. Karlen, eds.), pp. 199227. CRC Press, Boca Raton, FL (in press). Funderburk, J., Maruniak, J., Boucias, D., and Garcia-Candeo, A. (1992). Efficacy of baculovimses and their impact on pest management programs. In “Pest Management in Soybean” (L. G. Copping, M. B. Green, and R. T. Rees, eds.), pp. 88-97. Elsevier, London and New York. Geier, P, W. (1966). Management of insect pests. Annu. Rev. Enromol. 11,470-490. Georghiou, G. P., and Saito, T. (1983). “Pest Resistance to Pesticides.” Plenum, New York. Gilbert, N. ( 1989). “Biometrical Interpretation: Making Sense of Statistics in Biology.” Oxford Univ. Press, New York. Gillespie, A. T. (1988). Use of fungi to control pests of agricultural importance. In “Fungi in Biological Control Systems” (M. N. Burge, ed.). pp. 37-60. Manchester Univ. Press, ManChester, UK. Goodell, G. (1984). Challenges to internationalpest management research and extension in the Third World: Do we really want IPM to work? Bull. Enromol. Soc. Am. 30, 18-26. Gould, F. ( 1986). Simulation models for predicting durability of insect-resistantgermplasm: Hessian fly (Diptera: Cecidomyiidae)-resistant winter wheat. Environ. Enromol. 15, 11-23. Gould, F. (1987). Ecological genetics and integrated pest management. In “Agrcecology” (C. R. Carroll, J. H. Vandermeer, and P. M. Rosset, eds.), pp. 441-458. McGraw-Hill, New York. Hammond. R. B., and Pedigo, L. P. (1982). Determination of yield-loss relationships for two soy-

168 JOE FUNDERBURK, LEON HIGLEY, AND G. DAVID BUNTIN bean defoliators by using simulated insect-defoliation techniques. J . Econ. Enromol. 75, 102-107. Hammond, R. B., and Stinner, B. R. (1987). Seedcorn maggots (Diptera: Anthomyiidae) and slugs in conservation tillage systems in Ohio. J . Econ. Enromol. 80, 680-684. Higgins, R. A., Pedigo. L. P., and Staniforth, D. W. (1984). Effect of velvetleaf competition and defoliation simulating a green cloverworm (Lepidoptera: Noctuidae) outbreak in Iowa on indeterminate soybean yield, yield components, and economic decision levels. Environ. Enromol. 13,917-925. Higley, L. G. (1992). New understandings of soybean defoliation and their implications for pest management. In “Pest Management of Soybean” (L.G. Copping, M. 9. Green, and R. T. Rees, eds.), pp. 56-65. Elsevier, London and New York. Higley, L. G., and Pedigo, L. P. (1991). Soybean yield responses and intraspecific competition from simulated seedcorn maggot injury. Agron. J . 83, 135-139. Higley, L. G., and Wintersteen, W. K. (1992). A novel approach to environmental risk assessment of pesticides as a basis for incorporating environmental costs into economic injury levels. Am. Eniomol. 38, 34-39. Higley, L. G., Zeiss, M. R., Wintersteen, W. K., and Pedigo, L. P. (1992). National pesticide policy: A call for action. Am. Enromol. 38, 139- 146. Higley, L. G.,Browde, J. A., and Higley, P. M. (1993). Moving towards new understandings of biotic stress and stress interactions. In “International Crop Science I” (D. R. Buxton er a/., eds.). Crop Sci. Soc.Am., Madison, WI (in press). Howarth, F. G. (1991). Environmental impacts of classical biological control. Annu. Rev. Enromol. 36,485-509. Hoy, C. W. (1991). Variable intensity sampling for proportion of plant infested with pests. J. Econ. Enrornol. 84, 148-157. Hoy, C. W., Jennison, C., Shelton, A. M., and Andaloro, J. T. (1983). Variable intensity sampling: A new technique for decision making in cabbage pest management. J. Econ. Enromol. 76, 139- 143. Hoy, M. A., and Conley, J. (1989). Propargite resistance in Pacific spider mite (Acari: Tetranychidae): Stability and mode of inheritance. J . Econ. Enromol. 82, 11- 16. Hoy, M. A., Conley, J., and Robinson, W. (1988). Cyhexatin and fenbutatin-oxide resistance in Pacific spider mite (Acari: Tetranychidae): Stability and mode of inheritance. J. Econ. Enromol. 81-57-64. Huffaker, C. 9 . (1985). Biological control in integrated pest management: An entomological perspective. In “Biological Control in Agricultural IPM systems” (M. A. Hoy and D. C. Herzog, eds.), pp. 13-23. Academic Press, Orlando. FL. Huffaker, C. B., and Smith, R. F. (1980). Rationale, organization, and development of a national integrated pest management project. In “New Technology of Pest Control” (C. B. Huffaker, ed.), pp. 1-24. Wiley, New York. Hull, L. A., and Beers, E. H. (1985). Ecological selectivity: Modifying chemical control practices to preserve natural enemies. In “Biological Control in Agricultural IPM systems” (M. A. Hoy and D. C. Herzog, eds.), pp. 103-122. Academic Press, Orlando, FL. Hutchins, S . H., and Funderburk, J. E. (1991). Injury guilds: A practical approach for managing pest losses to soybean. Agric. Zoo/. Rev. 4, 1-21. Hutchins, S . H., Higley, L. G., and Pedigo, L. P. (1988). Injury equivalency as a basis for developing multiple-species economic injury levels. J. Econ. Enromol. 81, 1-8. Iwao, S . (1968). A new regression method for analyzing the aggregation pattern of animal populations. Res. Popul. Ecol. 10, 1-30. Iwao, S . (1975). A new method of sequential sampling to classify populations relative to a critical density. Res. Popul. Ecol. 16, 281-288. 8

ARTHROPOD PEST MANAGEMENT

169

Jensen, R. L., Newsom, L. D., and Gibbens, J. (1974). The soybean looper: Effects of adult nutrition on oviposition, mating frequency, and longevity. J. Econ. Enromol. 67,467-470. Karandinos. M. G. (1976). Optimum sample size and comments on some published formulae. Bull. Entomol. SOC.Am. 22,417-421. Khush, G. S., and Brar, D. S. (1991). Genetics of resistance to insects in crop plants. Adv. Agron. 45,223-274.

Klun, J. A., Tipton, C. L., and Brindley, T. A. (1967). 2,4-Dihydroxy-7-methyoxy-I ,Cbenzoxazin3-one (DIMBOA), an active agent in the resistance of maize to the European corn borer. J. Econ Enromol. 60,1529-1533. Knipling. E. F. (1979). The basic principles of insect population suppression and management. US., Dept. Agric.. Agric. Handb. 512. Kogan, M., and Herzog, D. C. (1980). “Sampling Methods in Soybean Entomology.” SpringerVerlag, New York. Kogan, M., and Turnipseed, S. G. (1987). Ecology and management of soybean arthropods. Annu. Rev. Enromol. 32, 507-538. Krysan, J. L., Foster, D. E., Branson, T. F., Ostlie, K. R., and Cranshaw, W. S. (1986). Two years before the hatch: Rootworms adapt to crop rotation. Bull. Enromol. SOC. Am. 32, 250-253. Kuno, E. (1991). Sampling and analysis of insect populations. Annu. Rev. Entomol. 36, 285-304. Lamp, W. 0. (1991). Reduced Empoasca fubae (Homoptera: Cicadellidae) density in oat-alfalfa intercrop systems. Environ. Enromol. 20, 118-126. Lance, D. R., and Sutter. G. R. (1990). Field-cage and laboratory evaluations of semiochemicalbased baits for managing western corn rootworm (Coleoptera: Chrysomelidae). J. Econ. Enromol. 83, 1085- 1090. Law, S. E., Cooper, S. C., and Oetting, R. D. (1992). Advances in air-assisted electrostatic crop spraying of conductive pesticides. Trans. ASAE 92-1062. Legg, D. E., Shufran, K. A., and Yeargan, K. V. (1985). Evaluation of two sampling methods for their influence on the population statistics of alfalfa weevil (Coleoptera: Curculionidae) larva infestations in alfalfa. J. Econ. Enromol. 78, 1468- 1473. Levine, E., and Oloumi-Sadeghi, H.(1991). Management of diabroticite motworms in corn. Annu. Rev. Entomol. 36, 229-256. Liss, W. J., Gut. L. J., Westigard, P. H., and Warren, C. E. (1986). Perspectives on arthropod community structure, organization, and development in agricultural crops. Annu. Rev. Enromol. 31,455-478.

Luck, R. F., van den Bosch, R., and Garcia, R. (1977). Chemical insect control-A troubled strategy. BioScience 27, 606-61 1. Maxwell, F. G., and Jennings, P. R. (1980). “Breeding Plants Resistant to Insects.” Wiley, New York . Metcalf, R. L. (1982). Insecticides in pest management. I n “Introduction to Insect Pest Management” (R. L. Metcalf and W. H. Luckman, eds.), 2nd ed., pp. 217-278. Wiley, New York. Metcalf, R. L., Ferguson, J. E., Lampman, R., and Andersen, J. F. (1987). Dry cucurbitacincontaining baits for controlling diabroticite beetles (Coleoptera: Chrysomelidae). J . Econ. Enromol. 80, 870-875.

Milne, A. (1959). The natural control of insect populations. Can. Enromol. 89, 193-213. Morgan, J.. Wilde, G., and Johnson, D. (1980). Greenbug resistance in commercial sorghum hybrids in the seedling stage. J . Econ. Enromol. 73, 510-514. Morris, R. F. (1955). The development of sampling techniques for insect defoliators, with particular reference to the spruce budworm. Can. J . Zool. 33,225-294. Morris, R. F. (1960). Sampling insect populations. Annu. Rev. Enromol. 5 , 243-264. National Academy of Sciences (NAS) (1986). “Pesticide Resistance: Strategies and Tactics for Management.” National Academy Press, Washington, DC.

170 JOE FUNDERBURK, LEON HIGLEY, AND G. DAVID B W I N Newsom, L. D., and Herzog, D. C. (1977). Trap crops for control of soybean pests. La.Agric. 20, 14-15. Nyrop, J. P., and Binns, M. R. (1991). Quantitative methods for designing and analyzing sampling programs for use in pest management. In “Handbook of Pest Management in Agriculture” (D. Pimentel, ed.), 2nd ed., pp. 67-132. CRC Press, Boca Raton, FL. Nyrop, J. P., Foster, R. E., and Onstad, D. W. (1986). Value of sample information in pest control decision making. J. Econ. Enromol. 79, 1421-1429. Nyrop, J. P.. Shelton, A. M., and Theunissen, J. (1989). Value of a control decision rule for leek moth infestations in leek. Entomol. Exp. Appl. 53, 167-176. Oloumi-Sadeghi, H., Zavaleta, L. R., Lamp, W. O., h b r u s t , E. J., and Kapusta, G. (1987). Interactions of the potato leafhopper (Homoptera: Cicadellidae)with weeds in an alfalfa ecosystem. Environ. Enromol. 16, 1175- 1180. Onstad, D. W., and Shoemaker, C. A. (1984). Managementof alfalfa and the alfalfa weevil (Hypera posrica): An example of systems analysis in forage production. Agric. Sysr. 14, 1-30. Ottens, R. J., and Todd, J. W. (1979). Effects of host plant on fecundity, longevity, and oviposition rate of a whitefringed beetle. Annu. Enromol. SOC.Am. 72, 873-879. Painter, R. H. (1951). “Insect Resistance in Crop Plants.” Univ. of Kansas Press, Lawrence. Panda, N. (1979). “Principles of Host-Plant Resistance in Insect Pests.” Allanheld, Osmun, and Universe Books, New York. Panda, N., and Heinrichs, E. A. (1983). Levels of tolerance and antibiosis in rice varieties having moderate resistance to the brown planthopper, Nilaparvata lugens (Still) (Hemiptera: Delphacidae). Environ. Enromol. 12, 1204- 1214. Patterson, F. L., Foster, J. E., Ohm, H. W., Hatchett, J. H., and Taylor, P. L. (1992). Proposed system of nomenclaturefor biotypes of Hessian fly (Diptera: Cecidomyiidae) in North America. J. Econ. Enromol. 85,307-31 1. Pedigo, L. P. (1989). “Entomology and Pest Management.” Macmillan, New York. Pedigo, L. P., and Higley, L. G. (1992). A new perspective on the economic injury level concept and environmental quality. Am. Enromol. 38, 12-21. Pedigo, L. P., and van Schaik, J. W. (1984). Time-sequential sampling: A new use of sequential probability ratio test for pest management decisions. Bull. Enromol. SOC. Am. 30, 32-36. Pedigo, L. P., Higgins, R. A., Hammond, R. B., and Bechinski, E. J. (1981). Soybean pest management. In “Handbook of Pest Management in Agriculture” (D. Pimentel, ed.), Vol. 3, pp. 417-537. CRC Press, Boca Raton, FL. Pedigo, L. P., Bechinski, E. J., and Higgins, R. A. (1983). Partial life tables of the green cloverworm (Lepidoptera: Noctuidae) in soybean and a hypothesis of population dynamics in Iowa. Environ. Enrornol. 12, 186- 195. Pedigo, L. P., Hutchins, S. H., and Higley, L. G. (1986). Economic injury levels in theory and practice. Annu. Rev. Enromol. 31, 341 -368. Pedigo, L. P., Higley, L. G., and Davis, P. M. (1989). Concepts and advances in economic thresholds for soybean entomology. In “Proceedings of the Fourth World Soybean Research Conference” (A. J. Pascale, ed.). Vol. 3, pp. 1487-1493. Realizaci6n. Buenos Aim, Argentina. Perkins, J. H. (1982). “Insects, Experts, and the Insecticide Crisis.” Plenum, New York. Peterson, R. K. D., Danielson, S . D., and Higley, L. G. (1992). Photosynthetic responses of alfalfa to actual and simulated alfalfa weevil (Coleoptera: Curculionidae) injury. Environ. Entomol. 21,501-507. Pike, K. S., Allison, D. W., Low, G., Bishop, G. W., Halbert, S. E., and Johnston, R. L. (1990). Cereal aphid vectors: A western regional (USA) monitoring system. In “World Perspectives on Barley Yellow Dwarf” (P. A. Burnett, ed.), pp. 282-285. CIMMYT, Mexico. Pimentel, D. (1986). Agroecology and economics. In “Ecological Theory and Integrated Pest Management Practice” (M. Kogan, ed.), pp. 299-319. Wiley, New York.

ARTHROPOD PEST MANAGEMENT

171

Poston, F. L.. Pedigo, L. P., and Welch, S. M. (1983). Economic injury levels: Reality and practicality. Bull. Enromol. SOC.Am. 29,49-53. Raupp, M. J., Davidson, J. A., Koehler, C. S., Sadof, C. S., and Reichelderfor. K. (1987). Decision-makingconsiderations for aesthetic damage caused by pests. Bull. Enromol. SOC.Am. 34,27-32. Ridgeway. R. L., and Vinson, S. B. (1977). “Biological Control by Augmentation of Natural Enemies.” Plenum, New York. Rindos, D. (1984). “The Origins of Agriculture: An Evolutionary Perspective.” Academic Press, Orlando, FL. Ripper, W. E. (1956). Effect of pesticides on the balance of arthropod populations. Annu. Rev. Enromol. 1, 403-436. Risch, S . J., Andow, D., and Altieri, M. A. (1983). Agroecosysterndiversity and pest control: Data, tentative conclusions, and new research directions. Environ. Enromol. 12, 625-629. Rousch, R. T., and McKenzie, J. A. (1987). Ecological genetics of insecticide and acaricide resistance. Annu. Rev. Enromol. 32, 361-380. Ruesink, W. G. (1980). Introduction to sampling theory. In “Sampling Methods for Soybean Entomology” (M. Kogan and D. C. Herzog, eds.), pp. 61-78. Springer-Verlag. New York. Russell, G. E. (1978). “Plant Breeding for Pest and Disease Resistance.” Buttenvorth. Boston. Schuster. D. J., Funderburk, J. E., and Stansley, P. A. (1993). Selected IPM projects. In “Biological Control and IPM: The Florida Experience” (D.Rosen, F. Bennett, and J. Capinera, eds.). Intercept Ltd., Andover, Hampshire, UK (in press). Shelton, A . M., and Trurnble, J. T. (1991). Monitoring insect populations. I n “Handbook of Pest Management in Agriculture” (D. Pirnentel, ed.), 2nd ed., pp. 45-62. CRC Press, Boca Raton, FL. Shelton, A. M., Nyrop, J. P., Seaman, A., and Foster, R. E. (1986). Distribution of European corn borer (Lepidoptera: Pyralidae) egg masses and larvae on sweet corn in New York. Environ. Enromol. 15, 50 I - 506. Shepard, M., Camer, G. R.. and Turnipseed, S. G. (1977). Colonization and resurgence of insect pests of soybean in response to insecticides and field isolation. Environ. Enromol. 6 , 501 -506. Smith, C. M. (1989). “Plant Resistance to Insects.” Wiley, New York. Smith, R. F. (1970). Pesticides: Their use and limitations in pest management. I n “Concepts of Pest Management.” (R. L. Rabb and F. E. Guthrie, eds.), pp. 103- 113. North Carolina State University, Raleigh. Southwocd, T. R. E. (1978). “Ecological Methods.” Chapman & Hall, London. Stansly, P. A., and Orellana, G. J. (1990). Field manipulation of Nomuraea rileyi (Moniliales: Moniliaceae): Effects on soybean defoliators in coastal Ecuador. J. Econ. Enromol. 83, 21932195. Stansly, P. D., Schuster, D. J., and Leibee, G. L. (1991). Management strategies for the sweetpotato whitefly. I n “Proceedings of the Florida Tomato Institute.” (C. S. Vavrina, ed.), Spec. Ser. VEC-01, pp. 20-42. Fla. Agric. Exp. Stn., Gainesville. Starks, K. J., Muniappan, 1. R., and Eikenbary, R. D. (1972). Interactions between plant resistance and parasitism against greenbug on barley and sorghum. Ann. Enromol. Sor. Am. 65, 650- 655. Stern, V. M., Smith, R. F., van den Bosch, R., and Hagen, K. S. (1959). The integrated control concept. Hilgardia 22, 8 1 - 101. Stone, J. D., nd Pedigo, L. P. (1972). Development and economic-injury level of the green cloverworm on soybean in Iowa. J. Econ. Enromol. 65, 197-201. Stoner, K. A. (1990). Glossy leaf wax and plant resistance to insects in Brassica olerarea under natural infestations. Environ. Enromol. 19, 730-739. Strickland. A. H. (1961). Sampling crop pests and their hosts. Annu. Rev. Enromol. 6 , 201-220.

172 JOE FUNDERBURK, LEON HIGLEY, AND G. DAVID BUNTIN Tammes, P. M. L. (1961). Studies of yield losses. 11. Injury as a limiting factor of yield. Tijdschr. Plantenziekten 67, 257-263. Taylor, L. R. (1961). Aggregation, variance, and the mean. Nature (London) 189,732-735. Taylor, N. L. (1956). Pubescence inheritance and leafhopper resistance relationships in alfalfa. Agron. J . 48, 78-81. Teigen. J. B., and Vorst, J. J. (1975). Soybean response to stand reduction and defoliation. Agron. J . 67,813-816. Turnipseed, S. G . (1977). Influence of trichome variations on populations of small phytophagous insects in soybean. Environ. Entomol. 6, 815-817. Wald, A. (1947). “Sequential Analysis.” Wiley, New York. Walker, J. K., Niles, G. A., Gannaway, J. R., Bradshaw, R. D., and Glodt, R. E. (1976). Narrow row plantings of cotton genotypes and boll weevil damage. J. Econ. Entomol. 69, 249-253. Walker, J. K.,Gannaway, J. R., and Niles, G. A. (1977). Age distribution of cotton bolls and damage from boll weevil. J . Econ. Entomol. 70, 5-8. Way, M. J., and Khoo, K. C. (1992). Role of ants in pest management. Annu. Rev. Entomol. 37, 479-504. Welter. S.C. (1989). Arthropod impact on plant gas exchange. In “Insect-Plant Interactions” (E. A. Bernays, ed.), pp. 135-150. CRC Press, Boca Raton, FL. Wilkerson, G. G., Modena, S. A., and Coble, H. D. (1991). HERB: Decision model forpostemergence weed control in soybean. Agron. J . 83,413-417. Wintersteen, W. K., and Higley. L. G. (1993). Advancing IPM systems in corn and soybeans. In “Successful Implementation of Integrated Pest Management for Agricultural Crops” (A. R. Leslie and G. W. Cuperus, eds.), Lewis Publ., Chelsea, MI, pp. 9-32. Zilberman, D., Schmitz, A., Casterline, G., Lichtenberg, E., and Siebert, J. B. (1991). The economics of pesticide usage and regulation. Science 253, 518-522.

ACCUMULATION OF CADMIUM IN CROP PLANTS AND ITS CONSEQUENCES TO HUMAN HEALTH George J. Wagner Plant Physiology/Biochemistry/MolecularBiology Program Department of Agronomy University of Kentucky Lexington, Kentucky 40546

I. Introduction 11. Occurrence of Cadmium in the Agricultural Environment 111. Accumulation of Cadmiurn: Whole-Plant Studies IV.Accumulation of Cadmium: Cellular and Subcellular Aspects V. Contribution of Agricultural Products to Human Intake of Cadmium VI. Health Consequences of Cadmium Intake in Humans VII. Attempts to Manipulate Plants to Reduce the Cadmium Content of Crops: Promise and Problems VIII. Summary References

I. INTRODUCTION Cadmium is principally dispersed in natural and agricultural environments through human activities. This trace metal is a pollutant and potential toxin that has no known function in any biological organism. Of primary concern is its transfer from vegetable products of agriculture to the human diet. It is widely concluded that vegetable foods contribute 270% of Cd intake in humans (Ryan et af., 1982). The principal goal of this review is to outline our current understanding of processes involved in the accumulation of the pollutant metal Cd in crop plants and how such accumulation relates to human health. Some emphasis will be placed on comparison of experimental studies that have used high-Cd exposure conditions versus low-level exposure more commonly found in natural and agricultural environments. Some speculation will be offered regarding the promise 173 Advances in A p n m n y , Volume 5J

Copyright 0 1993 by Academic Press, Inc. All rights of reproduction in any form reserved

174

GEORGE J. WAGNER

and problems associated with present and future attempts to manipulate plants to reduce the Cd content of the human diet. Each section will begin with summary statements concerning key aspects, followed by discussion of these aspects and others that are considered relevant.

11. OCCURRENCE OF CADMIUM IN THE AGRICULTURAL ENVIRONMENT Several accepted generalizations regarding Cd in the agricultural environment are: A. Agricultural soils are mainly contaminated with Cd from phosphatic fertilizer use, sewage sludge dispersal, and atmospheric deposition. B. Cd contamination of the agricultural environment and the human food chain is thought to be increasing.

Cadmium is a group IIB transition element that is relatively rare in the earth’s crust, being the 32nd most abundant element (Wood and Wang, 1983). This trace metal is generally associated with Zn in a ratio of about 1: 445. Though its ion classification is somewhat debated (Nieboer and Richardson, 1980; Andersen, 1984), Cd is generally considered a soft ion (class B) with a ligand or donor atom preference sequence for ligands of I->B ->Cl ->F- and a preference for donor atoms in ligands of S>N>O. Its strong preference for SH- containing ligands (particularly polythiols) is undoubtedly the principal basis for its toxicity in biological systems. The principal thiol targets of low levels of Cd are thought to be Zn metalloenzymes, membrane phospholipids, and perhaps oxidative phosphorylation (Vallee and Ulmer, 1972). Zn occurs in various metalloenzymes where it coordinates with amino acids to enhance catalysis at the active site. Indeed, Zn is the only trace metal that is known to be essential in the function of at least one enzyme of each of the enzyme classes established by the International Union of Biochemistry (Vallee and Auld, 1990). Well-known examples are carbonic anhydrase, carboxypeptidases, alcohol dehydrogenase, and RNA and DNA polymerases. Zn-finger proteins are known to play a key role in gene regulation (Kaptein, 1991). The relationships between Cd and Zn in plants regarding uptake, translocation, and remobilization of these ions are complex (Smith and Brennan, 1983). In natural soils, Cd content varies with parent rock type. It is high in soils on basaltic rock and low in soils on granitic rock (Ryan et al., 1982). Unpolluted soils generally contain 0.01 to 30 pg/Cd/gm dry wt with a mean (variously stated) of 0.06 to 0.5 pglgm fresh wt (Hopps, 1974; Page et af., 1981; Ryan et al., 1982). Higher levels can occur in carboniferous soils and as a result of

CADMIUM IN CROPS AND EFFECTS ON HUMAN HEALTH

175

contamination from mining or smelting activity. An example of the former is the detection of 30 pg Cd/gm in natural soils derived from Monterey shale in the coastal ranges of southern California (Page et al., 1981). An example of the latter is found in the Kempen area of The Netherlands, where a 350-km2 area contains levels of soil Cd in excess of 1.O pg/gm, with the most polluted areas containing > 10 pg/gm (Copius Peereboom-Stegman and Copius Peereboom, 1989). Dietary Cd intake alone is estimated to be increased in these Kempen areas by 2.6- and over 7-fold, respectively, over that in unpolluted areas of The Netherlands. Cadmium is a by-product metal produced mainly during Zn, Cu, and Pb primary-product extraction and refinement. Because it is obtained in ample supply and is inexpensive, little refined Cd has been recycled for economic reasons. Increasing environmental awareness is leading to greater effort to recycle this metal. The lack of recycling and dissipation through use of Cd-containing products and fuels contribute to contamination of general soils, principally via atmospheric fallout near primary air emission sources (mining and refining, municipal waste incinerators, fossil fuel combustion sources) and process waste disposal sites (Nriagu, 1980; Nriagu and Pacyna, 1988). As shown in Table I, on a global scale, environmental release of Cd from anthropogenic sources may be about 10-fold that predicted from natural sources (Nriagu, 1980; VogeliLange, 1989). Ayres (1992) has pointed out that similar or higher anthropogenicto-natural release ratios are apparent for Pb, Zn, Cu, and Sb. He argues that without pollution abatement through governmental regulation, serious global pollution trends will continue. In contrast, and somewhat heartening perhaps, is a recent report indicating that tropospheric contamination of Pb and Cd in the Northern Hemisphere has decreased since the late 1960s (Boutron et al., 1991). Worldwide production of Cd has grown from less than 10 tons/year in the early 1900s to over 17,000 tons/year in the late 1970s (Elinder, 1986). It is generally concluded that soil contamination with Cd will continue to increase as long as Cd is released into the environment (Ryan et al., 1982). The four principal Cd-consuming nations are the United States, Germany, Britain, and Japan. Its main uses are in electroplating, pigment manufacture (cadmium yellow and cadmium orange, CdS), stabilization of plastics (particularly PVC), and NiCd batteries (Nriagu, 1980; Yost, 1984). A potential future, large-scale use may be for solar collectors containing CdS. Major inputs of Cd to agricultural soils are due to application of phosphatic fertilizers, municipal sewage sludges as fertilizers/soil amenders, and atmospheric deposition (Ryan et al., 1982; Van Bruwaene et al., 1984). Cd content of commercial phosphate fertilizers depends on their geographical origin and can range from 0.1 to 200 pg/gm. Elinder (1986) reported that soils that had been fertilized with phosphate rock for 20 years experienced only a small increase in Cd content. However, since increased Cd accumulation in plants grown on these

GEORGEJ. WAGNER

176

Table I Estimated Worldwide Emissions of Cadmium to the Atmosphere from Natural and Anthropogenic sources in 1975"

Source Natural sources Windblown dust Forest fires Volcanogenic particles Vegetation Sea salt sprays Total Anthropogenic sources Mining, nonferrous metals Primary nonferrous metal production Cd

cu

Pb Zn Secondary nonferrous metal production iron and steel production Industrial applications Coal combustion Oil (incl. gasoline) combustion Wood combustion Waste incineration Manufacture, phosphate fertilizers Total

Annual emission (tons/year) 100 12 520 200 1 833

2 110 1600 200 2800 600

70 50

60 3 200 1400

210 7305

"Adapted from Nriagu (1980) as in Vogeli-Lange (1989).

soils was closely correlated with soil Cd, such small increases may be important to dietary Cd content. The observed correlation may relate to the time of fertilizer application relative to crop development, soil interactive factors, etc., and perhaps an enhancing effect of available phosphorous on Cd translocation from roots to shoots (Williams and David, 1976). The effect of phosphorus on Cd translocation may depend on the form of Cd in soil (Street et al., 1978). Reducing the Cd content of phosphatic fertilizers during manufacture may be the most economical means for reducing Cd input into the agricultural environment (Jones et al., 1992). A second major anthropogenically derived source of Cd for agricultural soils is sewage sludge (Davis, 1984). In highly populated countries, sewage treatment is necessary to protect water quality. The principal by-product of sewage treat-

CADMIUM IN CROPS AND EFFECTS ON HUMAN HEALTH

177

ment is solid sludge, which is most economically disposed of by dispersement on land. About 90% of Cd in raw sewage is separated into sludge (Davis, 1984) and certain sludges can contain very high Cd (and other metal) loads as a result of industrial inputs into sewage treatment systems (Yost, 1984). Because of economic considerations and because sludges provide plant nutrients and soil conditioning, sludge disposal on agricultural as well as nonagricultural lands is increasing (Davis, 1984; Mulchi et al., 1987). In the 1980s, Britain’s agricultural soils received about 50% of the sewage sludge produced in that country (Davis, 1984). The alternative to land disposal is incineration, a process that is practiced and results in substantial emission of Cd into the atmosphere (Nriagu and Pacyna, 1988). Cadmium content is the principal factor limiting use of sludge on agricultural land. The significance of Cd inputs into soil via sludge (or any source) depends principally on its availability for crop uptake. Cd uptake from sludge-amended soil is primarily dependent on Cd load, soil pH, crop type, soil type, the nature of sludge treatment, and mode (frequency) of application of a total Cd load (Ryan et al., 1982; Davis, 1984; Mulchi et al., 1987). Thus, the relationship between sludge disposal and Cd accumulation in crops grown on sludgeamended soil is highly complex and varied. Federal guidelines regulating sludge disposal have been established in many countries. In the United States, “highquality sludges” containing no more than 25pg/gm Cd, 1 mg/gm Pb, and 10 p g / gm polychlorinated biphenyls are concluded to be safe for annual application to acid soils for as many as 200 years before maximum cumulative Cd loading is reached (Naylor and Loehr, 1981). Estimated maximum loading is based on a calculated permissible total human intake not to exceed 70 p g Cd/day (Friberg et a!., 1986). The maximum sludge loading estimate level would approximately double Cd intake from food. The logic behind the 70 p g Cd/day estimate is conservative; it accounts for the individuals who smoke and, therefore, have higher Cd intake, and for those most susceptible to effects of Cd (Naylor and Loehr, 1981). But it is based on the concept that preventing a concentration of more than 200 pg/gm wet wt in kidney cortex will assure no negative health impact from Cd. As will be described in Section VI, some disagree with this premise. Regulations for applying sludge to agricultural land in the United States were promulgated by the EPA, were revised and modified by the FDA and USDA, and have been summarized in a unified federal policy released in 1979 (Naylor and Loehr, 1981; Mulchi et al., 1987). A revision of this statement is to be released soon. Aspects concerning Cd dispersal are not expected to change. A number of agricultural experiment stations in the United States have established their own guidelines based on N and P requirements of crops to be grown and on heavy metal content of sludge to be applied (Mulchi et al., 1987). Use of sludge on tobacco soils does not appear to be advisable under any circumstance. Ryan et al. (1982) have discussed methods employed in determining

178

GEORGEJ. WAGNER

effects on crop Cd accumulation after sewage sludge application to crop lands (CAST reports). Evidence that Cd has increased in the general agricultural environment since the 1800s is found in at least two studies. Kjellstrom et al. (1975) reported that the Cd content of Swedish wheat increased about three-fold between 1900 and 1980 (Fig. 1A). Jones et al. (1992) studied levels of Cd in herbage collected at a semirural, undisturbed site in England between 1860 and 1990 (Fig. 1B). Cd concentration appeared to double during this period. Herbage from a similar, but limed plot (liming began in 1916) followed a similar trend but contained about half the Cd concentration found from the unlimed control (data not shown). The main source of Cd input in both cases (Figs. 1A and B) was thought to be atmospheric Cd fallout. Jones et al. (1992) have emphasized the importance of not underestimating the importance of Cd deposition from atmospheric fallout. Where significant fallout occurs, much Cd on crops may be initially deposited on tissue surfaces.

A 0

100

Fall Wheat, Uppsala Fall Wheat, other areas of Sweden

50

90 3

_gj

1870

H

1860

1900

1940

1980

Herbage,Rothamsted,England 1900

1940

1980

Year Figure 1. Evidence for increased Cd accumulation in crops grown on the same plots from about

1860 to about 1980. (A) Replotted data of Kjellstrom e t a / . (1975) for Swedish wheat. (B)Data of Jones er al. (1992) representing the Cd content of herbage grown on an undisturbed plot at a semi-

rural site in southeast England. In both cases, Cd contamination of soils is thought to have occurred primarily via atmospheric deposition.

CADMIUM IN CROPS AND EFFECTS ON HUMAN HEALTH

179

111. ACCUMULATION OF CADMIUM: WHOLE-PLANT STUDIES It is generally recognized that: A. Accumulation of Cd by plants is principally from soil. The major factors regulating Cd uptake are soil Cd concentration and pH. B. Plants generally accumulate Cd from soil with a concentration factor of 5 1 0 on a dry weight basis. Tissue distribution can vary with plant type and variety. C. Levels of Cd in agricultural plants are generally below phytotoxic levels and therefore healthy crops can contribute substantial Cd to the human diet. As already noted, except in conditions of high atmospheric Cd fallout, Cd accumulation by higher plants is principally from soil. Foliar interception is considered to be minimal (Van Bruwaene e?al., 1984). Using IWCdas a radiotracer, it was found that only a few percent of Cd applied to the surface of plant leaves was taken up and translocated. In contrast to higher plants, mosses and lichens may absorb substantial Cd by a foliar route (Cox, 1986). Field plants generally contain <0.02 to 1 pg Cd/gm dry wt with a mean of about 0.3 pg/gm (Davis, 1984). The plant to soil Cd concentration ratio is generally about 1 1 0 on a dry weight basis under conditions of low to moderate Cd exposure (Frank er al., 1977; Cataldo and Wildung, 1978; Davis, 1984; Wagner and Yeargan, 1986). Major factors influencing Cd uptake by plant roots are soil pH and Cd concentration. Other influential factors in a given situation may be crop species, soil redox potential, soil type and cation-exchange capacity, presence of competing metals (i.e., Zn), microbial activity, and mycorrhizal status (Ryan er al., 1982; Davis, 1984; Adamu er al., 1989). Soil pH has a profound effect on Cd solubility, but the nature of the soil solid phase can have a substantial influence on the pH effect. It has been observed that in the presence of iron oxides, little soil sorption of Cd occurs below pH 4.5, but at pH 6, most Cd is absorbed. In contrast, organic matter may absorb substantial Cd even at pH 4 (Ryan et al., 1982). Cd availability appears to be related to the concentration of Cd2+in the soil solution (Bingham, 1979) and in the absence of evidence to the contrary, it is generally expected that it is the free ion that is transported across root membranes and those of other tissues. It is difficult to assess all factors that influence the availability of Cd in soil solution adjacent to the root but it is generally found that acid soils result in higher plant Cd uptake and liming generally reduces uptake (Bingham, 1979; Ryan er al., 1982; Mulchi er al., 1987). Bingham (1979) investigated the relationship between the extent of Cd accumulation in leaf and edible tissues in plants grown with soil Cd rates causing a

Carrot

=Tobacco

-1

A

Beetroot

-

French Bean

, Corn Zucchini

Edible tissue

Leaf

Wheat

r&i

Tomato tissua

Turnio .

Spinach

Kale

Cabbage

1

I

I

0

5

Lettuce I 10

I 15

ClgCdlg dry wt. Tobaccos Carrot Beetroot Radish

t

t1

FrenchBean Corn

t

t

0Tobacco'

1Tobaccon

Zucchini

t < 0.lpgCdlg

t

Wheat

-

Inorgank Cd Cd-sludge

Tomato Turnip

t

Spinach Swiss Chard

Cabbage

0

1

2

3

FgCdlg dry wt. Figure 2. Cd accumulation (dry weight basis) in leaf and edible tissues of potted plants grown in soils with (A) very high (69 pg Cd/gm soil) and (B) moderate (0. I pg Cd/gm soil) added Cd. Data of panel A are from studies of Davis (1984) using inorganic Cd. Data of panel B are from studies of Bingham (1979, Table 3) using 0.1 pg Cd/gm soil added as sludge and Page era/. (1981, Table 3.1) using 0.1 pg Cdlgm soil added as inorganic metal. Both studies included in panel B utilized a calcareous soil. At the top of panel B are data from two field studies of tobaccos (*-Frank er a / ., 1977; and **-Yeargan et al., 1992) grown on tobacco soils without additional Cd. Tobacco data were extrapolated to a soil Cd content of 0.1 pg Cd/gm soil to allow comparison with food crops.

CADMIUM IN CROPS AND EFFECTS ON HUMAN HEALTH

18 1

25% reduction in yield. Plants having low tolerance to Cd such as spinach and soybean were identified and leafy tissue was identified as high accumulating. This was also found in pot experiments where calcereous soil was amended with sludge to provide moderate (0.1 mg Cd/gm) and higher (5 p g Cd/gm) Cd loads. Again, leaf tissue was high accumulating; seed, fruit, and tuber tissue were lower accumulating. Somewhat similar results with respect to relative tissue accumulation were reported by Davis (1984) for pot trials when soils contained very high (69 mg/gm) Cd. Figure 2A compares Cd accumulation (leaf versus edible tissue) in various pot-grown crop plants grown in soil amended with very high Cd (69 pg/gm dry wt soil). As shown, leaf tissues were highest accumulating in every case. Grains, roots, and fruit tissues were lower accumulating under these exposure conditions. Also noteworthy is the very high Cd accumulation potential of the one tobacco variety included in this study. In Fig. 2B, Cd content in edible tissues of most plants represented in Fig. 2A is compared, but after lower level exposure. In these experiments, a moderate level (0.1 pg/gm dry wt soil) of Cd was supplied in the field, as either sludge or inorganic metal salt. Where comparison of sludge versus inorganic Cd effects is possible (comparative data are limited to leaf tissues), the form of Cd supplied did not appear to greatly affect the level of metal accumulated at this exposure. Both experiments utilized a calcareous soil. Except for carrot, nonleaf tissues were low accumulating, as in the high-Cd exposure case (Fig. 2A). Data obtained with field-grown tobacco from two other studies are included in Fig. 2B. These crops were field grown in tobacco soils containing about 0.4 pg/gm Cd dry wt soil. Values of leaf Cd content represented in the figure were adjusted to predict leaf accumulation from a soil Cd level of 0.1 pg/gm so that tobacco and other crop tissues could be compared. Tobacco (adjusted values) was apparently similar to other leaf crops in Cd accumulation potential under this lower-Cd exposure condition. Perhaps tobacco has an exceptional capacity to accumulate Cd under high exposure conditions (Fig. 2A) because of its relative tolerance to Cd (as compared with, e.g., spinach and soybean). Accumulation of high levels of Cd in tobacco can result from growth on soils amended with high levels of sewage sludge (Gutenmann et af., 1982). Figure 3 provides contrasting data for some of the same tissues grown on “nonpolluted” soils as summarized by Page et al. (1981). Under these natural conditions, differences in tissue Cd accumulation were less marked between, say, leaf versus grain versus fruit tissues. However, in general, leaf crops were somewhat higher accumulating. Arrows mark the mean levels of Cd in some of these crops as determined in FDA market-basket surveys (as reported by Yost et af., 1980). As shown, market-basket survey-derived values compare well in most cases with natural condition results summarized by Page et al. (1981). Comparison of Figs. 2A, 2B, and 3 points out the importance of experimental dose in uptake experiments. Even the moderate to low Cd-added field experiments of Fig. 2B differ substantially in Cd concentration and in Cd accumulation

182

GEORGE J. WAGNER Potato Beetroot Radish

-

Sovbean J

/

meat

Tomato

Turnip Spinach

Swiss Chard Cabbage

’

Lettuce

patterns among tissues from that found in consumer-availablecrops (Fig. 3). One must keep in mind, however, that data like that in Fig. 2 are limited and often represent different plant varieties and growth conditions (Page et al., 1981). Also, results of similar Cd-added experiments do not always agree, presumably because of differences in soil type, pH, and other factors (Page et al., 1981). We note that Fig. 2 data are on a dry weight basis whereas Fig. 3 data are as fresh weight and represent crop tissues that differ in moisture content. It will be emphasized later that the amount of a particular food consumed can be as important as its Cd concentration to daily Cd dietary intake. The generally high level of Cd in leaves of field-grown tobacco [ 1 to 3 pg/gm dry wt (Ryan et al., 1982)] relative to that of most plants (about 0.3 pg/gm) may be principally due to the high Cd content of tobacco soils (Frank et al., 1977; Yeargan et al., 1992), the tendency of tobacco to accumulate more Cd in leaves than in roots, and perhaps low or transiently low pH due to post-Nfertilization and nitrification in this highly fertilized crop. Also, tobacco is often grown repeatedly on the same plots and phosphatic fertilizer is often used in excess. Thus, tobacco soils contain elevated Cd loads [i.e., 0.4pg/gm (Frank et al., 1977)j. These factors together have led to the general, but unwritten, recommendation that sludge not be applied in the culture of tobacco. Detailed distribution of Cd within tissues of a plant has been studied relatively

CADMIUM IN CROPS AND EFFECTS ON HUMAN HEALTH

183

little. We examined Cd distribution in mature, flowering N. rabacum and N . rusrica exposed in solution culture to moderate (3 pM) Cd. Cd was greatest in oldest leaf lamina and least in youngest, suggesting gradual deposition with time. Distribution within the leaf blade was relatively uniform and midribs and stems contained lower Cd than leaves (Wagner and Yeargan, 1986). Jarvis ef al. (1976) observed that the shoot/root ratio of Cd was less than 1 in all but 3 of 23 crop species tested. We have observed substantial differences in shoot/root Cd ratios in several tobaccos grown in solution culture. Most cultivated varieties have a ratio close to 1 while two species, including N. rusrica, had ratios close to 0.2, making these low-leaf Cd accumulators (Wagner and Yeargan, 1986). Results of tissue distribution studies comparing N. rabacum and N . rustica suggest that the higher fine root mass of N. rusrica may account for the unusual root Cd binding capacity of this species (Wagner and Yeargan, 1986). The somewhat contradicting results regarding root/shoot Cd distribution in N. rabacum versus N . rustica reported by Mench et al. (1989) for soil-grown plants may be due to difficulty in recovering fine roots from soil. From these and other studies, a general rule emerges: Cd accumulation in plant tissues is generally leaves > fibrous roots > seeds = storage organs. Potato tubers (a stem tissue) are relatively high accumulating. Potatoes can contribute significant Cd to the human diet (see Section V). Cd concentrations encountered by plants in agriculture and in most natural environments, even under mild Cd pollution conditions, are likely to be below phytotoxic levels for most plants. Bingham (1979) observed a 25%reduction in the growth of bean plants grown in solution culture containing 0.44 pM Cd. All other plants tested required substantially higher exposure levels to cause growth depression. The most sensitive soil-grown plants tested in this study (spinach and soybean) required 4 to 5 pg Cd/gm soil to produce a 25% yield decrement. This level is 10 to 100 times that found in mildly contaminated and uncontaminated soils, respectively. Therefore, because of the general tolerance of plants to Cd, healthy crops grown in a Cd-polluted soil have potential to contribute substantial Cd to the human diet. It is difficult to estimate soil solution Cd concentration because of the many factors contributing to it. But we have done so to provide a contrast between exposure levels used in many experimental studies with whole plants in solution and soil culture and with tissue culture, with levels generally encountered by agricultural plants in the field (Wagner and Krotz, 1989). Mean Cd concentration of soils, as already discussed, has been estimated to be about 0.06 to 0.5 pg/gm (the lower figure being more often cited). If we assume a mean soil density of 1.3 g/cc and that at near neutral pH plants can access 5% of soil content, available Cd would be about 0.04 to 0.32 pM. This might be doubled under acid soil conditions. These concentrations are in contrast to ranges of 45 to lo00 pM Cd often used in tissue culture studies and 0.1 to 6 pM often used in solution culture experiments (Wagner and Krotz, 1989). This estimate of soil solution Cd in

184

GEORGEJ. WAGNER

unpolluted soil may be subject to error, but we argue that it can form the basis for discussion of the relevance of doses used in many experimental studies to exposures generally occurring in agricultural production. It is not uncommon to read in grant proposals and the introductions of publications (describing experimental studies using Cd levels an order of magnitude or more above those in agricultural environments) the inference that results obtained will be relevant to problems concerning Cd contamination of the environment and/or the human diet. High-level experimental studies can be interesting and important, but the concept of dose versus response should not be lost. This argument also applies to studies of Cd toxicology using animals (Spivey Fox, 1988). Several workers, including ourselves (Wagner and Krotz, 1989), have observed growth stimulation in solution-cultured plants exposed to low-to-moderate levels of Cd (Hirt et al., 1989). In this context, it is interesting to note that the marine diatom Thalassiosiru weissjlogii can substitute Cd for its Zn requirement in instances of Zn deprivation. This was recently discovered by Price and Morel (1990) when they were seeking a reason for surface depletion of Cd in coastal ocean waters. Because Cd is not considered to be required for growth of any biological organism, it was surprising to find that Cd was able to restore growth of Zn-depleted T. weissjlogii to 90% its maximum rate. Co would also substitute for Zn, but less efficiently than Cd. Growth of Chlorella is said to be stimulated by low levels of Cd, but inhibited at high levels (Pakaline et al., 1970; Vallee and Ulmer, 1972). It is noteworthy that Cd has been shown to stimulate certain Zn metalloenzyme activities in certain systems and inhibit them in others (Vallee and Ulmer, 1972). In view of these findings, the question of possible nutrient or substitute-nutrient status of Cd in higher plants may be worthy of systematic study. The recent explosion of work concerning the role of Zn (and the possible role of Zn metallothionein) in transcriptional regulation will undoubtedly invigorate studies of Cd interactions with nucleic acids and nucleoproteins (Kaptein, 1991). In this regard, Cd-substituted GAL4 (a well-characterized Zn-requiring transacting factor involved in transcriptional regulation of galactose utilization in yeast) was recently shown in crystallographic studies to have a similar secondary structure to the native Zn-substituted protein (Baleja et al., 1992).

IV. ACCUMULATION OF CADMIUM: CELLULAR AND SUBCELLULAR ASPECTS It is argued here that: A. There is little evidence to suggest the occurrence of significant accumulation of phytochelatins (PCs) in agricultural crop plants exposed to low, fieldlike Cd concentrations.

CADMIUM IN CROPS AND EFFECTS ON HUMAN HEALTH

185

B. PCs produced in response to high-Cd challenge appear to sequester accumulated Cd, principally in the vacuole. A role for these peptides in unstressed plants remains to be shown. C. Various mechanisms are suggested (by limited evidence) for Cd accumulation under low-level exposure conditions. These include a degree of metal binding in cell walls, chelation in the cytosol, cytosolic chelation and subsequent transfer to the vacuole, and transport of free ion into the vacuole followed by sequestration with organic acids and other ligands. Several reviews have considered cell-level mechanisms of Cd detoxification and tolerance to Cd toxicity in plants with focus on the possible role of so-called phytochelatins in these processes (Robinson and Jackson, 1986; Tomsett and Thurman, 1988; Rauser, 1990; Verkleij and Schat, 1990; Mehra and Winge, 1991; Steffens, 1991). No attempt will be made here to duplicate these very useful treatments. Rather, discussion will be focused on recent studies of Cdbinding peptides, organic acids, subcellular compartmentation, membrane transport of Cd, and speciation of Cd complexes occurring in v i v a as these may be involved in Cd accumulation in the agricultural environment. As just noted, much recent work on Cd responses in plants has focused on the role of Cd-binding peptides, variously called phytochelatins, cadystin, y-glutamyl peptides, (yEC).G, etc. (Rauser, 1990). While a wholly accurate nomenclature for these molecules is still being sought, the trivial name phytochelatins (PCs) has become recognized by most and will be used here. Detailed discussion of the history of discovery (Rauser, 1990), chemistry, isolation (Grill et ul., 1991, and other papers in that volume), and properties of PCs can be found in other reviews (noted in the preceding). The mode of synthesis of these peptides is a matter of some controversy and the subject of considerable current interest. The early report (Grill et ul., 1989) of an enzyme capable of synthesizing PCtermed PC synthase-has been followed by reports describing several possible routes of biosynthesis (Hayashi er al., 1991; Mehra and Winge, 1991). Animals and certain yeasts and fungi produce cysteine-rich, metal-binding proteins called metallothioneins. At least one yeast, Canidiu glubratu, produces metallothionein in response to Cu and PC in response to Cd (Mehra and Winge, 1991). Much of the work to characterize the properties and functions of PCs has been patterned after that to characterize metallothionein over the past 30 years. Metallothionein will not be discussed in detail here. Excellent reviews are found in several references (Hamer, 1986; Kagi, 1991; Mehra and Winge, 1991). My purpose here is to focus on current evidence for a role of PCs in fieldgrown agricultural crops. Particularly in this context it is important to critique several generalizations that have emerged concerning properties and suggested functions of these peptides, some of which, in this author’s opinion, are supported by substantial experimental data and others that are not. These questions relate to: the ubiquity of PCs in the plant kingdom, the ability of metals other

186

GEORGE J. WAGNER

than Cd to induce PC synthesis, the ability of PCs to bind metals other than Cd in vivo or in v i m , the constitutive nature of PCs in the absence of substantial metal challenge, their possible role in tolerance to Cd, their possible role in Zn and Cu homeostasis, and role in Cd transport across membranes. The ubiquitous nature of metal-inducible PCs is indicated primarily by one study of 200 plant species representing all major plant classes, orders, and families (Gekeler ef al., 1989). This survey was performed under conditions of Cd induction only and PCs were identified on the basis of their separation by HPLC as sulfur-rich components. The study shows that most plants are capable of synthesizing PCs in the presence of Cd challenge. However, in all studies relying on these methods of detection, care must be taken to distinguish PCs from lowmolecular-weight thiol-metal complexes. This may be particularly true in cases where PCs with low n numbers only are detected. PCs were first shown to be induced by Cd (Kondo ef al., 1984). Grill and coworkers (1987) undertook a limited survey of metal specificity in PC induction using Rauvolfia suspension cells. Single concentrations of the single metal salts were tested and relative values observed for percentage induction of PCs were: 100 pM Cd (loo%), 1000 pM Pb (56%), 1000 pM Zn (41%), 2000 pM Sb (41%), 50 pM Ag (40%), and 100 pM Ni (28%). The other metals and metalloids tested (again, as single salt species), Hg, As, Cu, Sn, Se, Au, Bi, Te, and W using from 10 to 100 pM salt concentrations, caused relative PC accumulations of 5 to 21%. Ca, Al, Fe, Mg, Mn, Na, Mo, Cs, Cr, U, and V (tested as different but single concentrations and salts) cause to induction. Effects on cell growth in these experiments were not reported. Only Cd induced significant amounts of PCs having higher (2n=4) n numbers. Deficiencies of this study were that only single metal-anion salts were used, and concentration dependence and effects on growth were not reported. It is interesting that ions tested that gave the strongest PC induction were, with the exception of Se, all borderline or intermediate class ions (Nieboer and Richardson, 1980). In Schizosaccharomyces pombe the pattern of PC induction by various metals was similar to that in Rauvolfia only in that Cd was the most effective inducer (Grill er al., 1986). We note that we were unable to detect significant PC induction by Zn (2000 pM) in tobacco cultured cells. BSO, an inhibitor of PC synthesis, increased sensitivity of cells exposed to 22 pM Cd but did not significantly affect cells exposed to 90 pM Cu or 1800 pM Zn (Reese and Wagner, 1987a). Although data are limited, common inducers of plant stress such as heat shock, cold stress, UV light, and fungal elicitors are not reported to induce PC. In contrast, animal metallothioneins are produced by various nonmetal, biotic, and abiotic factors (Kagi, 1991). The pattern of PC induction is, therefore, different in different systems tested to date, but Cd appears to be the most efficient inducer and limited data suggest little true ion specificity in PC induction. The most relevant aspect of PC induc-

CADMIUM IN CROPS AND EFFECTS ON HUMAN HEALTH

187

tion regarding agricultural crop production is that high levels of metal appear to be required for the process(es). However, Tukendorf and Rauser (1990) have reported occurrence of low levels of PC in corn roots exposed to low (0.05 pM) Cd in solution culture. Though the occurrence of constitutive PC has been suggested in several reports, this has not been shown conclusively. The ability of an ion to induce PC synthesis suggests that it will bind the ion in vivo and in vitro. To this author’s knowledge, only Cd and Cu are shown to coordinate with PC, and these only in a preliminary rigorous way [electronic spectra (see Rauser, 1990) and pH titration (Reese and Wagner, 1987b; Reese and Winge, 1988; Reese et al., 1988)]. CDM MCD, ESCA, NMR, and other techniques applied to detailed studies of metal binding in metallothionein (Schaffer, 1991) have not been applied to the study of PCs. Binding of Zn to PC, as reported in reconstitution experiments, is said to be weak (and stabilized by sulfide) and is only supported by weak electronic spectral evidence (Reese and Winge, 1988). Thus, the term PC, which implies general metal binding ability, has not yet been shown to be appropriate. It has been suggested that PCs are involved in micronutrient Cu and Zn homeostasis in a manner analogous to metallothionein in animals (Grill et al., 1987). The principal evidence for this conclusion is found in one study from experiments in which various plant cell cultures were transferred from nutrientdepleted medium to fresh medium containing only micronutrient levels of Zn and Cu. Concentration of Zn and Cu in media were found to decrease as PC levels increased. As Zn and Cu became depleted from the medium, PC levels decreased. It was not shown that Zn and Cu were bound to PC during these transitions. These results are most interesting. However, until the correlation can be shown to be direct, questions remain concerning a homeostatic role, particularly in the case of Zn (Rauser, 1990). As noted in the foregoing, Zn binding to PC in vivo or in vitro has not been unequivocally demonstrated. The argument has also been made recently that PC can serve to transfer Zn and Cu to apometalloproteins in vivo (Thurman et al., 1991). It was reported that Zn-PC and Cu-PC prepared in vitro (characterization of binding not reported) were able to transfer metal to apoproteins in vifro. But their efficiency of transfer was reported to be less than that of free ions. A similar observation has been made for the relative efficiency of free Zn and Cu versus Zn- or Cumetallothionein in metal transfer. Hamer (1986) has suggested that, in animals, liver Zn-metallothionein serves as a labile storage source of Zn that can be mobilized in early development and under stress conditions, and that Zn filling of apo-Zn enzymes occurs via free metal species. Binding of Cu to metallothionein is very strong (Hamer, 1986). Similarly, Cu is reported to bind to PC with high affinity (Reese et al., 1988), making both molecules poor candidates as CU shuttles. PCs have been suggested as possible sulfur carriers during sulfate reduction.

188

GEORGE J. WAGNER

This hypothesis, though advanced in 1986 (Steffens et al., 1986) and reiterated (Steffens, 1991), has received no support in the refereed literature. New speculation has been advanced concerning possible roles of PCs in sulfur metabolism (Steffens, 1991). Though interesting, these speculations await experimental evidence. In summary, the ubiquitous nature and Cd inducibility of PCs in higher plants have relatively strong support. Also, though evidence is limited to a few reports and systems, inducibility appears to extend to several class B and intermediate class metals and metaloids. But coordination of metals with PCs has not been rigorously studied, particularly for metals other than Cd. Very little data are available concerning possible roles for PC in micronutrient homeostasis or other physiological processes. Whatever the roles of these peptides may be in induced systems, significant PC occurrence under low-level exposure conditions as in agriculture may be questioned. Unless PCs are generally present in field crops at low levels and have key catalytic properties, for example, in membrane transport, they may not be significant players in agricultural production. Although tolerance of plants to toxic metals such as Cd is a phenomenon associated with high Cd exposure and is therefore not generally relevant to agriculture, it will be briefly discussed here to complete the preceding considerations of PCs, and also to introduce discussion to follow on the subcellular compartmentation and accumulation of Cd. Metal tolerance is a complex phenomenon undoubtedly based on multiple mechanisms. Metal binding to cell walls, chelation with peptides, organic anions, and inorganic anions, and active ion pumping into the vacuole have all been suggested as mechanisms in tolerance in higher plants (Baker, 1981; Woolhouse, 1983; Tomsett and Thurman, 1988; Cumming and Taylor, 1990; Verkleij and Schat, 1990; Verkleij er al., 1991). Though a good case may be made for a role of PCs in Cd detoxification under high-Cd exposure conditions, evidence for a role of these peptides in metal specific tolerance is insufficient (Rauser, 1990; Verkleij et d.,1990). The relative lack of metal specificity in PC induction, overproduction of PCs in sensitive as well as tolerant clones of Cd-challenged Agrostis (Rauser, 1990), and perhaps the limitation that most studies of PCs relating to this phenomenon to date have used tissue cultured cells argue against a direct role for PCs in metal tolerance. However, an interesting observation is that in both Darura cultured cells and Agrostis roots the rates of PC synthesis may be important to tolerance (Rauser, 1990). Rate of response is considered to be a key factor in a number of disease resistance responses in plants. As indicated earlier, binding to cell walls, transfer of metals from the cytoplasm to the vacuole, and chelation with peptides and organic and inorganic anions in the cytosol have all been suggested as mechanisms for tolerance. Some of these may also be relevant in Cd accumulation under low-level exposure conditions. Figure 4 describes various strategies that incorporate these suggested mechanisms. Mechanism 1 involves binding of metals in the cell wall.

CADMIUM IN CROPS AND EFFECTS ON HUMAN HEALTH

189

Figure 4. A model describing several speculative modes of control of Cd uptake/exclusion/ accumulation in plant cells. Arguments for and against each are made in the text. Modes are ( I ) cell wall binding; (2) limitation of Cd flux through the plasma membrane; (3) an active efflux pump mechanism as occurs in bacteria; (4) cytosolic accumulation/sequestration as various complexes; ( 5 ) transport of cytosolic Cd complexes or Cd from these into the vacuole and Cd sequestration therein (ligands of vacuolar sequestration complexes (X)may be transported into the vacuole independent of Cd); (6) a ApH-driven, CdlH antiport activity leading to Cd transport to the vacuole and subsequent sequestration therein; and (7)transport of cytosolic Cd complexes out of the cell.

Several studies have examined the distribution of Cd between buffer-soluble and -insoluble fractions of Cd-exposed tissues (i.e., Baker, 1978; Weigel and Jager, 1980; Cataldo et al., 1981; Wagner and Trotter, 1982; Wagner and Yeargan, 1986), but few have defined the nature of particulate Cd occurring in extracts (Verkleij and Schat, 1990). Pectic substances and hemicellulosic wall polymers undoubtedly bind Cd in vivo and their exchange capacity may vary (in time and with tissue type) with pectin methylation state and wall ion speciation. The binding capacity of cell walls under low-level exposure conditions may be highly significant, but this remains to be shown. Studies of Cd distribution in bean and tobacco plants exposed to low to moderate (0.005 to 7.6 pM) Cd reported significant Cd in insoluble fractions containing cell wall (Weigel and Jager, 1980; Wagner and Yeargan, 1986). However, soluble/insoluble Cd ratios were usually 2 I , suggesting that substantial Cd enters the cell even under low-level conditions in intact plants. Rapid entry of Cd into tissue cultured cells is also documented (Krotz et al., 1989). Mechanism 2 involves limitation of Cd movement across the plasmalemma.

190

GEORGE J. WAGNER

Because of characteristics of Cd binding to phospholipids, membranes are not viewed as an efficient barrier to Cd penetration (Vallee and Ulmer, 1972; Cumming and Taylor, 1990). We know nothing about Cd transport across plant plasmalemma. However, several ion porters and channels (not thought to be entirely ion specific) are known or anticipated (Poole, 1988). It has been suggested that Cd is transported across the plasmalemma of a pituitary cell line via a voltagegated Ca channel (Hinkle et al., 1987). However, in a similar study with the algae Criscospaera elongate, results proved inconclusive (Karez et al., 1990). Metal was found to rapidly equilibrate with the soluble fraction of tobacco cultured cells exposed to full-growth-permitting, moderate (for culture cells) Cd concentration (45 pM).This suggests the absence of an exclusion barrier or significant wall sequestration in this case (Krotz et al., 1989). Membrane transport characteristics of Cd in whole plant tissues is little characterized. Biphasic uptake characteristics have been observed with soybean roots (Cataldo and Wildung, 1978). K,’s of 0.22 and 1.67 pM suggest the existence of active and diffusional components. For the preceding reasons, Mechanism 2 is undoubtedly less significant than Mechanism 1 . There is no evidence at present for an anion efflux mechanism in plants such as that found in bacteria and described by Mechanism 3 (Cumming and Taylor, 1990). A plasmid-encoded, plasmalemma Cd-translocating, Cd-efflux ATPase is characterized in Staphylococcus (Nuciflora et al., 1989). Its gene sequence shows homology to E,E, cation-translocatingATPases found in both procaryotes and eucaryotes. A chelation strategy (Mechanism 4) involves sequestration in the cytosol via protein or peptide complexes, organic acids, or inorganic (phosphate, sulfide) complexes. Ernst and colleagues were the first to introduce a Zn-organic acid chelation mechanism when they suggested that Zn tolerance in plants might be a result of metal chelation with malic acid in the cytosol with subsequent transfer of metal to the vacuole for sequestration as Zn-oxalate (see Verkleij et al., 1991). Recent computer simulation studies of metal speciation in tobacco cultured cells suggest that citrate may be a better candidate as a cytosolic chelator of Cd (or Zn) than malate (Wang er al., 1991, 1992). The role of organic acids in chelating vacuolar Cd will be considered in a following discussion. PCs have been suggested as possible chelators of Cd in the cytosol and elsewhere. Under low-level exposure such as in agriculture, PCs may not be abundant and therefore may not be significant in Mechanism 4. After high-level exposure, available evidence suggests that the bulk of PCs and Cd are found in the vacuole (VogeliLange and Wagner, 1990). Another possible cytoplasmic Cd chelator may be plant metallothionein. This protein class has not yet been isolated from higher plants but genes having significant sequence homology to fungal and animal metallothioneins have been detected in several plant systems (Robinson et al., 1990; de Framond, 1991; Kawashima et al., 1991; Kille et al., 1991; Tommey

CADMIUM IN CROPS AND EFFECTS ON HUMAN HEALTH

191

et al., 1991). Metal binding regions of the predicted plant proteins are metallothioneinlike. Interestingly, constitutive plant metallothionein gene expression is highest in roots not exposed to metal in both pea and maize (de Framond, 1991). The relevance of these potential Cd chelators, if they exist as stable proteins, is not known. Mechanism 5 is an extension of Mechanism 4: Chelation of Cd in the cytosol by proteins/peptides and organic or inorganic anions (X in Fig. 4) and subsequent transfer of complexes or metal to the vacuole. The possibility that PC or plant metallothionein or their Cd complexes are transported into the vacuole has apparently not yet been tested. To repeat, the bulk of protoplast PC as well as Cd has been observed to be vacuolar in both a cultured cell and a leaf system (Krotz et al., 1989; Vogeli-Lange, 1989; Vogeli-Lange and Wagner, 1990). It should be noted, however, that there is little evidence to date that Cd is accumulated in the plant vacuole after low-level Cd exposure (Wagner, 1979; Krotz et al., 1989). However, some results suggest this to be the case (G. J. Wagner, unpublished data). Organic acids and certain amino acids have stability constants that are favorable for Cd binding (Krotz et al., 1989). Neither citrate nor malate was found to stimulate Cd uptake in tonoplast vesicles from oat roots (D. Salt and G. J. Wagner, unpublished data). Cu uptake in a hepatocytes is reported to be facilitated by amino acids, but co-transport is not evident (Balliatori, 1991). Recent evidence for an ABC-type membrane transporter that putatively transports PC-Cd-sulfide complexes across the tonoplast in high-Cd-exposed fission yeast will be discussed in Section VII. To summarize, there is no evidence in plants for complex facilitated transport of Cd into the vacuole of plants exposed to low or high levels of Cd. However, work in this area has just begun. The occurrence of CdS-PC crystallites in fission yeast exposed to high Cd concentrations is well studied (Mehra and Winge, 1991), however, their subcellular location is not established. Mechanism 6 involves transfer of free Cd2+ to the vacuole via a Cd/H antiporter energized by ApH generated by the well-established tonoplast V-type ATPase. We have observed this activity in tonoplast vesicles isolated from roots of oat seedlings grown in the presence or absence of Cd (Salt and Wagner, 1993). NO,-sensitive ATPase, total Mg ATPase, and ApH-dependent Cd transport of tonoplast membrane vesicle preparations equilibrated on linear sucrose gradients at 1.1 1 gm/cc, the density of tonoplast membrane. The ApH-dependent Cd accumulation observed showed saturation kinetics with a K, (app) of 5.5 pM, making it a candidate for Cd transport into the vacuole under high or low exposure conditions. Cd transport was also observed in membranes having higher density than tonoplast and characteristics of plasma membrane. We have suggested that under low- or high-Cd exposure conditions, Cd may enter the cell as free Cd2+via channels in response to the plasma membrane potential and then may be transported into the vacuole via a tonoplast Cd/H antiporter activity.

192

GEORGE J. WAGNER

Organic acids, PCs, sulfite, amino acids, phenolic acids, or other anions, where these may be present, may serve to trap/sequester vacuolar Cd. Compartmental flux analyses have suggested the possible importance of metal transport into the vacuole as a possible tolerance mechanism (Brookes et al., 1981). Computer simulation of vacuolar Cd speciation suggests that once inside the vacuole, Cd would be sequestered principally as Cd citrate or, if PC, sulfide, or metallothionein were present, as complexes with these ligands (Wang et al., 1991). It is not known if the newly discovered Cd/H antiport activity utilizes the established Ca/H antiporter of tonoplast (Bush and Sze, 1986) or a different protein. The last mechanism speculated on here (Mechanism 7) involves transfer of cytosolic Cd complexes outside the cell. Black deposits are often observed in high-Cd-exposed tissues and cultured cells (see Krotz et al., 1989; Rauser, 1990). These may be CdS or stress response metabolites inside or outside the protoplasm. Oxalate salts are often found in plant cell walls, intercullular spaces, and vacuoles. Oxalate has affinity for Cd (Wang et al., 1991) but identification of crystalline Cd oxalate inside or outside plant cells has not been reported. We argue that Mechanism 6 is the best candidate for an intracellular Cd accumulation mechanism in plants exposed to low or moderate Cd, circumstances where PC and especially sulfide formation is expected to be low. Cd may enter the cell principally via ion channels, energized by the plasma membrane potential. This may also be the case for Zn and other trace inorganic divalent cations. Tobacco cultured cells exposed to high (2000pM) Zn produce little PC and accumulate Zn principally in the vacuole (Reese and Wagner, 1987a; Krotz er al., 1989). We also note that PCs are reported to complex less than 1% of soluble Zn in roots of Acer and Silene plants grown on a Zn mine waste site (Grill et al., 1988a). The vacuole is the common store for the bulk of intracellular nonprotein anions and cations in plants, thus this compartment and its membrane are equipped for inward ion transport. The vacuole ion store has multiple functions in osmotic regulation, ion and solute storage, and sequestration. Its evolution constitutes one of the most fundamental adaptations made when plants invaded the terrestrial environment (Walter and Stadleman, 1968). In summary, considerable knowledge of cellular and subcellular aspects of Cd accumulation has been gained in the past decade. Clearly some impetus for renewed interest in a long-standing fascination with the metal tolerance phenomenon (Woolhouse, 1983; Verkleij et al., 1991) has resulted from the discovery and characterization of the PCs. In this author’s opinion, future studies must focus not only on PC biology and molecular biology, but also on other possible mechanisms for Cd accumulation if we are to efficiently reach the level of understanding necessary to allow plant manipulation to lower Cd (and other toxic metal) accumulation in agricultural crops and therefore the human diet. Another practical goal of research for the future is to provide plants that may be used for

CADMIUM IN CROPS AND EFFECTS ON HUMAN HEALTH

193

efficient bioremediation of Cd and other metals from wastes, or for extraction of low-grade ores to recover valuable mineral elements such as Ag and Au. It is easy to imagine that increasing PC accumulation in plants via the recombinant DNA approach might lead to development of super-accumulator plants for bioremediation. Work to that end will clearly proceed when the enzymology and genetics of PC synthesis are better understood.

V. CONTRIBUTION OF AGRICULTURAL PRODUCTS TO HUMAN INTAKE OF CADMIUM Major issues are: A. Various crop plants accumulate Cd differently, some being leaf accumulators, some root accumulators, etc. The basis of this tissue partitioning is not understood. Although Cd concentration is important, contribution of a crop to dietary intake of Cd is dependent on the amount of the crop consumed as food as well as the Cd concentration of the consumed tissue. B. The bioavailability of Cd from various foodstuffs is poorly understood. It can vary with crop type, diet composition, the nutritional status of the consuming animal, and possibly other factors.

As already discussed (Section 111), the plant tissue in which Cd is most accumulated can vary with species and variety (John and van Laerhoven, 1976; Page et al., 1981; Wagner and Yeargan, 1986). Some edible tissues are relatively low in Cd concentration, but dietary intake is a function of the amount of tissue consumed as well as its Cd concentration. For example, grain products generally have moderate Cd concentration, but are consumed in relatively large amounts (Yost er al., 1980) and therefore contribute substantial Cd to the diet (Table 11). Conversely, consumption of a small amount of a food containing relatively high Cd concentration (i.e., shellfish or visceral meats) may not greatly impact total cumulative Cd. Obviously consumption of large amounts of such high-Cd content foods or their use on a regular basis can impact greatly on body Cd burden (see Spivey Fox, 1988). Regarding vegetable foods, potatoes, leafy vegetables, and certain root vegetables often have relatively high concentrations of Cd (Table 11). Strict vegetarians (consuming no meat, dairy products, or eggs) may consume higher Cd because most animal products contribute less metal than do vegetables. A lacto-ovo-vegetarian (eating dairy and egg foods) consumes less grain, since they eat dairy products, and therefore would have a lower Cd intake than a total vegetarian. If a person with a normal diet substitutes potatoes for legume vegetables and fruits, they may increase Cd intake by as much as 45%

194

GEORGEJ. WAGNER Table I1

Food Consumption and Cadmium Intake for Various Diets Vegetarians” Food class Dairy products Meat, fish, poultry Grain and cereal Potatoes Leafy vegetables Legume vegetables Root vegetables Garden fruits Fruits Oils, fats, shortenings Sugars and adjuncts Beverages Totals

ppb Cda fresh wt 5.7 15.3 23.2 48.0 40.5 6.2 32.3 14.7 3.0 15.3 10.0 3.0

glday 584

P g Cdl day 3.3

-

-

203 43 252 166 284 107 110 600 2349

4.7 2.1 10.2 1.0 0.8 1.6 1.1 1.8 26.6

Regular diet glday 704 262 424 177 54 66 32 88 222 72 83 684 2868

Pi3 Cd/ day 4.0 4.0 9.9 8.5 2.2 0.4 1.0 1.3 0.7 1.1

0.8 2.1 36.0

“From FDA 1974 total diet studies. bBased on the response of 183 southern Californians in a food frequency questionnaire by the Department of Biostatistics and Epidemiology, Lorna Linda University School of Health, 1975. Leafy vegetable class includes garden fruit and root vegetable classes from the regular diet. The cadmium concentrations for the various food classes were assumed to be the same as in the FDA study. (From Naylor and Loehr 1981.)

without changing their caloric intake. In contrast, substituting fruits for potatoes could decrease total daily Cd intake by about 30% (Ryan et al., 1982). As noted, Cd content of muscle meats is relatively low. In contrast, visceral meats (kidney, liver, pancreas) are high because these organs in particular accumulate Cd in animals (Van Bruwaene er al., 1984). Aquatic food species (fish, crabs, oysters, etc.) bioconcentrate Cd and therefore can have high Cd concentrations. Milk, although having moderate Cd concentration (Table 11), is a major source of Cd for infants and children, contributing 50 and 25% of Cd intake in the first year for bottle-fed and breast-fed infants, respectively (Ryan er al., 1982). All types of tobacco use can contribute substantially to Cd intake because of the high Cd content of tobacco leaf and, in the case of smoking, efficient absorption of Cd via the inhalation route (Ryan er al., 1982). Factors influencing dietary Cd intake (also other trace contaminants and nutrients) include age, sex, race, residential region, ethnicity, and personal preference (Yost et al., 1980). In 1974 the USDA (Compliance Program Evaluation) estimated the mean Cd intake of 15- to 20-year-old males in the United States to

CADMIUM IN CROPS AND EFFECTS ON HUMAN HEALTH

195

be between 26 and 52 pg Cd/day (Yost et al., 1980; Ryan et al., 1982). This estimate was based on market-basket surveys using food samples often having Cd concentrations near the level of detection of the analytical methods used. Subsequent analyses have combined data on dietary habits to arrive at a mean value of about 38 pg Cd/day for a normal diet and 27 pg/day for the diet of the lacto-ovo-vegetarian (Yost et al., 1980; Naylor and Loehr, 1981; Travis and Etnier, 1982; see Table 11). Mean daily Cd intake from food in developed countries has been estimated to vary between a low of about 16 to a high of 60 pg Cd/day (Sherlock, 1984). Cd intake estimates based on estimated consumption from various foods and their Cd concentrations are generally two to three times higher than those based on analysis of human feces. The basis for this discrepancy is not known (Yost er al., 1980; Ryan et al., 1982). The method of data analysis influences the estimate of the percentage of individuals predicted to ingest Cd at a higher level than that recommended by the joint Food and Agriculture Organization and the World Health Organization (FAO/WHO) Expert Committee on Food Additives (70 pg Cd/day) (see Naylor and Loehr, 1981; Ryan et al., 1982; Sharma et al., 1983). Some reports predict that as much as 14% of the U.S. population may be at risk because they exceed this level of Cd intake (Yost et al., 1980). Other estimates are lower. In this context, it should be kept in mind that diet composition and therefore dietary Cd intake varies greatly with age group, sex, and other factors (Loenstein, 1978). Another relevant consideration here is the impact of changing eating habits on present and future Cd intake. Travis and Etnier (1982) estimated that increasing Cd concentration in foods in the United States between 1920 and 1945 elevated daily Cd intake over that period. Between 1950 and 1965 the continued rise in food Cd content was balanced by a decline in consumption rate for food items having the highest Cd concentrations (i.e., potatoes). From 1965 to 1975, average daily intake was estimated to have increased slightly. Overall, increasing environmental Cd pollution and changes in food consumption patterns are concluded to have caused approximately constant daily Cd intake in the United States between 1945 and 1982. Bioavailability of Cd from various foods is not well understood. Factors that may influence bioavailability include the chemical form of Cd in consumed tissue, content of competing ions and ligands in the diet, effects of food preparation methods, and the nutritional state of the consuming animal. It is thought that Cd-metallothionein accounts for most Cd in animal tissues and thus foods derived from animals (Spivey Fox, 1983), but this is not established. This protein is relatively heat stable and may be unaffected during food preparation. Other Cd-binding proteins have been observed in shellfish. Their characteristics are not well established (Spivey Fox, 1983). The form of Cd in vegetable foods may vary considerably. As already discussed, PCs are probably not abundant in agricultural crops. More common forms of Cd in leaf and root crops may be Cd-

196

GEORGEJ. WAGNER

citrate (Wang et af., 1991), Cd-pectin, Cd-hemicellulose (Spivey Fox, 1988), and Cd-membrane complexes. Metal citrate complexes can be very stable (Francis er af., 1992) and their fate in the gastrointestinal tract of animals is worthy of study. Phylates are known to avidly bind dietary Zn and may also chelate Cd (Spivey Fox, 1988). Zn deficiency in animals is often associated with excretion of Zn-phytate. Cd in grain has been found to be associated with the protein fraction in soybean and wheat (Cataldo et af., 1981; Wagner et af., 1984). The form of Cd in consumed plant tissues has been shown to influence its gastrointestinal absorption and subsequent tissue distribution (Spivey Fox, 1988). Comparisons have been made of the bioavailability of Cd associated with plant foods versus inorganic Cd salts. Results have been contradictory (see McKenna et af., 1992). The influence of Cd form on tissue deposition after absorption has also been studied in a number of cases. Results suggest that Cd ingested in an inorganic form has a similar fate in terms of organ distribution as does Cd accumulated in leaf or grain during plant growth (Wagner et af., 1984; Weigel et al., 1987; McKenna et af., 1992). McKenna et af. (1992) have reported that the Occurrence of accumulated Zn and Cd in lettuce and spinach fed to quail resulted in lower bioavailability of Cd than when plants had only accumulated Cd. They concluded that toxicological studies using Cd salts and Zn supplements do not assess Cd bioavailablity in animals ingesting crops containing both high Zn and high Cd. Interactions between Zn and Cd are clearly important in Cd uptake, translocation, and its subsequent bioavailability from plant foods. Other nutrients, including Fe, Cu, Se, Ca pyridoxine, ascorbic acid, and protein (quantity and type), are thought to influence Cd bioavailability. These nutrients are believed to protect against the effects of Cd and nutrient imbalance exacerbated by Cd (Spivey Fox, 1983, 1988).

VI. HEALTH CONSEQUENCES OF CADMIUM INTAKEINHUMANS It is generally accepted that: A. Renal tubular dysfunction and pulmonary emphysema are the principal pathological lesions attributed to Cd exposure in humans. Less excepted, but compelling, is recent evidence suggesting a causal effect of chronic Cd exposure in osteoporosis/osteomalaciain certain individuals. B. Cd has no known biological function. Its long half-life in humans and known acute toxicity make it a potential hazard of concern under low-level chronic as well as high-level exposure conditions. C. Current information indicates that, while most individuals in the general

CADMIUM IN CROPS AND EFFECTS ON HUMAN HEALTH

197

populations of most industrialized countries are probably unaffected by current Cd intake from food and tobacco consumption, certain segments of certain human populations may be at health risk. D. No Cd chelation treatment is presently available for reducing Cd body burden in humans. Several critical reviews discuss the consequences of Cd exposure to humans (Fassett, 1975; Carmichael et al., 1982; Yasumura er al., 1980; Ryan ef al., 1982; Samarawickram, 1983; Hallenbeck, 1984; Friberg et al., 1986; Gairola er al., 1992). All of these sources point to foods and tobacco consumption as the main sources of Cd exposure. Though scientific data defining the impact of chronic low-level Cd exposure (as from food and tobacco use) is far from complete, restriction and regulation of industrial Cd use and disbursement and of the allowable Cd content of agricultural products will undoubtedly increase as a result of the perceived danger of Cd intake to human health. This perception of danger is exemplified by the introduction into the solid waste bill before the U.S. Congress of a section that would ban all “nonessential” use of Cd. This bill was recently introduced by the ranking minority member of the Committee on Environmental and Public Works. The section was introduced because of, as concluded by R. A. Denison, of the New York-based Environmental Defense Fund, “the extreme toxicity” of Cd and because “it can lead to severe kidney problems as well as to problems with bone development, the liver and pancreas” (Goode, 1990). The basis of the Cd accumulation problem is that its biological half-life in humans is 10-30 years depending on age. Excretion rate of Cd is estimated to be 0.005% of body burden and food is a constant source of intake (Ryan et al., 1982). Excessive exposure to Cd has been associated with various illnesses in humans, including gastroenteritis, renal tubular dysfunction, hypertension, cardiovascular disease, pulmonary emphysema, cancer, and osteoporosis. However, current acceptance is that the two principal, presently recognized, pathologic lesions associated with Cd intake are renal tubular dysfunction and pulmonary emphysema (Ryan et al., 1982; Gairola er al., 1992). Strong, recent evidence suggests that a role of Cd in osteoporosis may be generally important for a susceptible segment of the human population (Bhattacharyya er al., 1988a,b; Krall and Dawson-Hughes, 1991; Mazess and Barden, 1991). Although mutagenic and carcinogenic effects of Cd have been reported in some experimental studies (Nordberg, 1974; Heinrich, 1988), and Cd in cigarette smoke is a suspected organ-specific carcinogen (Surgeon General, 1981), the evidence for the carcinogenic potential of cigarette smoke Cd and, for that matter, Cd from any nominal source is considered inadequate (Ryan er al., 1982; Hallenbeck, 1984; Gairola er al., 1992). There appears to be no evidence that human exposure to Cd causes heritable genetic change or that Cd is tumorigenic or teratogenic at ex-

198

GEORGE J. WAGNER

posure levels that are environmentally relevant (Carmichael et al., 1982; Ryan er al., 1982). However, individual reports continue to appear that conclude, for example, increased risk for developing aggressive prostate tumors due to increased Cd exposure from occupation, high dietary Cd, or smoking (Elghany et al., 1990). The suggestion that Cd has significant adverse effects on the cardiovascular system of humans is controversial (Ryan et al., 1982; Geiger er al., 1989; Gairola et al., 1992). Current evidence does not support a role for Cd in hypertension (Whittemore et af., 1991). However, there is evidence that elevated blood Cd due to smoking [can be up to four-fold that in nonsmokers (Ryan et al., 1982; Sharma er al., 1983)] may contribute to coronary vasospasm or atherscoleritic processes (Koop er al., 1980; Kannell, 1981; Templeton and Cherian, 1983; Sackett et al., 1986; Chiba and Masironi, 1992). Cadmium exposure is also said to enhance susceptibility to bacterial, protozoal, and viral infections and to result in impaired humoral and cell-mediated immune responses (Blakley and Rajpal, 1986). However, these results were obtained in high-dose experiments, so the possible impact of chronic, low-dose exposure remains unclear. It is widely thought that renal tubular dysfunction, once induced by Cd exposure, is generally irreversible and can constitute a significant adverse health effect (Ryan et al., 1982; Friberg et al., 1986). What is debated is the level of Cd that is required to produce disease (Ryan et al., 1982; Hallenbeck, 1984) and what chemical form of renal Cd is critical to dysfunction and disease (Nomiyama, 1980; Babich and Davis, 1981; Sharma er al.. 1983; Nomiyama and Nomiyama, 1986). It is generally considered that a level of 200 to 300 pg/gm wet wt Cd in kidney cortex is the level above which renal damage is imminentusually evidenced by proteinurea (Ryan et al., 1982; Samarawickram, 1983; Hallenbeck, 1984) as observed principally in men occupationally exposed to Cd (Buchet et al., 1990). It was on this basis that the World Health Organization set a maximum tolerable level of intake at 70 p g Cd/day for most adults. An approximate estimated intake from food in the United States is 38 pg/day with an additional intake of the dietary equivalent of up to 25 pg from heavy smoking (Ryan et al., 1982). Given the assumption that 200 pg/gm wet wt kidney cortex is the threshold level for kidney disease, it has been said that at least 95% of the U.S. and European middle-age populations are sufficiently below this limit in Cd level (Ryan et al., 1982). There are reports, however, that suggest that the percentage of individuals “at risk” may be greater (Yost et al., 1980). In a study of nearly 1700 subjects aged 20-80 years, sampled randomly from the general population of four areas of Belgium having various levels of industrially derived Cd pollution, it was concluded that about 10%of the Belgian population absorb sufficient Cd to cause slight renal dysfunction (Buchet et al., 1990). A possible increased risk of Cd-induced renal disease in diabetic patients was also newly discovered in this study. It is reported that since the beginning of the nineteenth

CADMIUM IN CROPS AND EFFECTS ON HUMAN HEALTH

199

century, environmental Cd pollution in Europe may have caused a 50-fold rise in the Cd content of human renal cortex (Drasch, 1983). A 1982 to 1988 survey of blood Cd and Pb levels in West Germany showed that, while blood Pb has decreased since 1979 and 1981 surveys (presumably due to increased control of industrial and municipal pollution), no comparable decrease in blood Cd is found. Unlike Pb, blood Cd was no lower in rural citizens than in urbanites. As commonly observed, smokers had three to four times higher blood Cd than nonsmokers (Ewers et al., 1990). These reports are perhaps alarming. Additional studies of general and subpopulations appear to be needed. The second most recognized, pathological lesion attributable to Cd is pulmonary emphysema. It would appear that Cd-induced emphysema is related only to the inhalation route (Ryan et al., 1982). Cd absorption by the lung, while dependent on chemical form, particle size, etc., is efficient (40 to 50%) relative to absorption (4 to 8%, also dependent on matrix factors) in the GI tract (Ryan et al., 1982; Friberg et al., 1986). The severity of emphysema, the accumulation of Cd in kidney cortex and liver, and the body burden of Cd in active smokers are reported to be proportional to the number of pack-years smoked (Snider et al., 1973). Though the mechanism of Cd-induced lung damage is not known, it is thought that Cd may be involved in release of lysosomal enzymes from pulmonary alveolar macrophages and leucocytes contributing to emphysema (Samarawickram, 1983). Cadmium is also known to be a potent inhibitor of critical processes in alveolar macrophages. Research indicates that Cd does not cause destruction of lung elastin (Snider et al., 1988). Cd-induced air space enlargement may rather occur as a result of the influence of fibrosis. Tobacco Cd may play a role in induced pulmonary inflammation in chronically smoke-exposed rats (Gairola, 1989). As noted in the beginning of this section, there is perhaps a growing case for effects of Cd in osteoporosis/osteomalacia.The work of Bhattacharyya er al. (1988a,b) suggests that Cd may contribute to bone demineralization, particularly in multiparous and postmenopausal females. Experimental work done in mouse, fetal rat, and in v i m bone culture systems shows that low levels of Cd can have dramatic effects on Ca loss. Because the Cd levels used in these studies were comparable to that in the blood of smokers, the authors suggested that Cd in cigarette smoke together with smoking-induced, decreased estrogen levels may lead to increased incidence of fractures and early tooth loss as observed in smoking women. This recent work is particularly interesting in that conclusions drawn from it regarding the role of Cd in bone demineralization in ovariectomized and multiparous (calcium-stressed) animals may explain the fact that 95% of the victims of Itai Itai disease (severe osteoporosis/osteomalaciaand renal tubular dysfunction) were postmenopausal women. Elevated levels of Cd in the diet and drinking water were concluded to be causative factors in the 1964 Occurrence of Itai Itai disease in the Toyoma Prefecture in Japan (Hallenbeck, 1984). The

2 00

GEORGEJ. WAGNER

source of the Cd was acid drainage from a Pb-Zn-Cd mine into the Jintsu River. Water from the river was used for drinking and irrigation of rice paddies. Cd intake from food and water was estimated at 300 to 600 pglday (Ryan er al., 1982). Interestingly, in this incident, the occurrence of proteinurea and glucoseurea, lesser symptoms of Cd poisoning, was also higher in women than in men. Several groups of individuals have been identified as being “at risk” from excess Cd exposure, including: persons having severe nutritional deficiencies (Fe, Ca, Zn, protein, vitamin D) that are aggravated by Cd; persons consuming more than normal levels of visceral meats, fish, and shellfish; pregnant and lactating females having a negative Ca balance; nursing infants; persons with kidney ailments; and multiparous, postmenopausal women (Babich and Davis, 1981; Ryan et al., 1982; Friberg et al., 1986; Nichols, 1988; Spivey Fox, 1988). In another study (Ward et al., 1987) placental element levels were monitored in relation to fetal development for obstetrically normal births. Of 37 elements tested, only Cd and Pb correlated with significantly lowered birth weight. The authors concluded that lower birth weights associated with cigarette smoking during pregnancy appeared to result partially from increased Cd from cigarettes. Although it is assumed that most growth decrements in newborns are reversible, effects on development of the nervous system may not be (Carmichael et al., 1982). Other work suggests that Cd may result in lowered Zn transfer in the placenta and this may lead to Zn deficiency in the developing fetus (Kuhnert et al., 1988). While a number of potential Cd chelators have been tested in experimental animals, there is currently no chelation therapy available for reducing Cd body burden in humans (Ryan et al., 1982). As an interesting note, horses are considered to be at higher risk, since adult horses have about five-fold greater renal Cd levels than adult humans (Piscator, 1981). In summary, it is not established that chronic exposure to Cd from dietary foods and/or smoking represents a health risk, particularly to individuals not considered at risk for the reasons listed in the foregoing. However, each new study leads to more debate about the appropriateness of standards and methods of measurement, and new information suggests better parameters for establishing tolerable levels. There is little doubt that continued study of general populations, particularly in industrialized countries having a history of Cd pollution, will add new information regarding the most appropriate tolerable levels of Cd exposure. Another result of such studies will be increased public awareness of the Cd content of agricultural products. This will undoubtedly lead to lower allowable Cd concentration standards for agricultural products, particularly in certain international markets. It is logical that since environmental contamination with this nonessential, potentially toxic element is increasing and since food and tobacco use are clearly significant sources for Cd intake in humans, we are wise to seek logical methods for lowering Cd content of foods and tobacco leaf.

CADMIUM IN CROPS AND EFFECTS ON HUMAN HEALTH

201

VII. ATTEMPTS TO MANIPULATE PLANTS T O REDUCE THE CADMIUM CONTENT OF CROPS: PROMISE AND PROBLEMS Some generalizations may be made, but work in this area is at a preliminary stage. A. Seed-transmissible expression of animal metallothionein in crop plants may offer promise for sequestering Cd into nonconsumed tissues to reduce dietary Cd. However, efforts testing this strategy are in their infancy and these studies have revealed some difficulties with the approach. B. Introduction of high Cd binding capacity into leaves may be possible by infection of susceptible plants with recombinant virus encoding metallothionein. This approach, though perhaps impractical in crop production, may be useful in a bioremediation strategy and warrants further study. C. Studies to enhance vacuolar Cd sequestration may have promise for entrapment of Cd to lower Cd content in crops or in a bioremediation strategy. However, if mechanisms involved are not metal-specific, nutrient (Zn, Cu) imbalances may result. Manipulation of Cd accumulation in plants for practical purposes was alluded to at the end of the previous section. Here, current work and some future avenues in this area will be briefly discussed in an effort to present one view of the promise and problems associated with plant manipulation to achieve the capacity to alter patterns of Cd accumulation. To this author’s knowledge, little effort has been made in the past to utilize plant breeding to reduce Cd accumulation in products of agricultural plants. One barrier to this approach is the lack of an efficient screen for Cd accumulation. Another is our lack of understanding of plant responses associated with Cd accumulation that might be useful as markers in a screening program. Also, unlike the case of pest resistance, primitive progenitors of cultivated plants are not likely to have favorable Cd accumulation characteristics, at least in terms of total Cd accumulation. Most primitive terrestrial environments would not have contained phytotoxic levels of Cd and thus would have lacked selective pressure for developing a mechanism for Cd exclusion. Therefore, germplasm for breeding favorable Cd accumulation may be limited. We know of no variety releases from breeding programs focused on reducing Cd accumulation in crops. We have attempted to transfer the very high root-low leaf Cd accumulation characteristic of Nicotiunu rusticu (Wagner and Yeargan, 1986) to high roothigh leaf accumulating, cultivated N. tubucum. Since the basis of the N. rusricu Cd accumulation trait is not understood, screening has been on the basis of root versus shoot total Cd accumulation analysis in seedlings. To date, we have gen-

2 02

GEORGE J. WAGNER

erated an addition line that has morphological characteristics of the N. rabacum parent but N. rustica-like root/shoot Cd accumulation. This line retains one additional chromosome and has low fertility (M. T. Nielsen and G. J. Wagner, unpublished data). Current efforts are directed toward inducing recombination of the relevant gene(s). If maximally successful, this approach could result in a 50% reduction of leaf Cd in N. tabacum. Breeding for low seed Cd in sunflower is apparently also under way. Several groups have conducted experiments utilizing the recombinant DNA approach to manipulate Cd accumulation in plants. Until recently, these efforts have focused on expression of genes encoding animal metallothioneins in an effort to increase Cd accumulation potential. In the first published study, Brassicu plants were transformed by infection with a recombinant cauliflower mosaic virus carrying a Chinese hamster metallothionein gene (Lefebvre et a f . , 1987). Expression of the protein was achieved after systemic infection of plants with the recombinant virus. Infected tissue had increased Cd binding capacity and tolerance to metal as found in a leaf disk assay. Metallothionein accounted for 0.5% of soluble protein in infected leaves. Though perhaps not practical from a crop production point of view, it would be interesting to know if this strategy could be used to enhance Cd accumulation in leaves of potato to reduce Cd accumulation in the tubers. It may be a useful strategy for preparing super-accumulating plants for use in bioremediation and in biomining of precious metals. In an effort to obtain seed containing an integrated metallothionein gene, we have performed Agrobacrerium-mediated transformation to introduce mouse metallothionein into several tobaccos. The objective of the transformations was to entrap Cd in roots as Cd-metallothionein and thereby reduce Cd accumulation in shoots. The study was based on earlier work that characterized Cd accumulation in N. tabucum cv. KY 14 and indicated that the root regulates Cd transfer to the shoot (Wagner er a f . , 1988). Expression of metallothionein in leaves and roots of transgenic N. rabacum cv. KY 14 plants was at a level of about 0.05 to 0.1% of soluble protein (Maiti et al., 1988). In most primary transformants, the marker chimeric gene segregated with a ratio of 3: 1, characteristic of a single dominant gene (Maiti et af., 1989). Very young transformed seedlings were more tolerant to growth with high Cd then were controls. To compare Cd accumulation characteristics, transformed and control seedlings were exposed for 15 days in liquid culture at an estimated field, soil-solution-likeCd concentration of 0.02 pM. After this exposure, transformed seedlings of N. tabucum cv. KY 14 contained about 24% lower Cd concentration in shoots and about 5% higher Cd concentration in roots than did control seedlings. Dry weights did not differ significantly (Maiti et al., 1989). In the field, mature transformed plants exposed only to endogenous soil Cd contained about 14% lower leaf lamina Cd concentration than did controls. Leaf dry weights did not differ significantly but transformed plants had 12% fewer leaves and were 9% shorter in plant height than

CADMIUM IN CROPS AND EFFECTS ON HUMAN HEALTH

203

controls. Copper concentrations were about 10% higher in the bottom portion of transformed plants, suggesting that reduced leaf number and height may have been caused by Cu toxicity or deficiency (Yeargan e f al., 1992). Alternatively, these abnormalities may have resulted from somaclonal variation during tissue culture steps used after transformation, or gene insertion position effects. Zn concentrations were unchanged. With another variety, cv. Petit Havana, liquidcultured, transformed seedlings contained no less Cd in shoots, 48% higher Cd in roots, and differed in dry weight as compared with nontransformed controls. In the field no differences in leaf Cd, plant height, or leaf number were observed with this variety, control versus transformed. Since it is difficult to recover fine roots (highest Cd accumulating) from soil, it was not possible to examine this tissue in field-grown plants. To conclude, although comparison of additional metallothionein-expressing tobaccos and other plants is needed, results obtained with cultivar KY 14 support the hypothesis that sequestration of Cd in roots as Cd-metallothionein may have potential for reducing Cd content of above-root tissues in certain plants. However, as found in our successful experiment with KY 14, other effects of transformation may occur. In our case, Cu metabolism may have been altered because metallothionein binds this metal, indeed more avidly than Cd (Hamer, 1986). Experiments studying transformation of plants to introduce new chelators of Cd may be considered to be at an early stage. Our current efforts are directed toward defining the correlation between the presence of metallothionein and tissue Cd, enhancing the level of expression of metallothionein, affecting tissue-specific expression (Maiti er al., 1991), and modifying the gene to reduce Cu binding in the recombinant protein. In a somewhat similar study to ours (Misra and Gedamu, 1989), tobacco and Brassica seedlings expressing the human metallothionein I1 gene were shown to be less affected by the presence of high levels of Cd than were controls. Segregation ratios for the chimeric marker gene and for tolerance to growth with Cd were both 3 : 1 . In this case, the level of MT expression was not quantified, Cd accumulation characteristics were not tested, and field tests were not reported. In another study, no differences were found in Cd content of a field-grown tobacco expressing metallothionein (Brandle et al., 1993). Ow and co-workers (Ortiz er al., 1992) have taken a somewhat different strategy to manipulate Cd accumulation. A gene designated hmrl was isolated from an S. pombe mutant deficient in accumulation of a sulfide-PC-Cd complex that is formed during growth in the presence of very high levels of Cd. The gene is thought to encode for a tonoplast protein having homology to eucaryotic and procaryotic ABC-type membrane transport proteins. ABC-type transporters are known to function in membrane transport of ions, peptides, and other small molecules in mammalian and bacterial cells (see Ortiz et al., 1992). Yeasts containing a hmrl -expressing, multicopy plasmid were shown to have increased Cd

2 04

GEORGEJ. WAGNER

accumulation and tolerance under high-level exposure conditions. Overexpression of the hmtl gene in plants may enhance the potential for vacuolar sequestration of Cd and, if expressed tissue specifically, may allow manipulation of tissue Cd partitioning in a manner analogous to that described for metallothionein. Whether the hmtl putative translocator transports Cd-PC or Cd, or sulfidePC-Cd, remains to be determined. Also, its effects on Zn and Cu homeostasis remain unknown. However, results with S. pombe suggest that this transporter may at least have value for increasing Cd accumulation in a bioremediation strategy where high-level Cd exposure results in PC and sulfide formation. We note that the Cd/H antiporter activity described in Section IV may also have potential for altering Cd tissue distribution if the protein responsible for this activity is found to be specific for Cd or heavy metals. If it is not, its overexpression may disturb Zn/Cu balance. The lack of absolute metal specificity of chelators such as metallothionein may be a major obstacle to their use for partitioning Cd into nonconsurned plant tissues or in bioremediation strategies. Similarly, lack of absolute ion specificity of membrane transporters may be problematic. Zn and Cu deficiencies are rare in agriculture; however, both are known to occur in high-pH soils (Marschner, 1983). Though Zn is rarely toxic to plants (toxic at soil solution >20 ppm), Cu toxicity is more common, occurring in soil solutions having about 0.1 ppm Cu. Mechanisms of Zn and Cu partitioning in plants are not understood.

VIII. SUMMARY Accumulation of the pollutant metal Cd in humans is largely a result of its uptake by plants from soil and its concentration in plant-derived components of the human diet. The principal sources of Cd contamination of agricultural soils are phosphatic fertilizers, sewage sludge, and atmospheric fallout from industrial and municipal activities. Cd contamination of the agricultural environments is generally thought to be increasing. Levels of this trace pollutant in crop plants are generally well below phytotoxic levels, so healthy plants can contribute substantial Cd to the human diet. Thus, concern about Cd is focused on the agricultural production phase of the human food chain. Reducing the Cd content of phosphatic fertilizers during manufacture may be the most economical and realistic means for reducing future Cd input into the agricultural environment and human food chain. Most studies of Cd uptake and accumulation in plants to date have used highlevel Cd exposure to elicit responses, levels far above those encountered in agricultural production. There is little evidence to suggest that certain extensively studied responses to high-Cd exposure (i.e., phytochelatin and sulfide formation)

CADMIUM IN CROPS AND EFFECTS ON HUMAN HEALTH

205

significantly relate to the low-exposure field case. Our understanding of Cd uptake and accumulation processes that are important in agricultural production may be quite limited. High-level-exposure experimental studies may be more relevant in efforts to develop “super-accumulator” plants for use in bioremediation or biomining of soils and water contaminated with Cd and related metals. Crop plants vary somewhat in the tissue distribution of accumulated metal and therefore certain crop foods have more potential to contribute high Cd to the diet than do others. However, total Cd intake is dependent on the amount of a food consumed as well as on foodstuff Cd concentration. An example is the case of grains that generally contain moderate Cd concentration but contribute significant Cd in many diets because large amounts of grain-derived foods are generally consumed. Bioavailability of Cd from different foods is a complex and little understood aspect of the transfer of Cd from agricultural crops to humans. There is no conclusive evidence regarding negative health impact from chronic exposure to low levels of Cd as occurs from diet and smoking. However, certain groups of individuals have been identified as being “at risk” to negative health impact from such exposure. Renal tubular dysfunction and pulmonary emphysema are considered to be the principal pathological lesions attributable to Cd in humans. Recent evidence suggests possible impact of chronic low-level Cd exposure on bone health and on low birth weight in certain groups of individuals. Some recent epidemiological studies of Cd effects on general populations (as opposed to past studies focused on individuals exposed to Cd in the workplace) perhaps suggest the need for additional analyses of the impact of general environmental and dietary Cd on human health, particularly in populations of heavily industrialized and urban regions. Since Cd in the human diet is largely due to crop plant uptake and accumulation of this pollutant and since at least a perceived danger about Cd in crops will undoubtedly continue to impact agricultural markets, we are wise to investigate possibilities for altering Cd accumulation in crop plants to reduce its accumulation in consumed tissues. Such studies are in their infancy.

REFERENCES Adamu, C. A . Mulchi, C. L . , and Bell, P. F. (1989). Relationships between soil pH, clay, organic matter and CEC and heavy metal concentrations in soils and tobacco. Tob. Sci. 3 3 , 9 6 - 100. Andersen, 0. (1984). Chelation of cadmium. Environ. Health Perspect. 54, 249-266. Ayres, R . V. (1992). Toxic heavy metals: Materials cycle optimization. Proc. Narl. Acad. Sci. U.S.A. 89, 815-820. Babich, H . , and Davis, D. L. (1981). Food tolerances and action levels: Do they adequately protect children? BioScience 31,429-438. Baker, A . J. M. (1978). Fkophysiological aspects of zinc tolerance in Silene rnuririrna. New Phyrol. 80,637-647.

206

GEORGE J. WAGNER

Baker, A. J. M. (1981). Accumulators and excluders-Strategies in the response of plants to heavy metals. J. Plant. Nurr. 3, 643-654. Baleja, J. D., Marmorstein, R., Harrison, S. C., and Wagner, G. J. (1992). Solution structure of the DNA-binding domain of Cd2-GAL4 from S. cerevisiae. Nature (London) 356,450-453. Balliatori, N. (1991). Mechanisms of metal transport across liver cell plasma membranes. Drug. Merab. Rev. 23, 83- 132. Bhattacharyya, M. H., Whelton, B. D., Stern, P. H., and Peterson, D. P. (1988a). Cadmium accelerates bone loss in ovariectomized mice and fetal rat limb bone in culture. Proc. Narl. Acad. S C ~U.S.A. . 85, 8761-8765. Bhattacharyya, M. H., Whelton, B. D., Peterson, D. P., Carnes, B. A., Moretti, E. S . , Toomey, J. M., and Williams, L. L. (1988b). Skeletal changes in multiparous mice fed a nutrient-sufficient diet containing cadmium. Toxicology 50, 193-204. Bingham, F. T. (1979). Bioavailability of Cd to food crops in relation to heavy metal content of sludge-amended soil. Environ. Health Perspect. 28, 39-43. Blakley, B. R., and Rajpal, S. T. (1986). The effect of cadmium on antibody responses to antigens with different cellular requirements. Int. J . Immunopharmucol. 8(8), 1009- 1015. Boutron. C. F., Gorlack, V.,Candelone, J., Bolshov, M. A., and Delmas, R. J. (1991). Decrease in anthropogenic lead, cadmium and zinc in Greenland snows since the late 1960’s. Nature (London) 353, 153-156. Brandle, J. E., Labb6, H.,Hattori, J., and Miki, B. L. (1993). Field performance and heavy metal concentrations of transgenic flue-cured tobacco expressing a mammalian metallothionein-pglucuronidase gene fusion. Genome (in press). Brookes, A., Collins, J. C., and Thurman, D. A. (1981). The mechanism of Zn tolerance in grasses. J. Plant Nurr. 3, 695. Buchet, J. P., Lauwerys, R., Roels, H., Bernard, A., Bruaux, P., Claeys, F.. Ducoffre, G., DePlaen, P., Staessen, J., Amery, A., Lijnen, P.,Thijs, L., Rondia, D., Sartor, F., Saint Remy, A., and Nick, L. (1990). Renal effects of Cd body burden of the general population. Lancet 336, 699-702, Bush, D. R., and Sze, H. (1986). Calcium transport in tonoplast and endoplasmic reticulum vesicles isolated from cultured carrot cells. Plant Physiol. 80, 549-555. Carmichael, N. G., Backhouse, B. L., Winder, C., and Lewis, P. D. (1982). Teratogenicity, toxicity and perinatal effects of Cd. Hum. Toxicol. 1, 159- 186. Cataldo, D. A., and Wildung, R. E. (1978). Soil and plant factors influencing the accumulation of heavy metals by plants. Environ. Health Perspect. 27, 149-159. Cataldo, D. A., Garland, T. R.,and Wildung, R. E. (1981). Cadmium distribution and chemical fate in soybean plants. Plant Physiol. 68, 835-839. Chiba, M., and Masironi. R. (1992). Toxic and trace elements in tobacco and tobacco smoke. Bull. W.H.O. 70,269-275. Copius Peereboom-Stegman, J. H. J., and Copius Peereboom, J. W. (1989). The intake of Cd in the kempen, an area in the south of the Netherlands. Ecotoxicol. Environ. Saj. 18, 93- 108. Cox, R. M. (1986). Contamination and effects of Cd in native plants. Experienria, Suppl. 50, 101-109. Cumming, J. R., and Taylor, G. J. (1990). Mechanisms of metal tolerance in plants: Physiological adaptations for exclusion of metal ions from cytoplasm. I n “Stress Responses in Plants: Adaptation and Acclimation Mechanisms,” pp. 329-356. Wiley-Liss, New York. Davis, R. D. (1984). Cadmium in sludges used as fertilizer. Experienria 40, 117- 126. de Framond, A. J. (1991). A metallothionein-likegene from maize. FEBS Leu. 290, 103- 106. Drasch, G . A. (1983). An increase of Cd body burden for this century. Sci. Total Environ. 26, 111 - 1 19.

CADMIUM IN CROPS AND EFFECTS ON HUMAN HEALTH

207

Elghany, N. A., Schumacher, M. C., Slattery, M. L., West, D. W., and Lee, J. S. (1990). Occupation, cadmium exposure, and prostate cancer. Epidemiology 1, 107- 115. Elinder, C. G. (1986). Cadmium: Uses, occurrence, and intake. I n “Cadmium and Health: A Toxicological and Epidemiological Appraisal” (L.Friberg, c . G. Elinder, T. Kjellstmm, and G . F. Nordberg, eds.), Vol. I , pp. 23-63. CRC Press, Boca Raton, FL. Ewers, U., Brockhaus, A., Dolgner, R., Freier, I., Turfeld, M., Engelke, R., and Jermann. E. (1990). Levels of Pb and Cd in blood of 55-66 year-old women living in different areas of North Rhine-Westphalia-Chronological trend 1982- 1988. Zenrrulbl. Hyg. 189,405-418. Fassett, D. W. (1975). Cadmium: Biological effects and occurrence in the environment. Annu. Rev. Pharmucol. 15,425-435. Francis, A. J., Dodge, C. J., and Gillow, J. B. (1992). Biodegradation of metal citrate complexes and implications for toxic-metal mobility. Nurure (London) 356, 140- 142. Frank, R.. Braun, H. E., Holdrinet, M., and Stonefield, K. I. (1977). Metal contents and insecticide residues in tobacco soils and cured tobacco leaves collected in southern Ontario. Tob. Sci. 21, 74-80. Friberg, L., Elinder, C., Kjellstrom, T., and Nordberg, G. F., eds. (1986). “Cadmium and Health: A Toxicological and Epidemiological Appraisal,” Vols. 1 and 2. CRC, Boca Raton, FL. Gairola, C. G. ( 1989). Cadmium-enriched cigarette smoke-induced cytological and biochemical alterations in rat lungs. J. Toxicol. Environ. Heulrh 27, 317-329. Gairola, C. G., Wagner, G. J., and Diana, J. N. (1992). Tobacco, Cd and health: A brief review. J. Smoking-Reluf.Dis. 3, 1-6. Geiger, H.. Bahner, U.. and Hiedland, A. (1989). Does cadmium contribute to the development of renal parenchymal hypertension? Inr. J . Arrific. Organs 12, 733-737. Gekeler, W., Grill, E., Winnacker, E., and Zenk, M. H. (1989). Survey of the plant kingdom for the ability to bind heavy metals through phytochelatin. Z . Nuturforsch., B: Chem. Sci. 448, 361-369. Goode. S. (1990). They draw the line at cadmium. Chem. Eng. News July, 60-61. Grill, E., Winnacker, E. L., and Zenk, M. H. (1986). Synthesis of seven different homologous phytochelatins in metal-exposed S.pombe cells. FEBS Len. 197, 115- 120. Grill, E., Winnacker, E. L., and Zenk, M. H. (1987). Phytochelatins, a class of heavy-metal-binding peptides from plants are functionally analogous to metallothioneins. Proc. Nurl. Acud. Sci. U.S.A. 84,439-443. Grill, E., Winnacker, E. L., and Zenk, M. H. (1988a). Occurrence of heavy metal binding phytochelatins in plants growing in a mining refuse area. Experienriu 44, 539-540. Grill, E., Thurman, J., Winnacker, E. L.. and Zenk, M. H. (1988b). Induction of heavy-metal binding phytochelating by inoculation of cell cultures in standard media. Plant Cell Rep. 7, 375-378. Grill, E., Loffler, S., Winnacker, E. L., and Zenk, M. H. (1989). Phytochelatins, the heavy-metalbinding peptides of plants are synthesized from glutathione by specific y-glutamyl cysteine dipeptidy1 transpeptidase (phytochelatin synthase). Proc. Nurl. Acud. Sci. U.S.A. 86, 6838-6842. Grill, E.. Winnacker, E. L., andZenk, M. H. (1991). Phytochelatins. In “Methods in Enzymology” (J. F. Riordan and B. L. Vallee, eds.), Vol. 205, pp. 333-341. Academic Press, San Diego. Gutenmann, W. H., Bache, C. A., and Lisk, D. J. (1982). Cadmium and nickel in smoke of cigarettes prepared from tobacco cultured on municipal sludge-amended soil. J . Toxicol. Environ. Health 10,423-431. Hallenbeck, W. H. (1984). Human health effects of exposure to cadmium. Experienriu 70, 136- 142. Hamer, D. H. (1986). Metallothionein. Annu. Rev. Biochem. 55,913-915. Hayashi, Y., Nakagawa. C. W., and Mutoh, N. (1991). Two pathways in the biosynthesis of cadystins (yEC).G in cell-free system of the fission yeast. Biochem. Cell. Biol. 69, 115-121.

GEORGE J. WAGNER

208

Heinrich, V. (1988). Cadmium and cancer. I n “Cadmium” (M. Stoepler and M. Piscator, eds.), pp. 13-36. Springer-Verlag. Berlin and New York. Hinkle, P. M., Kinsella, P. A,, and Osterhoudt, K. C. (1987). Cadmium uptake and toxicity via voltage-sensitive calcium channels. J . Biol. Chem. 262, 16333- 16337. Hirt, H., Casari, G., and Barta, A. (1989). Cadmium-enhanced expression in suspension-culture cells of tobacco. Planta 179,414-420. Hopps, H. C. (1974). Overview. I n “Geochemistry and the Environment,” Vol. I , pp. 3-21. Natl. Acad. Sci., Washington, DC. Jarvis, S. C., Jones, L. H., and Hopper, L. J. (1976). Cadmium uptake from solution and its transport from roots to shoots. Plant Soil 44, 179- 191. John, M. K.. and Van Laerhoven, C. J. (1976). Differential effects of cadmium on lettuce varieties. Environ. Pollut. 10, 163-173. John, M. K.. Van Laerhoven, C. J., and Chuah, H. H. (1972). Factors affecting plant uptake and phytotoxicity of Cd added to soil. Environ. Sci. Technol. 6, 1005- 1008. Jones, K. C., Jackson, A., and Johnston, A. E. (1992). Evidence for an increase in the Cd content of herbage since the 1860’s. Environ. Sci. Technol. 26, 834-836. Kagi, 1. H. R. (1991). Overview of metallothionein. I n “Methods in Enzymology” (J. F. Riordan and B. L. Vallee, eds.), Vol. 205, pp. 613-626. Academic Press, San Diego. Kannell. W. B. (1981). Update on the role of cigarette smoking in coronary artery disease. Am. Heart J . 101(3), 319-328. Kaptein, R. (1991). Zinc fingers. Curr. Opin. Strucr. Biol. 1, 63-70. Karez, C. S . , Allemand, D., DeRenzis. G., Gnassia-Barelli, M., Romeo, M., and Puiseux-Ozo, S. (1990). Ca-Cd interaction in the prynnesiophyte Cricosphaera elongata. Plant Cell Environ. 13,483-487.

Kawashima, I., Inokuchi, Y., Chino, M., Kimura, M.. and Shimizu, N. (1991). Isolation of a gene for a metallothionein-like protein from soybean. Plant Cell Physiol. 32, 913-916. Kille, P., Winge, D. P., Harwood, J. L., and Kay, J. (1991). A plant metallothionein produced in E. coli. FEES Lett. 295, 17 1 - 175. Kjellstrom, T., Lind, B., Linnman, L., and Elinder, C. (1975). Variation in Cd concentration in Swedish wheat and barley. Arch. Environ. Health 30, 321-328. Kondo, N., Imai, K., Isobe, M., Goto, T., Murasugi, A., Wanda-Nakagawa, C., and Itayashi, Y. (1984). Structure of cadystin, the unit peptide of Cd-binding peptides induced in fission yeast. Tetrahedron Lett. 25, 3869. Koop, S., Glonek, T.,and Erlanger, M. (1980). Altered metabolism and function of rat heart following chronic low level cadmiumllead feeding. J . Mol. Cell. Cardiol. 12, 1407- 1425. Krall, E., and Dawson-Hughes, B. (1991). Smoking and bone loss among postmenopausal women. J . Bone Miner. Res. 6, 33 1-338. Krotz, R. M., Evangelou, B. P., and Wagner, G. J. (1989). Relationships between Cd, Zn, Cd-peptide and organic acid in tobacco suspension cells. Plant Physiol. 91, 780-787. Kuhnert, B. R., Kuhnert, P. M., Lazobnik, N., and Erhard, P. (1988). The effect of maternal smoking on the relationship between maternal and fetal zinc status and infant birth weight. J . Am. Coll. Nutr. 7, 309-316. Lefebvre, D. D.. Miki, B. L., and Lalibertk, J. (1987). Mammalian metallothionein functions in plants. Bio Technology 5 , 1053- 1056. Loenstein, F. W. (1978). What foods are eaten, how often, by whom? I n “Geochemistry and the Environment” (K. C. Beeson, ed.), Vol. 3, Chapter V. Natl. Acad. Sci., Washington, DC. Maiti, 1. B., Hunt, A. G . , and Wagner, G. J. (1988). Seed-transmissable expression of mammalian metallothionein in transgenic tobacco. Biochem. Biophys. Res. Commun. 150, 640-647. Maiti, 1. B., Wagner, G. J., Yeargan, R., and Hunt, A. G . (1989). Inheritance and expression of the mouse metallothionein gene in tobacco. Plant Physiol. 91, 1020- 1024.

CADMIUM IN CROPS AND EFFECTS ON HUMAN HEALTH

209

Maiti, 1. B., Wagner, G. J., and Hunt, A. G. (1991). Light inducible and tissue-specific expression of a chimeric mouse metallothionein cDNA gene in tobacco. Plant Sci. 76, 99- 107. Marschner, H. (1983). General introduction to the mineral nutrition of plants. Encycl. Plant Physiol.. New Ser. 154, 5-49. Mazess, R., and Barden, H. (1991). Bone density in p-emenopausal women; Effects of age, dietary intake, physical activity, smoking and birth control pills. Am. J. Clin. Nutr. 53, 132- 142. McKenna, I. M., Chaney, R. L., Tao, S., Leach, R. M., Jr., and Williams, F. M. (1992). Interactions of plant Zn and plant species on the bioavailability of plant Cd to Japanese quail fed lettuce and spinach. Environ. Res. 57, 73-87. Mehra, R. K., and Winge, D. R . (1991). Metal ion resistance in fungi: Molecular mechanisms and their regulated expression. J. Cell. Biochem. 45, 30-40. Mench, M., Tiancogne, J., Gomez, and Juste, C. (1989). Cadmium bioavailability to N. tabucum, N . rusticu and Z. mays grown in soil amended or not amended with cadmium nitrate. Biol. Fertil. Soils 8, 48-53. Misra, S., and Gedamu, L. (1989). Heavy metal tolerant transgenic Brussicu and Nicoriunu plants. Theor. Appl. Genet. 78, 161-168. Mulchi. C. L., Bell, P. F., Adamu, C., and Heckman, J. R. (1987). Bioavailability of heavy metals in sludge-amended soils ten years after treatment. Recent Adv. Phyrochem. 21, 235-259. Naylor, L. M., and Loehr, R. C. (1981). Increase in dietary Cd as a result of application of sewage sludge to agricultural land. Environ. Sci. Techno/. 15, 881-886. Nichols, J. (1988). Pregnancy, cadmium and tobacco. J. R. Coll. Gen. Pruct. 38,310. Nieboer, E., and Richardson, D. H. S. (1980). The replacement of the nondescript term “heavy metals” by a biologically and chemically significant classification of metal ions. Environ. Pollut., Ser. B l, 3-26. Nomiyama, K. (1980). Recent progress and perspectives in cadmium health effects series. Sci. Total Environ. 14, 199-232. Nomiyama, K., and Nomiyama, H. (1986). Critical concentration of ‘‘unbound” cadmium in the rabbit renal cortex. Experientiu 42, 149. Nordbeg. G. (1974). Hazards of environmental cadmium pollution. Ambio 3, 55-65. Nriagu, J. 0. (1980). Cadmium in the atmosphere and in precipitation. I n “Cadmium in the Environment” (J. 0. Nriagu, ed.), Part I , pp. 35-1 14. Wiley. New York. and Pacyna, J. M. (1988). Quantitative assessment of worldwide contamination of Nriagu, J. 0.. air, water and soils by trace metals. Nature (London) 333, 134- 139. Nuciflora, G., Chu, L., Misra, T. K., and Silver, S. (1989). Cadmium resistance from S. uureus plasmid PI258 CadA gene results from a cadmium efflux ATPase. Proc. Nutl. Acud. Sci. U.S.A. 86, 3544-3548. Ortiz, D. F.,Kreppel. L., Speiser, D. M.,Scheel, G.,McDonald, G., and Ow, D. W. (1992). Heavy metal tolerance in fission yeast requires an ATP-binding cassette-type vacuolar membrane transporter. EMBOJ. 11, 3491-3499. Page. A. L.. Bingham, F. T., and Chang, A. C. (1981). Cadmium. I n “Effect of Heavy Metal Pollution on Plants” (N. W. Lepp, ed.), Vol. I , pp. 77- 109. Appl. Sci. Publ., London. Pakaline, D.. Nallendorf, A. F.. and Upites, V. (1970). Little investigated trace elements in Chlorellu cultures: Cadmium. Lam. PSR Zinat. Akad. Vestis 11, 16. Piscator, M. (1981). Toxic metals of clinical importance. Prog. Clin.Biol. Res. 77, 209-217. Poole. R. J. (1988). Plasma membrane and tonoplast. I n “Solute Transport in Plant Cells and Tissues’’ (J. L. Hall, ed.), pp. 83- 105. Longman, London. Price, N. M.. and Morel, M. M. (1990). Cadmium and cobalt substitution for Zn in a marine diatom. Nature (London) 344,658-660. Rauser, W. E. (1990). Phytochelatins. Annu. Rev. Plunr Physiol. 59,61-86. Reese, N. R., and Wagner, G. J. (1987a). Effects of buthionine sulfoximine on Cd-binding peptide

2 10

GEORGE J. WAGNER

levels in suspension-cultured tobacco cells treated with Cd, Zn or Cu. Plunr Physiol. 84, 574- 577.

Reese, N. R., and Wagner, G. J. (3987b). Properties of tobacco cadmium-binding peptide@). Biochem. J. 241,641-647. Reese, N. R., and Winge, D. R. (1988). Sulfide stabilization of Cd-y-glutanyl peptide complex of S. pombe. J. Biol. Chem. 263, 12832- 12835. Reese, N. R., Mehra, R. K., Bart Tarbet, E., and Winge, D. R. (1988). Studies on the y-glutamyl Cu-binding peptide from Schizosaccharomyces pombe. J . Biol. Chem. 263,4186-4192. Robinson, N. J., and Jackson, P. J. (1986). “Metallothionein-like” metal complexes in angiospersm; Their structure and function. Physiol. Planr. 67, 499-506. Robinson, N. J., Gupta, A., Fordham-Skelton, A. P., Coy, R. R. D., Whitton, B. A., and Huckle, J. W. ( 1990). Procaryotic metallothionein gene characterization and expression: Chromosome crawling by ligation-mediated PCR. Proc. R. SOC. London. Ser. B 242, 241-247. Ryan, J. A., Pahren, H. R., and Lucas, J. B. (1982). Controlling cadmium in the human chain: Review and rationale based on health effects. Environ. Res. 28, 251-302. Sackett, D. L., Gibson, R. W., Bross, 1. D. J., and Pickren (1986). Relation between aortic artherosclerosis and the use of cigarettes and alcohol. N . Engl. J . Med. 279, 1413- 1420. Salt, D., and Wagner, G.J. (1993). Transport of Cd in tonoplast of oat roots: Evidence for a Cd/H antiport activity. J . Biol. Chem. 268, 12297-12302. Samarawickram, G.(1983). Cadmium in animal and human health. In “Trace Elements in Health” (1. Rose, ed.), pp. 21-43. Butterworth, London. SchalTer, A. (1991). Absorption, circular dichroism, and magnetic circular dichroism spectroscopy of metallothionein. In “Methods in Enzymology” (J. F. Riordan and B. L. Vallee, eds.), Vol. 205, pp. 529-540. Academic Press, San Diego. Sharma, R. P., Kjellstrom, T., and McKenzie, J. M. (1983). Cadmium in blood and urine among smokers and non-smokers with high Cd intake via food. Toxicology 29, 163- 171. Sherlock, J. C. (1984). Cadmium in foods and diet. Experientia 40, 152-156. Smith, G. C., and Brennan, E. G. (1983). Cadmium-zinc interrelationships in tomato plants. Phyroparhology 73,879-882. Snider, G. L., Hayes, J. A., Korthy, A. L.,and Lewis, G. P. (1973). Centrilobular emphysema experimentally induced by cadmium chloride aerosol. Am. Rev. Respir. Dis. 108,40-48. Snider, G.L.. Lucey, E. C., Faris, B. Yangja, J.-L., Stone, P. J., and Franzblau, C. (1988). Cadmium-chloride-induced air-space enlargement with interstitial pulmonary fibrosis is not associated with destruction of lung elastin. Am. Rev. Respir. Dis. 137,918-923. Spivey Fox, M. R. (1983). Cadmium bioavailability. Fed. Proc., Fed. Am. SOC. Exp. Biol. 42, 1726-1729.

Spivey Fox, M. R. (1988). Nutritional factors that may influence bioavailability of Cd. J. Environ. Qua/. 17, 175-180.

Steffens, J. C. (1991). The heavy metal-binding peptides of plants. Annu. Rev. Planr Physiol. 41, 553- 575.

Steffens, J. C., Hunt, D. F., and William, B. G. (1986). Accumulation of non-protein metal-binding polypeptides (y-glutamyl-cystinyl).-glycine in selected Cd-resistant tomato cells. J . Biol. Chem. 261, 13879-13882.

Street, J. J., Sabey, B. R., and Lindsey, W. L. (1978). Influence of pH, phosphorus, cadmium, sewage sludge, and incubation time on the solubility and plant uptake of Cd. J. Environ. Quul. 7,286-290.

Surgeon General (1981). “The Health Consequences of Smoking: The Changing Cigarette,” A report to the Surgeon General, PHS, DHHS (PHS) 81-05156. U.S. Dept. of Health and Human Services, Washington, DC.

CADMIUM IN CROPS AND EFFECTS ON HUMAN HEALTH

2 11

Templeton, D. M., and Cherian, M. D. (1983). Cadmium and hypertension. Trends Pharmacol. S C ~4, . 501-503.

Thurman, J., Grill, E., Winnacker, E. L., and Zenk, M. H. (1991). Reactivation of metal-requiring apoenzymes by phytochelatin-metal complexes. FEES Len. 284, 66-69. Tommey. A. M.. Shi, J., Lindsay, W. P., Urwin, P. E., and Robinson, N. J. (1991). Expression of the pea gene PsMT, in E . coli. FEES Lett. 292.48-52. Tomsett, A. B.. and Thurman, D. A. (1988). Molecular biology of metal tolerances in plants. Plant, Cell Environ. 11,383-394. Travis, C. C., and Etnier, E. L. (1982). Dietary intake of Cd in the United States: 1920-1975. Environ. Res. 27, 1-7. Tukendorf, A., and Rauser, W. E. (1990). Changes in glutathione and phytochelatins in roots of maize seedlings exposed to Cd. Plant Sci. 70, 155- 166. Vallee, B. L., and Auld, D. S. 1990. Active site zinc ligands and activated H 2 0 of zinc enzymes. Proc. Narl. Acad. Sci. U.S.A. 87, 220-224. Vallee, B. L., and Ulmer, D. (1972). Biochemical effects of mercury, cadmium and lead. Annu. Rev. Eiochem. 41, 91-128. Van Bruwaene. R., Kirchmann, R., and Impens, R. (1984). Cadmium contamination in agriculture and zoo technology. Experientia 40,43-52. Verkleij, J. A. C., and Schat, H. (1990). Mechanisms of metal tolerance in higher plants. I n “Evolutionary Aspects of Heavy Metal Tolerance in Plants” (J. Shaw, ed.), pp. 179-193. CRC Press, Boca Raton, FL. Verkleij, J. A. C., Koevoets, P.. Van’t Riet, J., Bank, R., Nijdam, Y., and Emst, W. H. 0. (1990) Poly(y-glutamylcysteinyl)glycines or phytochelatins and their role in Cd tolerance in Silene vulgaris. Plant, Cell Environ. 13, 913-921. Verkleij, J. A. C.. Lolkems, P. C., DeNeeling, A. L., and Harmens, H. (1991). Heavy metal resistance in higher plants: Biochemical and genetic aspects. I n “Ecological Responses to Environmental Stresses” (J. Rozema and J. A. C. Verkleij, eds.), pp. 8-19. Kluwer Academic Publ.. The Netherlands. Vogeli-Lange, R. (1989). Studies of the role of Cd-binding peptides in tolerance to Cd in tobacco. Ph.D. Thesis, University of Kentucky, Lexington. Vogeli-Lange, R., and Wagner, G. J. (1990). Subcellular localization of cadmium and cadmiumbinding peptides in tobacco leaves-Implication of a transport function for cadmium-binding peptides. Plant Physiol. 92, 1086- 1093. Wagner, G. J. (1979). The subcellular site and nature of intracellular Cd in plants. Trace Subst. Environ. Health 13, 115-123. Wagner, G. J., and Krotz, R. M. (1989). Perspectives on Cd and Zn accumulations, accommodation and tolerance in plant cells: The role of Cd-binding peptide versus other mechanisms. I n “Metal Ion Homeostasis: Molecular Biology and Chemistry” (D. H. Hamer and D. R. Winge, eds.), pp. 325-336. A. R. Liss, New York. Wagner, G. J., and Trotter, M. M. (1982). An inducible Cd-binding complex in cabbage and tobacco leaves. Plant Physiol. 69, 804-809. Wagner, G . J., and Yeargan. R. (1986). Variation in Cd accumulation potential and tissue distribution of Cd in tobacco. Plant Physiol. 82, 274-279. Wagner, G . J., Nulty, E., and LeFevre, M. (1984). Cadmium in wheat grain: Its nature and fate after ingestion. J . Toxicol. Environ. Health 13,979-989. Wagner, G . J., Sutton, T. G . , and Yeargan, R. (1988). Root control of leaf cadmium accumulation in tobacco. Tob. Sci. 190(20), 64-68. Walter, H., and Stadleman, E. (1968). The physiological prerequisites for the transition of autotrophic plants from water to terrestrial life. BioScience 18, 694-701.

2 12

GEORGEJ. WAGNER

Wang, J., Evangelou, B. P., Nielsen, M. T., and Wagner, G. J. (1991). Computer simulated evaluation of possible mechanisms for quenching Cd activity in plant vacuoles. Plant Physiol. 97, 1154-1160. Wang, J., Evangelou, B.P., Nielsen. M. T., and Wagner, G. J. (1992). Computer-simulated evaluation of possible mechanisms for sequestering metal ion activity in plant vacuoles. 11. Zinc. Plant Physiol. 99,621 -626. Ward, N. I . , Watson, R.,and Bryce-Smith, D. (1987). Placental element levels in relation to fetal development for obstetrically “normal” births: A study of 37 elements. Evidence for effects of cadmium, lead and zinc on fetal growth, and for smoking as a source of cadmium. Inr. J . Biosocial Res. 9(1), 63-81. Weigel, H. J., and Jager, H . J. (1980). Subcellular distribution and chemical form of Cd in bean plants. Plant Physiol. 65,480-482. Weigel, H. J., Hge, D., Elmadja, I., and Jager, H. J. (1987). Availability and toxicological effects of low levels of biologically bound Cd. Arch. Environ. Contarn. Toxicol. 16, 85-93. Whittemore, A. S . , DiCiccio, Y., and Provenzeno, G. (1991). Urinary cadmium and blood pressure: Results from the NHANES I1 survey. Environ. Health Perspect. 91, 133- 140. Williams, C. H.,and David, D. J. (1976). The accumulation in soil of Cd residues from phosphate fertilizers and their effect on the Cd content of plants. Soil Sci. 121, 86-93. Wood,J. M., and Wang, H.-K. (1983). Microbial resistance to heavy metals. Environ. Sci. Technol. 17,582A-590A. Woolhouse, H. W. (1983). Toxicity and tolerance in the responses of plants to metals. In Encycl. Plant Physiol., New Ser. 12C, 245-300. Yasumura, S . , Vartsky, D., Ellis, K. J., and Cohn, S. H. (1980). Cadmium in human beings. In “Cadmium in the Environment” (J. 0. Nriagu, ed.), Part I , pp. 12-34. Wiley (Interscience), New York. Yeargan, R.,Maiti, I. B., Nielsen, M. T., Hunt, A. G., and Wagner, G. J. (1992). Tissue partitioning of Cd in transgenic tobacco seedlings and field grown plants expressing the mouse metallothionein 1 gene. Transgenic Res. 1, 261-267. Yost, K. J. (1984). Cadmium, the environment and human health: An overview. Experientia 40, 157-164. Yost, K. J., Miles, L. J., and Parsons, T. W. (1980). A method for estimating dietary intake of environmental trace contaminants: Cadmium-A case study. Environ. Int. 3,473-484.

Index A

individuals at risk, 200 input to agricultural soils, 175- 177 occurrence in agricultural environment, 174-178 phytochelatin role, 185- 188 properties, 174 soil content, 174- 175 solubility and soil pH. 179 transport across plant plasmalemma. 190 Calcium distribution in soil, 55-56 gypsum application effects. 9- 10, 13 Cardiovascular system. cadmium effects, I98 Chisel-plow system effect on residue cover, 59-60 surface water and, 40 Compaction, soil, 45-46 Cone index, gypsum effects, 23-25 Confidence interval method, 142 Conservation tillage adoption and suitability, 36-37 advantages and disadvantages, 37-38 definitions, 35-36 energy use, 44-45 evolution, 34-35 future needs and direction, 73 growing cover crops for mulch and nitrogen, 61-65 nutrient management fertilizers, 66-69 nitrogen, 65-66 pest management, 69-72 progress, 72-73 soil chemical properties, 52-57 erosion, 38-39 physical properties. 45-52 productivity, 42-43 surface mulch management, 57-65 groundwater quality, 39-42 Copper, homeostasis, I87 Cotton seedlings, phosphogypsum effect on emergence, 20, 22

Aggregation, soil, 46-48 Alfisol. dispersion, 20 Alleles A/-br. 111 01-dt, 110- I I I a l - m l . 95-97 al-m(papu), 103- 104

CI locus, 112-113

wxx-m9,93-94 Aluminum AI’+, co-sorption with SO:-. 16 gypsum application effects, 9- 10, 13 Antibiosis. 146 Antixenosis. 146- 147 Arthropods, see also Integrated pest management plant resistance. 145- 148 population regular, 154- 155 selection pressures, 155- 156

B Biological control, pests, 151 Bulk density, soil, 45-46

C Cadmium accumulation cellular and subcellular aspects, 184- 193 modes of control, 188- 189 plant manipulation to reduce, 201 -204 in plant vacuole, 191-192 whole-plant studies, 179- 184 binding to cell walls, 188- 190 bioavailability from foods, 195- 196 chelation strategy. 190- 191 crop uptake, 177- 178, 204-205 detoxification. phytochelatin role, 188 emissions, 175- 176 excessive exposure to, 197- 198 human intake health consequences. 196-200 maximum tolerable level. 198 from plants, 193- 196

213

2 14

INDEX

Cover crops adaptability, 61 -62 killing, 64-65 seeding, 62-64 self-reseeding, 64 Crop residues benefits, 58-59 tillage effect, 59-61 Crops cadmium uptake, 177- 178, 204-205 cover, see Cover crops damage curve, 132- 133 environmental factors, 134 responses to gypsum, 5- 1 I effects on aluminum and calcium, 9- 10 prediction, 11 - 12 rooting. 8-9 soil pH, 7-8 yield, 5-7. 9-10 responses to injury, 133- 134 arthropod injury, 134- 135 rotation, 149 stress management, 131 pest tolerance, 130- 135 Cultural management, integrated pest management. 148-151

D Damage curve, 132- 133 Diseases, pest management, 70-72 Drosophila, chromosomes, position effect, 91

E Ecosystems, impact of integrated pest management, 156-158 Emphysema, cadmium-induced, 199 Energy, analysis under tillage systems, 44-45 Entomopathogens, 153 Enzymatic changes, transposable elements, 111-1 14 Erosion, soil, see Soil, erosion Evaporation, soil water, 49-50

F Fertilizer application methods, 67-68 improving nitrogen efficiency, 68-69 management in conservation tillage, 66-69 Food, consumption and cadmium intake, 194- 195

G Genes A l , 92-93 Adh-I, 111-112 concept at time of McClintock experiments, 90-92 p h l , 91 unstable, pre-McClintock period concepts. 90 Genomes presence of transposable elements, 114- 115 yield components, 82-83 Germplasm, exotic, 8 I Groundwater, quality, conservation tillage and, 40-42 Gypsum as ameliorant for acid subsoils co-sorption of SOf- and All+, 16 crop responses, 5- 11 ion pair formation, 17 leaching studies, 14- I5 precipitation of solid phases, 16 self-liming effect, 15- 16 soil solution studies, 12- 14 as ameliorant for soil physical properties clay dispersion, 17-18 hardsetting soils, 22-23 hydraulic conductivity, 25-26 subsurface hard layers, 23-25 surface crusting, 17-22 applications, I composition, 2-3 industrial by-product materials, 2 mined, 2 properties, 2-4

H Hardsetting, soils, 22-23 Harvest, manipulation of date, pest management, 150, 153-154 Herbicides killing cover crops, 64 volatilization, tillage systems and, 41 Hydraulic conductivity, in soils, gypsum effects, 25-26

I Infiltration soil, 46-48 soil water, 48

INDEX Insecticides biological, 153 preventative and therapeutic roles, 152I53 resistant biotypes, 156 Insects, pest management, 70-72 Integrated pest management, 125- 166 based on ecological principles, 129 crop tolerance to injury, 130- I35 current status, 158- 159 decision making. 135- 139 economic injury level, 128. 135-139 equation, 136 incorporation of environmental risks, 138 limitations, 137 multiple species, I38 yield loss per pest, 137 economic threshold, 138- 139 ecosystem impact. 156- 158 focus, 165 improving efficiency. 159- 162 increased integration of tactics, 162- 164 lack of knowledge, 161 management actions biological, I5 I cultural, 148- 151 plant resistance to arthropods, 145- 148 therapeutic, 152- 154 objectives, 126- I27 pest interactions, 161 - 162 pests and test status. 127-128 philosophies for preventing pest damage, 128-130 population regulation, 154- 155 postinsecticide era, 129 sampling and monitoring, 139- 145 binomial sampling. 143- 144 confidence interval method, 142 field basis versus state or region basis, I40 level of precision, 141 program development, 140 sequential probability ratio test, 142- 143 sequential sampling, 141- 143 time-sequential sampling, 144 variable intensity sampling, 143- 144 selection pressures, 155- 156 Intercropping, 149 Ion pair, formation, gypsum application, 17

215

L Leaching pesticides, 41 studies with gypsum. 14- 15

M Magnesium, distribution in soil, 55-56 Maize genetic gain, 80-81 genome. 80-82 high density hybrids, 82 hybrids and stress, 81 phosphogypsum effects on yield, 9- 10 transposable elements. see Transposable elements, in maize yield response to gypsum, 5-7 Meat, cadmium content, 194 Metallothioneins, 185 expression of genes encoding. 202 Metals, plant tolerance. 188 Molecular changes, transposable elements, 111-114 Mulch growing cover crops for. 61 -62 surface, crop residue benefits, 37-38, 58-59 disadvantages, 37-38 tillage effects. 59-61

N Neoteny, 115- I16 Nitrogen distribution in soil, 55 fertilizer application methods, 67-68 improving efficiency. 68-69 growing cover crops for, 61-62 management, in conservation tillage, 65-66 utilization efficiency, in maize, 81 -82 NO1-, transport through soils, 40-41 No-tillage system definition, 36 surface water and, 40

0 Organic matter, in soil effects of conventional and conservation tillage, 56-57 soil productivity effects, 42 Osteomalacia, cadmium effects, 199

2 16

INDEX

Osteoporosis, cadmium effects, 199 Oxisol, dispersion, 20

P Pesticides crisis, 126 leaching, 41 persistence, 152 population resurgence and secondary outbreak, 155 preventive and therapeutic roles, 152- 153 Pesticide treadmill, definition, 156 Pest management, see also Integrated pest management diseases and insects, 70-72 issues, 130 weed control, 69-70 Pests definition, 128 economic injury level, I28 impact on agricultural systems, 127 resurgence, 155 pH, soil, 53-54 cadmium solubility and, 179 Phosphogypsum, 2 cotton seedling emergence effects, 20, 22 crust formation prevention, 18, 21 maize yield effects, 9- 10 Phosphorus, distribution in soil, 54-55 Phytochelatins cadmium detoxification role, 188 induction by cadmium, 186- 187 micronutrient homeostasis role, 187- 188 as sulfur carriers, 187- 188 transfer of zinc and copper to apometalloproteins. 187 Planting, date manipulation, pest management, 150

Plants cadmium accumulation, 179- 184 manipulation, 201 -204 resistance to arthropods, 145-148 Population density, 139 Potassium, distribution in soil, 55-56 Pseudoresistance, 146

R Ridge tillage system, definition, 36

S Seeding, cover crops, 62-64 self-reseeding, 64 Seedlings, cotton, see Cotton seedlings Self-liming effect, 15- 16 Sequential probability ratio test, 142- 143 Sludge, in soil, 177 Soil agricultural, cadmium input, 175- 177 cadmium content, 174- 175 chemical properties nutrient distribution, 54-56 organic matter, 56-57 pH, 53-54 degradation, gypsum effects, I8 erosion conservation tillage, 38-39 effect on soil productivity, 42-43 fertilizer distribution, 66-67 gypsum effects on physical properties hardsetting soils, 22-23 hydraulic conductivity, 25 - 26 subsurface hard layers, 23-25 surface crusting, 17-22 organic matter, effect on soil productivity, 42 pH, gypsum application effects, 7-8 physical properties, 45-52 aggregation and infiltration, 46-48 bulk density and compaction, 45-46 soil temperature, 50-52 soil water, 48-50 productivity, conservation tillage and, 42-43 solution studies, gypsum application, 12- 14 Solid phases, precipitation, gypsum application, 16 Stubble-mulch farming, 36 Subsoils, acid, gypsum as ameliorant for, 3- 17 Subsurface, hard layers, 23-25 Sulfate, co-sorption with All+, 16 Superphosphate, single and triple, cumulative application effects, 7-8 Surface crusting, gypsum effects, 17-22

T Temperature, soil, 50-52 Therapeutic management, arthropod pests, 152-154 Tillage, 36, see also Conservation tillage Time-sequential sampling, 144

INDEX Tobacco cadmium accumulation, 181- 183 human cadmium exposure. 197 metallothionein in, cadmium accumulation reduction, 202-203 Tolerance, plant role in arthropod resistance, 145-147 to toxic metals. 188 Transposable elements, in maize, 79- I16 bridge-breakage-fusion cycle, 86-88 discovery gene concepts in McClintock experiments, 90-92 McClintock experiments, 86-90 unstable genes. concepts in pre-McClintock period, 90 effects on gene expression enzyme and molecular changes, I I 1 - I 14 phenotypic changes, 110- I I 1 E n l , 96 En2.97-98 EnlSpm , 94- 99 En transposition, 105- 106 exceptional kernel, 88-89 expression, 84 genetic components, specificity, 92-93 genetic resolution, 100- 102 heterogeneity question, 79-80 identificationof parts, 107- 108 molecular TNPA and TNPD,93-99 M p , 103 neoteny and realignment of resources, 115-116

217

phenotype variegation, 83-86 position effect, 91 presence in genome, 114- 115 systems. 99- 100 Taml. 113 transposition footprints and altered products of excision, 108-110 gene alteration, 105, 107- 108 genetic detection, 101- 106 Uq,114- I15 variability induction, 105, 107- 108

W Water quality. conservation tillage, 39-42 soil evaporation, 49-50 infiltration, 48 surface, conservation tillage and, 39-40 Water-supplyingcapacity, effect on soil productivity, 43 Weed, control, pest management and, 69-70 Wheat, root growth, A1 and Ca saturation effect, 7

Y Yield components, 82-83

Z Zinc, homeostasis, 187